What is the deal with the crummy new drivers?????

[The PIT]

I've got to agree with Rags. Via need to do better testing with there drivers before releasing them. Until they sort out the Q/A problem they'll always be second rate board manufacturer.
It's shame as the Athlon for one is a great chip. It's the prospect of random blue screen of death, none working hardware thats put me off. Since I sold my last Via chipset the only time I see strange problems are on collegues home machines and the odd machine at work (All concerned have gone for a cheap option).

[Joel]

EXCUSE ME!!! I have two SS7 systems based on the VIA chipset and they are both running very stable with the last drivers, THANK YOU VERY MUCH.

Joel

Oh and PS: VIA is not a board manufacturer, all they do is produce the chipset.

[This message has been edited by Joel (edited 01 March 2000).]

[The PIT]

Okay Via make the chips. But you've got to admit the majority of queries on this board are about Via chipsets.
So the conclussion would be Via is the most common chipset. Since it isn't the only other conclussion is that it's not the most "Friendly of chipsets" and that the drivers tend to have more bugs than drivers than supplied for or by Intel (Excluding the latest Intel farce before someone brings that up).
By the way I'm not getting at the owners of Via chipsets just the chipset. So if yours works fine don't get hot under the coller.

[Gurm]

*sigh*

Well, I feel obliged to jump in on this topic at this time.

Point 1:

We ALL LIKE VIA. Even those of us that bash them WANT Intel to have competition. Ok?

Point 2:

We're not insulting you personally if you have had the incredible good fortune to never run into a problem with a VIA chipset. More power to ya! You're a lucky person!

Point 3:

Having stated points 1 and 2, I must now state point 3, which is that VIA IS UTTER CRAP FOR CHIPSETS.

Not in features, not in performance, but in compatibility. The only reason the MVP3 runs even VAGUELY stably now (and it's been 2 years since it was introduced, almost as long as the BX) is because special drivers have been developed by every major card manufacturer to bypass the broken parts of the VIA implementation. Period. End of story. That's just the way it is. VIA broke some stuff, and has never fixed it. They've gotten BETTER, certainly - the 133+ is a wonderful chipset compared to the MVP3. But it's not perfect. It's not 100% stable. It's not even CLOSE. Try running a TNT2 or GeForce on an MVP3... failure. Try it on a 133+... crapshoot. Put in on a BX... works! Wow!

Sorry to get all sarcastic, but it's just the plain ol' truth.

- Gurm

------------------
Listen up, you primitive screwheads! See this? This is my BOOMSTICK! Etc. etc.

[Joel]

We're not insulting you personally if you have had the incredible good fortune to never run into a problem with a VIA chipset. More power to ya! You're a lucky person!

I don't think good fortune or luck has anything to do with it.

Joel

[bmcgarvey]

merchant2112-i have an Athlon 600 on the MSI K7pro moboard. after working out some problems concerning my sound card (MX300) it seems to be working out pretty good. i am not using a AMD recommended power supply, case or memory but i haven't run into any problems. here is my entire configuration:

MSI K7 Pro mainboard
Athlon 600
128MB Siemens PC100
300watt power supply
Phoenix case (not sure of company name)
20.5GB IBM Deskstar 34GXP 7200RPM
Millennium G400MAX
Diamond Monster Sound MX300
Creative Encore 5x DVD (not using decoder)
Creative 4x2x24 CDRW
56k v90 external modem
Ext. Zip drive

[LAMFDTK]

What a Crummy thread

[EchoWars]

lol...

Interesting to note that my Mill G200, which would NEVER run at 2x on a BX chipset, does so on the 133a chipset. But only with the 5.30-5.41 mix of drivers.

As far as memory performance, the 133 and 133a are a few percent slower -AT THE SAME CLOCK SPEED-. Memory transfer rates on a BX at 100Mhz is nowhere near a 133a chipset at 133Mhz. Everything I have read on the web says this, and my own tests comfirm it. The difference is inconsequential at the same bus speed, unless all you do on your system is run benchmarks. If you look at some comparisons of the i820 running the MTH at 133Mhz VS the VIA at 133Mhz, the i820 gets slaughtered. So much for Intel's superiority. Again, this may not matter much in *real world* apps, but this time, the difference is a lot more than a few percent.

