One of the list owners asked me to explore a bit deeper into the "hidden
refresh" indicated as the cause of this problem by Microsoft's article.

  I intended to check it out further in the book produced by the makers of
TweakBIOS, but didn't get to it.
  However, a web search finds several FAQs from STB about possible problems
with some of their video cards and this option, which they indicate is also
sometimes called "decouple-refresh".

  And at "The AMI BIOS Survival Guide, Part Two"

http://hjs.geol.uib.no/amibios2.html

describes several refresh options supported by some AMI BIOSes; it recommends
"hidden refresh" as having least impact on system performance, but warns that
not all RAM will work with this option.  [And apparently, not all video or
sound cards....]


http://www.mindmachine.co.uk/Technologies/memory/refresh.htm

tells more than you could ever want to know about the different ways that
dynamic RAM refresh can be done.  In general, you're limited to the methods
the designers of your motherboard chose to implement.
  [Speaking of which, I also found a board on which a BIOS revision included
switching the default setting for "Hiden Refresh" from "Enabled" to
"Disabled" -- presumably, fewer customers were having this work than the
designers expected.]




David G

             Do you want to signoff PCSOFT or just change to
                    Digest mode - visit our web site:
                    http://nospin.com/pc/pcsoft.html