On Sun, 2006-11-19 at 13:24 +0000, Timothy Murphy wrote: > Alessandro Brezzi wrote: > > >> So I'll ask again if anyone is happily running the x86_64 version of FC-6 > >> on an Athlon 64 machine? > > > > Yes, I'm happily running FC6 x86_64 on my box Athlon 64 3200+. > > No problems with the installer (GUI). I've done a fresh install (my > > preferred method for upgrading). The MoBo is an ASUS, graphic card ATI > > Radeon, 2GB of DDR. Both NIC, the GigaBit integrated and a RealTek > > 10/100 are working. > > OK, thanks for the responses. > Evidently the x86_64 version of FC-6 works fine, in principle. > > I also have an Asus motherboard and ATI Radeon 9600 video card, > though I only have 512MB RAM. > > What I find puzzling is that my machine works perfectly > under the i386 version of FC-6 and under Windows XP. > Under the x86_64 version it can be guaranteed to crash within 1 hour > (usually within 1 minute), even if run in text mode, > and even with a compiled vanilla kernel. > > Also, after the crash it is not sufficient to re-boot > by pressing the power switch for 10 seconds - > one has to switch the machine off and on at the back. > I take it this shows some kind of memory corruption. > This description tells me that something is definitely not right with memory after the crash. Have you checked with memtest86 to see if that reports anything? You also might try reducing the speed of the memory and see if that has any affect. I built one machine recently that used DDR2-800 memory. Running at 800 it would test correctly with memtest, but was unstable in operation. I manually reduced the speed to 667 and it became very rock solid. Never did determine if that was bios or mobo, but I am sure the memory itself was good at the design speed. > It's a Targa machine, and I looked at the www.targa.com > to see if they had upgraded the CPU or BIOS, > but they are still shipping the same versions, > so I assume there has been no general failure. > > Could it be heat? > It seems improbable to me that one OS can create much more heat > than another. > > So I am genuinely baffled, and would like to hear > of any theory on this. > > At the same time, this has really become a theoretical question, > as I am perfectly happy to run the i386 version, > which seems to me to run at least as fast as the x86_64 version > ever did. What about cooling on the processor? Are you using the heat sink that was provided with the CPU or better? If so then it is not likely that it is heat related, although with inadequate cooling it is possible. > > -- > Timothy Murphy > e-mail (<80k only): tim /at/ birdsnest.maths.tcd.ie > tel: +353-86-2336090, +353-1-2842366 > s-mail: School of Mathematics, Trinity College, Dublin 2, Ireland >
