Don’t jump to conclusions… we’re using VIA-based motherboards in dozens of
NT machines in three states. They work great, and are very fast. The only
problem we’re seeing is (get ready for this) an inability to successfully
burn CD-R’s on an 867/133 MB/CPU/memory motherboard (?!?).
Naturally I’d blame it on the VIA’s EIDE bus, except that this is a SCSI
CD-R drive (Yamaha 4416S) on a PCI SCSI interface. We’ve tried two different
SCSI interfaces from two different vendors (Mylex and Adaptec)… two
separate CD-R software packages (Adaptec and Nero)… every combination you
can think of, and they all fail the same way. The CD-R burning software
believes it has been successful, but a manual comparison between the source
and target files reveals differences!
Everything works great on a 450/100 MB/CPU/memory platform. But if we
literally leave all the peripherals in place and simply swap the underlying
MB/CPU/memory, everything continues to work great - except we get bad CD’s.
Even through the VIA system is running its FSB at 133MHz, its PCI slots are
still at 33MHz so the SCSI interfaces shouldn’t see a difference. Likewise,
the CD-R drive is always seeing the SCSI interface and is entirely isolated
from the PCI slot, let alone the motherboard’s FSB. Thus the only thing
we’ve been able to think of is a HAL problem, since we’re not re-installing
NT when we switch motherboards. That’s why I asked about a VIA133-specific
HAL… it’s the only thing left that could explain this behavior.
Any ideas?
At 01:24 PM 10/04/2000 +0400, Maxim S. Shatskih wrote:
Thanks Richard, I will not buy the VIA chipset after this
only Intel
8xx.----- Original Message -----
> I’m seeing some strange behavior which seems to be linked to the use of
> VIA133 vs. Intel BX chipsets. Is there a different HAL between these two?
> I’m not aware of one, but it might exist. I would presume the HAL to be the
> same for both, but if not - where can I obtain more information on the
> correct HAL and how to install it on a running copy of NT4?