nForce woes?

[Peter Wemm]

On Monday 06 December 2004 01:55 pm, Scott Dodson wrote:
> Do we expect the nForce 4 chipset to alleviate many of the problems
> found in nForce 3 or should I steer clear of products using this
> chipset as well?

My experience is with the nForce3-Pro150 and the reference bioses for 
it.  The nForce3-Pro250 might be better.  The same goes for the 
nForce4.

There were two problems on the Pro150 boards, possibly one was an 
artifact of the other.

1) the bioses were really shoddy.  If I understand things correctly, 
this is because the reference bioses were shoddy and few motherboard 
makers bothered to do anything about it.

2) the APIC subsystem was very unreliable under FreeBSD.  It is 
distinctly possible that this was bugs/races in our code, but I believe 
it far more likely that it was genuine chipset bugs or malfunctions due 
to bad bios setup of the hardware.  The problems should exist on 
FreeBSD/i386 as well.  I verified that the easiest ones to reproduce on 
my boards do show up in 32 bit mode as well.

I was so gun-shy over the whole thing that I've avoided them like the 
plague.  Since I was buying them myself and not using sample boards or 
the like, I didn't want to risk wasting money on more junk.  For all I 
know, they've fixed bugs in the chipset.  But I'm not going to be the 
one to fork out $ to find out, I've already lost enough sleep over 
them.

-- 
Peter Wemm - peter at wemm.org; peter at FreeBSD.org; peter at yahoo-inc.com
"All of this is for nothing if we don't go to the stars" - JMS/B5