Have you seen me?

[Rob M.]

If you have, you've been victim of the matrox g400 bug. I don't know why but it just happened to me as it often does, but this time I could continue doing things.. So I quickly took a screenshot and saved the file

I'd just like to know, have you ever seen this? Personally I've seen 4 g400s, and every one displayed this problem which doesn't seem to be reproduceable if you live in a certain part of Montreal..

btw, to keep the image size down I had to compress very heavily(max setting), if you want another small pic of the screen that's not compressed as much, click here.

[Kruzin]

Looks like too much overclocking, or too much heat...or both...

[Rags]

Kruzin is correct, This is usually from too much overclocking/heat. I have also seen this happen in win2k when I had a bad stick of memory.

Rags

[Rob M.]

I've opened up my computer and had fans blowing on my g400 and had this happen. I've seen this on 4 seperate cards in 5 systems, none of them overclocked. That picture I took, while I had my goldfinger overclocking my cpu, the agp bus was at 66 mhz, at agp 1x. Amazingly enough it doesn't frustrate me as much as the icd problem. At 1x this is fairly rare.

[Wombat]

Sounds like you're making decisions about narrowing down the problem.

Do all these machines have good power supplies? Are you using a line noise suppressor? How many different sticks of RAM have you tried?

RF interference problems?

[Rob M.]

I've seen this problem with at least 5 power supplies, and at least 6 sticks of ram. I've seen this crash on my current 300w power supply, and my ram that runs stable at pc133 cas3, even when it was running at pc100 cas3.

I don't think it's RF or line noise. It's almost perfectly stable at 1x, but at times crash nonstop at 2x.

In fact, after living with this for almost a year now, my observations go totally against my knowledge that I've gathered when working as a technician. This is what appears to be happening:

First off, agp 2x will run fine for a few days. It will then suddenly garble. After I can reboot, and it might be good for an hour or so, but will garble again. Eventually I won't be able to finish booting up without my computer crashing. Even if I leave my computer off for a day to cool down, the longest it'll last is maybe another hour.
It immediately goes away once I enable 1x (except for the rare crash like the one today). In fact, after I leave it on AGP 1x for a while (maybe a few weeks) and then enable 2x, it'll be stable again... for another few days. Lather, rinse, repeat. Yuck. I've even noticed the exact same thing in win2k.

In fact, it's quite similar to the main problem I'm having with q3radiant. It'll work fine for a day or two, but will suddenly crash while I'm working with a curve. I'll get BSOD's that'll take down the OS in win9x. In win2k, q3radiant simply crashes. It'll continue crashing whenever I look at a curve, but if I turn off curves then it'll seem stable, for about another hour.
Eventually, I won't be able to load up a map at all without crashing. Disabling the g400icd.dll file (renaming it) and using software mode, q3radiant is stable for weeks on end. After some time, if I put the ICD back, q3radiant will be stable again for another day or two, at which point the cycle will repeat again.

DRIVING ME INSANE!

The 3 cards I've had all have done exactly the same thing to me, even when I tested them out in different systems.. My friend seems to only get this crash on rare occasions, like me in AGP 1x mode.

Maybe I'm really really unlucky and have got 3 really bad cards in a row?

[Greebe]

Rob, you mentioned using a GFD, might I suggest dropping it down a hair. This might sound odd to you but I've seen this problem all too many times while testing cpu's.

------------------
Designer of the first commercial Athlon mod.
Anyone up for an Athlon800+? Email me (epc@netdirect.net) for more information

3000 mips, 1.172 gflops @ 900 mHz 1:3 latency w/ active cooling
CPUmark99 75
FPU Winmark 4890

3Dmark 99 Max (8x6 16, 16, triple) PD5.41, Athlon@850mHz
8000
15181

[dZeus]

you told us you have 5 systems with this problem? I know 4 people in real live with a G400, so that makes 5 including me, and I have never, _ever_ seens such a odd anonymaly

[Electric Amish]

I'll have to agree with the BB'z..

The only time I've seen anything like that is an overheat or trying to push to much on the overclock of the G400..

amish

[Rob M.]

As I've said, I get this regardless of overclocking. I don't think I've even ever overclocked this video card. All 4 of the cards I've seen this on were the MAX models that have their built in HSF.

I've seen this on:
3celerons, a p3 and an athlon
abit BH6, BF6, BX6R2; Asus K7V
(I get the crash just as reliably with only the video card plugged into the system)
I also spent half a day in the store I worked at swapping the MB, ram, cpu, and power supply trying to see what combo would work.

I'd be amazed if no one else has seen this before, I can't avoid it. Do you all have 4x cards? The models I've had were all 2x.

[This message has been edited by Rob M. (edited 01 August 2000).]

[Maggi]

Hi Rob,

I might have missed it, but what MoBo do you use and how does it handle the FSB/AGP ratio, i.e. what is your FSB/AGP ratio set to ?

Cheers,
Maggi

[Rob M.]

the celerons ran at a 1:1 ratio, the p3 and athlon at 2:3. 66mhz across the board

[Greebe]

I think the point is moot now. He's tried everything, multiple G400's, power supplies, MB's, CPU's, memory, etc. Must be that "Proximity Effect" Kruzin and I have seen so many times before.

[Rob M.]

Well, I remember a few people when the G400s were first released claiming a problem that sounded like this. But if it is a proximity thing, then it's only my proximity to the g400. I've had no problems getting other cards running in my system.

[Vic]

My G200 would regularly do the very same thing with certain earlier driver releases.

Vic