PCI Latency bug is an issue. SB Live! cards are out. Otherwise, I really like my KT133A system.

Edit: I see what you mean. So the VIA chipset doesnt work with Live!? Do i have to only buy a nForce board? I would rather not have to buy an Audigy 1. I dont hate them, just dont want to spend more money. Does nForce2 work well with 9x?

PCI Latency and SB Live! problems come with the VIA 686B Southbridge boards. I like these boards bc they have an ISA slot. I suspect there are other Socket A boards that do not have the bug.

OH ok. So im sitll perfectly safe as long as i avoid those ones. Perfect, i would have never known this. Thanks for info. Thought i would be stuck with nForce2. Looks like the ASUS A7V8X-X uses VIA VT8235. I cant seem to find anything about that mobo having issues with SB Live!.

The issue has been exaggerated over the years. It only affects the 686b southbridge which means most KT133 and KT133A motherboards. other via chipsets were not affected. Looking up your board on google, it has a KT400 and should not have any issues with the soundblaster live.

I have a Kt133 motherboard (fic AZ11e) and back in the day I had that issue when using a soundblaster live. The issue with data loss only happens when you copy data between a hard drive on the primary and one on the sceondary port. Otherwise you just get some additional noise (crackling and chirping) on the sound output and thats it. Bios updates and later drivers significantly reduced those issues and then there was a unofficial patch that solved it completely, but I can't remember what it was called.

Well perfect then. Thank you a lot for confirming. Looks like im going for this ASUS board. Its very cheap and has AGP 8x.

I had a sblive in my Abit board for nearly 3 years before I upgraded to a P4 system and never had an issue with it. I even had it back in there a couple of years ago
and there were no issues. It's a ABIT KT7-RAID which is KT133, see my sig.
It's a 686A southbridge chip, so maybe they never had an issue.

Running an AthlonXP cpu here, though with a V3-3500 instead. It's my high performing 98 machine. Well... At least high performing to me (don't like highest performing possible).

It runs good, solid and all that. AthlonXP's are good performing 9x solutions. If you go the route of GF4, then you will have to do a lot of testing, in order to get the correctly driver combo. Takes more time to find the right drivers, and that makes it fun. 😜

Trank wrote:

Edit: I see what you mean. So the VIA chipset doesnt work with Live!? Do i have to only buy a nForce board? I would rather not have to buy an Audigy 1. I dont hate them, just dont want to spend more money. Does nForce2 work well with 9x?

My sample size is small, so my answer is not definitive, but in two cases in which I tried an nforce2 board with Windows 98 SE, I failed. Both boards were Asus, and they both freaked out as soon as I installed their official (the ones came with the board) w9x drivers. My CPUS were not Athlon XP but Athlon 64 (3200+ if I remember correctly). Any remedies like updated official (again) drivers did not work, either. Both boards were rock solid with Windows XP, Fedora and Pardus.

boxpressed wrote:

PCI Latency bug is an issue. SB Live! cards are out. Otherwise, I really like my KT133A system.

Wasn't there a fix for this latency bug? At least I remember it was 😉

kixs wrote:

boxpressed wrote:

PCI Latency bug is an issue. SB Live! cards are out. Otherwise, I really like my KT133A system.

Wasn't there a fix for this latency bug? At least I remember it was 😉

Do you mean this one? http://www.georgebreese.com/net/software/read … 0b21_readme.HTM

http://www.georgebreese.com/net/software/

Yes. I used to have Abit KT7A (KT133A) and SBlive! back then and remember always installing latency patch after reinstalling Windows - I did that quite a lot 😉 And it worked fine.

Running an XP 2400+ on a KT133A board together with a Ti4600. Using the 45.23 Detonator drivers and 4.42 VIA 4-in-1 chipset drivers. Never gave me any trouble whatsoever, except for having to recap the mobo. In fact, it feels even more stable and robust than my 440BX system. Works just as good in DOS too.

And yes, using the latest PCI latency patch from the page linked above.

Detailed specs here, if you're interested 😀

You can go as high as socket 754/939 w/o any issues in win98 as long as you use compatible chipsets (VIA K8T800 / ULi M1698 and nforce 3 to some extent). On the LGA 775 side there's the PT880 from VIA and the i865 from intel. Regarding the PCI latency bug, that can be fixed in bios on some boards (set "PCI Latency" to "64" and "PCI Wait for Pending Transaction" to "Enabled") or you can use cards that are unaffected like the Crystal 4xxx series, the Yamaha DS-XG or the Vortex 2.

I believe AM2 K8T890 boards have drivers for Windows 98.

gdjacobs wrote:

I believe AM2 K8T890 boards have drivers for Windows 98.

K8T890 has Windows 9x drivers, and quite a few of the AM2 boards support the 6 core PHenom II line, not that you really need a Phenom II X6 for 98SE

Didnt think you could go that far into the AMD line. I wont be going with a KT133A board anyways though. Looks like this KT400 do. With a Athlon 1600+ maybe.

Trank wrote:

Didnt think you could go that far into the AMD line. I wont be going with a KT133A board anyways though. Looks like this KT400 do. With a Athlon 1600+ maybe.

If using KT400 try an XP 3000/XP2800

I ran an AMD Athlon XP 3000+ on a Gigabyte GA-7VAX motherboard (KT400A chipset) for years as my main Windows XP PC (as I recall, I bought the motherboard in 2002 and only upgraded in 2008).
I started out with an Athlon XP 1800+ and upgraded to the 3000+ later on.

The motherboard eventually became unstable (due to a failed capacitor) and I replaced it with the GA-7VAXP motherboard. However, now that my soldering skills have improved a bit, I can probably just replace the capacitor.

Overall, I was quite happy with the service I got out of that PC.