Re: [CAD_CAM_EDM_DRO] DeskNC Windows version

I haven't looked into Linux and it's interrupt rates, but the way I would do
it is reprogram the hardware timer on the system board to a much higher rate
then add your own interrupt service routine to divide down the faster tick
rate back to the original rate for routines that depend on it.

That's the trick we used to do on DOS to get high interrupt rates. Then trap
INT 08 if memory serves.
I'm not sure what the minimum resolution was but it seems like it was around
.8 us.

The kernel or ring 0 processes could do this in Linux and probably Windows.
I've never looked into this for Windows though I'm confident there's a way
to do it. I know I did some high res timing code in the early days of
Windows, but things have changed a lot since then.


later,
Larry E.

----- Original Message -----
From: "Art Fenerty" <fenerty@...>
To: <CAD_CAM_EDM_DRO@yahoogroups.com>
Sent: Wednesday, April 18, 2001 10:19 PM
Subject: Re: [CAD_CAM_EDM_DRO] DeskNC Windows version


> I must check this out further. 30000 steps/second from linux? Anyone know
> how the linux knows to interrupt itseld 30000 times a second. I don't see
> how its possible. I can't think of any way to instruct a motherboard to
tell
> linux to interrupt at that rate, no matter how fast the processor.
> Interesting though. But I think that 10 motor turns per inch at 200 pulses
> per turn is 2000 pulses per inch. This would make it 180 * 2000 / 60 =
6000
> pulses per second in the example given. Encoders do not up the steps per
> second. They simply give a higher resolution to the step decision.
>
>
> My VXD can interrupt at 8192 and then time to the closest 2 nanosecond
> granularity to ensure smoothness. Windows can still cause instability, but
I
> haven't seen this cause any lost steps or problems yet. I have users who
> tell me that their speed is up to 3 times what they could get from DeskNC
or
> KellyCam. But there's no question that windoze will never be ideal.
> Certainly ideal enough for a good 50% of home CNC users though. P6 onboard
> ASIC's are only on the latest generation of processors, typically at
700MHz
> and up.(the ones I've tested anyway).
> I have no problem releasing the Controller section of Master5 for
people
> to write programs for themselves, I just have to get time to document the
> OLE call to control the buffers.
>
> Timing is hard to check for smoothness. Actual running is the
easiest.
> I have had several users report scoping my pulses, and seeing drop-outs ,
> yet commenting that the motors were running surprisingly smooth. This is
due
> to an enertial averaging at the motor. Any stepper seems to have a certain
> window of tolerance to irregularities in its pulse stream. Stay within
this
> window, and no problems seem to occur. I have my pulse engine try to
correct
> for these irregularities by subtimeing from interrupt to the closest
> nanosecond possible with the hardware. Seems to work well, but like I
said,
> Windoze will never be EMC.
>
> I'd like to hear from anyone actually getting more than 8000 pulses from
> linux though.
>
> (My head hurts)
>
> Art
>
>
>
>
>
>
> Welcome to CAD_CAM_EDM_DRO@...,an unmoderated list for the
discussion of shop built systems, for CAD, CAM, EDM, and DRO.
>
> Addresses:
> Post message: CAD_CAM_EDM_DRO@yahoogroups.com
> Subscribe: CAD_CAM_EDM_DRO-subscribe@yahoogroups.com
> Unsubscribe: CAD_CAM_EDM_DRO-unsubscribe@yahoogroups.com
> List owner: CAD_CAM_EDM_DRO-owner@yahoogroups.com, wanliker@...
> Moderator: jmelson@... timg@... [Moderator]
> URL to this page: http://groups.yahoo.com/group/CAD_CAM_EDM_DRO
> bill,
> List Manager
>
> FAQ: http://www.ktmarketing.com/faq.html
>
> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/
>
>