On 09/08/2015 09:30 PM, Chris Radek wrote:
> On Tue, Sep 08, 2015 at 08:30:13PM -0500, Jon Elson wrote:
>
>> Now, you COULD figure out the right number by modulo
>> division and work it that way.
> You know, I hadn't thought about this. The original design for
> spindle-synced motion was when we h
Is it not possible to make a circular counter that always overflows to zero
every rev? Don't angular axis already do this?
Brian
Sent from my iPad
--
Monitor Your Dynamic Infrastructure at Any Scale With Datadog!
Get re
On Tue, Sep 08, 2015 at 08:30:13PM -0500, Jon Elson wrote:
> Now, you COULD figure out the right number by modulo
> division and work it that way.
You know, I hadn't thought about this. The original design for
spindle-synced motion was when we had single precision floating
point positions, whic
On 09/08/2015 12:55 PM, EBo wrote:
> Well now. If I was going to try to top that (for realizes)
> at least you started with an 8080. I once built a CPU from
> wire-wrapped discrete logic. 7 functions and 8 bits
> wide... EBo --
Here's two!
A 32-bit CPU using AMD 2903/2910 bit slice chips, and a
On 09/08/2015 12:08 PM, Stuart Stevenson wrote:
> Why would a subsequent tapping operation need to use any index? Once the
> spindle is homed you should not need to look for index other than a
> convenient place to start the motion.
> You could then specify an angle to start.
> Easy to cut multilea
Hi Kenneth,
There's no fundamental limitation here, it's just a matter of how big we
make the queue structure. It's a circular buffer, so it could be 1000
segments longer if we could afford the space. I just made it 100 segments
because it seemed reasonable. For context, the stock queue is 2000 se
On 9/7/15 7:53 PM, sam sokolik wrote:
> Rob had some time to look at this branch - rebased it on 2.7 and made a
> few fixes
Very cool. My only comment is this: 2.7 is closed for new features like
this, please base this work on master instead.
--
Sebastian Kuzminsky
--
On Tue, Sep 08, 2015 at 12:08:19PM -0500, Stuart Stevenson wrote:
> Why would a subsequent tapping operation need to use any index?
Only because that's how it's already implemented.
--
___
On 8 September 2015 at 18:55, EBo wrote:
> This reminds me of some friends playing the game "Ah that's nothing,
> when I was young all we had to compute with was an abacus..
I am allowed to do this, I _am_ a Yorkshireman,
https://www.youtube.com/watch?v=Xe1a1wHxTyo
--
atp
If you can't fix it,
On Sep 8 2015 12:26 PM, andy pugh wrote:
> On 8 September 2015 at 18:55, EBo wrote:
>> This reminds me of some friends playing the game "Ah that's nothing,
>> when I was young all we had to compute with was an abacus..
>
> I am allowed to do this, I _am_ a Yorkshireman,
>
> https://www.youtube.com
On Sep 8 2015 10:26 AM, Jon Elson wrote:
> On 09/08/2015 07:49 AM, andy pugh wrote:
>> On 8 September 2015 at 13:43, Kenneth Lerman
>> wrote:
>>> (Written by a man who owned a computer with 16K of memory --
>>> although he
>>> expanded it to 48K.)
>> My first computer had 1k of memory.
>>
>>
> I
Why would a subsequent tapping operation need to use any index? Once the
spindle is homed you should not need to look for index other than a
convenient place to start the motion.
You could then specify an angle to start.
Easy to cut multilead threads then.
--
On 09/08/2015 07:49 AM, andy pugh wrote:
> On 8 September 2015 at 13:43, Kenneth Lerman wrote:
>> (Written by a man who owned a computer with 16K of memory -- although he
>> expanded it to 48K.)
> My first computer had 1k of memory.
>
>
I can top that. I built an 8008 CPU with only 256 bytes of
On 09/08/2015 07:43 AM, Kenneth Lerman wrote:
> Why is there a 100 step limit? If it is because that's way more than anyone
> would ever need, that's fine.
>
> Memory is cheap. At 1000 bytes per step, storing 10 thousand steps is only
> 10 meg. That's not much in a machine with a gigabyte or more.
At work, I used an Intel 4004 which had damned if I remember memory.
But the first computer I used was an IBM 650 which I started programming in
decimal and then coded in assembler. It had 2000 ten digit words of drum
memory.
It also had an optimizing assembler and a FORTRAN compiler (a variant
c
I realize that, but CAM generated gcode might have hundreds of steps per
inch.
Ken
On Tue, Sep 8, 2015 at 12:25 PM, Jon Elson wrote:
> On 09/08/2015 07:43 AM, Kenneth Lerman wrote:
> > Why is there a 100 step limit? If it is because that's way more than
> anyone
> > would ever need, that's fin
On Tuesday 08 September 2015 11:34:47 andy pugh wrote:
> On 8 September 2015 at 16:22, Chris Radek wrote:
> > Tapping/threading will cause additional index resets later on - this
> > won't hurt orient, right?
