Re: [CAD_CAM_EDM_DRO] Re: Fwd: Emc > 0 C.

Dave Engvall wrote:

> On 6/11/01 10:17 AM, "Jon Elson" <elson@...> wrote:
>
> > Ray wrote:
> >
> >> ---------- Forwarded Message ----------
> >> Subject: Emc > 0 C.
> >> Date: Sun, 10 Jun 2001 21:24:21 -0700
> >> From: Dave Engvall <dengvall@...>
> >>
> >> Hi Ray,
> >> A couple of days ago I bit the bullet and moved my desktop machine (K6/300,
> >> 128 Mb, etc) to the shop to sub for my Boeing surplus P200 that seemed to be
> >> having real problems with freezing. Prelim results would seem to indicate
> >> that the switch was a good one. No crashes yet but will keep testing. I can
> >> edit from xemc, or vi and not get a freeze which is really nice.
> >
> > This sounds good. How many hours have you had it running without trouble?
> > Which version of EMC are you running? Does TkEMC work, too, or does it
> > freeze up?
> After reading this I tired tkemc and it was really flaky. X couldn't even
> display it correctly. Will try again today and try to accum some stats.
> Stats of ones are non-existent.
> >

OK, this is the same as the results I get with my 333 MHz Pentium II (the one I had
up at NAMES) with the 15-Mar-2000 EMC. I'm pretty sure I get that same
thing with my reliable 100 MHz Pentium classic that is on my milling machine.
I CAN use TkEMC reliably on the 100 MHz machine with the 20-Dec-1999
version of EMC. Maybe this is a clue. (I think one of the differences is that
the RT section is periodically rescheduled on 15-Mar-2000, but the interval
rescheduled RT is used on the 20-Dec-1999. I don't know if that is significant,
though.) I think there were also some major changes internal to TkEMC, and that
may be closer to the problem.

> I may have to try the upgrade on XF86 yet in hopes of having something work.
> It should give me access to a couple more video boards. Will stay in touch
> as data develops.

Hmmm, I don't know if this will bear fruit or not. I think some people who are using
late versions of Red Hat (like 6.2) have the same problems. (There are even newer
XF86's than on the RH 6.2, though.) But, as Ray seems pretty convinced it is not just
a simple X problem, I don't know if that will help.

Thanks,

Jon