Re: [Emc-users] Latency Warning Messages

[Gene Heskett]

On Tuesday 17 December 2019 12:44:47 Dave Matthews wrote:

> On Tue, Dec 17, 2019 at 9:22 AM Peter C. Wallace  
wrote:
> > You can get really bad base jitter if you try to run the base thread
> > faster than the machine is capable of responding (the latency-tests
> > default 40 KHz base thread is too fast for some systems so you might
> > get considerably better reported latency at 20 KHz for example)
> >
> >
> > Peter Wallace
> > Mesa Electronics
>
> For those of us in the cheap seats how does the base thread frequency
> relate to the maximum number of words per second sent to the parallel
> port?  I figure 20k words / second would work out to 375 inches /
> minute of travel on my homebuilt CNC router.  Far faster than I would
> need and thus in the range of not needing to worry about it all that
> much.
>
> Dave
>
There is also the effect of variable latency to concider, and its very 
very important with software stepping.  Imagine when moving that near 
your calculated speed limit, and something disturbs the timeing of the 
steps.

The motor effectively stops, then the high speed signals resume.  But 
they resume at the current speed without the acceleration rampup in 
speed that accompany's the speed change.

Unless you have motors are wound on air cores, and the copper in the 
windings is also zero weight, a massless rotor in effect, the motor is 
not capable of resumeing that speed so it locks in place until the step 
rate received is low enough to allow it to restart.  But because it will 
restart at any of it 200 positions, lcnc has lost the reference point to 
the axis only and your part is wrecked. 

This is your REAL speed limit, and because the table has varying drag, 
increaseing as your approach ends of travel, your need additional head 
room.

The end result can often be 10% or less of what your calculations show.

Generally speaking that is still faster than the spindle horsepower 
provides for, and as those limits apply only to the rapid moves, and 
will probably be cutting air.

And its a lesson that cost me some broken tools and wrecked parts to 
learn.

And since every system involving springs has a resonant frequency that is 
often well below this speed limit, the motors magnetic bounce will cause 
the stalling to manifest itself at even lower speeds, so 
torsional "dampers" will often result in much hugher speeds as they 
dampen the resonance, in one case here a 435 oz motor as the vertical 
(Z) motor on a baby hf mill, went from an 8 ipm speed limit, to 34 ipm. 
That axis was not a ball screw, but a double nutted NOOK acme that I 
could adjust for about .0005" of backlash in a rotating nut design. You 
can see pix of these dampers on my web page. Those are alternating big 
steel fender washers with powdered (talcum) rubber disks cut out of 1/8 
thick rubber but the center holes in the rubber are smaller than the 
spools hub, so the center of the washer when installed is few thou 
fatter.  So the steel to rubber slips and the slippage is the resonance 
absorber. There are of coarse other designs. Choose your poison.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page 


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency Warning Messages

[Gene Heskett]

On Tuesday 17 December 2019 13:41:09 Dave Matthews wrote:

> On Tue, Dec 17, 2019 at 12:55 PM andy pugh  wrote:
> > On Tue, 17 Dec 2019 at 17:47, Dave Matthews  wrote:
> > > For those of us in the cheap seats how does the base thread
> > > frequency relate to the maximum number of words per second sent to
> > > the parallel port?
> >
> > The system can send a maximum of one pulse per base period.
>
> And that pulse contains the information for all of the motors being
> controlled.  Correct?
>
> Dave
>
Afaik, yes.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page 


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency Warning Messages

[Gene Heskett]

On Tuesday 17 December 2019 13:18:10 Eric Keller wrote:

> In the past, I have had a reasonable amount of luck reducing real time
> delays by building a custom kernel.  Not sure if those kinds of gains
> are still possible.  And it's a lot of work unless you are using
> realtime preempt.
>
> As far as Gene's problem with the beep, they are probably using the
> 8051 on the system management bus for that.  I wonder if a sound card
> would have the same problem. I think most beeps can be routed through
> the sound card. Eric Keller
> Boalsburg, Pennsylvania
>
Probably could be done, but as has been said, this is my house machine, 
and newly built and will never cut metal.  I've to had no chance to fine 
tune yet, as I am just this afternoon back from a couple days in the 
shop, getting new Aortic valve in the top of my heart. It seems to be 
working well.

For those that observe Christmas, have a Merry one.
>
> On Mon, Dec 16, 2019 at 9:19 PM John Dammeyer 
>
> wrote:
> > > > For example.  The board in the machine at the moment generates a
> > > > Servo Jitter of 33217 and Base Jitter of 131983.
> > >
> > > Thats pretty slow john, I'd take that machine to the office and
> > > bring out the office machine.  It might just be faster. :)
> >
> > Yes Gene,
> > I sometimes wonder if there is something else going on.  Something
> > the BIOS settings that makes these two machines so slow.  Processor
> > and memory wise they should be more than adequate.  They weren't
> > expensive as they were used.  But what's to say if I went to COSTCO
> > and bought a new one that I'd be any better off.
> >
> > John
> >
> >
> >
> >
> > ___
> > Emc-users mailing list
> > Emc-users@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/emc-users
>
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users


Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page 


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency Warning Messages

[Dave Matthews]

On Tue, Dec 17, 2019 at 2:00 PM John Dammeyer  wrote:
>
> Actually I think it's at one pulse per two base periods.  One to set the step 
> signals high, one to set them low.  The document referenced by "bari",
>
> How to choose your base period:
> http://wiki.linuxcnc.org/cgi-bin/wiki.pl?TweakingSoftwareStepGeneration
>
> explains that in detail.
> John Dammeyer
>
Thank you, I hadn't seen that bit of documentation.  Looks like I can
do about 285 in/min with my current setup without any tweaking.  Half
of that is a realistic speed for my Gatton CNC to do rapids.

Dave


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency Warning Messages

[andy pugh]

On Tue, 17 Dec 2019 at 19:00, John Dammeyer  wrote:
>
> Actually I think it's at one pulse per two base periods.  One to set the step 
> signals high, one to set them low.

The parport driver offers a "reset" function that sets selected pins
back to zero. You can run this later in the thread than the "write"
function and it either resets the pins immediately if the step time
has expired, or waits a few µS for that to happen then resets.

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1912


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency Warning Messages

[John Dammeyer]

Actually I think it's at one pulse per two base periods.  One to set the step 
signals high, one to set them low.  The document referenced by "bari",

How to choose your base period:
http://wiki.linuxcnc.org/cgi-bin/wiki.pl?TweakingSoftwareStepGeneration

explains that in detail.
John Dammeyer


> -Original Message-
> From: Dave Matthews [mailto:n36...@gmail.com]
> Sent: December-17-19 10:41 AM
> To: Enhanced Machine Controller (EMC)
> Subject: Re: [Emc-users] Latency Warning Messages
> 
> On Tue, Dec 17, 2019 at 12:55 PM andy pugh  wrote:
> >
> > On Tue, 17 Dec 2019 at 17:47, Dave Matthews  wrote:
> >
> > > For those of us in the cheap seats how does the base thread frequency
> > > relate to the maximum number of words per second sent to the parallel
> > > port?
> >
> > The system can send a maximum of one pulse per base period.
> >
> 
> And that pulse contains the information for all of the motors being
> controlled.  Correct?
> 
> Dave
> 
> 
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency Warning Messages

[John Dammeyer]

> From: andy pugh [mailto:bodge...@gmail.com]
> On Mon, 16 Dec 2019 at 23:42, John Dammeyer 
> wrote:
> 
> > For example.  The board in the machine at the moment generates a Servo
> Jitter of 33217 and Base Jitter of 131983.
> 
> That's fairly terrible.
> 
> Out of interest, how well does that machine work with Mach3?
> 

That's a good question so I just had to go down and try it since it's been a 
while since I did that sort of exercise of moving windows with photos around 
while the machine is moving an axis.

WIN-XP, Onboard (slower) video, 1024x768 to LCD monitor, MACH3, 60kHz Parallel 
Port setting.

Machine Y axis:
0.2" pitch lead screw.
4:1 pulley from DC Servo
DC Servo max RPM 3000, Encoder 250 lines (1000 PPR)

Therefore 4000 pulses to move 0.2" or 20,000 pulses to move 1".  
Target speed 120ipm or 2ips which is 40kHz stepping pulses.

Machine X axis:
0.2" pitch lead screw.
3:1 pulley from DC Servo
DC Servo max RPM 3000, Encoder 250 lines (1000 PPR)

Therefore 3000 pulses to move 0.2" or 15,000 pulses to move 1".  
Target speed 180ipm or 3ips which is 45kHz stepping pulses.

Doing other activities while moving at those high rates does not appear to make 
the system lose steps or stop or declare faults.

Having said that, the reason I switched to a USB Smooth Stepper on my CNC 
Router with MACH3 was that on an older, probably single core PC, doing anything 
like resizing a window while the system was running would cause loss of motion 
control due to motor lockup.  I suspect because
  a) the processor wasn't fast enough
  b) an interruption and then catch up in step pulses to a stepper motor 
results in rapid changes in motor speed control that the motor itself cannot 
handle since the step pulses translate directly into winding current.

With the DC Servo run with the HP_UHU drive and Henrik Olsson's dsPIC upgrade 
processor module, the slight interruptions to periodic step pulses don't change 
the overall motion with respect to closed loop control monitoring the encoder 
so the DC Servo doesn't get confused.

And moving 6 inches at 180 ipm (3ips) is really fast.  Kind of scary in itself. 
 A lot can happen in 2 seconds.

So in summary, no issues with MACH3 and parallel port stepping.  I'd still 
upgrade to an Ethernet Smooth Stepper simply to remove that parallel port 
issue.  Especially since I'd rather run WIN-7 on that PC and not WIN-XP.

But I haven't decided yet between LinuxCNC and MACH3.  There are aspects of the 
MACH3 user interface I like way more than the LinuxCNC one.  There are also 
parts of LinuxCNC I like more.

John Dammeyer



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency Warning Messages

[Dave Matthews]

On Tue, Dec 17, 2019 at 12:55 PM andy pugh  wrote:
>
> On Tue, 17 Dec 2019 at 17:47, Dave Matthews  wrote:
>
> > For those of us in the cheap seats how does the base thread frequency
> > relate to the maximum number of words per second sent to the parallel
> > port?
>
> The system can send a maximum of one pulse per base period.
>

And that pulse contains the information for all of the motors being
controlled.  Correct?

Dave


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency Warning Messages

[Eric Keller]

In the past, I have had a reasonable amount of luck reducing real time
delays by building a custom kernel.  Not sure if those kinds of gains are
still possible.  And it's a lot of work unless you are using realtime
preempt.

As far as Gene's problem with the beep, they are probably using the 8051 on
the system management bus for that.  I wonder if a sound card would have
the same problem. I think most beeps can be routed through the sound card.
Eric Keller
Boalsburg, Pennsylvania



On Mon, Dec 16, 2019 at 9:19 PM John Dammeyer 
wrote:

> > >
> > > For example.  The board in the machine at the moment generates a Servo
> > > Jitter of 33217 and Base Jitter of 131983.
>
>
> > Thats pretty slow john, I'd take that machine to the office and bring out
> > the office machine.  It might just be faster. :)
> >
>
> Yes Gene,
> I sometimes wonder if there is something else going on.  Something the
> BIOS settings that makes these two machines so slow.  Processor and memory
> wise they should be more than adequate.  They weren't expensive as they
> were used.  But what's to say if I went to COSTCO and bought a new one that
> I'd be any better off.
>
> John
>
>
>
>
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>

___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency Warning Messages

[andy pugh]

On Tue, 17 Dec 2019 at 17:47, Dave Matthews  wrote:

> For those of us in the cheap seats how does the base thread frequency
> relate to the maximum number of words per second sent to the parallel
> port?

The system can send a maximum of one pulse per base period.

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1912


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency Warning Messages

[Dave Matthews]

On Tue, Dec 17, 2019 at 9:22 AM Peter C. Wallace  wrote:
>
> You can get really bad base jitter if you try to run the base thread faster 
> than
> the machine is capable of responding (the latency-tests default 40 KHz base
> thread is too fast for some systems so you might get considerably better
> reported latency at 20 KHz for example)
>
>
> Peter Wallace
> Mesa Electronics
>
For those of us in the cheap seats how does the base thread frequency
relate to the maximum number of words per second sent to the parallel
port?  I figure 20k words / second would work out to 375 inches /
minute of travel on my homebuilt CNC router.  Far faster than I would
need and thus in the range of not needing to worry about it all that
much.

Dave


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency Warning Messages

[Peter C. Wallace]

On Tue, 17 Dec 2019, andy pugh wrote:



Date: Tue, 17 Dec 2019 11:32:22 +
From: andy pugh 
Reply-To: "Enhanced Machine Controller (EMC)"

To: "Enhanced Machine Controller (EMC)" 
Subject: Re: [Emc-users] Latency Warning Messages

On Mon, 16 Dec 2019 at 23:42, John Dammeyer  wrote:

For example.  The board in the machine at the moment generates a
Servo Jitter of 33217 and Base Jitter of 131983.



That's fairly terrible.



Out of interest, how well does that machine work with Mach3?


You can get really bad base jitter if you try to run the base thread faster than 
the machine is capable of responding (the latency-tests default 40 KHz base 
thread is too fast for some systems so you might get considerably better 
reported latency at 20 KHz for example)



Peter Wallace
Mesa Electronics

(\__/)
(='.'=) This is Bunny. Copy and paste bunny into your
(")_(") signature to help him gain world domination.



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency Warning Messages

[andy pugh]

On Mon, 16 Dec 2019 at 23:42, John Dammeyer  wrote:

> For example.  The board in the machine at the moment generates a Servo Jitter 
> of 33217 and Base Jitter of 131983.

That's fairly terrible.

Out of interest, how well does that machine work with Mach3?

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1912


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency Warning Messages

[Gene Heskett]

On Monday 16 December 2019 23:12:20 bari wrote:

> On 12/16/19 9:02 PM, Gene Heskett wrote:
> > intel cpus are better than equ dated amd's, my old quad core phenom
> > was phenominally bad at latency-test. 100+ microseconds. New 9th gen
> > 6 core i5, 4 or 5 microseconds. Till kmail beeps the chassis
> > speaker, then its 200 u-secs.
>
> BIOS issues likely since Phenom and Phenom II's were used for most the
> RTAI work from '12-'14 and we had them in the low single digit
> microseconds.
>
I had the latest available on an Asus M2N-SLI Deluxe. Several of the 
linuxcnc kernels ran as daily drivers over the years since I built it in 
2006. It was stuck in the 100 microsecond region. Original phenom, only 
aavailable for a month or so when I bought it, ordered slower 2.1 GHz 
version and never really pushed as it was a room heater anyway. This 6 
core i5 runs 30C cooler than the phenom, and its doing it at 3.7GHz. 

By '12 or '14, the phenoms had been optimized enough there likely could 
be considered a new family, die shrinks etc had pulled the fastest ones 
down to 95 watts. I inquired at Asus in about that time frame and was 
told the later ones wouldn't run on that nearly $300 board.

Then that board burned up starting at one of the usb headers about 3 
weeks back so I had to rebuild it. So this time I spent more on the cpu 
and less on the board.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page 


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency Warning Messages

[bari]

On 12/16/19 9:02 PM, Gene Heskett wrote:
> intel cpus are better than equ dated amd's, my old quad core phenom was 
> phenominally bad at latency-test. 100+ microseconds. New 9th gen 6 core 
> i5, 4 or 5 microseconds. Till kmail beeps the chassis speaker, then its 
> 200 u-secs.
BIOS issues likely since Phenom and Phenom II's were used for most the
RTAI work from '12-'14 and we had them in the low single digit microseconds.


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency Warning Messages

[Gene Heskett]

On Monday 16 December 2019 21:17:40 John Dammeyer wrote:

> > > For example.  The board in the machine at the moment generates a
> > > Servo Jitter of 33217 and Base Jitter of 131983.
> >
> > Thats pretty slow john, I'd take that machine to the office and
> > bring out the office machine.  It might just be faster. :)
>
> Yes Gene,
> I sometimes wonder if there is something else going on.  Something the
> BIOS settings that makes these two machines so slow.  Processor and
> memory wise they should be more than adequate.  They weren't expensive
> as they were used.  But what's to say if I went to COSTCO and bought a
> new one that I'd be any better off.
>
> John

At costco? That would be tossing quarters, and hoping they come back down

General rules of thumb I have observed:

intel cpus are better than equ dated amd's, my old quad core phenom was 
phenominally bad at latency-test. 100+ microseconds. New 9th gen 6 core 
i5, 4 or 5 microseconds. Till kmail beeps the chassis speaker, then its 
200 u-secs.

Nvidia video cards being driven by nvidia drivers are horrible, returning 
2 or 3 milliseconds, the nouveau driver may not be as fast for video, 
but it doesn't lock out the irq's either, so its actually quite a bit 
better. But linux drivers pushing an ATI card were usually best. ATI is 
now part of AMD but haven't checked the numbers there in several years.

So far, nothing beats the intel atom powered D-525-MW mobo for good 
latency and more than enough cojones even w/o a mesa card to software 
step.  But they've been discoed for quite a few years now, so the only 
place to get them is ebay, and I'm a little shy about paying 20 to 50 
bucks more than new for used. I have mesa 5i25 cards in both of mine 
now, with 7i76D daughter cards.

However, I am pleasantly surprised at how well an rpi4b is running my 
sheldon lathe.

>
>
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users


Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page 


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency Warning Messages

[bari]

On 12/16/19 5:40 PM, John Dammeyer wrote:
>
> For example.  The board in the machine at the moment generates a Servo Jitter 
> of 33217 and Base Jitter of 131983.   
> For the MESA and PP configuration I have:
> BASE_PERIOD = 24000
> SERVO_PERIOD = 100
>
> If I understood the document link above correctly I should have set the base 
> to at least 132000 which would limit my step rate to about 3kHz.  The video 
> board is a ow profile card that has a DVI connector out and can run to two 
> VGA monitors.  Really just designed for desktop PC's that need multiple 
> monitors to display government forms or whatever.  The Lenovo PC and card 
> came from the government surplus.
>
> Now if I remove the video card and run the latency test for the on board 
> video I get a Servo Jitter of 351232 and Base Jitter of 418333.
>
> Setting the .ini BASE_PERIOD to 420,000 results in no longer getting the 
> initial servo message.  Of course now step rates are so severely restricted 
> that jog on the X axis results in an immediate following error.
>
Yes it shounds like you are understanding the docs.

Have you turned off all the power management in BIOS along with any
speed stepping any virtualization? All should off and not used.

If you have all these off then the motherboard and CPU chosen is not a
low latency combo. Often the BIOSes are broken and even though you have
them off they may not really be off.




___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency Warning Messages

[John Dammeyer]

> >
> > For example.  The board in the machine at the moment generates a Servo
> > Jitter of 33217 and Base Jitter of 131983.


> Thats pretty slow john, I'd take that machine to the office and bring out
> the office machine.  It might just be faster. :)
> 

Yes Gene,
I sometimes wonder if there is something else going on.  Something the BIOS 
settings that makes these two machines so slow.  Processor and memory wise they 
should be more than adequate.  They weren't expensive as they were used.  But 
what's to say if I went to COSTCO and bought a new one that I'd be any better 
off.

