Re: [Emc-users] Max Jitter

Przemek Klosowski wrote: > > > I don't know---but the project goal is to get it around 10us. The > rt-linux people have a tool called cyclictest > > http://people.redhat.com/williams/latency-howto/rt-latency-howto.txt > > as well as several other interesting tools, including a way to measure > SMI

Re: [Emc-users] Max Jitter

On Thu, Apr 12, 2012 at 10:00 PM, Jon Elson wrote: > Przemek Klosowski wrote: >> The good news are here: RT integration in the mainline kernel keeps >> moving ahead, and the latest 3.x kernels need less and less patches at >> least for Thomas Gleixner's RT subsystem. >> >> http://www.h-online.com/

Re: [Emc-users] Max Jitter

Eric Keller wrote: > before rt-linux went commercial and then who knows what happened, the idea > behind hal was to make it so you could run EMC either on rt-linux or rtai. > I really haven't been motivated to keep up with the developments of the RT > kernel to know what needs to be done to make HA

Re: [Emc-users] Max Jitter

Przemek Klosowski wrote: > The good news are here: RT integration in the mainline kernel keeps > moving ahead, and the latest 3.x kernels need less and less patches at > least for Thomas Gleixner's RT subsystem. > > http://www.h-online.com/open/features/Kernel-Log-real-time-kernel-goes-Linux-3-0-13

Re: [Emc-users] Max Jitter

On Thu, Apr 12, 2012 at 3:42 PM, Przemek Klosowski < przemek.klosow...@gmail.com> wrote: > > This of course is not RTAI, so it doesn't help EMC at the moment, but > it seems to me that in the long term either RTAI will benefit from the > RT integration, or, if it falls behind in development, the m

Re: [Emc-users] Max Jitter

On Sat, Mar 31, 2012 at 7:19 AM, Jan de Kruyf wrote: > IF and WHEN the software can handle the mainline kernel RT effort, I > personally would change earlier rather than later. > > As an example: the whole Beagle board saga that Jon went through would not > have happened if at the time the vanill

Re: [Emc-users] Max Jitter

indirectly. Start the System monitor. It should show more than one CPU. Start several, processes, or a build. If one CPU stays at no load, you have an isolated CPU. -m Am 31.03.2012 um 20:42 schrieb Jeshua Lacock: > > On Mar 30, 2012, at 1:59 PM, Viesturs Lācis wrote: > >> My googling results

Re: [Emc-users] Max Jitter

On Mar 30, 2012, at 1:59 PM, Viesturs Lācis wrote: > My googling results tell, that You should edit /etc/default/grub > For that You will need root privileges. You can do that by running in > terminal this command: > sudo gedit > > Text editor will be opened, click "open" and find the file. > >

Re: [Emc-users] Max Jitter

Good people, I was maybe a bit snappy last night, as I came just out of one of those african music and drama things that take all of you (I participated) Off course the cnclinux group has discovered some excellent boards over time. And off course the RTAI modified kernel is probably the fastest ar

Re: [Emc-users] Max Jitter

On Saturday, March 31, 2012 01:44:21 AM Jon Elson did opine: > Jan de Kruyf wrote: > > Do you get those errors with servos Gene? or with steppers! > > > > I am interested, since Jon was praising that board so much. > > > >> But I am encountering following errors after the motor has reached > >>

Re: [Emc-users] Max Jitter

I thought I saw a D510 on that page in a different rack. The graph for the D510 did not look as good as the graph for the 3rd board down in the first rack. That is a gigabyte board I believe. I wasn't sure how they came up with the test numbers, I was comparing curves relative to each other.

Re: [Emc-users] Max Jitter

Jan de Kruyf wrote: > Do you get those errors with servos Gene? or with steppers! > > I am interested, since Jon was praising that board so much. > > >> >> But I am encountering following errors after the motor has reached speed >> and moved a good distance at that speed, but don't know what to

Re: [Emc-users] Max Jitter

Jan de Kruyf wrote: > That is very impressive Jon, > I will certainly look out for an Intel D510MW being hooked up in the OSADL > test rack, > and see what it does under their exact conditions! > And how many days did you run that board with an RTAI kernel and got that > result? > Whoops, sorry,

Re: [Emc-users] Max Jitter

On 3/30/2012 5:41 PM, Jan de Kruyf wrote: > Do you get those errors with servos Gene? or with steppers! > > I am interested, since Jon was praising that board so much. > > j. > > > Jan: Such numbers seem to be typical broadly for boards equipped with Atom D510/D525 cpus and NM10 chipsets. See, fo

