Tyan Trinity S1598C2 Motherboards

[TA152H]

George,

Well, I never said the Tyan was better, I said that I prefer it. I would need way more information before I could say something like one is better than the other.

The IDE controller for the EPOX is an add-on chip, whereas the Tyan uses a newer 686 Southbridge and does not need it, so the data transfer rates can differ. It could also have been a driver issue, a lot of times that is what causes big differences with data transfer. Not sure though, I hate IDE and always use SCSI.

As far as problems with the video cards, I have had none at all, but maybe because I am not using the same ones you are. I only buy ATI, because I never have had problems with them (except for the initial AIW Radeon, which works fine with the drivers that come with it, but goes nuts if you use newer drivers. So much for progress ).

One problem that you may have had is an old MVP3, the CD version, which caused all sorts of problems. Unfortunately, Tyan used these chipsets way too long even after it was discovered they were problematic and the CE version was out. Luckily, the ones I got were all CE version so are not a problem, but if yours were CD versions I suspect you would have far more problems than someone with a CE. I suspect your Epox is a CE, mine is.

The newer revisions of the MVP3G5 may be able to run at 550 MHz fine, but mine will not. The newer S1598C2 may run with fewer problems than the ones you have (or maybe you have a new one), so it may just be the revisions of the motherboards more than the companies themselves. Having said that, Tyan using the CD version of the chipset for so long is certainly not something they should be proud of, so if you got stuck with one of these lemons they deserve your acrimony. They should never have been used after they had better choices.

If you get a chance to look, or remember, let me know what revision your chipset was. It would be interesting to see if that is the cause of a lot of the problems.

[georgep1]

I sold the two Tyan C2 systems with Voodoo3 cards in them to friends for their kids. They work fine for that.

As for the revisions, I'm not sure as I don't have them, and don't have access to them. But that may explain the AGP issues, if not the IDE issues.

I don't think it was driver issues at all for the IDE. I tried both the Microsoft default version (recommended for MVP3 chipsets), and the Via IDE drivers. Didn't really make a difference in the performance.

Epox released a BIOS revision in April 2000 that increased IDE speed considerably. My thinking is, is that Tyan could have, but in their typical fashion, decided the MVP3 motherboards were old enough that they didn't want to devote any more BIOS development for them. They did the same thing with the Apollo 133 chipset mobos, releasing a few BIOS revisions over a 6 month period of time, then quitting. Very different from Epox or ASUS or most other motherboard makers that release BIOS updates 1.5 years or more after the motherboard is first released.

[TA152H]

George,

Hmmm, your post makes me think it could have been a driver issue, because you really wanted the MVP4 drivers for the Tyan, since it uses the newer Southbridge and not the one normally used by the MVP3. To my knowledge, Tyan was the only one to do this; everyone else used the standard 586 Southbridge and threw in another controller for ATA 66. That may also be why they needed another BIOS update. Hard to say though.

My experience with Tyan has been fairly dreadful too; I would never get a motherboard of theirs again unless it were a special case like this. I do like this motherboard though, but as I mentioned before their insistence on using CD versions of the MVP3 Northbridge reveal a distinct lack of quality. These things were really buggy, and even back then I remember how everyone avoided them when the bugs were discovered. That Tyan went on using them so long afterwards is, to me, unforgivable. I personally like Aopen the best, as I have never had any problems with their products. All things considered, my EPOX is fine too, it just is not as elegant as the Tyan and does not run at 550 MHz. No real problems though.

[georgep1]

The Microsoft default IDE driver is the one that is recommended for MVP3 or MVP4 chipset motherboards with Win98 (which is what I had for those mobos). I.E. that is the one that installs by default in Windows. The Via IDE Miniport driver was the possible replacement but when installed, did nothing for performance improvement. Since the installation of either driver is done with no user input during the process, there is no way to choose which southbridge/chipset it is being installed for. The installs should detect that automatically, if needed.

It's possible the revision of the motherboard did make a difference. But in all the time I hung out on the usenet Tyan newsgroups trying to figure it out or work around it, by far the most problems posted by users were about IDE performance.

[Lee]

I just sent an e-mail to find out if anymore are available.

Great Post!!!

Thank You Very Much TA152H!!!

Lee

P.S. I hope I'm not too late!!!

[Lee]

Hi All,

I just got a reply about the availability of these Motherboards.

He has 30+ left!

I had the e-mail address wrong, (I use external mail), TA152H helped me out:

Mitsubishi0uk@hotmail.com

(That's a Zero not a Capital O)

Thanks Again TA152H!!!

Lee

P.S. He gave me a price of £20 each with £25 for shipping to the
U.S.A. Midwest, ymmv. Good Luck.

[thetempleoflove]

Hi Y'All,
Which pins are what - if I want to connect the USB at the front of the case to the mobo?
Cheers,
Ges.

[TA152H]

Just a follow up post regarding these motherboards and a K6-III+ 400 ATZ I bought a while back.

