Athlon 1200 system crashes?

G

Guest

Guest
I posted this to Opinions/Faulty Products but am crossposting here in hopes of getting a better response. This is driving me nuts; I think its incompatible video or memory but I don't have spare parts to swap. Thoughts are appreciated!

------

Here's one for ya guys:
Got a brand-spanking-new Athlon 1200/266bus upgrade for my system and everything runs fine until I use it.

System is:
Inwin full-tower case
AGI "Athlon-approved" 300W power supply
Gigabyte GA-7DX rev3.0 motherboard, F1 BIOS
Athlon 1200/266Mhz FSB
Cooler Master DP5-6H11 fan
128MB PC2100 DDR SDRAM
Hercules 3D Prophet DDR-DVI
D-Link DFE-503TX+ ethernet
Adaptec PCI SCSI-2 - model unknown
Maxtor 80GB ATA100 drive
Mitsumi 3.5 floppy
Toshiba 12x DVD, IDE
Smart & Friendly 12x4x32 CD-RW on SCSI, ID=6
HP 6250Cxi scanner on SCSI, ID=2
Windows 98 SE (2222 A) with IE 5.5 and all critical updates
Latest drivers for all hardware

I have the distributed.net client running as a service on this machine, so the CPU is at 100% all the time. Gigabyte's SIV reports CPU temp at 63-67degC. I can leave the machine at the network login all night and not have any problems, but as soon as I start using it interactively, it will crash (usually bluescreen, sometimes IPF) within 10 minutes. Most of the crashes are page faults and invalid instructions (05/06/0E) and every one seems to be different. Will post the addresses and modules if anyone would find them useful.

Have tried formatting the hard drive and reinstalling everything to no avail. Also tried installing Win2K Pro but get STOP 0x0000000A (IRQL_NOT_LESS_OR_EQUAL) during file copy.

One final note: The processor, motherboard, HSF, and memory replaced the following:
Athlon 950/200FSB
Asus A7V motherboard
128MB of generic PC133 SDRAM
ThermalTake Super Orb cooler
- and that system, with all other hardware identical, has been running just fine the last 4 months.

Sorry for the rambling. Your thoughts are appreciated.
 
G

Guest

Guest
Thanks for the site - I did find some missing VxD files and Windows 98 seems a bit quicker now, and crashes a little less. I'll brainmark that technique. (Hint for others doing this: System File Checker can extract a file from the install cabs more easily, especially if you don't know which one its in. Start/Run "sfc".)

However, this does not explain why Windows 2000 won't even install (clean install, partition isolated w/ System Commander).

Still looking for ideas here...
 
G

Guest

Guest
63C, that kinda hot, WHAT HSF you ussing. Actually, more importantly, do you OC it. IF you are keeping it at 1.2, TEMP shouldn't be your problem.(maybe, maybe not) but just in-case, you can try TESTING it with case open, and bring it near an OUTSIDE cool AIR window. BUT if you have it at default of 1.2 i don't think that the problem. 63C does seem HOT, my 1.2 is around MID-HIGH 40C LOAD.

ONE conclusion i got from a few experiment i tried one my cpu is that, IF if too hot, IT won't be Overclockable, and STable. I'm on of the unlucky ass, who got ripp off with an OLD 1.2, anyway, I couldn't GET it stable over 1.24 with CASE close, but i could get the 1.2 to be stable at 1.35 with OPEN case and open WINDOW, It will post at 1.4 but NO good boot in to W2K. so if you hadn't had a proble with your 900mhz, i think it might be the HEAT.

WHEN the CPU get TOO hot, IT might be workable, but there might be a Slight miscalculation. and just that slight mis-calc. will bring down an OS. so give it a try, HEAT my be your problem.
 
G

Guest

Guest
See my original post... its a Cooler Master DP5-6H11.

I have a feeling Gigabyte's SIV is boosting the figures about 10 degrees (phenomenon has been reported at HardOCP and others)... in the BIOS, case open, the CPU temp is reported at 50 degC. Closing the case causes a temperature rise of 3-4 degrees.

I'm not overclocking and won't even try until the system is stable at 1200/266.

Would you recommend a different HSF?