!!! LET'S START AN OFFENSIVE AGAINST THE VIA-MATROX - CONFLICT !!!

[Surfwienix]

you're right, guys. i'm going to buy a new mobo/cpu soon, but i won't go with a bx, because they are pretty old now and i would also prefer an athlon (it's much cheaper!!!), but show me a fast and stable athlon-mobo.

ss7 is not the best, i know, i bought it because of the price. what me makes me angry is, that all the time there were no no no problems with it, and suddenly with the 5.41 it appeared. it couldn't be such a big thing!

[This message has been edited by Surfwienix (edited 24 May 2000).]

[Joel]

Why is it that no one will state what the specific problem is? I am by the way a beta tester for Matrox's video drivers for the G400 and I have a VIA based SS7 system to boot. I use the latest 4.20 4-in-1 VIA drivers along with the 4.02 AGP driver. I can try to check out the problems and then report it directly to Matrox, if need be, but I need specifics.

Joel

Also for y'all running an Intel CPU on a VIA chipset based system of course you are going to have problems, but it is neither VIA's or Matrox's problem it is Intel's. Just think about it for a minute. Intel wants to be the monoploy so why should they develope their CPUs to work with any other chipset but their own.

[Electric Amish]

Joel-

The only problem I'm having is that I couldn't play Q3. I tried lots of things. I ended up with the 4.17 4in1's from VIA and the latest (5.52) drivers from Matrox. It seems to be working OK, now. I still get the occasional BSOD, but I'm only using the Q3 Demo without any patches. I haven't tried HL. The other games I play, Asheron's Call and Rogue Spear all seem to work ok.

This is all on an Athlon 700 (not overclocked) and an Abit KA7 motherboard.

I put together the same system for a friend, except I used the 4.20 VIA drivers and he has an Elsa GeForce GTS2 and it started right up with Q3...

amish

[Brama]

Joel, the problem is well known by Matrox.

But anyway this is the situation: with Via Apollo Pro 133A and VIAGART.SYS rev. 4.02 you can not use Quake 3, unless you disable GL extensions in quake 3 config file.
The only Matrox driver that works is 5.30 but it is very old and DirectX performances are very poor.

If you want to play Quake 3 with PD 5.52, you have to install VIAGART.VXD rev. 3.56, the one that is contained in VIA 4in1 4.17. Unfortunately this VIA AGP driver doesn't enable DirectX AGP and it is very slow with Asus P3V4X MB.

I think big companies like Matrox and VIA could solve the problem in hours and not, like now, trying to give responsability one against the other for months.

In this moment, if I have my wallet full, I would go to the nearer store and would buy a Geforce2.

It seems that Matrox is trying to lose its last home customers.

[Joel]

IMO part of y'alls problems are that you are using a driver pack that VIA does not recommend that you use. This is what they state on their website. "(Install at your own discretion. We do not endorse this driver, but have reposted in response to numerous requests)" Try the 4.16 or 4.18 driver pack and see if that makes a difference.

Joel

[Electric Amish]

Joel-

Where do you find those driver packs??

All I've seen are the 4.20's (that didn't work with Quake) and the 4.17's that did work with Quake...

amish

[IceStorm]

If you've a Super7 board, you should retire it and move up to an Athlon or Celeron II/PIII.

As to the driver problem, since all of VIA's AGP chipsets use the SAME DRIVER, it would behoove VIA to fix the problems with said driver.

Look, like it or not the VIA Apollo Pro 133A is here to stay. It's the ONLY choice for PIII users who don't want to be limited by Intel's 440BX (i820 isn't an option). i815 will not be produced en mass. We're all stuck until Intel releases their next chipset, or the whole world moves to AMD.

VIA's AGP port and Matrox's drivers have not allowed all features of the Matrox drivers to function since PD 5.30. Do a search for my past posts and you'll see the threads where I pointed this out.

At this point, I no longer care. A V5-5500 is within my grasp, and I'll be retiring the G400 MAX to the 440BX board it is on now (Asus P3B-F), while placing the V5-5500 on an Asus P3V4X (provided the drivers do not totally suck).

Hopefully, things are fixed in time for the G800.

Just my $0.02

[This message has been edited by IceStorm (edited 24 May 2000).]

[Taz]

I have three systems here at the moment using chipsets from three different manufacturers plus two Matrox video cards, G200 & G400.

System 1 - Iwill XA100plus - ALi Alladin V Chipset - Runs both G200 & G400 no probs with and without TurboGL drivers installed as shipped.

System 2 - Iwill BDB100 - Intel BX - Runs both G200 & G400 fine again with and without TurboGL using standard drivers.

System 3 - Abit KA7 - Via KX133 - Runs G400, (have'nt tried G200 yet), Ok but no TurboGL, drivers modified to get them to work.

Now from this I can draw only one conclusion, the problem is VIA's and not Matrox's.

Due to Intel's balls up with the i820 & i840 chipsets and ALi's lack of Slot1 or SlotA chipsets we are stuck using VIA's which would be fine it the worked without hours of messing around.