I started this thread in the hopes that the Matrox people who read posts here might know that these problems exist. I did not intend to have to defend my choice in hardware (unless I bought a PCChips board with a SIS chipset..lol). I read the threads here, and there are a lot of people running BX chipsets who cannot upgrade their drivers beyond 5.3 or 5.41. Especially if they are running a G200.
I also understand the loyalty for Matrox here. But when someone says that are having a problem running a vid card on a *very mainstream* piece of hardware, it is a cop-out to point to the MB or chipset, and say 'ya shoulda bought sumpthin' else'. Functional drivers CAN be written (NV10 and TNT2 run great with the newest drivers on the 133a chipset, and framerates are equal to or better than on a BX). If Matrox has issues with *all* their cards on the 133 and 133a chipset, or with G200's on the BX platform, a *known issues* in the Readme file would go a long way towards eliminating a great deal of frustration in screwing with drivers trying to find a bloody release that actually WORKS. *whew*
Would just like to see some progress made here. Drivers should get better and more compatable, not less-stable and chipset-specific.

[Compton]

EchoWars, I agree with you. I have an Apollo Pro133 based board and I haven't had any issues with it other than the Matrox G200 drivers. (I'm glad to hear that Matrox is trying to resolve this problem. Look here http://forums.murc.ws/ubb/Forum5/HTML/007980.html ) Via gets a bad rap because there's board makers out there putting out shit (FIC and Abit's VA6) and some people are dumb enough to buy it.

[paulcs]

"NV10 and TNT2 run great with the newest drivers on the 133a chipset, and framerates are equal to or better than on a BX"

I don't believe this to be true at all. I have never heard anyone make this claim before, and I've certainly have never seen any data that backs this up.

In addition, there seems to be enormous variance from board to board:
http://www.anandtech.com/showdoc.html?i=1182&p=26

Now, I doubt Anand is using the newest drivers. However, the new 4in1's appear to be a disaster for many people. We've seen quite a few people on the MURC forums return to ver. 4.17, and their problems with the new drivers didn't appear to have anything to do with their video boards.

In addition, if your information is correct, this seems to me a pyrrhic victory. Finally catching up with a 3 year old chipset just doesn't seem to me to be that great an accomplishment.

Paul
paulcs@flashcom.net


[This message has been edited by paulcs (edited 01 March 2000).]

[EchoWars]

**sigh**

My point is lost.

Paulcs, if you want to read benchmarks, I can dig 'em up. But you are still missing the point. You and others here are focusing on excuses for Matrox. The proof is in the pudding...others have drivers that work ***really goddamn good*** on the new VIA chipsets. Why not Matrox?


[This message has been edited by EchoWars (edited 01 March 2000).]

[paulcs]

No, you missed my point. I did not mention Matrox *once* in my post and focused exclusively on VIA and one statement you made that I found specious. Even if it is true, I still think VIA is being damned with faint praise.

I think Matrox's VIA support certainly could be better. I've been hearing similar complaints since the release of the G200. I also think VIA could provide better support and should stop releasing drivers that cause more problems then they solve. I just don't think you can point the finger exclusively at Matrox.

I'd like to see those benchmarks. If you could point me at a link, I'd appreciate it. I assume these numbers were generated by the new Gigabyte board. I have to say, other than the Abit board, it isn't easy to find boards based on the 133A locally.

Paul
paulcs@flashcom.net

[chip]

the amd athlon with msi 6167 is simply SUPERIOR to ANY INTEL JUNK.period.

------------------
IT'S NEVER FAST ENOUGH!

[EchoWars]

The review I read, which pitted a VIA Apollo Pro 133A directly agenst a BX..well...I can't find it (I know what this sounds like, so spare me..) However, Tom's Hardware compared the KX133 Athlon chipset with the i840. If you agree that the i840 has AT LEAST as good of an AGP implementation as the BX (which I'm sure it does, hell it might be better), and can accept that the VIA 133A uses the SAME AGP implimentation as the KX133 (development was done at pretty much the same time), then you might find this enlightening.
VIA KX133 vs i840
I agree that VIA has a bit of their act to get together, but I think the next few months should prove to be very educational for VIA-bashers. It's bloody hard to compete with Chipzilla...

[merchant2112]

waxling could you give us some numbers. im going to be building a k7 system for a buddy.

how do you like your k7pro board. (im curently useing the msi 6167 my self and love it. give us the low down

noel

------------------
msi 6167 mobo k7 500 wk41 now at 650. 256 meg ram ,addtronics case w 250watt sp power supply, matrox g400, maxtor diammax 2500+ 10gig hd,10x aopen slot dvd, 3com 10/100 nic and 56k ext modem, sb live xgamer sound card, lots of controlers (joystick throttle rudder raceing wheel), 19in ctx monitor, microsoft inteli mouse explorer(5 buttons, and klipsch promedia v2-400 speakers.

noel
Don't son that gun is loaded.