John




___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency Warning Messages

[Gene Heskett]

On Monday 16 December 2019 18:40:35 John Dammeyer wrote:

> > -Original Message-
> > From: bari [mailto:bari00...@gmail.com]
> > Sent: December-16-19 10:51 AM
> > To: emc-users@lists.sourceforge.net
> > Subject: Re: [Emc-users] Latency Warning Messages
> >
> > It depends on how you configure and run Linuxcnc with your intended
> > hardware.
> >
> > In the .INI file there is a section EMCMOT
> > http://linuxcnc.org/docs/2.6/html/config/ini_config.html#sub:EMCMOT-
> > section
> >
> > The settings you place in BASE_PERIOD and SERVO_PERIOD set the
> > length of each of these task periods for Linuxcnc.
> >
> > BASE_PERIOD = 5 - the Base task period in nanoseconds.
> >
> > SERVO_PERIOD = 100 - This is the "Servo" task period in
> > nanoseconds.
> >
> > If either thread run by the processor takes longer than the times in
> > these settings you will see a latency message for it taking too
> > long. This can be caused by things like trying to step too quickly
> > (moving the motors faster) using the base period with a LPT port or
> > by using too low a setting for the task periods to begin with based
> > on your hardware and BIOS settings.
> >
> > How to choose your base period:
> > http://wiki.linuxcnc.org/cgi-bin/wiki.pl?TweakingSoftwareStepGenerat
> >ion
> >
> > The reason that you stopped having the latency messages when you
> > changed
> > from software stepping on the LPT port to hardware stepping on the
> > MESA 7i92H is that you changed from using the Base Period (very
> > short, fast) to the Servo Period (much slower) from the CPU's
> > perspective. This allows you to use a PC with longer latency or
> > higher jitter.
> >
> > On 12/16/19 12:03 PM, John Dammeyer wrote:
> > > That's what I've found.  When I run the test software to determine
> >
> > machine latency the on board video provides the smallest number. 
> > Any of the add on boards I tried all gave much worse numbers.
>
> For example.  The board in the machine at the moment generates a Servo
> Jitter of 33217 and Base Jitter of 131983.
Thats pretty slow john, I'd take that machine to the office and bring out 
the office machine.  It might just be faster. :)

> For the MESA and PP 
> configuration I have:
> BASE_PERIOD = 24000
> SERVO_PERIOD = 100
>
> If I understood the document link above correctly I should have set
> the base to at least 132000 which would limit my step rate to about
> 3kHz.  The video board is a ow profile card that has a DVI connector
> out and can run to two VGA monitors.  Really just designed for desktop
> PC's that need multiple monitors to display government forms or
> whatever.  The Lenovo PC and card came from the government surplus.
>
> Now if I remove the video card and run the latency test for the on
> board video I get a Servo Jitter of 351232 and Base Jitter of 418333.
>
> Setting the .ini BASE_PERIOD to 420,000 results in no longer getting
> the initial servo message.  Of course now step rates are so severely
> restricted that jog on the X axis results in an immediate following
> error.
>
> I believe that the DC Servos are more forgiving if the step pulses are
> occasionally held back since the internal step counter is compared
> against the encoder counter and a following error fault by the motor
> only happens if it reads more encoder counts than received on the step
> input.  Probably why with the video board installed and the initial
> error that I don't get faults.
>
> And none of that is an issue with the MESA card.
> John
>
> > > Also, if I understand how this works, if the fastest speed
> > > required for
> >
> > stepping is not to high then you won't get the messages.  But if
> > your step rates are up into 50kHz and higher then you get that
> > warning once when you start up LinuxCNC.
> >
> > > But I may have it all wrong.   For my current testing I get that
> > > message
> >
> > when I use the parallel port but it doesn't appear to affect the
> > step rates to the drives.  I just dispose of the message.  But
> > that's because I don't care at the moment.  I'll switch to the MESA
> > once the testing and development is done.
> >
> > > John
> > >
> > >> -Original Message-
> > >> From: N [mailto:nicklas.karlsso...@gmail.com]
> > >> Sent: December-16-19 9:38 AM
> > >> To: Enhanced Machine Controller (EMC)
> > >> Subject: Re: [Emc-users] Latency Warning Messages
> > >

Re: [Emc-users] Latency Warning Messages

[John Dammeyer]

> -Original Message-
> From: bari [mailto:bari00...@gmail.com]
> Sent: December-16-19 10:51 AM
> To: emc-users@lists.sourceforge.net
> Subject: Re: [Emc-users] Latency Warning Messages
> 
> It depends on how you configure and run Linuxcnc with your intended
> hardware.
> 
> In the .INI file there is a section EMCMOT
> http://linuxcnc.org/docs/2.6/html/config/ini_config.html#sub:EMCMOT-
> section
> 
> The settings you place in BASE_PERIOD and SERVO_PERIOD set the length of
> each of these task periods for Linuxcnc.
> 
> BASE_PERIOD = 5 - the Base task period in nanoseconds.
> 
> SERVO_PERIOD = 100 - This is the "Servo" task period in nanoseconds.
> 
> If either thread run by the processor takes longer than the times in
> these settings you will see a latency message for it taking too long.
> This can be caused by things like trying to step too quickly (moving the
> motors faster) using the base period with a LPT port or by using too low
> a setting for the task periods to begin with based on your hardware and
> BIOS settings.
> 
> How to choose your base period:
> http://wiki.linuxcnc.org/cgi-bin/wiki.pl?TweakingSoftwareStepGeneration
> 
> The reason that you stopped having the latency messages when you
> changed
> from software stepping on the LPT port to hardware stepping on the MESA
> 7i92H is that you changed from using the Base Period (very short, fast)
> to the Servo Period (much slower) from the CPU's perspective. This
> allows you to use a PC with longer latency or higher jitter.
> 
> On 12/16/19 12:03 PM, John Dammeyer wrote:
> > That's what I've found.  When I run the test software to determine
> machine latency the on board video provides the smallest number.  Any of
> the add on boards I tried all gave much worse numbers.

For example.  The board in the machine at the moment generates a Servo Jitter 
of 33217 and Base Jitter of 131983.   
For the MESA and PP configuration I have:
BASE_PERIOD = 24000
SERVO_PERIOD = 100

If I understood the document link above correctly I should have set the base to 
at least 132000 which would limit my step rate to about 3kHz.  The video board 
is a ow profile card that has a DVI connector out and can run to two VGA 
monitors.  Really just designed for desktop PC's that need multiple monitors to 
display government forms or whatever.  The Lenovo PC and card came from the 
government surplus.

Now if I remove the video card and run the latency test for the on board video 
I get a Servo Jitter of 351232 and Base Jitter of 418333.

Setting the .ini BASE_PERIOD to 420,000 results in no longer getting the 
initial servo message.  Of course now step rates are so severely restricted 
that jog on the X axis results in an immediate following error.

I believe that the DC Servos are more forgiving if the step pulses are 
occasionally held back since the internal step counter is compared against the 
encoder counter and a following error fault by the motor only happens if it 
reads more encoder counts than received on the step input.  Probably why with 
the video board installed and the initial error that I don't get faults.

And none of that is an issue with the MESA card.
John



> >
> > Also, if I understand how this works, if the fastest speed required for
> stepping is not to high then you won't get the messages.  But if your step
> rates are up into 50kHz and higher then you get that warning once when you
> start up LinuxCNC.
> >
> > But I may have it all wrong.   For my current testing I get that message
> when I use the parallel port but it doesn't appear to affect the step rates to
> the drives.  I just dispose of the message.  But that's because I don't care 
> at
> the moment.  I'll switch to the MESA once the testing and development is
> done.
> >
> > John
> >
> >> -Original Message-
> >> From: N [mailto:nicklas.karlsso...@gmail.com]
> >> Sent: December-16-19 9:38 AM
> >> To: Enhanced Machine Controller (EMC)
> >> Subject: Re: [Emc-users] Latency Warning Messages
> >>
> >>> I bought 3 different low profile video cards to try and get rid of that
> latency
> >> message.  The ultimate and best solution was to buy the MESA 7i92H.
> Does
> >> away with the parallel port completely.
> >>> John Dammeyer
> >> The parrallel port cause latency messages?
> >>
> >>
> >> ___
> >> Emc-users mailing list
> >> Emc-users@lists.sourceforge.net
> >> https://lists.sourceforge.net/lists/listinfo/emc-users
> >
> >
> > ___
> > Emc-us

Re: [Emc-users] Latency Warning Messages

[bari]

It depends on how you configure and run Linuxcnc with your intended
hardware.

In the .INI file there is a section EMCMOT
http://linuxcnc.org/docs/2.6/html/config/ini_config.html#sub:EMCMOT-section

The settings you place in BASE_PERIOD and SERVO_PERIOD set the length of
each of these task periods for Linuxcnc.

BASE_PERIOD = 5 - the Base task period in nanoseconds.

SERVO_PERIOD = 100 - This is the "Servo" task period in nanoseconds.

If either thread run by the processor takes longer than the times in
these settings you will see a latency message for it taking too long.
This can be caused by things like trying to step too quickly (moving the
motors faster) using the base period with a LPT port or by using too low
a setting for the task periods to begin with based on your hardware and
BIOS settings.

How to choose your base period:
http://wiki.linuxcnc.org/cgi-bin/wiki.pl?TweakingSoftwareStepGeneration

The reason that you stopped having the latency messages when you changed
from software stepping on the LPT port to hardware stepping on the MESA
7i92H is that you changed from using the Base Period (very short, fast)
to the Servo Period (much slower) from the CPU's perspective. This
allows you to use a PC with longer latency or higher jitter.

On 12/16/19 12:03 PM, John Dammeyer wrote:
> That's what I've found.  When I run the test software to determine machine 
> latency the on board video provides the smallest number.  Any of the add on 
> boards I tried all gave much worse numbers.
>
> Also, if I understand how this works, if the fastest speed required for 
> stepping is not to high then you won't get the messages.  But if your step 
> rates are up into 50kHz and higher then you get that warning once when you 
> start up LinuxCNC.
>
> But I may have it all wrong.   For my current testing I get that message when 
> I use the parallel port but it doesn't appear to affect the step rates to the 
> drives.  I just dispose of the message.  But that's because I don't care at 
> the moment.  I'll switch to the MESA once the testing and development is done.
>
> John
>
>> -Original Message-
>> From: N [mailto:nicklas.karlsso...@gmail.com]
>> Sent: December-16-19 9:38 AM
>> To: Enhanced Machine Controller (EMC)
>> Subject: Re: [Emc-users] Latency Warning Messages
>>
>>> I bought 3 different low profile video cards to try and get rid of that 
>>> latency
>> message.  The ultimate and best solution was to buy the MESA 7i92H.  Does
>> away with the parallel port completely.
>>> John Dammeyer
>> The parrallel port cause latency messages?
>>
>>
>> ___
>> Emc-users mailing list
>> Emc-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/emc-users
>
>
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency Warning Messages

[Gene Heskett]

On Monday 16 December 2019 12:38:18 N wrote:

> > I bought 3 different low profile video cards to try and get rid of
> > that latency message.  The ultimate and best solution was to buy the
> > MESA 7i92H.  Does away with the parallel port completely. John
> > Dammeyer
>
> The parrallel port cause latency messages?
>
Some motherboards route an "out" thru a lot of junk to get there.
>
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users


Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
If we desire respect for the law, we must first make the law respectable.
 - Louis D. Brandeis
Genes Web page 


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency Warning Messages

[John Dammeyer]

That's what I've found.  When I run the test software to determine machine 
latency the on board video provides the smallest number.  Any of the add on 
boards I tried all gave much worse numbers.

Also, if I understand how this works, if the fastest speed required for 
stepping is not to high then you won't get the messages.  But if your step 
rates are up into 50kHz and higher then you get that warning once when you 
start up LinuxCNC.

But I may have it all wrong.   For my current testing I get that message when I 
use the parallel port but it doesn't appear to affect the step rates to the 
drives.  I just dispose of the message.  But that's because I don't care at the 
moment.  I'll switch to the MESA once the testing and development is done.

John

> -Original Message-
> From: N [mailto:nicklas.karlsso...@gmail.com]
> Sent: December-16-19 9:38 AM
> To: Enhanced Machine Controller (EMC)
> Subject: Re: [Emc-users] Latency Warning Messages
> 
> > I bought 3 different low profile video cards to try and get rid of that 
> > latency
> message.  The ultimate and best solution was to buy the MESA 7i92H.  Does
> away with the parallel port completely.
> > John Dammeyer
> 
> The parrallel port cause latency messages?
> 
> 
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency Warning Messages

[N]

> I bought 3 different low profile video cards to try and get rid of that 
> latency message.  The ultimate and best solution was to buy the MESA 7i92H.  
> Does away with the parallel port completely.
> John Dammeyer

The parrallel port cause latency messages?


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency Warning Messages

[andy pugh]

On Mon, 16 Dec 2019 at 05:47, Thomas D. Dean  wrote:

> Looks like the on-board video adapter shares main memory with the CPU.
> Linuxcnc system requirements says this is a no=no.

That advice might be outdated, I have many systems that seem perfectly
happy with this.

> Question:  Will installing a PCI video card likely to reduce the latency
> and eliminate the warnings?

It might, but there are probably other solutions.

If the "Stretch" drive can be considered disposable, you could try
installing a 64-bit Debian Buster on there, and then try out the
experimental 64-bit RTAI packages in http://www.linuxcnc.org/temp/

(This assumes that your PC can run 64-bit linux. Some D2500 can't
(depending on motherboard) but it sounds like 64-bit stretch is
working)

-- 
atp
"A motorcycle is a bicycle with a pandemonium attachment and is
designed for the especial use of mechanical geniuses, daredevils and
lunatics."
— George Fitch, Atlanta Constitution Newspaper, 1912


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency Warning Messages

[John Dammeyer]

I bought 3 different low profile video cards to try and get rid of that latency 
message.  The ultimate and best solution was to buy the MESA 7i92H.  Does away 
with the parallel port completely.
John Dammeyer


> -Original Message-
> From: Thomas D. Dean [mailto:tomd...@wavecable.com]
> Sent: December-15-19 9:46 PM
> To: 'Enhanced Machine Controller (EMC)'
> Subject: [Emc-users] Latency Warning Messages
> 
> My CNC computer was purchased several years ago from Sherline with CNC
> installed.
> 
> I have two disk drives.  One has linuxcnc-2.7.14-wheezy.iso and the
> other has linuxcnc-stretch-uspace-amd64.iso.
> 
> On both OS's, I see 'unexpected latency warning messages', infrequent on
> 'wheezy' and 99% of the time at linuxcnc startup on 'stretch'.  The
> latency test shows jitter around 40us for 'wheezy' and 100us for 'stretch'.
> 
> The cpu is Intel Atom CPU D2500   @ 1.86GHz
> It has 1G memory.
> Lots of free disk space.
> 
> lspci shows:
> 00:02.0 VGA compatible controller: Intel Corporation Atom Processor
> D2xxx/N2xxx Integrated Graphics Controller (rev 0b)
> 
> and lspci -v shows:
> 00:02.0 VGA compatible controller: Intel Corporation Atom Processor
> D2xxx/N2xxx Integrated Graphics Controller (rev 0b) (prog-if 00 [VGA
> controller])
>  Subsystem: Intel Corporation Atom Processor D2xxx/N2xxx
> Integrated Graphics Controller
>  Flags: bus master, fast devsel, latency 0, IRQ 29
>  Memory at 4010 (32-bit, non-prefetchable) [size=1M]
>  I/O ports at 20d0 [size=8]
>  [virtual] Expansion ROM at 000c [disabled] [size=128K]
>  Capabilities: [d0] Power Management version 2
>  Capabilities: [b0] Vendor Specific Information: Len=07 
>  Capabilities: [90] MSI: Enable+ Count=1/1 Maskable- 64bit-
>  Kernel driver in use: gma500
>  Kernel modules: gma500_gfx
> 
> Looks like the on-board video adapter shares main memory with the CPU.
> Linuxcnc system requirements says this is a no=no.
> 
> Question:  Will installing a PCI video card likely to reduce the latency
> and eliminate the warnings?
> 
> Tom Dean
> 
> 
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users



___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Latency Warning Messages

[Thomas D. Dean]

My CNC computer was purchased several years ago from Sherline with CNC 
installed.


I have two disk drives.  One has linuxcnc-2.7.14-wheezy.iso and the 
other has linuxcnc-stretch-uspace-amd64.iso.


On both OS's, I see 'unexpected latency warning messages', infrequent on 
'wheezy' and 99% of the time at linuxcnc startup on 'stretch'.  The 
latency test shows jitter around 40us for 'wheezy' and 100us for 'stretch'.


The cpu is Intel Atom CPU D2500   @ 1.86GHz
It has 1G memory.
Lots of free disk space.

lspci shows:
00:02.0 VGA compatible controller: Intel Corporation Atom Processor 
D2xxx/N2xxx Integrated Graphics Controller (rev 0b)


and lspci -v shows:
00:02.0 VGA compatible controller: Intel Corporation Atom Processor 
D2xxx/N2xxx Integrated Graphics Controller (rev 0b) (prog-if 00 [VGA 
controller])
Subsystem: Intel Corporation Atom Processor D2xxx/N2xxx 
Integrated Graphics Controller

Flags: bus master, fast devsel, latency 0, IRQ 29
Memory at 4010 (32-bit, non-prefetchable) [size=1M]
I/O ports at 20d0 [size=8]
[virtual] Expansion ROM at 000c [disabled] [size=128K]
Capabilities: [d0] Power Management version 2
Capabilities: [b0] Vendor Specific Information: Len=07 
Capabilities: [90] MSI: Enable+ Count=1/1 Maskable- 64bit-
Kernel driver in use: gma500
Kernel modules: gma500_gfx

Looks like the on-board video adapter shares main memory with the CPU. 
Linuxcnc system requirements says this is a no=no.


Question:  Will installing a PCI video card likely to reduce the latency 
and eliminate the warnings?


Tom Dean


___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] latency excursions

[Peter C. Wallace]

On Sun, 23 Jul 2017, dave wrote:


Date: Sun, 23 Jul 2017 10:37:09 -0700
From: dave 
Reply-To: "Enhanced Machine Controller (EMC)"

To: "Enhanced Machine Controller (EMC)" 
Subject: Re: [Emc-users] latency excursions



On 07/23/2017 10:14 AM, Peter C. Wallace wrote:

On Sun, 23 Jul 2017, dave wrote:


Date: Sun, 23 Jul 2017 08:48:19 -0700
From: dave 
Reply-To: "Enhanced Machine Controller (EMC)"

To: "Enhanced Machine Controller (EMC)" 
Subject: [Emc-users] latency excursions

dave@linuxcnc:~$ uname -a
Linux linuxcnc 3.4-9-rtai-686-pae #1 SMP PREEMPT Debian 3.4.55-4linuxcnc 
i686 i686 i386 GNU/Linux


task: 103857488 cycles, min=0.49, max=0.474214, avg=0.010993, 86 
latency excursions (> 10x expected cycle time of 0.01s)


task: 25439189 cycles, min=0.49, max=0.205898, avg=0.010989, 17 
latency excursions (> 10x expected cycle time of 0.01s)


I've done nothing to tune this cpu and as far as I can tell the motion is 
ok.


Still it leave me uncomfortable.

What do I need to change to clean this uo?


TIA

Dave


These are not real-time latency issues but rather delays in task
so are not a actual problem unless they are so long that they cause
a disruption the the non-real time tasks linuxCNC is doing (user interface 
etc)


These delays will not be reduced by any real time latency tuning of the 
system ( and may well be made worse by things that slow the userland 
performance like isocpus )



Peter Wallace
Mesa Electronics

(\__/)
(='.'=) This is Bunny. Copy and paste bunny into your
(")_(") signature to help him gain world domination.

Thanks Peter,
I suppose I should have picked up on the 'task'. I'm going to use age as an 
excuse. ;-)


I re-tuned the x axis yesterday for some improvement; following error is 
somewhere around 2E-5.


Dave


You can do a little experiment with delaying task by starting LinuxCNC from 
the command line and then suspending it while running gcode with control Z. 
This does not disrupt motion in progress, as long as the delay is not too

long, and if it is too long you will get:

USRMOT: ERROR: command timeout


(after the control Z, you can resume task/user-interface with "fg 1")



-- 
Check out the vibrant tech community on one of the world's most

engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users



--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users



Peter Wallace
Mesa Electronics

(\__/)
(='.'=) This is Bunny. Copy and paste bunny into your
(")_(") signature to help him gain world domination.


--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] latency excursions

[dave]

On 07/23/2017 10:14 AM, Peter C. Wallace wrote:

On Sun, 23 Jul 2017, dave wrote:


Date: Sun, 23 Jul 2017 08:48:19 -0700
From: dave 
Reply-To: "Enhanced Machine Controller (EMC)"

To: "Enhanced Machine Controller (EMC)" 


Subject: [Emc-users] latency excursions

dave@linuxcnc:~$ uname -a
Linux linuxcnc 3.4-9-rtai-686-pae #1 SMP PREEMPT Debian 
3.4.55-4linuxcnc i686 i686 i386 GNU/Linux


task: 103857488 cycles, min=0.49, max=0.474214, avg=0.010993, 86 
latency excursions (> 10x expected cycle time of 0.01s)


task: 25439189 cycles, min=0.49, max=0.205898, avg=0.010989, 17 
latency excursions (> 10x expected cycle time of 0.01s)


I've done nothing to tune this cpu and as far as I can tell the 
motion is ok.


