Very Computer

While trying to install the Slackware 3.0 distribution on a new clone mobo
with the AMD DX4/120MHz 'xxxxxSV8B' 'Enhanced Plus' CPU, with AMD 7/95 BIOS,
we experienced multiple segmentation faults, and were ultimately unsuccessful.
 Installation was flawless using my older mobo, a DX2/66MHz using an Intel CPU
with an older ('93?) AMI BIOS.  

For both cases we used the 'net' boot diskette and the 'color.gz' root
diskette.

BIOS cache setting changes and mobo bus clock changes resulted in bogomip
numbers of 2.29 and 2.38 with internal cache enabled, and 50.08 with internal
cache disabled and the mobo running at 33MHz vice 40MHz.  We also tried
various combinations of wait states to slow the cpu/board down as much as
possible.

There did not seem to be any consistent pattern to the occurence of
segmentation faults.

The DX4/120 mobo/BIOS worked fine for Wfwg 3.11.

If anyone has any ideas regarding this mobo/cpu setup, we are anxious to hear
them.

TIA,

Frank

There can be a very simple answer to this problem....

Check your Motherboard and see if it can handle different chip voltages.
The AMD chips are typically UNDERRATED for power... a 3V chip (on avg - but not
always) really takes 3.3V or 3.45V.  This is fine and the chip can handle it.
I was having the same seg fault problem until I learned of this and upped the
voltage to the chip (same chip too!) and it resolved my probs.

-Chris

: While trying to install the Slackware 3.0 distribution on a new clone mobo
: with the AMD DX4/120MHz 'xxxxxSV8B' 'Enhanced Plus' CPU, with AMD 7/95 BIOS,
: we experienced multiple segmentation faults, and were ultimately unsuccessful.
:  Installation was flawless using my older mobo, a DX2/66MHz using an Intel CPU
: with an older ('93?) AMI BIOS.  

: For both cases we used the 'net' boot diskette and the 'color.gz' root
: diskette.

: BIOS cache setting changes and mobo bus clock changes resulted in bogomip
: numbers of 2.29 and 2.38 with internal cache enabled, and 50.08 with internal
: cache disabled and the mobo running at 33MHz vice 40MHz.  We also tried
: various combinations of wait states to slow the cpu/board down as much as
: possible.

: There did not seem to be any consistent pattern to the occurence of
: segmentation faults.

: The DX4/120 mobo/BIOS worked fine for Wfwg 3.11.

: If anyone has any ideas regarding this mobo/cpu setup, we are anxious to hear
: them.

: TIA,

: Frank