[Alfie]

I'll have to side with Joel here.
WHAT SPECIFIC PROBLEMS ARE YOU HAVING.
You see Joel,perhaps shouting will get an answer!
Unless this is the "Let's complain" Forum!
Yes there can be problems with ss7,
however having used it for the past 2 years the problems I've encountered hav been minimal.TGL doesn't work?QuakeIII doesn't work?
Why is then,if you delete Viagart.vxd.they do work?
I'm serious about developing a bat.file.
Instead of complaining why not figure out how to do it!
Or upgrade to a better platform!
Don't let Gurm get you worked up,he's been pissed off since he couldn't get his PcChips mobo to work and his Maxtor hard drive crashed!

[This message has been edited by Alfie (edited 25 May 2000).]

[Surfwienix]

i never said that it is only matrox' fault, i think too that it's maybe on via's side to produce a new more stable agp-driver.

but i wonder why other graphic-chips doesn't have such problems with via-chipsets????? - correct me when i'm wrong - and matrox didn't have, too, before 5.41. that makes me thinking that it is on matrox' side.

now it's not very important on which side the fault is but it's time to figure out this problem and via and matrox should do this TOGETHER. i'm not sure if they work TOGETHER. but that's what they have to!!!

[Gurm]

MJ12:

Grow up. We've been over this issue a hundred times. There are enlightened dissenters, but for the most part all are in agreement:

NO MATTER WHAT ANYONE SAYS THE VIA SS7 CHIPSETS WERE REALLY REALLY REALLY REALLY REALLY REALLY REALLY REALLY POORLY DESIGNED.

And it really pisses me off to see people continue to bash Matrox over something that just categorically ISN'T THEIR FAULT.

Everyone I've known who has problems with an SS7 has switched to a nice little BX board and POOF! their problems have magically disappeared.

I don't give a rat's ass if Intel has a monopoly... as long as they keep building the ONLY stable hardware (consistently, reliably stable) I will keep buying it. As soon as VIA releases something stable with mature drivers that don't require a 4-in-1 tweak every other weekend, I'll consider it. The 133A is almost there... but only almost. Not quite, at least not yet.

- Gurm

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

[Gurm]

Surfwienix:

Let's see...

ATI: Horrible graphical corruption with VIA

3dfx: Doesn't really use AGP therefore doesn't count...

S3: Nasty ugly problems with VIA, corrected by cutting speed in half on VIA boards

nVidia: Took A YEAR AND A HALF to make TNT boards not crash with VIA, and even then it was at the cost of 15-20% performance.

Matrox: Last couple driver revs. have been picky with Via, no speed loss.

So how, exactly, is Matrox WORSE than the other guys?

- Gurm

[EchoWars]

Damn...look at all these posts!

Mark F-
Only way to overclock on the Tyan is to fool the MB into thinking that you have an honest-to-gosh 133MHz part. With a FC-PGA processor, this means a slotket adapter. With a Slot 1, this means taping off pin a14 so it no longer makes contact in the CPU slot.

Gurm-
Most of the time I respect your intellegent replies and useful information. This is an exception...
We *ALL* know your opinion on VIA chipsets, and while I agree IN PART, sometimes you might consider that if you don't have anything positive to contribute to the discussion, either tone down the attitude, or forgo replying at all...

I say this in all due respect...

[Gurm]

Echo:

Thanks for the advice. I'll consider it. I just get frosted when people start DEMANDING that Matrox fix a problem that isn't theirs to fix... or that at the very least, they can't fix without sacrificing performance.

Let VIA fix their buggy chipset. It's not my opinion of VIA - it's a fact. The 133A has already had more patches than the BX ever had. Let's not get into this argument again, shall we? The problem here is that we can't solve Surf's issue if he won't TELL us the issue, ok?

- Gurm

[EchoWars]

When you're a 'new driver junkie' like a lot of us, it is disheartening to have to wait so long for a 'fix'. Everyone wants their hardware to perform at peak levels, and the hope is always that 'the next driver release is gonna fix ALL my problems', whether those problems are real, or simply an imagined performance deficit.

Since Matrox is not really known as the maker of 'cutting-edge' performance cards, seems to me that whatever they would need to do to enable their next drivers to operate as fast and reliably as possible (even at the loss of some performance on alternative chipsets) should be a priority. No whining, no finger-pointing. VIA is at least 75% of the problem (their new company slogan **VIA- We suck, but not as bad as we used to**).

What amazed me today (similar related note...) was my taking out the G200 Mill from my dad's computer (flags at half-mast around the EchoWars compound) and installing a TNT2 Ultra today. Q3 was, for all practical purposes, unplayable even at 640x480 with a lot of stuff turned off on his Epox EP-MVP3G-M and K6-2 500 with the G200. After the Ultra went in, I discovered that AGP 2x finally works, and Q3 is beautifully smooth at 800x600x16 with all the goodies turned on. Didn't mean to be a traitor, but dads machine needed a new card able to play Q3, and had to go with the most 'bang-for-the-buck'. Bottom line-- whatever else can be said about Nvidia, they have some of the best software developers in the biz, and it shows in their ability to distribute drivers that function as advertised.

Matrox should send a few recuiters down to MIT and snag a couple of eggheads for some outragous salary and show what their hardware can do. On any platform.

Myself...I'm just bummed that, after using 5.41 for so long, I had to actually drop back to 5.30 after buying System Shock 2, at a noticeable loss in performance.

"Here's hoping the next driver release solves all our problems..."
***The technicians prayer***

hehe