Still it leave me uncomfortable.

What do I need to change to clean this uo?


TIA

Dave


These are not real-time latency issues but rather delays in task
so are not a actual problem unless they are so long that they cause
a disruption the the non-real time tasks linuxCNC is doing (user 
interface etc)


These delays will not be reduced by any real time latency tuning of 
the system ( and may well be made worse by things that slow the 
userland performance like isocpus )



Peter Wallace
Mesa Electronics

(\__/)
(='.'=) This is Bunny. Copy and paste bunny into your
(")_(") signature to help him gain world domination.

Thanks Peter,
I suppose I should have picked up on the 'task'. I'm going to use age as 
an excuse. ;-)


I re-tuned the x axis yesterday for some improvement; following error is 
somewhere around 2E-5.


Dave
-- 


Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users



--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] latency excursions

[Gene Heskett]

On Sunday 23 July 2017 11:48:19 dave wrote:

> dave@linuxcnc:~$ uname -a
> Linux linuxcnc 3.4-9-rtai-686-pae #1 SMP PREEMPT Debian
> 3.4.55-4linuxcnc i686 i686 i386 GNU/Linux
>
> task: 103857488 cycles, min=0.49, max=0.474214, avg=0.010993, 86
> latency excursions (> 10x expected cycle time of 0.01s)
>
> task: 25439189 cycles, min=0.49, max=0.205898, avg=0.010989, 17
> latency excursions (> 10x expected cycle time of 0.01s)
>
> I've done nothing to tune this cpu and as far as I can tell the motion
> is ok.
>
> Still it leave me uncomfortable.
>
> What do I need to change to clean this uo?
>
>
> TIA
>
> Dave
>
How many cores does that cpu have? And if its an intel, is hyperthreading 
turned on? Thats a bit hard on the latencytest results. Turn it off in 
the bios.

If 2 or more cores, it may help to add to the grub line that loads the 
kernel, isolcpus=lastcore, where lastcore could be 2, 4, or even 8 if 
you've a really new barn burner machine.

With that kernel, your dmesg should contain a line:

Brought up 2 CPUs

The 2 will vary depending on the cpu in that machine, so in this case the 
kernel loading line in /boot/grub/grub.cfg might have "isolcpus=1" 
appended to it, without the dbl-quotes.

This isolates that core from the scheduler, linuxcnc sees it and uses 
that core for the realtime part, having it all to itself. But a monitor 
such as top or htop, can't tell you how hard that cpu core is running.

I have one machine still doing software stepping, and that line in its 
grub has "isolcpus=1 idle=poll", which seemed to help the latency there.

And just for my own enjoyment, I usually remove the splash and quiet 
options. Splash covers up any boot msgs with an ugly graphic, and quiet 
turns the boot msgs off, which makes the boot faster when it doesn't 
have to do all the screen stuffs. That way I can scroll back and check 
that everything looks ok. Or, if mount decides to fsck the drive, you at 
least have something on screen to explain the delayed boot.

I don't normally care how long it takes to boot, only that it boots 
normally, doing what it has to do to make it run as usual for that 
machine.

I HTH Dave.

Cheers, Gene Heskett
-- 
"There are four boxes to be used in defense of liberty:
 soap, ballot, jury, and ammo. Please use in that order."
-Ed Howdershelt (Author)
Genes Web page 

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] latency excursions

[Peter C. Wallace]

On Sun, 23 Jul 2017, dave wrote:


Date: Sun, 23 Jul 2017 08:48:19 -0700
From: dave 
Reply-To: "Enhanced Machine Controller (EMC)"

To: "Enhanced Machine Controller (EMC)" 
Subject: [Emc-users] latency excursions

dave@linuxcnc:~$ uname -a
Linux linuxcnc 3.4-9-rtai-686-pae #1 SMP PREEMPT Debian 3.4.55-4linuxcnc i686 
i686 i386 GNU/Linux


task: 103857488 cycles, min=0.49, max=0.474214, avg=0.010993, 86 latency 
excursions (> 10x expected cycle time of 0.01s)


task: 25439189 cycles, min=0.49, max=0.205898, avg=0.010989, 17 latency 
excursions (> 10x expected cycle time of 0.01s)


I've done nothing to tune this cpu and as far as I can tell the motion is ok.

Still it leave me uncomfortable.

What do I need to change to clean this uo?


TIA

Dave


These are not real-time latency issues but rather delays in task
so are not a actual problem unless they are so long that they cause
a disruption the the non-real time tasks linuxCNC is doing (user interface 
etc)


These delays will not be reduced by any real time latency tuning of the system 
( and may well be made worse by things that slow the userland performance like 
isocpus )




--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users



Peter Wallace
Mesa Electronics

(\__/)
(='.'=) This is Bunny. Copy and paste bunny into your
(")_(") signature to help him gain world domination.


--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] latency excursions

[Nicklas Karlsson]

For me this or something similar i happening one of my computers at every 
startup. I think it is running smootly in between now but this have not always 
been the case, I removed a graphics card a while ago  and might have changed 
things or simply a newer kernel.

On Sun, 23 Jul 2017 08:48:19 -0700
dave  wrote:

> dave@linuxcnc:~$ uname -a
> Linux linuxcnc 3.4-9-rtai-686-pae #1 SMP PREEMPT Debian 3.4.55-4linuxcnc 
> i686 i686 i386 GNU/Linux
> 
> task: 103857488 cycles, min=0.49, max=0.474214, avg=0.010993, 86 
> latency excursions (> 10x expected cycle time of 0.01s)
> 
> task: 25439189 cycles, min=0.49, max=0.205898, avg=0.010989, 17 
> latency excursions (> 10x expected cycle time of 0.01s)
> 
> I've done nothing to tune this cpu and as far as I can tell the motion 
> is ok.
> 
> Still it leave me uncomfortable.
> 
> What do I need to change to clean this uo?
> 
> 
> TIA
> 
> Dave
> 
> --
> Check out the vibrant tech community on one of the world's most
> engaging tech sites, Slashdot.org! http://sdm.link/slashdot
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] latency excursions

[dave]

dave@linuxcnc:~$ uname -a
Linux linuxcnc 3.4-9-rtai-686-pae #1 SMP PREEMPT Debian 3.4.55-4linuxcnc 
i686 i686 i386 GNU/Linux


task: 103857488 cycles, min=0.49, max=0.474214, avg=0.010993, 86 
latency excursions (> 10x expected cycle time of 0.01s)


task: 25439189 cycles, min=0.49, max=0.205898, avg=0.010989, 17 
latency excursions (> 10x expected cycle time of 0.01s)


I've done nothing to tune this cpu and as far as I can tell the motion 
is ok.


Still it leave me uncomfortable.

What do I need to change to clean this uo?


TIA

Dave

--
Check out the vibrant tech community on one of the world's most
engaging tech sites, Slashdot.org! http://sdm.link/slashdot
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency problem with debian 7 and not with ubuntu 10.04

[Ricardo Moscoloni]

​Hi pierre
the same ​was observed in my p4 asus 775i65G, really bad latencys on debian
pae kernels, dont know if its a kernel issue or anything, switch back to
10.04 agh.
regards
rck

2016-05-24 15:41 GMT-03:00 Forum Deswysen :

> Hello,
>
> i86 pentium 4 Fujitsu siemens scenic E600 :
>
> with ubuntu 10.04 and 2.7 linuxcnc
> *latency +/- 22,000 max*
>
> I format the pc and i install debian 7 and 2.7 linuxcnc
> *latency +/- 65,000 max*
>
> Why do you have an idea?
>
> Best regards,
>
> Pierre
>
> --
> Mobile security can be enabling, not merely restricting. Employees who
> bring their own devices (BYOD) to work are irked by the imposition of MDM
> restrictions. Mobile Device Manager Plus allows you to control only the
> apps on BYO-devices by containerizing them, leaving personal data
> untouched!
> https://ad.doubleclick.net/ddm/clk/304595813;131938128;j
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
--
Mobile security can be enabling, not merely restricting. Employees who
bring their own devices (BYOD) to work are irked by the imposition of MDM
restrictions. Mobile Device Manager Plus allows you to control only the
apps on BYO-devices by containerizing them, leaving personal data untouched!
https://ad.doubleclick.net/ddm/clk/304595813;131938128;j
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Latency problem with debian 7 and not with ubuntu 10.04

[Forum Deswysen]

Hello,

i86 pentium 4 Fujitsu siemens scenic E600 :

with ubuntu 10.04 and 2.7 linuxcnc
*latency +/- 22,000 max*

I format the pc and i install debian 7 and 2.7 linuxcnc
*latency +/- 65,000 max*

Why do you have an idea?

Best regards,

Pierre
--
Mobile security can be enabling, not merely restricting. Employees who
bring their own devices (BYOD) to work are irked by the imposition of MDM
restrictions. Mobile Device Manager Plus allows you to control only the
apps on BYO-devices by containerizing them, leaving personal data untouched!
https://ad.doubleclick.net/ddm/clk/304595813;131938128;j
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Latency?

[Todd Zuercher]

I was testing the latency of this computer and I thought that it was doing 
pretty good, but over night there were some pretty significant latency spikes. 
I was afraid this was going to happen, it is an HP-Compaq dc5750 and the bios 
has virtually zero user settable power settings (the only thing I can find is a 
setting for idle fan speed). Any suggestions for any work-a-rounds? I think I'm 
still going to use it anyway, because I have an application only using 
positioning of a step/dir servo, so other than step generation, there shouldn't 
be any real real-time demands. 

-- 

 

Todd Zuercher 
mailto:zuerc...@embarqmail.com 

 
--
Don't Limit Your Business. Reach for the Cloud.
GigeNET's Cloud Solutions provide you with the tools and support that
you need to offload your IT needs and focus on growing your business.
Configured For All Businesses. Start Your Cloud Today.
https://www.gigenetcloud.com/___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency issues; machinekit Kernel

[Andreas Pettersson]

Ah well yeah the GPU was an issue, running the onboard in the i5 - but 
giving it more resources helped alot.

For I/O i ran a create and move file around and delete script that uses 
/dev/urandom and create, random sized file, gzips it, moves it a couple 
of times to random folder and then deletes it. (same i use for grading 
webservers when doing node.js work).

Did not do the youtube thing - but ran glxgears x 5, and gzip -c 
/dev/urandom /dev/null. But i imagine some more stuff needs to be 
tested.. the APM is turned off so is also the CPU C instructions for 
throttling.

Thanks for help anyway. =)

// Andreas

John Kasunich skrev den 2015-02-27 23:37:
> On Fri, Feb 27, 2015, at 02:13 PM, Andreas Pettersson wrote:
>> If its stable running after 15min of basicly 100% cpu load - do i need
>> to bother running it any longer?
>>
> Is if 15 minutes of doing the same "load" thing over and over?  Or doing a lot
> of different things?
>
> CPU loading usually doesn't cause latency issues.  It's things like poorly 
> done
> power management (in the BIOS or the kernel), or graphics drivers, or .
>
> We had one computer that had great latency until you dragged a window
> around on the screen.  Then the GPU would take hold of the bus in order
> to redraw the window, and everything else screeched to a halt for hundred
> of microseconds.
>
> Stress it in a variety of ways - transfer large files on the network, drag
> windows around, play a youtube video, run glxgears, walk away until the
> screensaver and/or power management kicks in, etc.
>
>


--
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the 
conversation now. http://goparallel.sourceforge.net/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency issues; machinekit Kernel

[Sebastian Kuzminsky]

On 2/27/15 3:37 PM, John Kasunich wrote:
>
> Stress it in a variety of ways - transfer large files on the network, drag
> windows around, play a youtube video, run glxgears, walk away until the
> screensaver and/or power management kicks in, etc.

Also try disk I/O and USB hotplug events.


-- 
Sebastian Kuzminsky

--
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the 
conversation now. http://goparallel.sourceforge.net/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency issues; machinekit Kernel

[John Kasunich]

On Fri, Feb 27, 2015, at 02:13 PM, Andreas Pettersson wrote:
> If its stable running after 15min of basicly 100% cpu load - do i need 
> to bother running it any longer?
> 

Is if 15 minutes of doing the same "load" thing over and over?  Or doing a lot
of different things?

CPU loading usually doesn't cause latency issues.  It's things like poorly done
power management (in the BIOS or the kernel), or graphics drivers, or .

We had one computer that had great latency until you dragged a window 
around on the screen.  Then the GPU would take hold of the bus in order
to redraw the window, and everything else screeched to a halt for hundred
of microseconds.

Stress it in a variety of ways - transfer large files on the network, drag 
windows around, play a youtube video, run glxgears, walk away until the
screensaver and/or power management kicks in, etc.


-- 
  John Kasunich
  jmkasun...@fastmail.fm

--
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the 
conversation now. http://goparallel.sourceforge.net/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency issues; machinekit Kernel

[Andreas Pettersson]

If its stable running after 15min of basicly 100% cpu load - do i need 
to bother running it any longer?

// Andreas

Sebastian Kuzminsky skrev den 2015-02-27 19:49:
> On 2/27/15 11:44 AM, Andreas Pettersson wrote:
>> So im guessing the step, where you do ./configure -a in the debian
>> folder is void and null when running the RT_PREEMPT kernel, because it
>> states there is no realtime kernel and refuses to configure.
>> In effect i cant figure out which packages i need to install other than
>> running the other configure and see what it crashes on..   (a step that
>> ive got stuck on at several times).
> The docs on building LinuxCNC are a bit scattered and probably
> incomplete & out of date, but try this page:
>
> http://wiki.linuxcnc.org/cgi-bin/wiki.pl?Uspace
>
>


--
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the 
conversation now. http://goparallel.sourceforge.net/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency issues; machinekit Kernel

[Andreas Pettersson]

yeah i figured, i got it working and are stressing it now, i wrote this 
for my own head - but will most likely google index pretty fast as well..

http://www.roughedge.se/blogg/2015/02/27/installing-linuxcnc-2-7-uspace-on-debian-wheezy/

Servothread 11347, Base thread 12665 while running
5 glxgears, gzip -c /dev/urandom /dev/null, stress -c 16 -l 16 -m 16 -t1 
4400

So i think im in the clear now. =)

Uspace build could handle alot more punishment. =)

// Andreas

Sebastian Kuzminsky skrev den 2015-02-27 19:49:
> On 2/27/15 11:44 AM, Andreas Pettersson wrote:
>> So im guessing the step, where you do ./configure -a in the debian
>> folder is void and null when running the RT_PREEMPT kernel, because it
>> states there is no realtime kernel and refuses to configure.
>> In effect i cant figure out which packages i need to install other than
>> running the other configure and see what it crashes on..   (a step that
>> ive got stuck on at several times).
> The docs on building LinuxCNC are a bit scattered and probably
> incomplete & out of date, but try this page:
>
> http://wiki.linuxcnc.org/cgi-bin/wiki.pl?Uspace
>
>


--
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the 
conversation now. http://goparallel.sourceforge.net/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency issues; machinekit Kernel

[Sebastian Kuzminsky]

On 2/27/15 11:44 AM, Andreas Pettersson wrote:
> So im guessing the step, where you do ./configure -a in the debian
> folder is void and null when running the RT_PREEMPT kernel, because it
> states there is no realtime kernel and refuses to configure.
> In effect i cant figure out which packages i need to install other than
> running the other configure and see what it crashes on..   (a step that
> ive got stuck on at several times).

The docs on building LinuxCNC are a bit scattered and probably 
incomplete & out of date, but try this page:

http://wiki.linuxcnc.org/cgi-bin/wiki.pl?Uspace


-- 
Sebastian Kuzminsky

--
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the 
conversation now. http://goparallel.sourceforge.net/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency issues; machinekit Kernel

[Andreas Pettersson]

So im guessing the step, where you do ./configure -a in the debian 
folder is void and null when running the RT_PREEMPT kernel, because it 
states there is no realtime kernel and refuses to configure.
In effect i cant figure out which packages i need to install other than 
running the other configure and see what it crashes on..   (a step that 
ive got stuck on at several times).

// Andreas

Sebastian Kuzminsky skrev den 2015-02-27 18:26:
> On 2/27/15 9:20 AM, Andreas Pettersson wrote:
>> But stock wheezy is 64bit and with the RT_PREEMPT kernel, you managed to
>> compile linuxcnc with that? All i get when trying that is "No realtime
>> kernel present" error..  :/
> Yes, LinuxCNC runs fine on 64-bit RT-Preempt.  You need the "uspace"
> flavor, which lives in a different place in the deb archive, details here:
>
> http://linuxcnc.org/docs/2.7/html/getting-started/index.html#_alternate_install_methods
>
>
> If you're building linuxcnc from source, configure with
> "--with-realtime=uspace".
>
>


--
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the 
conversation now. http://goparallel.sourceforge.net/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency issues; machinekit Kernel

[Andreas Pettersson]

Ah i see, i was considering rebuilding the kernel, so if this isnt 
working i will do that.
But this could be useful to learn as well. - just reinstall for the 30th 
time this day.. *lmao*

// Andreas

Sebastian Kuzminsky skrev den 2015-02-27 19:21:
> On 2/27/15 11:17 AM, Andreas Pettersson wrote:
>> Sweet im gonna try that, having issues with PAE not claiming one of the
>> memory modules. =)
> This is a bug in our "pae" kernel - it actually has pae turned off, so
> you can only use the first 3.5 gigs or so.
>
> I punished the guy responsible for that bug...  (it was me)
>
> Give rt-preempt a try, someone who actually knows what they were doing
> built that kernel, and it works fine (as long as it meets you latency
> requirements on your hardware).
>
>


--
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the 
conversation now. http://goparallel.sourceforge.net/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency issues; machinekit Kernel

[Sebastian Kuzminsky]

On 2/27/15 11:17 AM, Andreas Pettersson wrote:
> Sweet im gonna try that, having issues with PAE not claiming one of the
> memory modules. =)

This is a bug in our "pae" kernel - it actually has pae turned off, so 
you can only use the first 3.5 gigs or so.

I punished the guy responsible for that bug...  (it was me)

Give rt-preempt a try, someone who actually knows what they were doing 
built that kernel, and it works fine (as long as it meets you latency 
requirements on your hardware).


-- 
Sebastian Kuzminsky

--
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the 
conversation now. http://goparallel.sourceforge.net/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency issues; machinekit Kernel

[Andreas Pettersson]

Sweet im gonna try that, having issues with PAE not claiming one of the 
memory modules. =)

Thanks!

// A

Sebastian Kuzminsky skrev den 2015-02-27 18:26:
> On 2/27/15 9:20 AM, Andreas Pettersson wrote:
>> But stock wheezy is 64bit and with the RT_PREEMPT kernel, you managed to
>> compile linuxcnc with that? All i get when trying that is "No realtime
>> kernel present" error..  :/
> Yes, LinuxCNC runs fine on 64-bit RT-Preempt.  You need the "uspace"
> flavor, which lives in a different place in the deb archive, details here:
>
> http://linuxcnc.org/docs/2.7/html/getting-started/index.html#_alternate_install_methods
>
>
> If you're building linuxcnc from source, configure with
> "--with-realtime=uspace".
>
>


--
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the 
conversation now. http://goparallel.sourceforge.net/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency issues; machinekit Kernel

[Sebastian Kuzminsky]

On 2/27/15 9:20 AM, Andreas Pettersson wrote:
> But stock wheezy is 64bit and with the RT_PREEMPT kernel, you managed to
> compile linuxcnc with that? All i get when trying that is "No realtime
> kernel present" error..  :/

Yes, LinuxCNC runs fine on 64-bit RT-Preempt.  You need the "uspace" 
flavor, which lives in a different place in the deb archive, details here:

http://linuxcnc.org/docs/2.7/html/getting-started/index.html#_alternate_install_methods


If you're building linuxcnc from source, configure with 
"--with-realtime=uspace".


-- 
Sebastian Kuzminsky

--
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the 
conversation now. http://goparallel.sourceforge.net/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency issues; machinekit Kernel

[Andreas Pettersson]

Ah that worked quite well, but only utilized a fraction of the ram 
properly.. Got the latency down to 3500 with that, its an acceptable speed..

im little bit worried about the ram tho..

anyway, gonna try to get this 6i25 card running next..