I am setting up a LAN so intended to use these motherboards for two machines, one with the K6-III+ and the other with a K6-2 550 I had in my EPoX MVP3G5. Neither the K6-2 550 or K6-III+ would run at 550 MHz in the EPoX.

So far, one motherboard has had problems that I think are indicative of it being defective. It sort of runs Windows ME, but I have way too many problems to be happy with it, and have tried to determine the cause. Right now, I am not sure, but I think the motherboard is the cause. Eventually, the machine would just crash on boots, so I replaced the motherboard and tried again. It booted up but crashed on some internet sites. Since the nature of the crashes changed, it may indicate something with the motherboard, but I will know more later and post it when I do. So far, a nightmare.

On this motherboard, however, I was able to determine the "overclockability" of this particular K6-III+. It is rated at 400 MHz and 1.6v. It runs at 450 MHz at 1.6v but will not run at 500 MHz. At 1.8v it runs at 500 MHz, at 2.0v it runs at 550 MHz. Just so you do not come to the obvious conclusion that it was overclocking the processor that caused the problems, when I was installing everything it was all done at 1.6v and 400 MHz, which makes me fairly certain the processor is not the problem (since it ran stably at 450 MHz at the same voltage.

The other motherboard I put my K6-2 550 in and it ran perfectly, and at 550 MHz. The only problem I had was with Windows Update, which did not like the date being so far off (I forgot to change it), but after that I had no problems at all. That it can run my K6-2 and K6-III+ at 550 MHz is a nice contrast to my EPoX MVP3G5 which simply would not. I am guessing the later revisions of that board do though.

If I can figure out the exact nature of my problems with the first setup, I will post them. The Radeon 9000 AIW is a suspect too, but it is supposed to be compatible with this setup. Hopefully, I will find out for sure and let you know.

[TA152H]

Well, I have some information that might be helpful with this setup.

In short, the PCI bus was the problem, because it was not able to switch quickly enough from one device to another because the settings in the chipset were wrong. After hours of research I was finally able to determine that the utility vlatency was made for this problem. It seems SoundBlaster cards are particularly troublesome, but since I have known for years to avoid them, I did not have them. My sound card had the same problem, and I was even having problems with my DVD player and hard drive. The Radeon AIW 9000 caused the problems, or really aggravated the existing problem. Once it was removed, no problems at all. With newer drivers, the IDE problems seemed to go away, but no luck at all with the sound, when using ATI's DVD player (oddly, the one that came with with the drive worked fine).

After I used the vlatency utility, it all cleared up. It fixes quite a few problems, and not just with the MVP3 but with newer chipsets too.

Hope this helps some of you that have weird problems, particularly clipping in audio.

With this software and the K6-III+ running at 550 MHz, the machine is quite fast. It runs DVDs perfectly, and even plays Dungeon Siege II very well.

Let me know if you need clarification on anything. I will do more research on this tool in the near future and hopefully be better able to describe what it does. I just wanted to get this down as quickly as possible so people would know there is a solution if they run into this problem.

[TA152H]

Some follow-up information.

Someone earlier in this thread mentioned how bad the hard disk performance was on this motherboard. The Vlatency utility cleared that all up, and boosted beformance on the Sandra benchmark by over 60% on the hard drive I used.

If your BIOS does not already do it, get a utility that interleaves the memory. It makes a big difference in the memory throughput.

For the Tyan motherboard this thread is about, set the memory speed to fast or turbo, not 8 ns. The 8 ns sets the timings to x-3-3, whereas fast or turbo sets it to x-2-2. The SDRAM cycle time determines what x is (2 or 3).

Some USB devices severely eat up memory bandwidth, and are probably best avoided, particularly on a K6-2. I had an Ethernet device that attached to the USB port and it chewed up memory bandwidth in a most inconsiderate and disagreeable manner. My USB modem also seriously degrades performance, so for MVP3 chipsets, you are probably better off with other choices. While the numbers given by the synthetic benchmark were dramatic, I am more basing them on the feel of the machine. The USB devices definitely slow them down. On this machine (which I am testing at 366/66), the modem lowered my memory bandwidth (MMX) from 121 to 19 MB/s while it was downloading a file. Not too hard to notice that . On my K6-III+, running at 550 MHz, the Ethernet device lowered the memory from 170 MB/s to 124 when connected to a USB 2.0 port, and to 148 when connected to a USB 1.1 port. So, it is substantial.

One last remark about the S1598C2. There is a 1.06e revision that Tyan never released, but is not too hard to get. You may want to try it out if you have problems you are not able to solve with the utilities. I saw no big difference between it and 1.06c, but you may. Ooops, forgot to mention that despite the fact Tyan says it only can see 37 GB disks, I have a 60 GB disks in it and it has no problem seeing it even with 1.06c. Last, but not least, www.esupport.com sells BIOS upgrades for just about all motherboards, and they do have one for the S1598C2, but it only allows it to see larger fixed disks. It costs $60 for the upgrade, which seems like a lot since you can probably buy an intelligent controller for less and deal with the problem that way, in particular a SATA controller.

Anyway, hope this is helpful to someone.