In my quest to find out what VIA’s official stand on the issues recently discovered with the 686 series Southbridge when using two IDE hard disks and a Creative SoundBlaster Live (SBL) PCI card, I sent off two emails. One was to a Technical Marketing Engineer at VIA Tech here in the USA, whom I’d dealt with in the past while he was employed at mainboard distributor I had dealings with, the other was to someone involved with VIA Cyrix Global Support, whom I had met at a trade fair. Both replied back to me promptly, both acknowledged there was an issue but did not have (or offer) any other details, and both also offered to get back to me with more info.
I never did hear directly back from either, but I did receive an email out of the blue from a Sales and Product Support Manager at VIA Cyrix in Plano, TX. Now, due to the sensitive nature of any issues that might show up over time with a product, and VIA’s relationship with mainboard manufacturers, it did not surprise me that my two contacts did not or could not supply me with more info. But the Sales and Product Support Manager did supply me (and apparently a few others) with the following as an official comment from VIA:
Regarding the “VIA 686B data corruption” issue. The issue should be the description as following , In fact , it is a specific problem . It only occurs on ” K7 series north-bridge with 686B ” on a specific Creative SoundBlaster Live PCI Sound card . Only when executing copy compare with a big file (more than 100 MB) , the system would hang ,not corruption . This should be a compatible issue. VIA will release a patch driver to fix the issue soon .
So let’s take a look at what is being said here…
- It’s only a K7 Series Northbridge & 686B issue when using a SBL and large files. So that means to me VIA feels it could be an issue with either a VIA KT133 or AMD 760 Northbridge if using the 686B Southbridge, but not with a Socket 370 Northbridge.
- Only large (100MB or more in size) files when copying.
- It’s not a matter of data errors but system errors (hang, lock-up, blue screen,…).
- VIA will release a patch soon to solve the problem.
Issue #1 – KT133(A) and 686B seems to be where most reports are coming from, but I did manage to also have errors with a 686A, and also to a limited extent a 694X & 686B. I also did not have any errors with the AMD 760 & 686B combo I tested. I think I’ll take a second look at the 686A & KT133, and also the 694X (Dean has a nifty new program called ‘Disk Blaster’ that works very well at finding data errors or causing the system to lockup due to disk problems).
Issue #2 – Well, I did some testing. With the Disk Blaster program, I could cause disk errors with files even as small as 8MB. So it is not just large files (but errors seem more likely to occur with them). I would say its got more to do with the type of disk access than file size.
Issue #3 – I would tend to agree that the issue causes system errors rather than data errors. The reason I believe so is that the Disk Blaster software tracks errors, but it reported none. What did happen is that the system would lock up or I would get a ‘Blue Screen’ error message stating there was an error writing to one of the disks (for some reason it always reported C drive). So, at this time I would not be concerned with data being corrupt on the disk, with the exception of currently open files when the system crashes.
Issue #4 – Although I have not received any patch officially, nor have I seen it on an official VIA site, there is a version that appears to be a complete usable patch at www.viahardware.com (no connection with VIA, it is an independent site). The patch is called ‘VIA PFD’ and is dated just over a week ago (5/4/2001). So VIA is working on it (be patient, it takes time to not only find exactly what the issue is but also find a solution and test it).
Be the first to discuss this article!