// A

Peter C. Wallace skrev den 2015-02-27 17:31:
> On Fri, 27 Feb 2015, Andreas Pettersson wrote:
>
>> Date: Fri, 27 Feb 2015 17:20:53 +0100
>> From: Andreas Pettersson 
>> Reply-To: "Enhanced Machine Controller (EMC)"
>>  
>> To: "Enhanced Machine Controller (EMC)" 
>> Subject: Re: [Emc-users] Latency issues; machinekit Kernel
>>
>> But stock wheezy is 64bit and with the RT_PREEMPT kernel, you managed to
>> compile linuxcnc with that? All i get when trying that is "No realtime
>> kernel present" error..  :
>
> The standard LinuxCNC wheezy ISO uses a RTAI kernel (and you can test latency
> by just booting from a USB stick, no need to install)
>
>
>
>
>> // A
>>
>> Peter C. Wallace skrev den 2015-02-27 17:15:
>>> On Fri, 27 Feb 2015, Andreas Pettersson wrote:
>>>
>>>> Date: Fri, 27 Feb 2015 17:09:10 +0100
>>>> From: Andreas Pettersson 
>>>> Reply-To: "Enhanced Machine Controller (EMC)"
>>>>   
>>>> To: "Enhanced Machine Controller (EMC)" 
>>>> Subject: Re: [Emc-users] Latency issues; machinekit Kernel
>>>>
>>>> Are you running a standard Wheezy AMD64 install and the master buildbot
>>>> installations as well then??
>>>> I can always reinstall, i rather have a 64bit OS than the PAE version
>>>> anyday.  =)
>>>>
>>>> // A
>>> Stock wheezy ISO works fine, I would try latency with that first
>>>
>>>
>>>> Peter C. Wallace skrev den 2015-02-27 16:20:
>>>>> On Fri, 27 Feb 2015, Andreas Pettersson wrote:
>>>>>
>>>>>> Date: Fri, 27 Feb 2015 15:58:08 +0100
>>>>>> From: Andreas Pettersson 
>>>>>> Reply-To: "Enhanced Machine Controller (EMC)"
>>>>>>
>>>>>> To: "Enhanced Machine Controller (EMC)" 
>>>>>> Subject: [Emc-users] Latency issues; machinekit Kernel
>>>>>>
>>>>>> So i had some issued with drivers, but got around that by installing the
>>>>>> machinekit-rtai kernel
>>>>>> and then manually compile linuxcnc ontop of wheezy..
>>>>>>
>>>>>> Well i dont think its optimal.
>>>>>>
>>>>>> The setup is a Core i5, 8gb ram and running SSD drivers in a raid 0
>>>>>> configuration.
>>>>>> Lets say it boots like its on a ramdrive basicly..
>>>>>>
>>>>>> Funny thing tho. Max Jitter: 815055 so far.. At a Max interval of 
>>>>>> 1815032.
>>>>>>
>>>>>> I find these numbers extremly high? Shouldnt they be alot.. well lower?
>>>>>>
>>>>>> // Andreas
>>>>> Thats pretty bad, do you have all power management/fan control off in the
>>>>> BIOS?
>>>>>
>>>>>
>>>>> I have a newer MB (ASRock H97M with a Pentium G3258 and built in Intel NIC
>>>>> =21x) that has about 4 usec latency with the stock Wheezy RTAI kernel
>>>>>
>>>>>
>>>>> I didnt try the intel 21x MAC with the RTAI kernel but it works fine with 
>>>>> the
>>>>> stock wheezy Preemt-RT kernel (3.2.something)
>>>>>
>>>>>
>>>>>
>>>>>> --
>>>>>> Dive into the World of Parallel Programming The Go Parallel Website, 
>>>>>> sponsored
>>>>>> by Intel and developed in partnership with Slashdot Media, is your hub 
>>>>>> for all
>>>>>> things parallel software development, from weekly thought leadership 
>>>>>> blogs to
>>>>>> news, videos, case studies, tutorials and more. Take a look and join the
>>>>>> conversation now. http://goparallel.sourceforge.net/
>>>>>> ___
>>>>>> Emc-users mailing list
>>>>>> Emc-users@lists.sourceforge.net
>>>>>> https://lists.sourceforge.net/lists/listinfo/emc-users
>>>>>>
>>>>> Peter Wallace
&g

Re: [Emc-users] Latency issues; machinekit Kernel

[Peter C. Wallace]

On Fri, 27 Feb 2015, Andreas Pettersson wrote:

> Date: Fri, 27 Feb 2015 17:20:53 +0100
> From: Andreas Pettersson 
> Reply-To: "Enhanced Machine Controller (EMC)"
> 
> To: "Enhanced Machine Controller (EMC)" 
> Subject: Re: [Emc-users] Latency issues; machinekit Kernel
> 
> But stock wheezy is 64bit and with the RT_PREEMPT kernel, you managed to
> compile linuxcnc with that? All i get when trying that is "No realtime
> kernel present" error..  :


The standard LinuxCNC wheezy ISO uses a RTAI kernel (and you can test latency 
by just booting from a USB stick, no need to install)




>
> // A
>
> Peter C. Wallace skrev den 2015-02-27 17:15:
>> On Fri, 27 Feb 2015, Andreas Pettersson wrote:
>>
>>> Date: Fri, 27 Feb 2015 17:09:10 +0100
>>> From: Andreas Pettersson 
>>> Reply-To: "Enhanced Machine Controller (EMC)"
>>>  
>>> To: "Enhanced Machine Controller (EMC)" 
>>> Subject: Re: [Emc-users] Latency issues; machinekit Kernel
>>>
>>> Are you running a standard Wheezy AMD64 install and the master buildbot
>>> installations as well then??
>>> I can always reinstall, i rather have a 64bit OS than the PAE version
>>> anyday.  =)
>>>
>>> // A
>> Stock wheezy ISO works fine, I would try latency with that first
>>
>>
>>> Peter C. Wallace skrev den 2015-02-27 16:20:
>>>> On Fri, 27 Feb 2015, Andreas Pettersson wrote:
>>>>
>>>>> Date: Fri, 27 Feb 2015 15:58:08 +0100
>>>>> From: Andreas Pettersson 
>>>>> Reply-To: "Enhanced Machine Controller (EMC)"
>>>>>   
>>>>> To: "Enhanced Machine Controller (EMC)" 
>>>>> Subject: [Emc-users] Latency issues; machinekit Kernel
>>>>>
>>>>> So i had some issued with drivers, but got around that by installing the
>>>>> machinekit-rtai kernel
>>>>> and then manually compile linuxcnc ontop of wheezy..
>>>>>
>>>>> Well i dont think its optimal.
>>>>>
>>>>> The setup is a Core i5, 8gb ram and running SSD drivers in a raid 0
>>>>> configuration.
>>>>> Lets say it boots like its on a ramdrive basicly..
>>>>>
>>>>> Funny thing tho. Max Jitter: 815055 so far.. At a Max interval of 1815032.
>>>>>
>>>>> I find these numbers extremly high? Shouldnt they be alot.. well lower?
>>>>>
>>>>> // Andreas
>>>> Thats pretty bad, do you have all power management/fan control off in the
>>>> BIOS?
>>>>
>>>>
>>>> I have a newer MB (ASRock H97M with a Pentium G3258 and built in Intel NIC
>>>> =21x) that has about 4 usec latency with the stock Wheezy RTAI kernel
>>>>
>>>>
>>>> I didnt try the intel 21x MAC with the RTAI kernel but it works fine with 
>>>> the
>>>> stock wheezy Preemt-RT kernel (3.2.something)
>>>>
>>>>
>>>>
>>>>> --
>>>>> Dive into the World of Parallel Programming The Go Parallel Website, 
>>>>> sponsored
>>>>> by Intel and developed in partnership with Slashdot Media, is your hub 
>>>>> for all
>>>>> things parallel software development, from weekly thought leadership 
>>>>> blogs to
>>>>> news, videos, case studies, tutorials and more. Take a look and join the
>>>>> conversation now. http://goparallel.sourceforge.net/
>>>>> ___
>>>>> Emc-users mailing list
>>>>> Emc-users@lists.sourceforge.net
>>>>> https://lists.sourceforge.net/lists/listinfo/emc-users
>>>>>
>>>> Peter Wallace
>>>> Mesa Electronics
>>>>
>>>> (\__/)
>>>> (='.'=) This is Bunny. Copy and paste bunny into your
>>>> (")_(") signature to help him gain world domination.
>>>>
>>>>
>>>> --
>>>> Dive into the World of Parallel Programming The Go Parallel Website, 
>>>> sponsored
>>>> by Intel and developed in partnership with Slashdot Media, is your hub for 
>>>> all
>>>> things parallel software developm

Re: [Emc-users] Latency issues; machinekit Kernel

[Andreas Pettersson]

But stock wheezy is 64bit and with the RT_PREEMPT kernel, you managed to 
compile linuxcnc with that? All i get when trying that is "No realtime 
kernel present" error..  :/

// A

Peter C. Wallace skrev den 2015-02-27 17:15:
> On Fri, 27 Feb 2015, Andreas Pettersson wrote:
>
>> Date: Fri, 27 Feb 2015 17:09:10 +0100
>> From: Andreas Pettersson 
>> Reply-To: "Enhanced Machine Controller (EMC)"
>>  
>> To: "Enhanced Machine Controller (EMC)" 
>> Subject: Re: [Emc-users] Latency issues; machinekit Kernel
>>
>> Are you running a standard Wheezy AMD64 install and the master buildbot
>> installations as well then??
>> I can always reinstall, i rather have a 64bit OS than the PAE version
>> anyday.  =)
>>
>> // A
> Stock wheezy ISO works fine, I would try latency with that first
>
>
>> Peter C. Wallace skrev den 2015-02-27 16:20:
>>> On Fri, 27 Feb 2015, Andreas Pettersson wrote:
>>>
>>>> Date: Fri, 27 Feb 2015 15:58:08 +0100
>>>> From: Andreas Pettersson 
>>>> Reply-To: "Enhanced Machine Controller (EMC)"
>>>>   
>>>> To: "Enhanced Machine Controller (EMC)" 
>>>> Subject: [Emc-users] Latency issues; machinekit Kernel
>>>>
>>>> So i had some issued with drivers, but got around that by installing the
>>>> machinekit-rtai kernel
>>>> and then manually compile linuxcnc ontop of wheezy..
>>>>
>>>> Well i dont think its optimal.
>>>>
>>>> The setup is a Core i5, 8gb ram and running SSD drivers in a raid 0
>>>> configuration.
>>>> Lets say it boots like its on a ramdrive basicly..
>>>>
>>>> Funny thing tho. Max Jitter: 815055 so far.. At a Max interval of 1815032.
>>>>
>>>> I find these numbers extremly high? Shouldnt they be alot.. well lower?
>>>>
>>>> // Andreas
>>> Thats pretty bad, do you have all power management/fan control off in the
>>> BIOS?
>>>
>>>
>>> I have a newer MB (ASRock H97M with a Pentium G3258 and built in Intel NIC
>>> =21x) that has about 4 usec latency with the stock Wheezy RTAI kernel
>>>
>>>
>>> I didnt try the intel 21x MAC with the RTAI kernel but it works fine with 
>>> the
>>> stock wheezy Preemt-RT kernel (3.2.something)
>>>
>>>
>>>
>>>> --
>>>> Dive into the World of Parallel Programming The Go Parallel Website, 
>>>> sponsored
>>>> by Intel and developed in partnership with Slashdot Media, is your hub for 
>>>> all
>>>> things parallel software development, from weekly thought leadership blogs 
>>>> to
>>>> news, videos, case studies, tutorials and more. Take a look and join the
>>>> conversation now. http://goparallel.sourceforge.net/
>>>> ___
>>>> Emc-users mailing list
>>>> Emc-users@lists.sourceforge.net
>>>> https://lists.sourceforge.net/lists/listinfo/emc-users
>>>>
>>> Peter Wallace
>>> Mesa Electronics
>>>
>>> (\__/)
>>> (='.'=) This is Bunny. Copy and paste bunny into your
>>> (")_(") signature to help him gain world domination.
>>>
>>>
>>> --
>>> Dive into the World of Parallel Programming The Go Parallel Website, 
>>> sponsored
>>> by Intel and developed in partnership with Slashdot Media, is your hub for 
>>> all
>>> things parallel software development, from weekly thought leadership blogs 
>>> to
>>> news, videos, case studies, tutorials and more. Take a look and join the
>>> conversation now. http://goparallel.sourceforge.net/
>>> ___
>>> Emc-users mailing list
>>> Emc-users@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/emc-users
>>
>> --
>> Dive into the World of Parallel Programming The Go Parallel Website, 
>> sponsored
>> by Intel and developed in partnership with Slashdot Media, is your hub for 
>> all
>> things parallel software development, from weekly thought leadership blogs to
>> news, videos, case studies, tutorials and more. Take a look and join the
&

Re: [Emc-users] Latency issues; machinekit Kernel

[Peter C. Wallace]

On Fri, 27 Feb 2015, Andreas Pettersson wrote:

> Date: Fri, 27 Feb 2015 17:09:10 +0100
> From: Andreas Pettersson 
> Reply-To: "Enhanced Machine Controller (EMC)"
> 
> To: "Enhanced Machine Controller (EMC)" 
> Subject: Re: [Emc-users] Latency issues; machinekit Kernel
> 
> Are you running a standard Wheezy AMD64 install and the master buildbot
> installations as well then??
> I can always reinstall, i rather have a 64bit OS than the PAE version
> anyday.  =)
>
> // A

Stock wheezy ISO works fine, I would try latency with that first


>
> Peter C. Wallace skrev den 2015-02-27 16:20:
>> On Fri, 27 Feb 2015, Andreas Pettersson wrote:
>>
>>> Date: Fri, 27 Feb 2015 15:58:08 +0100
>>> From: Andreas Pettersson 
>>> Reply-To: "Enhanced Machine Controller (EMC)"
>>>  
>>> To: "Enhanced Machine Controller (EMC)" 
>>> Subject: [Emc-users] Latency issues; machinekit Kernel
>>>
>>> So i had some issued with drivers, but got around that by installing the
>>> machinekit-rtai kernel
>>> and then manually compile linuxcnc ontop of wheezy..
>>>
>>> Well i dont think its optimal.
>>>
>>> The setup is a Core i5, 8gb ram and running SSD drivers in a raid 0
>>> configuration.
>>> Lets say it boots like its on a ramdrive basicly..
>>>
>>> Funny thing tho. Max Jitter: 815055 so far.. At a Max interval of 1815032.
>>>
>>> I find these numbers extremly high? Shouldnt they be alot.. well lower?
>>>
>>> // Andreas
>> Thats pretty bad, do you have all power management/fan control off in the
>> BIOS?
>>
>>
>> I have a newer MB (ASRock H97M with a Pentium G3258 and built in Intel NIC
>> =21x) that has about 4 usec latency with the stock Wheezy RTAI kernel
>>
>>
>> I didnt try the intel 21x MAC with the RTAI kernel but it works fine with the
>> stock wheezy Preemt-RT kernel (3.2.something)
>>
>>
>>
>>> --
>>> Dive into the World of Parallel Programming The Go Parallel Website, 
>>> sponsored
>>> by Intel and developed in partnership with Slashdot Media, is your hub for 
>>> all
>>> things parallel software development, from weekly thought leadership blogs 
>>> to
>>> news, videos, case studies, tutorials and more. Take a look and join the
>>> conversation now. http://goparallel.sourceforge.net/
>>> ___
>>> Emc-users mailing list
>>> Emc-users@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/emc-users
>>>
>> Peter Wallace
>> Mesa Electronics
>>
>> (\__/)
>> (='.'=) This is Bunny. Copy and paste bunny into your
>> (")_(") signature to help him gain world domination.
>>
>>
>> --
>> Dive into the World of Parallel Programming The Go Parallel Website, 
>> sponsored
>> by Intel and developed in partnership with Slashdot Media, is your hub for 
>> all
>> things parallel software development, from weekly thought leadership blogs to
>> news, videos, case studies, tutorials and more. Take a look and join the
>> conversation now. http://goparallel.sourceforge.net/
>> ___
>> Emc-users mailing list
>> Emc-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/emc-users
>
>
> --
> Dive into the World of Parallel Programming The Go Parallel Website, sponsored
> by Intel and developed in partnership with Slashdot Media, is your hub for all
> things parallel software development, from weekly thought leadership blogs to
> news, videos, case studies, tutorials and more. Take a look and join the
> conversation now. http://goparallel.sourceforge.net/
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>

Peter Wallace
Mesa Electronics

(\__/)
(='.'=) This is Bunny. Copy and paste bunny into your
(")_(") signature to help him gain world domination.


--
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the 
conversation now. http://goparallel.sourceforge.net/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency issues; machinekit Kernel

[Andreas Pettersson]

Are you running a standard Wheezy AMD64 install and the master buildbot 
installations as well then??
I can always reinstall, i rather have a 64bit OS than the PAE version 
anyday.  =)

// A

Peter C. Wallace skrev den 2015-02-27 16:20:
> On Fri, 27 Feb 2015, Andreas Pettersson wrote:
>
>> Date: Fri, 27 Feb 2015 15:58:08 +0100
>> From: Andreas Pettersson 
>> Reply-To: "Enhanced Machine Controller (EMC)"
>>  
>> To: "Enhanced Machine Controller (EMC)" 
>> Subject: [Emc-users] Latency issues; machinekit Kernel
>>
>> So i had some issued with drivers, but got around that by installing the
>> machinekit-rtai kernel
>> and then manually compile linuxcnc ontop of wheezy..
>>
>> Well i dont think its optimal.
>>
>> The setup is a Core i5, 8gb ram and running SSD drivers in a raid 0
>> configuration.
>> Lets say it boots like its on a ramdrive basicly..
>>
>> Funny thing tho. Max Jitter: 815055 so far.. At a Max interval of 1815032.
>>
>> I find these numbers extremly high? Shouldnt they be alot.. well lower?
>>
>> // Andreas
> Thats pretty bad, do you have all power management/fan control off in the
> BIOS?
>
>
> I have a newer MB (ASRock H97M with a Pentium G3258 and built in Intel NIC
> =21x) that has about 4 usec latency with the stock Wheezy RTAI kernel
>
>
> I didnt try the intel 21x MAC with the RTAI kernel but it works fine with the
> stock wheezy Preemt-RT kernel (3.2.something)
>
>
>
>> --
>> Dive into the World of Parallel Programming The Go Parallel Website, 
>> sponsored
>> by Intel and developed in partnership with Slashdot Media, is your hub for 
>> all
>> things parallel software development, from weekly thought leadership blogs to
>> news, videos, case studies, tutorials and more. Take a look and join the
>> conversation now. http://goparallel.sourceforge.net/
>> ___
>> Emc-users mailing list
>> Emc-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/emc-users
>>
> Peter Wallace
> Mesa Electronics
>
> (\__/)
> (='.'=) This is Bunny. Copy and paste bunny into your
> (")_(") signature to help him gain world domination.
>
>
> --
> Dive into the World of Parallel Programming The Go Parallel Website, sponsored
> by Intel and developed in partnership with Slashdot Media, is your hub for all
> things parallel software development, from weekly thought leadership blogs to
> news, videos, case studies, tutorials and more. Take a look and join the
> conversation now. http://goparallel.sourceforge.net/
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users


--
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the 
conversation now. http://goparallel.sourceforge.net/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency issues; machinekit Kernel

[Peter C. Wallace]

On Fri, 27 Feb 2015, Andreas Pettersson wrote:

> Date: Fri, 27 Feb 2015 15:58:08 +0100
> From: Andreas Pettersson 
> Reply-To: "Enhanced Machine Controller (EMC)"
> 
> To: "Enhanced Machine Controller (EMC)" 
> Subject: [Emc-users] Latency issues; machinekit Kernel
> 
> So i had some issued with drivers, but got around that by installing the
> machinekit-rtai kernel
> and then manually compile linuxcnc ontop of wheezy..
>
> Well i dont think its optimal.
>
> The setup is a Core i5, 8gb ram and running SSD drivers in a raid 0
> configuration.
> Lets say it boots like its on a ramdrive basicly..
>
> Funny thing tho. Max Jitter: 815055 so far.. At a Max interval of 1815032.
>
> I find these numbers extremly high? Shouldnt they be alot.. well lower?
>
> // Andreas

Thats pretty bad, do you have all power management/fan control off in the 
BIOS?


