Pentium 933 0 flakey with BSOD
A friend has a problem with A Pentium 933.
I told him to run WCPUID on it.
Here's the results.
Everything looks OK (I'm not a Pentium person), except.......
A 58C reading for the CPU core??????
That sounds real fishy.
Here's what he sent me, any comments welcome:
<snip>
I've spent a few more hours experimenting on the system now.
Here are the results of running WCPUID. (Incidentally, I can't believe
this utility doesn't have an option to save the info as text, only as a
graphics bitmap. Quite bizarre... I had to copy all this down by hand.)
----------
WCPUID Information:
Processor: Pentium III Brand: 2
Platform: Socket 370 (PGA370 Socket)
Vendor String: GenuineIntel
CPU Type: Original OEM Processor 0
Family: 6 Model: 8 Stepping ID: 6 (Standard)
Internal Clock: 930.95 MHz System Clock: 132.99 MHz
System bus: 132.99 MHz Multiplier: 7.0
L1 I-Cache: 16K L2 Cache: 256K
L1 D-Cache: 16K Full
L2 Speed: 930.95 MHz
MMX: Supported MMX+: Not supported
SSE: Supported 3DNow!: Not supported
SSE2: Not supported 3DNow!+: Not supported
Windows 98 Version 4.10.2222 A
Chipset Info:
Host Bridge:
Vendor 1106 Device 0691 Revision C4 SubSystem 00000000
VIA Apollo Pro133A
South Bridge:
Vendor 1106 Device 0686 Revision 22 Subsystem 06861106
VIA VT82C686A
VGA Device:
Vendor 102B Device 0525 Revision 04 SubSystem 9010102B
Matrox Millenium G400/G450
AGP Spec 2.0 Enabled
Data Rate 1x 2x 4x Command 2x
Fast Write Not supported
4 Gb Addressing Not supported
Side Band Addressing Supported (Enabled)
Graphics Aperture Size: 256 Mb
BIOS Information:
American Megatrends INc
Version String 62710
Release Date 03/15/2001
ROM Size: 256 Kb
Motherboard information:
Gigabyte Technology
Product String 6VX7-4X / 6VX7B-4X
Version String: 1.0
D:*.* S:2.3 V:F
-----------------------------------
I've installed the latest BIOS, video drivers, motherboard drivers,
and IDE bus-mastering drivers from Via (the 4.29 four-in-one driver set).
The current problem is that after a couple of boots, the system will get
itself into a state where it bombs out with a blue screen reporting
an exception at 0028:C00743E5 in VMM. Once this occurs, it's difficult to
get it to vanish on subsequent reboots. I have a saved partition image
which I can restore, and this appears to get rid of it for another couple
of boots. However, I haven't had time to confirm this definitively (each
iteration takes around 20-30 mins to test properly).
I thought I had noticed a pattern where the blue screens happened when
there was a CD-ROM in the drive during bootup, and not when the drive was
empty (I did seven boots in a row that followed this pattern). However,
I've since seen the blue screen even on a bootup with the drive empty, so
I've discounted this for now.
It seems possible, even likely, that the bus mastering drivers are somehow
at fault here. Unfortunately, I have to use them -- this system is for
video editing, and the drives don't operate fast enough in non-bus-master
mode to keep up with incoming video when capturing to disk.
Installed cards are:
- 3COM 3C905C 10/100 ethernet
- Matrox G400 video card + RT2000 capture card
- Soundblaster 16 PCI
- SCSI card for scanner (disabled in Device Manager)
Motherboard audio and modem are disabled in the BIOS.
Installed IDE devices are:
- Dual IBM 60 Gb 75GXP drives running on IDE Primary bus (ATA-66)
- Pioneer DVD and Plextor 8/4/32 CD-RW running on IDE Secondary bus
(ATA-33)
Memory is 512 Mb. (Tried dropping down to 256 Mb without much success.)
I've considered internal temperature as a possible cause of the boot
failures. However, checking the BIOS CPU health screen on a regular basis
shows essentially no variation regardless of whether the system is working
or not: CPU temperature is 58'C while system temperature is 37'C.
Any thoughts welcome. Meanwhile, I'll keep plugging away.
Eddy
<snip>
PS: He's in Ireland, so communication is slow.
Thanks for any insights or help.
Bob S
------------------
Bob S