Re: [Emc-users] Max Jitter

On Friday, March 30, 2012 07:29:30 PM Jan de Kruyf did opine: > Do you get those errors with servos Gene? or with steppers! > > I am interested, since Jon was praising that board so much. > > j. Steppers. Servo's are way out of my budget's reach. I am retired, 77yo with SS and a little consu

Re: [Emc-users] Max Jitter

On Friday, March 30, 2012 07:18:08 PM Viesturs Lācis did opine: > 2012/3/31 gene heskett : > > It is better to put it as a file in /etc/grub.d/, named > > as "07_rtai" as this way it can maybe survive a kernel update. > > It does not really make sense, because kernel updates are strongly not > re

Re: [Emc-users] Max Jitter

2012/3/31 Jan de Kruyf : > Do you get those errors with servos Gene? or with steppers! > > I am interested, since Jon was praising that board so much. > Jon is not the only one... I also feel like intel has really deserved a few good words (I kind of feel like amd fan) about that D510MO/D525MW lin

Re: [Emc-users] Max Jitter

Do you get those errors with servos Gene? or with steppers! I am interested, since Jon was praising that board so much. j. 2012/3/30 gene heskett > On Friday, March 30, 2012 10:12:25 AM Jan de Kruyf did opine: > > > Yes you are right on the dot there. > > > > j. > > > > On Fri, Mar 30, 2012 a

Re: [Emc-users] Max Jitter

That is very impressive Jon, I will certainly look out for an Intel D510MW being hooked up in the OSADL test rack, and see what it does under their exact conditions! And how many days did you run that board with an RTAI kernel and got that result? Keep the bright side up! j. On Fri, Mar 30, 201

Re: [Emc-users] Max Jitter

2012/3/31 gene heskett : > > It is better to put it as a file in /etc/grub.d/, named > as "07_rtai" as this way it can maybe survive a kernel update. It does not really make sense, because kernel updates are strongly not recommended - LinuxCNC will not accept any other kernel but the one it is bui

Re: [Emc-users] Max Jitter

On Friday, March 30, 2012 04:56:17 PM Jeshua Lacock did opine: > On Mar 30, 2012, at 1:20 AM, Viesturs Lؤپcis wrote: > >> Does anyone know of some known settings to try disabling in CMOS to > >> possibly get lower numbers? My 1.0ms Max Jitter is under 9300, but > >> if possible I would like to get

Re: [Emc-users] Max Jitter

On Mar 30, 2012, at 1:59 PM, Viesturs Lācis wrote: > 2012/3/30 Jeshua Lacock : >> >> And, that is very interesting! When I run the CPU hog, my max jitter goes >> down to about 1,000! > > Wow! How long did You run the latency test? Did You try to put some load to > PC? I let it run for about

Re: [Emc-users] Max Jitter

2012/3/30 Jeshua Lacock : > > And, that is very interesting! When I run the CPU hog, my max jitter goes > down to about 1,000! Wow! How long did You run the latency test? Did You try to put some load to PC? > I would like to set the CPU as instructed, but I have never used grub before. > I assu

Re: [Emc-users] Max Jitter

On Mar 30, 2012, at 1:20 AM, Viesturs Lācis wrote: >> Does anyone know of some known settings to try disabling in CMOS to possibly >> get lower numbers? My 1.0ms Max Jitter is under 9300, but if possible I >> would like to get it even lower. >> > > IMHO it is very good number already. You hav

Re: [Emc-users] Max Jitter

Jan de Kruyf wrote: > It is the combination that matters Viesturs. > The ASUS board in the first rack is impressive. > Impressively mediocre! It has jitter events out to 40 us! For about $150 you can build an Intel D510MW system that will do almost TEN TIMES better, a good reliable 5 us latency

Re: [Emc-users] Max Jitter

Jan de Kruyf wrote: > Look here for fast motherboard - cpu combinations: > > https://www.osadl.org/Individual-system-data.qa-farm-data.0.html#c4366 > > unfortunately we cant use the osadl kernel patch as yet (that will be my > next gripe, after I am over the present one. > So dont expect results an

Re: [Emc-users] Max Jitter

The board in rack 1 slot 2 is even more impressive! Dave On 3/30/2012 6:07 AM, Jan de Kruyf wrote: > It is the combination that matters Viesturs. > The ASUS board in the first rack is impressive. > > j. > > > On Fri, Mar 30, 2012 at 1:00 PM, Viesturs > Lāciswrote: > > >> 2012/3/30 Jan de Kru