I have a newer MB (ASRock H97M with a Pentium G3258 and built in Intel NIC 
=21x) that has about 4 usec latency with the stock Wheezy RTAI kernel


I didnt try the intel 21x MAC with the RTAI kernel but it works fine with the 
stock wheezy Preemt-RT kernel (3.2.something)



>
> --
> Dive into the World of Parallel Programming The Go Parallel Website, sponsored
> by Intel and developed in partnership with Slashdot Media, is your hub for all
> things parallel software development, from weekly thought leadership blogs to
> news, videos, case studies, tutorials and more. Take a look and join the
> conversation now. http://goparallel.sourceforge.net/
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>

Peter Wallace
Mesa Electronics

(\__/)
(='.'=) This is Bunny. Copy and paste bunny into your
(")_(") signature to help him gain world domination.


--
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the 
conversation now. http://goparallel.sourceforge.net/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Latency issues; machinekit Kernel

[Andreas Pettersson]

So i had some issued with drivers, but got around that by installing the 
machinekit-rtai kernel
and then manually compile linuxcnc ontop of wheezy..

Well i dont think its optimal.

The setup is a Core i5, 8gb ram and running SSD drivers in a raid 0 
configuration.
Lets say it boots like its on a ramdrive basicly..

Funny thing tho. Max Jitter: 815055 so far.. At a Max interval of 1815032.

I find these numbers extremly high? Shouldnt they be alot.. well lower?

// Andreas

--
Dive into the World of Parallel Programming The Go Parallel Website, sponsored
by Intel and developed in partnership with Slashdot Media, is your hub for all
things parallel software development, from weekly thought leadership blogs to
news, videos, case studies, tutorials and more. Take a look and join the 
conversation now. http://goparallel.sourceforge.net/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency problem

[Jon Elson]

Dave wrote:
> You won't have any issues with the D525MW and an SSD on that Pico power 
> supply.  I have used several and never had  a failure yet.
>
> Some have 3+ years on them now..running 24x7.
>   
Yes, the D525MW is supposed to have a normal power draw of 13 W,
while a worst-case maximum test was able to draw something over 20 W.
The SSDs generally draw less power when running full blast than many
spinning hard drives do when spun down.  So, the total draw of the system
should be well under 20 W.

Jon

--
Own the Future-Intel(R) Level Up Game Demo Contest 2013
Rise to greatness in Intel's independent game demo contest. Compete 
for recognition, cash, and the chance to get your game on Steam. 
$5K grand prize plus 10 genre and skill prizes. Submit your demo 
by 6/6/13. http://altfarm.mediaplex.com/ad/ck/12124-176961-30367-2
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency problem

[Dave]

You won't have any issues with the D525MW and an SSD on that Pico power 
supply.  I have used several and never had  a failure yet.

Some have 3+ years on them now..running 24x7.

Dave

On 3/28/2013 2:53 PM, bjørn wrote:
> snip
> /"I keep mulling over this "5-6 hours" observation. I know we recommend
> running the latency test for a long time but I don't think we understand
> why these spiking events show up rarely. It's all well and good to blame
> power saving features in modern CPUs but they are functioning over much
> smaller time scales. If they are involved in 5-6 hour spikes, then it
> would seem we're experiencing some sort of "perfect storm" phenomenon."/
>
> You are probably right, this happens after to long a time, and not the
> same each time either. But i am still suspicious  to the blank screen.
> All screensaver functions are already set to "never" thats a bit of a
> mystery.
>
> Now i have updated BIOS but no new options to turn of ACPI and still the
> same latency. so thats a dead end.
>
> The Picopsu is rated 80w as you say, but the "12v brick" that came with
> it is only 60w. and i see in the manual that G0 = Full power>  30 W
> but G1 = 5 W<  power<  52.5 W then with the SSD it could be that it's to
> smal and running hot.
> I will test with a "real powersupply" after the hollydays.
>
> To be continued..
>
> Tusen takk for hjelpen.
>
> Regards Bjørn.
>
>


--
Own the Future-Intel(R) Level Up Game Demo Contest 2013
Rise to greatness in Intel's independent game demo contest. Compete 
for recognition, cash, and the chance to get your game on Steam. 
$5K grand prize plus 10 genre and skill prizes. Submit your demo 
by 6/6/13. http://altfarm.mediaplex.com/ad/ck/12124-176961-30367-2
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency problem

[Jason Burton]

On Mar 28, 2013 2:58 PM, "bjørn"  wrote:
>
> snip
> /"I keep mulling over this "5-6 hours" observation. I know we recommend
> running the latency test for a long time but I don't think we understand
> why these spiking events show up rarely. It's all well and good to blame
> power saving features in modern CPUs but they are functioning over much
> smaller time scales. If they are involved in 5-6 hour spikes, then it
> would seem we're experiencing some sort of "perfect storm" phenomenon."/
>
> You are probably right, this happens after to long a time, and not the
> same each time either. But i am still suspicious  to the blank screen.
> All screensaver functions are already set to "never" thats a bit of a
> mystery

If it's an LCD monitor, perhaps it has to do with two way handshaking
between the monitor and the unix box.

Is there a power save mode built into the monitor itself?

Anything special about the monitor?

As an aside, turn off (in BIOS) any ports you do not need. I've seen Dell
desktops get good latency figures over 24 hour testing with unused ports
disabled. Those hosts used Mesa cards instead of the parallel ports. If I
recall correctly, the rs232 serial ports were disabled as well.

I'll be putting one of them into service on my first retrofit soonish. If
there was interest I could test disabling the ports separately.

Best,
Jason
--
Own the Future-Intel(R) Level Up Game Demo Contest 2013
Rise to greatness in Intel's independent game demo contest. Compete 
for recognition, cash, and the chance to get your game on Steam. 
$5K grand prize plus 10 genre and skill prizes. Submit your demo 
by 6/6/13. http://altfarm.mediaplex.com/ad/ck/12124-176961-30367-2
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency problem

[bjørn]

snip
/"I keep mulling over this "5-6 hours" observation. I know we recommend 
running the latency test for a long time but I don't think we understand 
why these spiking events show up rarely. It's all well and good to blame 
power saving features in modern CPUs but they are functioning over much 
smaller time scales. If they are involved in 5-6 hour spikes, then it 
would seem we're experiencing some sort of "perfect storm" phenomenon."/

You are probably right, this happens after to long a time, and not the 
same each time either. But i am still suspicious  to the blank screen.
All screensaver functions are already set to "never" thats a bit of a 
mystery.

Now i have updated BIOS but no new options to turn of ACPI and still the 
same latency. so thats a dead end.

The Picopsu is rated 80w as you say, but the "12v brick" that came with 
it is only 60w. and i see in the manual that G0 = Full power > 30 W
but G1 = 5 W < power < 52.5 W then with the SSD it could be that it's to 
smal and running hot.
I will test with a "real powersupply" after the hollydays.

To be continued..

Tusen takk for hjelpen.

Regards Bjørn.

-- 
Med vennlig hilsen Bjørn Kristiansen.
www.spesmek.no

--
Own the Future-Intel(R) Level Up Game Demo Contest 2013
Rise to greatness in Intel's independent game demo contest. Compete 
for recognition, cash, and the chance to get your game on Steam. 
$5K grand prize plus 10 genre and skill prizes. Submit your demo 
by 6/6/13. http://altfarm.mediaplex.com/ad/ck/12124-176961-30367-2
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency problem

[Kent A. Reed]

On 3/28/2013 5:23 AM, bjørn wrote:
> that I bought the power supply together with the D525MW board. here are a
> link -
> http://www.ebay.com/itm/160721043666?ssPageName=STRK:MEWAX:IT&_trksid=p3984.m1438.l2649
> it says 60w but i don't know the quality of it. I will test another one
> when i get it in a week.

Actually the "80" in its model number signifies 80w rating. It's a 
lightweight power supply but its specifications 
(http://www.logicsupply.com/products/picopsu_80?gclid=CNjTooGsn7YCFdGd4Aod6lIA-g)
 
suggests it is adequate for the D525MW itself (see table 30, 
http://downloadmirror.intel.com/19123/eng/D525MW_D525MWV_TechProdSpec.pdf). 
I don't know how much additional load you have with your SSD drive, etc.

I don't know how well these little guys filter spikes on the power mains 
but that can be tested by powering it from an UPS or power conditioner. 
(This is just good housekeeping. I haven't had problems with powerline 
noise in my home but it was a problem in my lab at work. It can come 
from unexpected sources. Years ago, a new copier machine in a nearby 
office generated enough EMI/RFI to drive my experimental apparatus crazy.)

These PicoPSUs don't have much thermal mass. You probably should make 
sure it is staying cool.

>
> I to have seen the problem regarding USB and latency on my mill and
> plasma table. I told the guys in the shop to transfer files from USB
> thumbdrive to desktop and remove USB thumbdrive
> before running Linuxcnc
>
> at first i started out with a wireless mouse and keypad with a USB
> dongle, but replaced it with USB when the spikes appeared. i have now
> tested using a ps2 mouse, i do not have a ps2 keyboard so that's still
> USB. but i have tried to disconnect and connect it while the
> latency-test where running and no spikes i left it out and no other USB
> devices connected, let the test run for 5-6 hours but still spikes at
> 5 - 6.
I'd say you've exercised due diligence:-)

I keep mulling over this "5-6 hours" observation. I know we recommend 
running the latency test for a long time but I don't think we understand 
why these spiking events show up rarely. It's all well and good to blame 
power saving features in modern CPUs but they are functioning over much 
smaller time scales. If they are involved in 5-6 hour spikes, then it 
would seem we're experiencing some sort of "perfect storm" phenomenon.

>
> just a tought, what about SATA and SSD disk could that have some
> influence on the latency?

This seems unlikely but device substitution is the usual quick-n-dirty 
test. You could try running the LinuxCNC/LiveCD distribution from a USB 
thumbdrive.  I've never heard anyone raise issues regarding SATA itself.

>
> But anyway i still think i have some powersave funktion running. I
> noticed that the screen was black when i come back to the office, witch
> i havent seen before.

Ubuntu 10.04LTS offers a screen-saver mode that can be disabled via the 
menu bar. Select System.../Preferences.../Power Management and select 
"never" for the display sleep function. If this function is already 
disabled then I'm out of ideas.

> I think i have to go back to check my bios version to see if there are
> an uppgrade or alternative.

It looks like 0130 is the current version 
(http://downloadcenter.intel.com/Detail_Desc.aspx?agr=Y&DwnldID=22377&lang=eng&wapkw=d525mw+bios).
 
I have to say I'm skeptical that your BIOS is the cause of this 
mysterious spiking but I suppose upgrading is worth a try (some say 
upgrading the BIOS falls into the "if it ain't broke, don't fix it" 
category).

> Regards Bjørn

"Lykke til!"

Regards,
Kent


--
Own the Future-Intel® Level Up Game Demo Contest 2013
Rise to greatness in Intel's independent game demo contest.
Compete for recognition, cash, and the chance to get your game 
on Steam. $5K grand prize plus 10 genre and skill prizes. 
Submit your demo by 6/6/13. http://p.sf.net/sfu/intel_levelupd2d
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency problem

[bjørn]

Message: 1 Date: Mon, 25 Mar 2013 15:25:31 -0400 From: "Kent A. Reed" 
 Subject: Re: [Emc-users] Latency problem To: 
"Enhanced Machine Controller (EMC)"  
Message-ID: <5150a4ab.6050...@gmail.com> Content-Type: text/plain; 
charset=ISO-8859-1; format=flowed On 3/25/2013 9:56 AM, Bj?rn wrote:
>>  now i have tested: boot with isolcpu(big 
>> improvement in general latency but still spikes)
>> turn of smi, (improvement in general latency but still spikes)
>> 'idle=poll' and 'nohlt'(little or no effect)
>>
>> acpi = off and api=off (little or no effect)
>>
>> For the moment i have been running the system for 1,5 hour and have - 
>> servothread interval 999655 and max jitter 4495 ns - base thread interval 
>> 32069 max jitter 7190.
>>
>> witch i am very happy with, but if i run for several hours i still get 
>> spikes up to 5 and above.
>> i suspect Michael are right, the problem is related toidle/powersave states.
>>
>> I have no settings in bios to turn of ACPI
>> my bios is: MWNT10N.86A.0083.2011.0524.1600
>> is this the same as others are running? and do you have option to turn off 
>> ACPI from bios?
> Bj?rn:
>
> I don't have an Intel D525MW in my stable of motherboards but I do have
> an ASUS AT5NM10-I whose numbers I posted last year to the table of
> latency-test results in the Wiki. Your "pre-spike" numbers are
> reasonably consistent with the numbers I got. The "best" numbers I
> posted were obtained with hyperthreading turned off in the BIOS and
> isolcpus=1 set in the boot parameters. Nothing heroic. I did my tests
> long before list discussions brought out the importance of other
> parameter settings like idle=poll. In recent shortterm tests, I also
> have not seen dramatic improvements manipulating these parameters.
>
> Testing this ASUS board, I haven't seen the kinds of spikes you are
> experiencing. (Just for the record, I was testing LinuxCNC over 10.04LTS
> with the RTAI kernel).
>
> I'm curious to know what all is attached to your motherboard, especially
> anything attached via USB. I once had a motherboard that started
> misbehaving when I switched from a PS/2 mouse to a USB mouse and a
> number of us have seen the simple act of plugging in a USB thumbdrive
> cause bursts in latency numbers. Is it possible you have a peripheral
> interface/device that is waking up on the same time scale as the
> occurrence of your spikes?
>
> My system guy never failed to ask if the computer were plugged to a live
> outlet as his first step in diagnosing problems. Obviously your computer
> is plugged in but how are the power supply and the AC mains? Adequate,
> stable, clean?
>
> Regards,
> Kent
>
>
I bought the power supply together with the D525MW board. here are a 
link - 
http://www.ebay.com/itm/160721043666?ssPageName=STRK:MEWAX:IT&_trksid=p3984.m1438.l2649
it says 60w but i don't know the quality of it. I will test another one 
when i get it in a week.

I to have seen the problem regarding USB and latency on my mill and 
plasma table. I told the guys in the shop to transfer files from USB 
thumbdrive to desktop and remove USB thumbdrive
before running Linuxcnc

at first i started out with a wireless mouse and keypad with a USB 
dongle, but replaced it with USB when the spikes appeared. i have now 
tested using a ps2 mouse, i do not have a ps2 keyboard so that's still 
USB. but i have tried to disconnect and connect it while the 
latency-test where running and no spikes i left it out and no other USB 
devices connected, let the test run for 5-6 hours but still spikes at 
5 - 6.

just a tought, what about SATA and SSD disk could that have some 
influence on the latency?

But anyway i still think i have some powersave funktion running. I 
noticed that the screen was black when i come back to the office, witch 
i havent seen before.
I think i have to go back to check my bios version to see if there are 
an uppgrade or alternative.

Regards Bjørn

--
Own the Future-Intel® Level Up Game Demo Contest 2013
Rise to greatness in Intel's independent game demo contest.
Compete for recognition, cash, and the chance to get your game 
on Steam. $5K grand prize plus 10 genre and skill prizes. 
Submit your demo by 6/6/13. http://p.sf.net/sfu/intel_levelupd2d
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency problem

[Kent A. Reed]

On 3/25/2013 9:56 AM, Bjørn wrote:
>  now i have tested: boot with isolcpu(big 
> improvement in general latency but still spikes)
>turn of smi, (improvement in general latency but still spikes)
> 'idle=poll' and 'nohlt'(little or no effect)
>
> acpi = off and api=off (little or no effect)
>
> For the moment i have been running the system for 1,5 hour and have - 
> servothread interval 999655 and max jitter 4495 ns - base thread interval 
> 32069 max jitter 7190.
>
> witch i am very happy with, but if i run for several hours i still get spikes 
> up to 5 and above.
> i suspect Michael are right, the problem is related toidle/powersave states.
>
> I have no settings in bios to turn of ACPI
> my bios is: MWNT10N.86A.0083.2011.0524.1600
> is this the same as others are running? and do you have option to turn off 
> ACPI from bios?

Bjørn:

I don't have an Intel D525MW in my stable of motherboards but I do have 
an ASUS AT5NM10-I whose numbers I posted last year to the table of 
latency-test results in the Wiki. Your "pre-spike" numbers are 
reasonably consistent with the numbers I got. The "best" numbers I 
posted were obtained with hyperthreading turned off in the BIOS and 
isolcpus=1 set in the boot parameters. Nothing heroic. I did my tests 
long before list discussions brought out the importance of other 
parameter settings like idle=poll. In recent shortterm tests, I also 
have not seen dramatic improvements manipulating these parameters.

Testing this ASUS board, I haven't seen the kinds of spikes you are 
experiencing. (Just for the record, I was testing LinuxCNC over 10.04LTS 
with the RTAI kernel).

I'm curious to know what all is attached to your motherboard, especially 
anything attached via USB. I once had a motherboard that started 
misbehaving when I switched from a PS/2 mouse to a USB mouse and a 
number of us have seen the simple act of plugging in a USB thumbdrive 
cause bursts in latency numbers. Is it possible you have a peripheral 
interface/device that is waking up on the same time scale as the 
occurrence of your spikes?

My system guy never failed to ask if the computer were plugged to a live 
outlet as his first step in diagnosing problems. Obviously your computer 
is plugged in but how are the power supply and the AC mains? Adequate, 
stable, clean?

Regards,
Kent


--
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_mar
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency

[Peter C. Wallace]

On Mon, 25 Mar 2013, Dave wrote:

> Date: Mon, 25 Mar 2013 13:52:28 -0500
> From: Dave 
> Reply-To: "Enhanced Machine Controller (EMC)"
> 
> To: emc-users@lists.sourceforge.net
> Subject: Re: [Emc-users] Latency
> 
>
> I missed that board and the other integrated Celeron boards.Thanks
> for pointing those out.
>
> Let us know what you find.  That Gigabyte board has a lot to like.
>
> What AMD MITX boards are known to work well?
>
> Dave
>
>> I think I'll order this one to try out today:
>>
>> http://www.newegg.com/Product/Product.aspx?Item=N82E16813128585
>>
>> Whether ubuntu 10.04 will run on this is unknown however
>>
>> I am not a great fan of the Atoms as they are quite slow and have hidden
>> latency problems(They have great latency if you dont do any I/O). The AMD
>> APU's run circles around the Atoms, the Intel Celeron 847s (and 743s) should
>> as well.
>>

This ones a possible:

http://www.newegg.com/Product/Product.aspx?Item=N82E16813128561

Not MITX but know to work well and fanless:

http://www.newegg.com/Product/Product.aspx?Item=N82E16813131875

Many more options if 1x PCIE is OK


>>
>>> On 3/25/2013 6:38 AM, Anders Wallin wrote:
>>>
>>>> On Mon, Mar 25, 2013 at 2:36 AM, Dave   wrote:
>>>>
>>>>
>>>>
>>>>> Has anyone tried running the latency test on the Intel H61, H77, or Z77
>>>>> design motherboards using a CPU with the integrated HD graphics?
>>>>>
>>>>>
>>>>>
>>>> Intel DH61AG  + i3 2120T 2.6 GHz LGA1155
>>>> http://www.anderswallin.net/2012/12/latency-histogram/
>>>>
>>>> It would be great if there was a latency-test that would automatically
>>>> upload results + hardware/software-configuration to a web-hosted 
>>>> database...
>>>>
>>>> Anders
>>>> --
>>>> Everyone hates slow websites. So do we.
>>>> Make your web apps faster with AppDynamics
>>>> Download AppDynamics Lite for free today:
>>>> http://p.sf.net/sfu/appdyn_d2d_mar
>>>> ___
>>>> Emc-users mailing list
>>>> Emc-users@lists.sourceforge.net
>>>> https://lists.sourceforge.net/lists/listinfo/emc-users
>>>>
>>>>
>>>>
>>>
>>> --
>>> Everyone hates slow websites. So do we.
>>> Make your web apps faster with AppDynamics
>>> Download AppDynamics Lite for free today:
>>> http://p.sf.net/sfu/appdyn_d2d_mar
>>> ___
>>> Emc-users mailing list
>>> Emc-users@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/emc-users
>>>
>>>
>> Peter Wallace
>> Mesa Electronics
>>
>> (\__/)
>> (='.'=) This is Bunny. Copy and paste bunny into your
>> (")_(") signature to help him gain world domination.
>>
>>
>> --
>> Everyone hates slow websites. So do we.
>> Make your web apps faster with AppDynamics
>> Download AppDynamics Lite for free today:
>> http://p.sf.net/sfu/appdyn_d2d_mar
>> ___
>> Emc-users mailing list
>> Emc-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/emc-users
>>
>>
>
>
> --
> Everyone hates slow websites. So do we.
> Make your web apps faster with AppDynamics
> Download AppDynamics Lite for free today:
> http://p.sf.net/sfu/appdyn_d2d_mar
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>

Peter Wallace
Mesa Electronics

(\__/)
(='.'=) This is Bunny. Copy and paste bunny into your
(")_(") signature to help him gain world domination.


