>I am no hardware (or even software) guru, but the fact that I have
>this problem in Linux but not Win95 on the same box makes me skeptical
>that the problem (in my case) is hardware. Glad you got it resolved
>though!
Yeah..same here, nothing wrong with the board running win (both
flavours), NT4(supp_3+hotfixes...groan..), or even RHL-3 & 4...you'd think
that conclusive in itself. Seems not...but perhaps the case is that all I
have done is altered my hardware profile, to suit my software's settings
or likings...weird magic if this is so, I might add.
I personally suspect broken/strange BIOS in my case (haha). In point,
think of this pondering. I *thought* I read somewhere, that the linux
kernel was largely BIOS independant, wherein all the system really need to
know, was where that crucial bootsector of your harddrive was. (I think
this is in the boxedset installbook actually)From there on, it would take
care of all the issues courtesy of your kernel profile & remake...rc
scripts... lala&so-on. Now, however wise this is, I think it stands in
contradiction to what the other_lot are doing in this respect..ie;
implementing bios extensions for PnP & suchlike, which makes this end of
the deal more complex...bios config interfaces have become a maze, or, as
in later boards with onboard video, bells, scsi..blahblah, the bios is
locked away.
I mentioned in previous posting, bios numbers of award that
end with -00 being bad magic. Not all, but most of these, where 'lots' of
bios runs done OEM to a number of asian manufactures. You end up finding
this information out by the way, trying to chase jumper-settings for the
thing, (cause it comes cheap with no book), so based only on the bios
'signature' number, you can try and chase down who made it, do they still
exist (as a company) blahblah. IF you luck_out here, then the best
recourse is to cross-reference the actual chipset & support chipset/s
against a manufacturer who still carries reference to the same config. If
you're lucky, you'll find a match where all the jumpers exist, just in
different locations. I've had a number of successes in this regard, and
one exception.
That board, is a umc8881/8886 (8883/8887 supp) chipset good
up to 586/133. 4ISA/3PCI...do yu think I can convince the bios to STOP
trying to update ESCD every reboot..takes 90seconds?!? Try and put
anything BUT a PnP enabled ne2000 in it, and it simply won't boot.., and I
later found out, if you stuck a PCI videocard in it, the ESCD would update
successfully in a flash..weird.
It was mentioned to me that these small OEM outfits, had made batch run
based on liasons with other local manufactures of cards and stuff, and
whole lots were sold as 'business package deals' with defined hardwares,
and bios' were tailored to suit the installations. I dunno how true this
is but it's what I have heard and the things I find seem to point that
way.
What exactly does 'largely bios independant' mean anyhow, when we all
know the bios can nail you to the wall if incorrectly set (;
cheers
Db
--
PLEASE read the Red Hat FAQ, Tips, Errata and the MAILING LIST ARCHIVES!
http://www.redhat.com/RedHat-FAQ /RedHat-Errata /RedHat-Tips /mailing-lists
To unsubscribe: mail [EMAIL PROTECTED] with
"unsubscribe" as the Subject.
