More on vic/vat long term testing

[Gregory W Cook <gcook@ecn.purdue.edu>]

A vic/vat update:

1.  I ran another long term test on vic.  I am using a Pentium III/Windows 2000
machine and it failed at about 18 hours 15 minutes after I started it.  Failure
in this case just means that it completely froze.  When I killed the vic window
the error message was:

The instruction at "0x100130ca" referenced memory at "0x00000008".  The
memory could not be "read".

It *seems* like it is a memory leak or pointer problem.  I am running
another test today except that I am insuring that at least one
receiver is also running.  I am not running in "scuba" mode, BTW.  
(Does anyone think this would make a difference?)


2.  To recap my problems with vat, my previous error was that it died
unceremoneously after about 5 seconds without an error message -- it just quits
without ever opening a window.  I am using an Osprey video card with an on-board
audio digitizer.  When I installed mash on a similar machine with an Osprey
video card that did not have on-board audio, vat worked fine.  Unfortunately,
this particular machine was a Windows NT 4.0, so it is not a completely perfect
test.  Anyway, all signs right now point to vat having problems if there is more
than one audio device installed.  Has anyone else seen this problem?

Greg Cook
cook@purdue.edu