--
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_mar
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency

[Dave]

I missed that board and the other integrated Celeron boards.Thanks 
for pointing those out.

Let us know what you find.  That Gigabyte board has a lot to like.

What AMD MITX boards are known to work well?

Dave

> I think I'll order this one to try out today:
>
> http://www.newegg.com/Product/Product.aspx?Item=N82E16813128585
>
> Whether ubuntu 10.04 will run on this is unknown however
>
> I am not a great fan of the Atoms as they are quite slow and have hidden
> latency problems(They have great latency if you dont do any I/O). The AMD
> APU's run circles around the Atoms, the Intel Celeron 847s (and 743s) should
> as well.
>
>
>> On 3/25/2013 6:38 AM, Anders Wallin wrote:
>>  
>>> On Mon, Mar 25, 2013 at 2:36 AM, Dave   wrote:
>>>
>>>
>>>
 Has anyone tried running the latency test on the Intel H61, H77, or Z77
 design motherboards using a CPU with the integrated HD graphics?


  
>>> Intel DH61AG  + i3 2120T 2.6 GHz LGA1155
>>> http://www.anderswallin.net/2012/12/latency-histogram/
>>>
>>> It would be great if there was a latency-test that would automatically
>>> upload results + hardware/software-configuration to a web-hosted database...
>>>
>>> Anders
>>> --
>>> Everyone hates slow websites. So do we.
>>> Make your web apps faster with AppDynamics
>>> Download AppDynamics Lite for free today:
>>> http://p.sf.net/sfu/appdyn_d2d_mar
>>> ___
>>> Emc-users mailing list
>>> Emc-users@lists.sourceforge.net
>>> https://lists.sourceforge.net/lists/listinfo/emc-users
>>>
>>>
>>>
>>
>> --
>> Everyone hates slow websites. So do we.
>> Make your web apps faster with AppDynamics
>> Download AppDynamics Lite for free today:
>> http://p.sf.net/sfu/appdyn_d2d_mar
>> ___
>> Emc-users mailing list
>> Emc-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/emc-users
>>
>>  
> Peter Wallace
> Mesa Electronics
>
> (\__/)
> (='.'=) This is Bunny. Copy and paste bunny into your
> (")_(") signature to help him gain world domination.
>
>
> --
> Everyone hates slow websites. So do we.
> Make your web apps faster with AppDynamics
> Download AppDynamics Lite for free today:
> http://p.sf.net/sfu/appdyn_d2d_mar
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
>


--
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_mar
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency

[Luis Bellot]

I'm running the linuxCNC in a P8H61-M LX motherboard with a 6i25 mesa card
toghether a 7i77 to control a 3 servo motor machine without big latency
problems the average is:

Max Jitter servo thread 8840 and base thread 9177 ms. But I always get a
very bad result when plug some usb devices like pendrives or loading the
VitualBox to run XP.

Regards,

Luis Bellot


2013/3/25 Peter C. Wallace 

> On Mon, 25 Mar 2013, Dave wrote:
>
> > Date: Mon, 25 Mar 2013 11:01:45 -0500
> > From: Dave 
> > Reply-To: "Enhanced Machine Controller (EMC)"
> > 
> > To: "Enhanced Machine Controller (EMC)"  >
> > Subject: Re: [Emc-users] Latency
> >
> > Unless I am reading that incorrectly, that appears to be fine for a 1 ms
> > servo thread, but would not work well for direct stepper control.
> >
> > The D525MW boards are really nice.   I have used a lot of them over the
> > last year plus and I have a bit of stockpile built up right now, but the
> > Intel 525 CPU is being phased out.  The Intel D525MW board is already
> > discontinued but
> > they have extended production probably because the D525MW replacements
> > have so many problems.  The newer Intel D2700MUD, 2500, etc boards video
> > chip is only Windows 7 32 bit compatible which is really silly.
> > The Intel manager who made that decision must have been using medical
> > marijuana at the time!
> >
> > Jetway has a "long term available" D525 board but it is about $200 now
> > and there is no assurance that they will not be phased this year also.
> >
> > Also PCI slots are getting rarer on the MITX boards.  So when you take
> > all of the available MITX boards, and eliminate the newer Intel CPUs,
> > and the boards without PCI slots, you end up with very few offerings.
> >
> > That's my take on the current Motherboard situation..
> >
> > Hence my question about the newer Intel Design MITX boards.
> >
> > I'd like to hear other viewpoints if you have differing opinions.
> >
> > Thanks,
> >
> > Dave
>
>
> I think I'll order this one to try out today:
>
> http://www.newegg.com/Product/Product.aspx?Item=N82E16813128585
>
> Whether ubuntu 10.04 will run on this is unknown however
>
> I am not a great fan of the Atoms as they are quite slow and have hidden
> latency problems(They have great latency if you dont do any I/O). The AMD
> APU's run circles around the Atoms, the Intel Celeron 847s (and 743s)
> should
> as well.
>
> >
> > On 3/25/2013 6:38 AM, Anders Wallin wrote:
> >> On Mon, Mar 25, 2013 at 2:36 AM, Dave  wrote:
> >>
> >>
> >>> Has anyone tried running the latency test on the Intel H61, H77, or Z77
> >>> design motherboards using a CPU with the integrated HD graphics?
> >>>
> >>>
> >> Intel DH61AG  + i3 2120T 2.6 GHz LGA1155
> >> http://www.anderswallin.net/2012/12/latency-histogram/
> >>
> >> It would be great if there was a latency-test that would automatically
> >> upload results + hardware/software-configuration to a web-hosted
> database...
> >>
> >> Anders
> >>
> --
> >> Everyone hates slow websites. So do we.
> >> Make your web apps faster with AppDynamics
> >> Download AppDynamics Lite for free today:
> >> http://p.sf.net/sfu/appdyn_d2d_mar
> >> ___
> >> Emc-users mailing list
> >> Emc-users@lists.sourceforge.net
> >> https://lists.sourceforge.net/lists/listinfo/emc-users
> >>
> >>
> >
> >
> >
> --
> > Everyone hates slow websites. So do we.
> > Make your web apps faster with AppDynamics
> > Download AppDynamics Lite for free today:
> > http://p.sf.net/sfu/appdyn_d2d_mar
> > ___
> > Emc-users mailing list
> > Emc-users@lists.sourceforge.net
> > https://lists.sourceforge.net/lists/listinfo/emc-users
> >
>
> Peter Wallace
> Mesa Electronics
>
> (\__/)
> (='.'=) This is Bunny. Copy and paste bunny into your
> (")_(") signature to help him gain world domination.
>
>
>
> --
> Everyone hates slow websites. So do we.
> Make your web apps faster with AppDynamics
> Download AppDynamics Lite for free today:
> http://p.sf.net/sfu/appdyn_d2d_mar
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
--
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_mar
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency

[Peter C. Wallace]

On Mon, 25 Mar 2013, Dave wrote:

> Date: Mon, 25 Mar 2013 11:01:45 -0500
> From: Dave 
> Reply-To: "Enhanced Machine Controller (EMC)"
> 
> To: "Enhanced Machine Controller (EMC)" 
> Subject: Re: [Emc-users] Latency
> 
> Unless I am reading that incorrectly, that appears to be fine for a 1 ms
> servo thread, but would not work well for direct stepper control.
>
> The D525MW boards are really nice.   I have used a lot of them over the
> last year plus and I have a bit of stockpile built up right now, but the
> Intel 525 CPU is being phased out.  The Intel D525MW board is already
> discontinued but
> they have extended production probably because the D525MW replacements
> have so many problems.  The newer Intel D2700MUD, 2500, etc boards video
> chip is only Windows 7 32 bit compatible which is really silly.
> The Intel manager who made that decision must have been using medical
> marijuana at the time!
>
> Jetway has a "long term available" D525 board but it is about $200 now
> and there is no assurance that they will not be phased this year also.
>
> Also PCI slots are getting rarer on the MITX boards.  So when you take
> all of the available MITX boards, and eliminate the newer Intel CPUs,
> and the boards without PCI slots, you end up with very few offerings.
>
> That's my take on the current Motherboard situation..
>
> Hence my question about the newer Intel Design MITX boards.
>
> I'd like to hear other viewpoints if you have differing opinions.
>
> Thanks,
>
> Dave


I think I'll order this one to try out today:

http://www.newegg.com/Product/Product.aspx?Item=N82E16813128585

Whether ubuntu 10.04 will run on this is unknown however

I am not a great fan of the Atoms as they are quite slow and have hidden 
latency problems(They have great latency if you dont do any I/O). The AMD 
APU's run circles around the Atoms, the Intel Celeron 847s (and 743s) should 
as well.

>
> On 3/25/2013 6:38 AM, Anders Wallin wrote:
>> On Mon, Mar 25, 2013 at 2:36 AM, Dave  wrote:
>>
>>
>>> Has anyone tried running the latency test on the Intel H61, H77, or Z77
>>> design motherboards using a CPU with the integrated HD graphics?
>>>
>>>
>> Intel DH61AG  + i3 2120T 2.6 GHz LGA1155
>> http://www.anderswallin.net/2012/12/latency-histogram/
>>
>> It would be great if there was a latency-test that would automatically
>> upload results + hardware/software-configuration to a web-hosted database...
>>
>> Anders
>> --
>> Everyone hates slow websites. So do we.
>> Make your web apps faster with AppDynamics
>> Download AppDynamics Lite for free today:
>> http://p.sf.net/sfu/appdyn_d2d_mar
>> ___
>> Emc-users mailing list
>> Emc-users@lists.sourceforge.net
>> https://lists.sourceforge.net/lists/listinfo/emc-users
>>
>>
>
>
> --
> Everyone hates slow websites. So do we.
> Make your web apps faster with AppDynamics
> Download AppDynamics Lite for free today:
> http://p.sf.net/sfu/appdyn_d2d_mar
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>

Peter Wallace
Mesa Electronics

(\__/)
(='.'=) This is Bunny. Copy and paste bunny into your
(")_(") signature to help him gain world domination.


--
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_mar
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency

[Dave]

Unless I am reading that incorrectly, that appears to be fine for a 1 ms 
servo thread, but would not work well for direct stepper control.

The D525MW boards are really nice.   I have used a lot of them over the 
last year plus and I have a bit of stockpile built up right now, but the 
Intel 525 CPU is being phased out.  The Intel D525MW board is already 
discontinued but
they have extended production probably because the D525MW replacements 
have so many problems.  The newer Intel D2700MUD, 2500, etc boards video 
chip is only Windows 7 32 bit compatible which is really silly.
The Intel manager who made that decision must have been using medical 
marijuana at the time!

Jetway has a "long term available" D525 board but it is about $200 now 
and there is no assurance that they will not be phased this year also.

Also PCI slots are getting rarer on the MITX boards.  So when you take 
all of the available MITX boards, and eliminate the newer Intel CPUs, 
and the boards without PCI slots, you end up with very few offerings.

That's my take on the current Motherboard situation..

Hence my question about the newer Intel Design MITX boards.

I'd like to hear other viewpoints if you have differing opinions.

Thanks,

Dave

On 3/25/2013 6:38 AM, Anders Wallin wrote:
> On Mon, Mar 25, 2013 at 2:36 AM, Dave  wrote:
>
>
>> Has anyone tried running the latency test on the Intel H61, H77, or Z77
>> design motherboards using a CPU with the integrated HD graphics?
>>
>>  
> Intel DH61AG  + i3 2120T 2.6 GHz LGA1155
> http://www.anderswallin.net/2012/12/latency-histogram/
>
> It would be great if there was a latency-test that would automatically
> upload results + hardware/software-configuration to a web-hosted database...
>
> Anders
> --
> Everyone hates slow websites. So do we.
> Make your web apps faster with AppDynamics
> Download AppDynamics Lite for free today:
> http://p.sf.net/sfu/appdyn_d2d_mar
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
>


--
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_mar
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency problem

[Bjørn]

Am 20.03.2013 um 08:48 schrieb Bj?rn:

> I bougt this.
> http://www.ebay.com/itm/200542103383?ssPageName=STRK:MEWAX:IT&_trksid=p3984.m1438.l2649
> running the latest version of linuxcnc.

specifically which version?

I have a D525MW and get much better results

> Hypertreading off in bios,
> ISOLCPU is done.
>
> when runnnig latencytest and 3 x glxgears + some video i can run for 30
> min and get:
> servo thread (1.0ms) max interval (ns) 1006129 max jitter 11104
> Base thread (25.0ps) max interval (ns) 42342 max jitter 17463
>
> i think i could do with this but if i run overnight the same way i get
> servo thread (1.0ms) max interval (ns) 1020138 max jitter 24978
> Base thread (25.0ps) max interval (ns) 83039 max jitter 58160
>
> does anyone have any idea of hvat this can be? is there a way to se any
> debugging information?

usually that points to CPU idle/powersave states; it has turned out that wakeup 
from idle/powersave states is rather nondeterministic and a likely cause of 
latency spikes

to isolate whether this is the case, try to boot the kernel with the option 
'idle=poll'

if things improve, idle/powersave states are very likely the cause

-m


> And another question, is it possible to dedicate one cpu to run gui and
> other non realtime funktions?
>
> Regards Bjorn
>
I am running 10.04 LTS and LinuxCNC 2.5.2

when you say, boot the kernel with the option 'idle=poll' is this done
the same way as isolcpu?

Regards Bjorn Kristiansen.

yes

try both 'idle=poll' and 'nohlt'

see section 4 of http://wiki.linuxcnc.org/cgi-bin/wiki.pl?XenomaiKernelPackages
this applies to RTAI just alike

- Michael

now i have tested: boot with isolcpu(big improvement in general latency but 
still spikes)  
  turn of smi, (improvement in general latency but still spikes)
'idle=poll' and 'nohlt'(little or no effect)

acpi = off and api=off (little or no effect)

For the moment i have been running the system for 1,5 hour and have - 
servothread interval 999655 and max jitter 4495 ns - base thread interval 32069 
max jitter 7190.

witch i am very happy with, but if i run for several hours i still get spikes 
up to 5 and above.
i suspect Michael are right, the problem is related toidle/powersave states.

I have no settings in bios to turn of ACPI
my bios is: MWNT10N.86A.0083.2011.0524.1600
is this the same as others are running? and do you have option to turn off ACPI 
from bios?

-- 
Regards: Bjørn Kristiansen

--
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_mar
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency

[Anders Wallin]

On Mon, Mar 25, 2013 at 2:36 AM, Dave  wrote:

> Has anyone tried running the latency test on the Intel H61, H77, or Z77
> design motherboards using a CPU with the integrated HD graphics?
>

Intel DH61AG  + i3 2120T 2.6 GHz LGA1155
http://www.anderswallin.net/2012/12/latency-histogram/

It would be great if there was a latency-test that would automatically
upload results + hardware/software-configuration to a web-hosted database...

Anders
--
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_mar
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency

[Peter C. Wallace]

On Sun, 24 Mar 2013, Dave wrote:

> Date: Sun, 24 Mar 2013 19:36:47 -0500
> From: Dave 
> Reply-To: "Enhanced Machine Controller (EMC)"
> 
> To: emc-users@lists.sourceforge.net
> Subject: [Emc-users] Latency
> 
> Has anyone tried running the latency test on the Intel H61, H77, or Z77
> design motherboards using a CPU with the integrated HD graphics?
>
> A number of companies make them and they all seem to be socket 1155 boards.
>
> The boards cost $50-$130 each and a 35 watt CPU runs about $80.
>
> Thanks,  Dave
>
I've been meaning to try one (also the celeron 847 which is ~a dual core I3 at 
1GHz)

>
>
> --
> Everyone hates slow websites. So do we.
> Make your web apps faster with AppDynamics
> Download AppDynamics Lite for free today:
> http://p.sf.net/sfu/appdyn_d2d_mar
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>

Peter Wallace
Mesa Electronics

(\__/)
(='.'=) This is Bunny. Copy and paste bunny into your
(")_(") signature to help him gain world domination.


--
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_mar
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Latency

[Dave]

Has anyone tried running the latency test on the Intel H61, H77, or Z77 
design motherboards using a CPU with the integrated HD graphics?

A number of companies make them and they all seem to be socket 1155 boards.

The boards cost $50-$130 each and a 35 watt CPU runs about $80.

Thanks,  Dave



--
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_mar
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency problem

[Michael Haberler]

>> 
>> Regards Bjorn
>> 
> I am running 10.04 LTS and LinuxCNC 2.5.2
> 
> when you say, boot the kernel with the option 'idle=poll' is this done 
> the same way as isolcpu?

yes

try both 'idle=poll' and 'nohlt'

see section 4 of http://wiki.linuxcnc.org/cgi-bin/wiki.pl?XenomaiKernelPackages 
this applies to RTAI just alike

- Michael


--
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_mar
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency problem

[Bjørn]

Am 20.03.2013 um 08:48 schrieb Bj?rn:

> I bougt this.
> http://www.ebay.com/itm/200542103383?ssPageName=STRK:MEWAX:IT&_trksid=p3984.m1438.l2649
> running the latest version of linuxcnc.

specifically which version?

I have a D525MW and get much better results

> Hypertreading off in bios,
> ISOLCPU is done.
>
> when runnnig latencytest and 3 x glxgears + some video i can run for 30
> min and get:
> servo thread (1.0ms) max interval (ns) 1006129 max jitter 11104
> Base thread (25.0ps) max interval (ns) 42342 max jitter 17463
>
> i think i could do with this but if i run overnight the same way i get
> servo thread (1.0ms) max interval (ns) 1020138 max jitter 24978
> Base thread (25.0ps) max interval (ns) 83039 max jitter 58160
>
> does anyone have any idea of hvat this can be? is there a way to se any
> debugging information?

usually that points to CPU idle/powersave states; it has turned out that wakeup 
from idle/powersave states is rather nondeterministic and a likely cause of 
latency spikes

to isolate whether this is the case, try to boot the kernel with the option 
'idle=poll'

if things improve, idle/powersave states are very likely the cause

-m


> And another question, is it possible to dedicate one cpu to run gui and
> other non realtime funktions?
>
> Regards Bjorn
>
I am running 10.04 LTS and LinuxCNC 2.5.2

when you say, boot the kernel with the option 'idle=poll' is this done 
the same way as isolcpu?

Regards Bjorn Kristiansen.



--
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_mar
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency problem

[Michael Haberler]

Am 20.03.2013 um 08:48 schrieb Bjørn:

> I bougt this.
> http://www.ebay.com/itm/200542103383?ssPageName=STRK:MEWAX:IT&_trksid=p3984.m1438.l2649
> running the latest version of linuxcnc.

specifically which version? 

I have a D525MW and get much better results

> 
> Hypertreading off in bios,
> ISOLCPU is done.
> 
> when runnnig latencytest and 3 x glxgears + some video i can run for 30 
> min and get:
> servo thread (1.0ms) max interval (ns) 1006129 max jitter 11104
> Base thread (25.0ps) max interval (ns) 42342 max jitter 17463
> 
> i think i could do with this but if i run overnight the same way i get
> servo thread (1.0ms) max interval (ns) 1020138 max jitter 24978
> Base thread (25.0ps) max interval (ns) 83039 max jitter 58160
> 
> does anyone have any idea of hvat this can be? is there a way to se any 
> debugging information?

usually that points to CPU idle/powersave states; it has turned out that wakeup 
from idle/powersave states is rather nondeterministic and a likely cause of 
latency spikes

to isolate whether this is the case, try to boot the kernel with the option 
'idle=poll'

if things improve, idle/powersave states are very likely the cause

-m 


> 
> And another question, is it possible to dedicate one cpu to run gui and 
> other non realtime funktions?
> 
> Regards Bjorn
> 
> 
> --
> Everyone hates slow websites. So do we.
> Make your web apps faster with AppDynamics
> Download AppDynamics Lite for free today:
> http://p.sf.net/sfu/appdyn_d2d_mar
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users


--
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_mar
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Latency problem

[Bjørn]

I bougt this.
http://www.ebay.com/itm/200542103383?ssPageName=STRK:MEWAX:IT&_trksid=p3984.m1438.l2649
running the latest version of linuxcnc.