Re: [Emc-users] Max Jitter

On Friday, March 30, 2012 10:12:25 AM Jan de Kruyf did opine: > Yes you are right on the dot there. > > j. > > On Fri, Mar 30, 2012 at 1:23 PM, Viesturs Lؤپcis wrote: > > 2012/3/30 Jan de Kruyf : > > > It is the combination that matters Viesturs. > > > > Well, I think that there are many facto

Re: [Emc-users] Max Jitter

Yes you are right on the dot there. j. On Fri, Mar 30, 2012 at 1:23 PM, Viesturs Lācis wrote: > 2012/3/30 Jan de Kruyf : > > It is the combination that matters Viesturs. > > Well, I think that there are many factors that matter. > My goal was to explain Jeshua and probably other new users that

Re: [Emc-users] Max Jitter

On Friday, March 30, 2012 07:21:25 AM Viesturs Lācis did opine: > 2012/3/30 gene heskett : > > One mistake in the above Viesturs, in your 40 kilohertz base_thread > > loop description, linuxcnc has been able to issue a complete step > > pulse per loop iteration for something like 3, maybe 4 years

Re: [Emc-users] Max Jitter

2012/3/30 Jan de Kruyf : > It is the combination that matters Viesturs. Well, I think that there are many factors that matter. My goal was to explain Jeshua and probably other new users that for stepper machines (meaning all the machines, which control motor drives with step/dir signals and do not

Re: [Emc-users] Max Jitter

It is the combination that matters Viesturs. The ASUS board in the first rack is impressive. j. On Fri, Mar 30, 2012 at 1:00 PM, Viesturs Lācis wrote: > 2012/3/30 Jan de Kruyf : > > Look here for fast motherboard - cpu combinations: > > > > https://www.osadl.org/Individual-system-data.qa-farm-d

Re: [Emc-users] Max Jitter

2012/3/30 Jan de Kruyf : > Look here for fast motherboard - cpu combinations: > > https://www.osadl.org/Individual-system-data.qa-farm-data.0.html#c4366 > And that page is very good at showing that more cores and higher cpu clock rates do not necessarily mean less jitter. For example, quad-core i7

Re: [Emc-users] Max Jitter

2012/3/30 gene heskett : > > One mistake in the above Viesturs, in your 40 kilohertz base_thread loop > description, linuxcnc has been able to issue a complete step pulse per loop > iteration for something like 3, maybe 4 years now.  It does this by timing > a 2 or 3 microsecond pulse out and clear

Re: [Emc-users] Max Jitter

Look here for fast motherboard - cpu combinations: https://www.osadl.org/Individual-system-data.qa-farm-data.0.html#c4366 unfortunately we cant use the osadl kernel patch as yet (that will be my next gripe, after I am over the present one. So dont expect results any time soon.) cheers, j. O

Re: [Emc-users] Max Jitter

On Friday, March 30, 2012 06:05:34 AM Viesturs Lācis did opine: > 2012/3/30 Jeshua Lacock : > > I am familiar with the latency test, that is exactly what I was > > referring to. It reports two Max Jitters there and I did not know for > > certain that one was for servos and the other for steppers.

Re: [Emc-users] Max Jitter

2012/3/30 Jeshua Lacock : > > I am familiar with the latency test, that is exactly what I was referring to. > It reports two Max Jitters there and I did not know for certain that one was > for servos and the other for steppers. Thank you. No, no, no!!! Based on the second sentence, You could not

Re: [Emc-users] Max Jitter

On Mar 29, 2012, at 12:59 AM, Viesturs Lācis wrote: > 1) If You mean: which jitter number from latency test results to take > into account? > 25us figure represents jitter in base thread. > 1ms figure represents jitter in servo thread. > > Which figure to use? It depends... Will You run anything

Re: [Emc-users] Max Jitter

2012/3/29 Jeshua Lacock : > > Hi, > > I am not sure to use the 1.0ms or 25us figure. Can anyone shed some light on > which jitter number I would use for the latency test? Based on the wording You used: "which jitter number I would use for the latency test", I am not sure that I understand, what e

Re: [Emc-users] Max Jitter

It depends what you are setting up for. If you are doing software stepping probably the 25us figure. On Wed, Mar 28, 2012 at 7:39 PM, Jeshua Lacock wrote: > > Hi, > > I am not sure to use the 1.0ms or 25us figure. Can anyone shed some light > on which jitter number I would use for the latency te

[Emc-users] Max Jitter

Hi, I am not sure to use the 1.0ms or 25us figure. Can anyone shed some light on which jitter number I would use for the latency test? Thanks! Cheers, Jeshua Lacock Founder/Engineer 3DTOPO Incorporated Phone: 208.462.4171