>
> I would think that things were badly broken if it did. The spindle
> index really ou
On 8 September 2015 at 16:11, EBo wrote:
> so we had to code it in hex... I kid you not
When I was 15 I had never even heard of assemblers, I thought that Hex
was how you were meant to write machine code. I used to be able to
disassemble hex dumps in my head. (well, enough to see which bits I
ne
On Tuesday 08 September 2015 11:11:29 EBo wrote:
> On Sep 8 2015 8:24 AM, Gene Heskett wrote:
> > On Tuesday 08 September 2015 08:49:57 andy pugh wrote:
> >> On 8 September 2015 at 13:43, Kenneth Lerman
> >>
> >> wrote:
> >> > (Written by a man who owned a computer with 16K of memory --
> >> > al
On Tuesday 08 September 2015 10:49:28 sam sokolik wrote:
> https://youtu.be/3aYaHxT6ZnQ
That's very impressive Sam.
[...]
While I agree that Rob is a genius, that almost borders on magic. kewl
even.
Cheers, Gene Heskett
--
"There are four boxes to be used in defense of liberty:
soap, ballo
On 8 September 2015 at 16:22, Chris Radek wrote:
> Tapping/threading will cause additional index resets later on - this
> won't hurt orient, right?
I would think that things were badly broken if it did. The spindle
index really ought to be in the same place every revolution.
I guess that with a
On Tue, Sep 08, 2015 at 03:18:42PM +0100, andy pugh wrote:
>
> My proposal is that the homing code would search the INI file for a
> [SPINDLE*] stanza (which already exists in many configs) and if it
> finds one that contains a homing sequence number it would assert the
> spindle-index-enable outp
On Sep 8 2015 8:24 AM, Gene Heskett wrote:
> On Tuesday 08 September 2015 08:49:57 andy pugh wrote:
>
>> On 8 September 2015 at 13:43, Kenneth Lerman
>> wrote:
>> > (Written by a man who owned a computer with 16K of memory --
>> > although he expanded it to 48K.)
>>
>> My first computer had 1k of
https://youtu.be/3aYaHxT6ZnQ
On 9/8/2015 7:43 AM, Kenneth Lerman wrote:
> Why is there a 100 step limit? If it is because that's way more than anyone
> would ever need, that's fine.
>
> Memory is cheap. At 1000 bytes per step, storing 10 thousand steps is only
> 10 meg. That's not much in a machin
On Tuesday 08 September 2015 08:49:57 andy pugh wrote:
> On 8 September 2015 at 13:43, Kenneth Lerman wrote:
> > (Written by a man who owned a computer with 16K of memory --
> > although he expanded it to 48K.)
>
> My first computer had 1k of memory.
Gotcha beat Andy, my first one, a Cosmac Elf,
In a recent feature request I proposed adding spindle homing to the
homing sequence.
https://sourceforge.net/p/emc/feature-requests/138/
The use case is mainly for toolchangers, but it might be useful for
lathes with a C-axis capability.
If the spindle.index-enable pin was asserted by default at
In the old days, we always wrote completely tangential paths for WEDM.
Not hard to do, kept you aware of the minimum radii that the wire kerf
could actually fit in. If its tangential ( never line to line
transitions ), how can there be any blending? I'd think only velocity
blending might occur, no
On 8 September 2015 at 13:43, Kenneth Lerman wrote:
> (Written by a man who owned a computer with 16K of memory -- although he
> expanded it to 48K.)
My first computer had 1k of memory.
Someone managed to write a playable chess game.
https://en.wikipedia.org/wiki/1K_ZX_Chess
--
atp
If you can
Why is there a 100 step limit? If it is because that's way more than anyone
would ever need, that's fine.
Memory is cheap. At 1000 bytes per step, storing 10 thousand steps is only
10 meg. That's not much in a machine with a gigabyte or more.
(Written by a man who owned a computer with 16K of mem
On Sep 8 2015 4:01 AM, andy pugh wrote:
> On 8 September 2015 at 02:53, sam sokolik
> wrote:
>> Currently the reverse run is exact stop - rob would like to add
>> basic
>> blending.
>
> I guess this is a problem if the forward run was blended then the
> reverse run might take a fractionally diff
I am also following the gEDA discussion and someone where talked about using
gschem for structural design of VHDL. Configuration of linuxcnc may also be
seen as connecting together different components with nets so in case someone
is thinking about graphical configuration gEDA symbols together w
On 8 September 2015 at 02:53, sam sokolik wrote:
> Currently the reverse run is exact stop - rob would like to add basic
> blending.
I guess this is a problem if the forward run was blended then the
reverse run might take a fractionally different path.
However, the likely application is wire spar
32 matches
Mail list logo