Hypertreading off in bios,
ISOLCPU is done.

when runnnig latencytest and 3 x glxgears + some video i can run for 30 
min and get:
servo thread (1.0ms) max interval (ns) 1006129 max jitter 11104
Base thread (25.0ps) max interval (ns) 42342 max jitter 17463

i think i could do with this but if i run overnight the same way i get
servo thread (1.0ms) max interval (ns) 1020138 max jitter 24978
Base thread (25.0ps) max interval (ns) 83039 max jitter 58160

does anyone have any idea of hvat this can be? is there a way to se any 
debugging information?

And another question, is it possible to dedicate one cpu to run gui and 
other non realtime funktions?

Regards Bjorn


--
Everyone hates slow websites. So do we.
Make your web apps faster with AppDynamics
Download AppDynamics Lite for free today:
http://p.sf.net/sfu/appdyn_d2d_mar
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency test

[gene heskett]

On Saturday, March 31, 2012 01:32:39 PM Kirk Wallace did opine:

> On Sat, 2012-03-31 at 08:00 +0200, Alwyn McLeod wrote:
> > When I run the latency test, I get Base thread jitter of 932659. In
> > the config file the max. is set at 50,000.
> > 
> > When I load EMC2 I get a message "Unexpected real time delay: check
> > dmesg for details' as well " RTAPI: ERROR:Unexpected realtime delay
> > on task 1. This message will only display once per session. Run
> > latency test and resolve before continuing.
> > 
> > Being a novice as far as Ubuntu is concerned, I am at a loss as to how
> > I can resolve these issues. Any assistance will be appreciated.
> 
> The jitter you measured is too high, so you'll need to make some
> configuration changes to make it better. If your PC where sitting in
> front of me, I would first check on what slots it has, the processor
> type and chipset, what video port is being used as well as the monitor.
> This would give me an idea of what changes can be made to keep the video
> system from interfering with latency. Also I would watch the latency
> test for a while to see when the really bad numbers come up. If the
> numbers jump right a way, my guess is the video driver, if some event
> like a disk access or plugging in or accessing a USB device, maybe a
> related BIOS setting may be needed. If it happens and repeats after a
> couple of minutes, it might be the SMI issue mentioned in the trouble
> shooting section on the wiki. When the latency numbers start getting
> pretty good from the changes you have made, it would be best to run the
> test over night to make sure something else that operates on a long time
> base doesn't slip through.
> 
> Without knowing all of the above, my most common solution tends to be
> getting the xorg.conf file settings to work. On a new install, this file
> doesn't exist. So one will need to be made. If you have an xorg.conf
> file, make a copy of it so you can reuse it if needed.
> 
> I usually start a terminal window (from the Desktop; Applications /
> Accessories / Terminal), type on the command line "sudo
> gedit /etc/X11/xorg.conf". This will give you root privileges so you can
> edit a system file, then gEdit will create the new file and you can fill
> in what is needed. Usually I add something similar to:
> ~
> Section "Device"
>   Identifier  "Configured Video Device"
> # Driver  "nvidia"
>   Driver  "vesa"
> EndSection
 
This does generally help considerably.  Unforch, the vesa driver, specs 
written back in 4/3 CRT display days, has not kept pace with the available 
monitors, which are all only available in 16/9 formats, at least here in 
the states, so while the vesa driver will work at its maximum 1024x768 
resolution, on modern monitors you no longer have 'square pixels, meaning 
round objects aren't round, but oval.

Occasionally a 3rd party driver will work well.  On the box I just retired 
because the motherboard caps were failing, I had an ATI 9200SE card in it, 
and switching to the ati driver as above, which in turn loads the correct 
ATI driver for that card, I was amazed to find that latencies had not taken 
a big hit (unlike the nvidia drivers which are truly horrible to the 
latency figures) and that I once again had square pixels on a 16/9 monitor.
In fact it worked amazingly well.

But as in all things video vs linuxcnc, YMMV.


> Section "Monitor"
> Identifier  "Configured Monitor"
> Option  "DPMS"
> Horizsync   30-66
> Vertrefresh 50-130
> EndSection
> 
> Section "Screen"
>   Identifier  "Default Screen"
>   Monitor "Configured Monitor"
>   Device  "Configured Video Device"
>   Defaultdepth24
>   SubSection "Display"
>   Viewport0 0
>   Depth   24
>   Modes   "1280x1024"
>   EndSubSection
> EndSection
> 
> Section "ServerLayout"
>   Identifier  "Default Layout"
>   screen  "Default Screen"
> EndSection
> ~~
> 
> The # comments out the current video driver if this line exists, then I
> load a generic driver (vesa) that plays well with latency. Sometimes
> there is an open source version of a proprietary driver, which can work
> well and have more features than vesa.

The nouveau driver seems to be working very well when it works, but isn't 
100% bulletproof _yet_.  I just ran it here for a couple weeks.  Text 
rendering is better than nvidia, but I had several outright crashes, 
usually the reboot screen being the first hint that things have gone face 
down in the oatmeal.  No traces in the logs, just the normal reboot 
sequence.
 
> The monitor section lets the computer know what frequencies and
> resolutions are valid for the monitor. My equipment tends to be very old
> and Xorg has a hard time getting the settings on its own. One thing I do
> first is to bring up System / Preferences / Monit

Re: [Emc-users] Latency test

[Eric Keller]

On Sat, Mar 31, 2012 at 2:00 AM, Alwyn McLeod wrote:

> When I run the latency test, I get Base thread jitter of 932659. In the
> config file the max. is set at 50,000.
>
>
> Being a novice as far as Ubuntu is concerned, I am at a loss as to how I
> can resolve these issues. Any assistance will be appreciated.
>
> although you have gotten some good advice, it's really hard to help
without knowing the details of the system.  Motherboard, processor, ram,
video card in particular

Eric
--
This SF email is sponsosred by:
Try Windows Azure free for 90 days Click Here 
http://p.sf.net/sfu/sfd2d-msazure
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency test

[Kirk Wallace]

On Sat, 2012-03-31 at 08:00 +0200, Alwyn McLeod wrote:
> When I run the latency test, I get Base thread jitter of 932659. In the
> config file the max. is set at 50,000.
> 
> When I load EMC2 I get a message "Unexpected real time delay: check dmesg
> for details' as well " RTAPI: ERROR:Unexpected realtime delay on task 1.
> This message will only display once per session. Run latency test and
> resolve before continuing.
> 
> Being a novice as far as Ubuntu is concerned, I am at a loss as to how I
> can resolve these issues. Any assistance will be appreciated.

The jitter you measured is too high, so you'll need to make some
configuration changes to make it better. If your PC where sitting in
front of me, I would first check on what slots it has, the processor
type and chipset, what video port is being used as well as the monitor.
This would give me an idea of what changes can be made to keep the video
system from interfering with latency. Also I would watch the latency
test for a while to see when the really bad numbers come up. If the
numbers jump right a way, my guess is the video driver, if some event
like a disk access or plugging in or accessing a USB device, maybe a
related BIOS setting may be needed. If it happens and repeats after a
couple of minutes, it might be the SMI issue mentioned in the trouble
shooting section on the wiki. When the latency numbers start getting
pretty good from the changes you have made, it would be best to run the
test over night to make sure something else that operates on a long time
base doesn't slip through.

Without knowing all of the above, my most common solution tends to be
getting the xorg.conf file settings to work. On a new install, this file
doesn't exist. So one will need to be made. If you have an xorg.conf
file, make a copy of it so you can reuse it if needed.

I usually start a terminal window (from the Desktop; Applications /
Accessories / Terminal), type on the command line "sudo
gedit /etc/X11/xorg.conf". This will give you root privileges so you can
edit a system file, then gEdit will create the new file and you can fill
in what is needed. Usually I add something similar to:
~
Section "Device"
Identifier  "Configured Video Device"
#   Driver  "nvidia"
Driver  "vesa"
EndSection

Section "Monitor"
Identifier  "Configured Monitor"
Option  "DPMS"
Horizsync   30-66
Vertrefresh 50-130
EndSection

Section "Screen"
Identifier  "Default Screen"
Monitor "Configured Monitor"
Device  "Configured Video Device"
Defaultdepth24
SubSection "Display"
Viewport0 0
Depth   24
Modes   "1280x1024"
EndSubSection
EndSection

Section "ServerLayout"
Identifier  "Default Layout"
screen  "Default Screen"
EndSection
~~

The # comments out the current video driver if this line exists, then I
load a generic driver (vesa) that plays well with latency. Sometimes
there is an open source version of a proprietary driver, which can work
well and have more features than vesa.

The monitor section lets the computer know what frequencies and
resolutions are valid for the monitor. My equipment tends to be very old
and Xorg has a hard time getting the settings on its own. One thing I do
first is to bring up System / Preferences / Monitors and see if the
monitor is recognized and the proper resolutions listed. If not, I add
the Monitor and Screen sections to xorg.conf. If you add the monitor
data, it is very important you use the actual data that matches the
make, model and version of your monitor. Setting these wrong can make
you monitor unreadable or go blank, and for a Linux novice can be very
hard to get out of gracefully. By the way, if this happens, Alt F1 may
get you to a terminal session. If Xorg does get some of these settings
correctly, then you can leave those parts out. Getting this sorted out
can take hours and a lot of rebooting. Oh, and when you save the new
settings, just log out and back in (use the power icon and log out
instead of power off) in order to activate the new settings or reboot.

If the latency problem is not video related, you may need to study the
troubleshooting and related sections in the wiki and other
documentation.

Don't be afraid to ask questions, good luck.

-- 
Kirk Wallace
http://www.wallacecompany.com/machine_shop/
http://www.wallacecompany.com/E45/index.html
California, USA


--
This SF email is sponsosred by:
Try Windows Azure free for 90 days Click Here 
http://p.sf.net/sfu/sfd2d-msazure
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency test

[Erik Friesen]

The simple explanation is that is a lot of jitter.  Look here to debug -
http://wiki.linuxcnc.org/cgi-bin/wiki.pl?TroubleShooting

I had a graphics card that ran at 200,000 when using the manufacturer 3d
driver, but down to 40,000 with software 3d.

On Sat, Mar 31, 2012 at 2:40 AM, Viesturs Lācis wrote:

> 2012/3/31 Alwyn McLeod :
> > When I run the latency test, I get Base thread jitter of 932659. In the
> > config file the max. is set at 50,000.
>
> What You set in INI file is the length of the particular thread. I
> guess that length of base thread is set at 50'000 ns and length of
> servo thread is 1'000'000 ns.
>
> You do not set a jitter value anywhere. And 932'659 ns delay in a
> 50'000 thread means that interrupt was almost 19 consecutive thread
> periods. The only way I know, how such numbers can be obtained, is
> running LinuxCNC and Ubuntu in a virtual machine. I have never heard
> of anything close to this, when LinuxCNC is ran in "normal" way.
>
> I think You should share some information, what hardware do You have,
> which version of LinuxCNC on which version of Ubuntu do You use and
> probably paste Your config and preferably paste the screenshot with
> the latency test results (let latency test run for some 30 min, try to
> load the pc in a meantime by pressing alt+f2 and typing 'glxgears' and
> hitting 'enter', open firefox, browse some web page, if You have
> access, move that firefox window across the screen, copy some files,
> do whatever else You can think of to load the pc) in picpaste.com or
> similar place.
>
> Viesturs
>
>
> --
> This SF email is sponsosred by:
> Try Windows Azure free for 90 days Click Here
> http://p.sf.net/sfu/sfd2d-msazure
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
--
This SF email is sponsosred by:
Try Windows Azure free for 90 days Click Here 
http://p.sf.net/sfu/sfd2d-msazure
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency test

[Viesturs Lācis]

2012/3/31 Alwyn McLeod :
> When I run the latency test, I get Base thread jitter of 932659. In the
> config file the max. is set at 50,000.

What You set in INI file is the length of the particular thread. I
guess that length of base thread is set at 50'000 ns and length of
servo thread is 1'000'000 ns.

You do not set a jitter value anywhere. And 932'659 ns delay in a
50'000 thread means that interrupt was almost 19 consecutive thread
periods. The only way I know, how such numbers can be obtained, is
running LinuxCNC and Ubuntu in a virtual machine. I have never heard
of anything close to this, when LinuxCNC is ran in "normal" way.

I think You should share some information, what hardware do You have,
which version of LinuxCNC on which version of Ubuntu do You use and
probably paste Your config and preferably paste the screenshot with
the latency test results (let latency test run for some 30 min, try to
load the pc in a meantime by pressing alt+f2 and typing 'glxgears' and
hitting 'enter', open firefox, browse some web page, if You have
access, move that firefox window across the screen, copy some files,
do whatever else You can think of to load the pc) in picpaste.com or
similar place.

Viesturs

--
This SF email is sponsosred by:
Try Windows Azure free for 90 days Click Here 
http://p.sf.net/sfu/sfd2d-msazure
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Latency test

[Alwyn McLeod]

When I run the latency test, I get Base thread jitter of 932659. In the
config file the max. is set at 50,000.

When I load EMC2 I get a message "Unexpected real time delay: check dmesg
for details' as well " RTAPI: ERROR:Unexpected realtime delay on task 1.
This message will only display once per session. Run latency test and
resolve before continuing.

Being a novice as far as Ubuntu is concerned, I am at a loss as to how I
can resolve these issues. Any assistance will be appreciated.

-- 
Alwyn McLeod
Cell 073 500 6992
Fax 086 510 8765
--
This SF email is sponsosred by:
Try Windows Azure free for 90 days Click Here 
http://p.sf.net/sfu/sfd2d-msazure
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency - was Re: Which video card/driver for LinuxCNC?

[Eric Keller]

On Tue, Feb 21, 2012 at 1:26 PM, Erik Friesen  wrote:
> It would be nice to know enough to debug latency issues, instead it feels
> like kludging around with no idea what is the cause.  I suppose rtai has
> some tools if you know what you are doing.
Do they?  It's been a while since I have followed development, but I
had never seen any information about that kind of tool.  While
tracking down latency problems is tricky, and sometimes impossible, it
is deterministic enough that you can feel confident you fixed things
when you are done.
Eric

--
Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-d2d
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency - was Re: Which video card/driver for LinuxCNC?

[Erik Friesen]

It would be nice to know enough to debug latency issues, instead it feels
like kludging around with no idea what is the cause.  I suppose rtai has
some tools if you know what you are doing.

On Tue, Feb 21, 2012 at 1:12 PM, Kirk Wallace
wrote:

> On Tue, 2012-02-21 at 03:47 -0800, charles green wrote:
> > the allocation of computing resources seems to be a recurring theme
> > with pc based machine controllers.
> >
> > while i have wished for an accesible pocket calculator gui more than
> > one time while standing in front of a machining center console, i have
> > almost never wanted to surf the web or send emails while doing so.
> >
> > the enhanced machine controller spans a disparity between its intended
> > function and the common usage of it's supporting hardware.
>
> I tend to think the CNC PC should only have CNC software on it, it seems
> more grown up to have a purpose built machine that requires a certain
> amount of discipline to join the club, but the more I use my LinuxCNC
> PC's the more I also find I use the generic programs. While tuning, I
> need to access the Net to find what I or FF does. While setting up a new
> piece of hardware, I need to find a pin-out or spec., or download grpn
> so I can calculate a component value. (By the way, I've found the
> default gcalctool has a better Hex<->Dec feature, so I usually have both
> on the screen.) I often have my VFD PDF manual on the screen. Inkscape
> and gEdit are often running for documentation purposes. gFTP is up too,
> so I can store and share my work. It all comes in handy, and in reality,
> none of it detracts from the core LinuxCNC function.
>
> My biggest frustration now is in finding PC hardware that plays well
> with Ubuntu and doesn't kill latency. The choice here is, develop new
> purpose built hardware and maybe rebuild LinuxCNC, or spend a day or two
> swapping out hardware until I get something that works well. Aimlessly
> fiddling with PC hardware for a couple days is painful, but months and
> months of development is not one of my options.
>
> --
> Kirk Wallace
> http://www.wallacecompany.com/machine_shop/
> http://www.wallacecompany.com/E45/index.html
> California, USA
>
>
>
> --
> Keep Your Developer Skills Current with LearnDevNow!
> The most comprehensive online learning library for Microsoft developers
> is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
> Metro Style Apps, more. Free future releases when you subscribe now!
> http://p.sf.net/sfu/learndevnow-d2d
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
--
Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-d2d
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency - was Re: Which video card/driver for LinuxCNC?

[Kirk Wallace]

On Tue, 2012-02-21 at 03:47 -0800, charles green wrote:
> the allocation of computing resources seems to be a recurring theme
> with pc based machine controllers.
>  
> while i have wished for an accesible pocket calculator gui more than
> one time while standing in front of a machining center console, i have
> almost never wanted to surf the web or send emails while doing so.
>  
> the enhanced machine controller spans a disparity between its intended
> function and the common usage of it's supporting hardware.

I tend to think the CNC PC should only have CNC software on it, it seems
more grown up to have a purpose built machine that requires a certain
amount of discipline to join the club, but the more I use my LinuxCNC
PC's the more I also find I use the generic programs. While tuning, I
need to access the Net to find what I or FF does. While setting up a new
piece of hardware, I need to find a pin-out or spec., or download grpn
so I can calculate a component value. (By the way, I've found the
default gcalctool has a better Hex<->Dec feature, so I usually have both
on the screen.) I often have my VFD PDF manual on the screen. Inkscape
and gEdit are often running for documentation purposes. gFTP is up too,
so I can store and share my work. It all comes in handy, and in reality,
none of it detracts from the core LinuxCNC function.

My biggest frustration now is in finding PC hardware that plays well
with Ubuntu and doesn't kill latency. The choice here is, develop new
purpose built hardware and maybe rebuild LinuxCNC, or spend a day or two
swapping out hardware until I get something that works well. Aimlessly
fiddling with PC hardware for a couple days is painful, but months and
months of development is not one of my options.

-- 
Kirk Wallace
http://www.wallacecompany.com/machine_shop/
http://www.wallacecompany.com/E45/index.html
California, USA


--
Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-d2d
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency - was Re: Which video card/driver for LinuxCNC?

[James Louis]

Although this is off topic by now, I agree with Charles.  A CNC machine should 
be just that, and have a reliable stand alone PC at its core.  A coffee machine 
makes coffee, and a CNC cuts metal.  The thing I love about LinuxCNC is that 
when paired with Pico or Mesa hardware it provides an affordable alternative 
for machine builders.  By remaining compatible with standard g-code a LinuxCNC 
control has capabilities close to that of a Fanuc or Haas.  This, in itself, is 
amazing!  All the other complaints about what it doesn't do, or what it's 
called for that matter, don't even bother me.  Thanks to the developers of 
LinuxCNC, I can cut metal with the accuracy of a $40,000 machine on a $4000 
machine.  How can you not love that?

-Original Message-
From: charles green [mailto:xxzzb...@yahoo.com]
Sent: Tuesday, February 21, 2012 5:47 AM
To: Enhanced Machine Controller (EMC)
Subject: Re: [Emc-users] Latency - was Re: Which video card/driver for LinuxCNC?

the allocation of computing resources seems to be a recurring theme with pc 
based machine controllers.

while i have wished for an accesible pocket calculator gui more than one time 
while standing in front of a machining center console, i have almost never 
wanted to surf the web or send emails while doing so.

the enhanced machine controller spans a disparity between its intended function 
and the common usage of it's supporting hardware.  maybe there is a workaround? 
 when i installed linux cnc packages, one of the steps was drive partitioning, 
in case there was another OS already there, with some kind of neutral zone for 
exchange of data when switching the pc from one dedication of control to a 
different dedication of control.  why not do the same thing with an enhanced 
machine controller?  in one mode, the pc has the sole function of machine 
control and user interface for machine control.  in the other mode, the pc has 
the common functions like internet and anything else in an OS's routine employ.

i only have a very basic understanding of pc construction, so there might be 
some sort of fatal flaw in the idea of a dedicated machine control operating 
mode for a pc.  however, my suspicion is that such operation has not been 
pursued for reasons of convenience of having a do-it-all pc operation mode.  
this may not always be advantagious.  for example, imagine waking up in the 
morning, turning on the coffee maker, showering, and then finding that the 
coffee pot is empty because the coffee machine had a problem downloading the 
most recent bean roast sensing algorithm.


