RE: TKemc

["jguenther-at-loudoun.com" <jguenthe-at-loudoun.com>]

Thanks Ray,

I tried this last night, and as it turns out anything between 0.250 and
0.300 seems to work ok.  If I go over 0.300 the system will crash, as in
immediate reboot.  So until I find a faster and perhaps smaller machine to
run EMC on I will be running with a CYCLE_TIME setting of 0.290 which I
think will be safe.  Space is at an extreme premium in my shop so fast and
small are the only way I can do CNC in my shop area.

Thanks for the help

John Guenther

> -----Original Message-----
> From: emc-at-nist.gov [emc-at-nist.gov]On Behalf Of Ray
> Sent: Tuesday, February 27, 2001 12:16 PM
> To: Multiple recipients of list
> Subject: Re: TKemc
>
>
>
>
> Good job, John
>
> Tkemc is demanding.  I haven't tried it on anything less than 266 Mhz P2.
> But your performance may be about what could be expected.  You could try
> reducing the refresh rate for tkemc.  This would make the
> backplot a little
> more rounded in places but you could always run it with sim.
>
> ; Cycle time, in seconds, that display will sleep between polls
> CYCLE_TIME =            0.250
>
> or
>
> CYCLE_TIME =            0.300
>
> If you go much beyond 0.30 you will see a lot of flicker.
>
> If you want, we could build a stripped down display.  I've been working on
> this along with the Tcl/Tk library routines.
>
> For freqmod, you would need to slow down the period variable a whole bunch
> and then gradually raise it until you encounter the mouse problems.
>
> Ray
>
>
> On Tue, 27 Feb 2001, John wrote:
> > Hi everyone,
> >
> > Well with a little help from WIll S. and others I now have a running EMC
> > system.
> <S>
> > I am experiencing a couple of problems.
> >
> > 1.  TKemc works fine when running sim.run but is impossible to use when
> > running emc.run.  Mouse movement is so slow you really can't do any
> > thing productive.  Xemc is works ok, mouse is a little slow but useable
> >
> > 2.  Freqmod locks up the entire system, requiring power off reset.
> >
> > Any suggestions would be greatly appreciated.
>
>