My guess that it might be a bridge or RAM issue.
Do you mean the north bridge / southbridge stuff on my MSI board. I've written on another message about the memory. I don't mind buy better if that is what it takes, but don't want to buy unnecessaryily.
Have you tried other memory-intensive things - like 'make -j 100' in /usr/src/linux? (That's a joke -- actually don't go that high, unless you have *gobs* of RAM.I have 512MB of DDR ram, but sorry I don't understand the rference, make -j 100
I'm going to be running mprime in windblows later today, will post the results.
John
Later =====
Well it seems we are all wrong.
I didn't think there was any point in running FSB=100 in W2K So here is the result of my W2K mprime test,
FSB=133 ======= Beginning a continueous self-test to check your computer. Please read stress.txt. Choose Test/Stop to end this test. Test1,400 Lucas-Lehmer iterations of M19922945 using 1024 FT length Test2,400 Lucas-Lehmer iterations of M19922943 using 1024 FT length Test3,400 Lucas-Lehmer iterations of M18874369 using 1024 FT length Test4,400 Lucas-Lehmer iterations of M18874367 using 1024 FT length Test5,400 Lucas-Lehmer iterations of M17825793 using 1024 FT length It actually got to Test6 but hung there.
Now M9.1 hung right at the beginning at the start of test1.In effect it didn't get anywhere in FSB=133, so this is a better result than in M9.1, but to be honest, not that much better, not as as much as I fondly thought, which shows the importance of having realistic demanding tests to work with. Impressions can be both accurate and misleading.
So I'm left thinking that something, whether memory or cpu works somewhat better in W2K on current settings, but that these may well not be the best and most appropriate settings for the equipement. Anyway I've got a few other things to do for a while now. Back later.
John
-- John Richard Smith [EMAIL PROTECTED]
Want to buy your Pack or Services from MandrakeSoft? Go to http://www.mandrakestore.com