--- On Mon, 2/20/12, Jon Elson  wrote:


From: Jon Elson 
Subject: Re: [Emc-users] Latency - was Re: Which video card/driver for LinuxCNC?
To: "Enhanced Machine Controller (EMC)" 
Date: Monday, February 20, 2012, 9:54 AM


Roland Jollivet wrote:
>
> My question is; if one built a headless system(no video) and disabled
> drivers wherever possible, would such a system be fast enough to run DC
> servo's and read the encoders on the parallel port?
>
You have to figure this out for each case.  What is the maximum encoder
RPM at the desired rapid
feed rate?  (Don't forget belt reduction.)  What is the base thread rate?

For example, 60 IPM, 1000 cycle/rev encoder and a 2:1 belt reduction to
a 5 TPI
ballscrew.  So, that is 60 * 5  * 2 * 1000 * 4 (encoder quadrature) / 60
secs/min =
40,000 counts/second.  Well, that is a bit iffy, a count every 25 us.
And, remember
that if the encoder exceeds the rate at which the computer can count the
pulses,
you get a servo runaway, which is a lot worse than just losing position with
a stepper.

Now, if 30 IPM is OK, and you have 250 cycle/rev encoders directly coupled
to the 5TPI screws, it should work fine, but that is a lot of compromises to
avoid a better interface.

I generally think that running steppers via software step generation is
not a
great thing to do, and running servos via software encoder counting is far
WORSE!

So, now you are suggesting using TWO computers instead of using
purpose-built
hardware to interface to the motion system!  Now, you have to have TWO
hard drives to back up, and TWO computers to boot and shut down cleanly.
There may be times such an arrangement is desirable, but I really think
it is the wrong approach.

Jon

--
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-dev2
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users
--
Keep Your Developer Sk

Re: [Emc-users] Latency - was Re: Which video card/driver for LinuxCNC?

[charles green]

the allocation of computing resources seems to be a recurring theme with pc 
based machine controllers.
 
while i have wished for an accesible pocket calculator gui more than one time 
while standing in front of a machining center console, i have almost never 
wanted to surf the web or send emails while doing so.
 
the enhanced machine controller spans a disparity between its intended function 
and the common usage of it's supporting hardware.  maybe there is a 
workaround?  when i installed linux cnc packages, one of the steps was drive 
partitioning, in case there was another OS already there, with some kind of 
neutral zone for exchange of data when switching the pc from one dedication of 
control to a different dedication of control.  why not do the same thing with 
an enhanced machine controller?  in one mode, the pc has the sole function of 
machine control and user interface for machine control.  in the other mode, the 
pc has the common functions like internet and anything else in an OS's routine 
employ.
 
i only have a very basic understanding of pc construction, so there might be 
some sort of fatal flaw in the idea of a dedicated machine control operating 
mode for a pc.  however, my suspicion is that such operation has not been 
pursued for reasons of convenience of having a do-it-all pc operation mode.  
this may not always be advantagious.  for example, imagine waking up in the 
morning, turning on the coffee maker, showering, and then finding that the 
coffee pot is empty because the coffee machine had a problem downloading the 
most recent bean roast sensing algorithm.
 

--- On Mon, 2/20/12, Jon Elson  wrote:


From: Jon Elson 
Subject: Re: [Emc-users] Latency - was Re: Which video card/driver for LinuxCNC?
To: "Enhanced Machine Controller (EMC)" 
Date: Monday, February 20, 2012, 9:54 AM


Roland Jollivet wrote:
>
> My question is; if one built a headless system(no video) and disabled
> drivers wherever possible, would such a system be fast enough to run DC
> servo's and read the encoders on the parallel port?
>   
You have to figure this out for each case.  What is the maximum encoder 
RPM at the desired rapid
feed rate?  (Don't forget belt reduction.)  What is the base thread rate?

For example, 60 IPM, 1000 cycle/rev encoder and a 2:1 belt reduction to 
a 5 TPI
ballscrew.  So, that is 60 * 5  * 2 * 1000 * 4 (encoder quadrature) / 60 
secs/min =
40,000 counts/second.  Well, that is a bit iffy, a count every 25 us.  
And, remember
that if the encoder exceeds the rate at which the computer can count the 
pulses,
you get a servo runaway, which is a lot worse than just losing position with
a stepper.

Now, if 30 IPM is OK, and you have 250 cycle/rev encoders directly coupled
to the 5TPI screws, it should work fine, but that is a lot of compromises to
avoid a better interface.

I generally think that running steppers via software step generation is 
not a
great thing to do, and running servos via software encoder counting is far
WORSE! 

So, now you are suggesting using TWO computers instead of using 
purpose-built
hardware to interface to the motion system!  Now, you have to have TWO
hard drives to back up, and TWO computers to boot and shut down cleanly.
There may be times such an arrangement is desirable, but I really think
it is the wrong approach.

Jon

--
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-dev2
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users
--
Keep Your Developer Skills Current with LearnDevNow!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-d2d
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency - was Re: Which video card/driver for LinuxCNC?

[Jon Elson]

Roland Jollivet wrote:
>
> My question is; if one built a headless system(no video) and disabled
> drivers wherever possible, would such a system be fast enough to run DC
> servo's and read the encoders on the parallel port?
>   
You have to figure this out for each case.  What is the maximum encoder 
RPM at the desired rapid
feed rate?  (Don't forget belt reduction.)  What is the base thread rate?

For example, 60 IPM, 1000 cycle/rev encoder and a 2:1 belt reduction to 
a 5 TPI
ballscrew.  So, that is 60 * 5  * 2 * 1000 * 4 (encoder quadrature) / 60 
secs/min =
40,000 counts/second.  Well, that is a bit iffy, a count every 25 us.  
And, remember
that if the encoder exceeds the rate at which the computer can count the 
pulses,
you get a servo runaway, which is a lot worse than just losing position with
a stepper.

Now, if 30 IPM is OK, and you have 250 cycle/rev encoders directly coupled
to the 5TPI screws, it should work fine, but that is a lot of compromises to
avoid a better interface.

I generally think that running steppers via software step generation is 
not a
great thing to do, and running servos via software encoder counting is far
WORSE! 

So, now you are suggesting using TWO computers instead of using 
purpose-built
hardware to interface to the motion system!  Now, you have to have TWO
hard drives to back up, and TWO computers to boot and shut down cleanly.
There may be times such an arrangement is desirable, but I really think
it is the wrong approach.

Jon

--
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-dev2
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency - was Re: Which video card/driver for LinuxCNC?

[Viesturs Lācis]

2012/2/20 Roland Jollivet :
>
> My question is; if one built a headless system(no video) and disabled
> drivers wherever possible, would such a system be fast enough to run DC
> servo's and read the encoders on the parallel port?

As Andrew mentioned, for normal speed with normal resolution encoders
it is not going to work.
I do not know prices for Pico Systems products (more info here:
http://pico-systems.com/motion.html), but prices for Mesa FPGA cards
start at 80-90$, which I find affordable. Even more - if You are
thining about "headless" PC for realtime tasks and another PC for
operator, I am sure that also Pico's solutions would be less cost than
second PC.
And FPGA card will provide also a lot more I/O pins than You have in 1
or even 2 parports.

Viesturs

--
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-dev2
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency - was Re: Which video card/driver for LinuxCNC?

[Andrew]

2012/2/20 Roland Jollivet 

> So the norm is that a stepper system can run off the parallel port, but for
> DC servo's, extra hardware is required to sample the encoders.
>
> My question is; if one built a headless system(no video) and disabled
> drivers wherever possible, would such a system be fast enough to run DC
> servo's and read the encoders on the parallel port?
>

It is possible of course. But with low speed and/or low accuracy.
It depends on your requirements. With 100ppr encoders at 3000rpm it's 50
kHz, might be too much even for very low latency.
And common 2500ppr encoders are hardly compatible at all.

Andrew
--
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-dev2
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Latency - was Re: Which video card/driver for LinuxCNC?

[Roland Jollivet]

On 20 February 2012 06:14, Jon Elson  wrote:

> Dropout wrote:
> > A quick question, does latency matter for servo drives?
> >
> Yes, absolutely.  But, it is not as critical.  For software stepping,
> the base
> thread is typically around 20 us, and so a jitter of even 5 us is more than
> noticeable.  But, on a typical servo system (or a stepper with external
> step generating hardware) there is no base thread and the servo thread
> runs at 1000 us, so the effect of a 5us jitter is insignificant.  As long
> as
> the worst-case jitter never exceeds 20 us, servos should run flawlessly.
>
> Jon
>
>
So the norm is that a stepper system can run off the parallel port, but for
DC servo's, extra hardware is required to sample the encoders.

My question is; if one built a headless system(no video) and disabled
drivers wherever possible, would such a system be fast enough to run DC
servo's and read the encoders on the parallel port?

Regards
Roland
--
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-dev2
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency - was Re: Which video card/driver for LinuxCNC?

[Jon Elson]

Dropout wrote:
> A quick question, does latency matter for servo drives?
>   
Yes, absolutely.  But, it is not as critical.  For software stepping, 
the base
thread is typically around 20 us, and so a jitter of even 5 us is more than
noticeable.  But, on a typical servo system (or a stepper with external
step generating hardware) there is no base thread and the servo thread
runs at 1000 us, so the effect of a 5us jitter is insignificant.  As long as
the worst-case jitter never exceeds 20 us, servos should run flawlessly.

Jon

--
Try before you buy = See our experts in action!
The most comprehensive online learning library for Microsoft developers
is just $99.99! Visual Studio, SharePoint, SQL - plus HTML5, CSS3, MVC3,
Metro Style Apps, more. Free future releases when you subscribe now!
http://p.sf.net/sfu/learndevnow-dev2
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency - was Re: Which video card/driver for LinuxCNC?

[Dropout]

Thanks.

On 2/19/2012 2:51 PM, Peter C. Wallace wrote:
> On Sun, 19 Feb 2012, Dropout wrote:
>
>> Date: Sun, 19 Feb 2012 13:45:41 -0500
>> From: Dropout
>> Reply-To: "Enhanced Machine Controller (EMC)"
>>  
>> To: "Enhanced Machine Controller (EMC)"
>> Subject: [Emc-users] Latency - was Re: Which video card/driver for LinuxCNC?
>>
>> A quick question, does latency matter for servo drives?
>
> Dropout.
>
> Well at some amount of latency it matters but servo systems will tolerate a
> lot more latency than software step generation will. 50 - 100 uSec of latency
> on a servo system will do little more than add a little (less than .0002"
> at normal cutting speeds) noise to the control loop, but the same latency on a
> system using software step generation will likely cause stalls during rapids.
>
> Peter Wallace
> Mesa Electronics
>
>
> --
> Virtualization&  Cloud Management Using Capacity Planning
> Cloud computing makes use of virtualization - but cloud computing
> also focuses on allowing computing to be delivered as a service.
> http://www.accelacomm.com/jaw/sfnl/114/51521223/
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
>

--
Virtualization & Cloud Management Using Capacity Planning
Cloud computing makes use of virtualization - but cloud computing 
also focuses on allowing computing to be delivered as a service.
http://www.accelacomm.com/jaw/sfnl/114/51521223/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency - was Re: Which video card/driver for LinuxCNC?

[Peter C. Wallace]

On Sun, 19 Feb 2012, Dropout wrote:

> Date: Sun, 19 Feb 2012 13:45:41 -0500
> From: Dropout 
> Reply-To: "Enhanced Machine Controller (EMC)"
> 
> To: "Enhanced Machine Controller (EMC)" 
> Subject: [Emc-users] Latency - was Re: Which video card/driver for LinuxCNC?
> 
> A quick question, does latency matter for servo drives?

Dropout.

Well at some amount of latency it matters but servo systems will tolerate a 
lot more latency than software step generation will. 50 - 100 uSec of latency 
on a servo system will do little more than add a little (less than .0002" 
at normal cutting speeds) noise to the control loop, but the same latency on a 
system using software step generation will likely cause stalls during rapids.

Peter Wallace
Mesa Electronics


--
Virtualization & Cloud Management Using Capacity Planning
Cloud computing makes use of virtualization - but cloud computing 
also focuses on allowing computing to be delivered as a service.
http://www.accelacomm.com/jaw/sfnl/114/51521223/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Latency - was Re: Which video card/driver for LinuxCNC?

[Dropout]

A quick question, does latency matter for servo drives?

Dropout.

On 2/19/2012 1:36 PM, Viesturs Lācis wrote:
> 2012/2/19 Alan Browning:
>> Thanks for the quick response. So you build your machines from the ground up 
>> with new componenets? Due to budget limitations I have to build things on 
>> the cheap.
>
> 2 of those machines were completely new builds, 2 were retrofits, one
> of the retrofits is my own machine, remaining are commercial services
> for customers.
> I prefer new components, where possible, but I am also very concerned
> about being as cost-effective as possible (that is why I delivered
> used monitor with one of those newly built machines). And that is why
> D525 is so attractive - for<70 EUR (<100 USD) it has mainboard with
> 2-core CPU and also built-in video card. Add RAM, HDD and PSU to get
> working PC.
> Of course, if You already have unused PC and are building machine for
> Your own use, it is obvious to use existing hardware for multitasking
> - not only collect dust in its case, but also drive a cnc machine.
>
> I think that best thing to do at the moment is installing Ubuntu +
> LinuxCNC on that PC and runing latency test to find out, how is Your
> particular PC behaving and what realtime performance You can get out
> of it.
>
> Viesturs
>
> --
> Virtualization&  Cloud Management Using Capacity Planning
> Cloud computing makes use of virtualization - but cloud computing
> also focuses on allowing computing to be delivered as a service.
> http://www.accelacomm.com/jaw/sfnl/114/51521223/
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>
>

--
Virtualization & Cloud Management Using Capacity Planning
Cloud computing makes use of virtualization - but cloud computing 
also focuses on allowing computing to be delivered as a service.
http://www.accelacomm.com/jaw/sfnl/114/51521223/
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Latency test [was: 3 threads, realtime errors]

[Kent A. Reed]

Gentle persons:

Andy Pugh's aside to Tom Easterday

"I found that having a webcam tab causes realtime errors in my
otherwise well-behaved D510MO setup."


should be a reminder to all of us that the latency-test results we post 
for different systems are not immutable. They may be helpful when 
selecting a motherboard and cpu but always measure the latency of the 
finished system as-it-is-actually-configured before using the numbers 
for seting up emc2.

It's always useful when folks tells us their "exceptions to the rule" 
like Andy did so we can avoid them.

Regards,
Kent


--
vRanger cuts backup time in half-while increasing security.
With the market-leading solution for virtual backup and recovery, 
you get blazing-fast, flexible, and affordable data protection.
Download your free trial now. 
http://p.sf.net/sfu/quest-d2dcopy1
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] latency calculation

[Jeff Epler]

The latency-test runs the following realtime component:
http://git.linuxcnc.org/gitweb?p=emc2.git;a=blob;f=src/hal/components/timedelta.comp

Each time a realtime thread runs, the time interval del=(now-last) is computed.
The minimum and maximum del ever seen are computed in min_ and max_.
The jitter is the bigger of max_-period or period-min_.

period is the promised realtime interval in nanoseconds, and
rtapi_get_time() is the measured wall time also in nanoseconds.

The timedelta component also measures long term error in 'err' and
'avg_err'.  These values are not shown in the latency-test GUI but can
be used to determine if there are long-term differences between expected
and actual intervals.

Jeff

--
ThinkGeek and WIRED's GeekDad team up for the Ultimate 
GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the 
lucky parental unit.  See the prize list and enter to win: 
http://p.sf.net/sfu/thinkgeek-promo
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] latency calculation

[Dave Caroline]

calculate? it is measuring your computer
and you can always read the source

Dave Caroline

On Fri, Jun 11, 2010 at 10:06 AM, ferdi andika  wrote:
> I need know how to calculate latency in EMC2's latency-test.
>
> could anyone tell me the latency test's algorithm?
> --
> ThinkGeek and WIRED's GeekDad team up for the Ultimate
> GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the
> lucky parental unit.  See the prize list and enter to win:
> http://p.sf.net/sfu/thinkgeek-promo
> ___
> Emc-users mailing list
> Emc-users@lists.sourceforge.net
> https://lists.sourceforge.net/lists/listinfo/emc-users
>

--
ThinkGeek and WIRED's GeekDad team up for the Ultimate 
GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the 
lucky parental unit.  See the prize list and enter to win: 
http://p.sf.net/sfu/thinkgeek-promo
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] latency calculation

[ferdi andika]

I need know how to calculate latency in EMC2's latency-test.

could anyone tell me the latency test's algorithm?
--
ThinkGeek and WIRED's GeekDad team up for the Ultimate 
GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the 
lucky parental unit.  See the prize list and enter to win: 
http://p.sf.net/sfu/thinkgeek-promo
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

[Emc-users] Latency and laptop.

[Slavko Kocjancic]

Hello...

I know that laptop's are not great for EMC, but I have few of them lying
around and wish to use it.

HP NX9005 work's now for longer time without some troubles. It has some
strange thing. The layency in 1'st few minutes (up to 5) is high(in
range 1ms). I don't know why but after that they are low (around 15uS)
and stay there all the day. more hour's machining isn't problematic. (I
didn't disabled SMI) I expect that is caused by death baterry(removed).

Now I try to wake up the IBM ThinkPad T23
Pentium III 1GHz
640MB/RAM
The 1'st issue was SMI. The 200uS latency every 60 seconds. I disabled
SMI and got latency in range 15uS. I leave just latency test running one
hour to be safe. (I was surprised as CPU fan is turned on and off few
times during that test with SMI off). After that I run glxgears and
leave another hour. all just fine. But then I maximized glxgear window
and horror of 2 miliseconds appear.
I try to open openoffice and was just ok. I can write / open / save no
bad latency. But the resizing window (no mather what window) does bad
latency. I discovered if I just run latency test and maximize it's
window the bad think happens?

What can be cause? Video? (S3 Graphich SuperSavage/IXC1014 reported
within WindowsXP)

Any cure?

Slavko.


--
ThinkGeek and WIRED's GeekDad team up for the Ultimate 
GeekDad Father's Day Giveaway. ONE MASSIVE PRIZE to the 
lucky parental unit.  See the prize list and enter to win: 
http://p.sf.net/sfu/thinkgeek-promo
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency test numbers confusing

[Jon Elson]

Neil Baylis wrote:
> OK, so I started up the computer, then waited 1/2 hour. Then I ran the
> latency test for 1/2 hour with no "abuse", and the latency was below 7
> microseconds. I have plenty of other computers, so I certainly don't
> need to be doing anything else on my EMC box when it's running EMC.
>   
The problem is EMC, ITSELF, abuses the computer a bit by running a 3-D 
display
in the Axis GUI.  That takes considerable resources, and unless you will be
running the GUI on another computer, or won't be using Axis, you need to
simulate that load to get an accurate reading.  Also, network interfaces can
cause DMA block transfers and CPU interrupts.  Clearly, you don't need
video players, music players, etc. for CNC.

Jon

--
Throughout its 18-year history, RSA Conference consistently attracts the
world's best and brightest in the field, creating opportunities for Conference
attendees to learn about information security's most important issues through
interactions with peers, luminaries and emerging and established companies.
http://p.sf.net/sfu/rsaconf-dev2dev
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users

Re: [Emc-users] Latency test numbers confusing

[Neil Baylis]

On Thu, Jan 14, 2010 at 1:00 PM, Andy Pugh  wrote:
> 2010/1/14 Neil Baylis :
>>
>> Ah OK. The penny drops. When I get home tonight, I'll follow the
>> directions to disable SMI and see what happens.
>
> It doesn't sound like SMI. That normally happens periodically. (in my
> case it was every 64 seconds)

You were correct. I followed the instructions for disabling SMI, and
it made no difference.

However, it seems reliably true that once you get past the two spikes,
the latency is tolerable.

I'm going to run that way for the time being, and see how it goes. If
I notice problems, I'll investigate further.

Neil

--
Throughout its 18-year history, RSA Conference consistently attracts the
world's best and brightest in the field, creating opportunities for Conference
attendees to learn about information security's most important issues through
interactions with peers, luminaries and emerging and established companies.
http://p.sf.net/sfu/rsaconf-dev2dev
___
Emc-users mailing list
Emc-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/emc-users