Galvin's Problems

[slyfox]

Hey Galvin,

After being such an ass around here do you really believe that the people around are going to believe you no matter what you say you did. You already ran off one of your best sources for information and help. And if I'm not mistaken isn't Rags a beta tester for Matrox? Which would mean that he has probably seen alot more driver versions than you have. And would it not also be his agreement with Matrox to report any problems he finds? Don't you think he would have checked your story out by now and is only offering you advice based on that observation and experience?

So why don't you just do what the guys suggest? If it don't work then at the very least you have at least cleaned your system up.

slyfox

[This message has been edited by slyfox (edited 15 December 2000).]

[Galvin]

Originally posted by slyfox:
Hey Galvin,

After being such an ass around here do you really believe that the people around are going to believe you no matter what you say you did. You already ran off one of your best sources for information and help. And if I'm not mistaken isn't Rags a beta tester for Matrox? Which would mean that he has probably seen alot more driver versions than you have. And would it not also be his agreement with Matrox to report any problems he finds? Don't you think he would have checked your story out by now and is only offering you advice based on that observation and experience?

So why don't you just do what the guys suggest? If it don't work then at the very least you have at least cleaned your system up.

slyfox

[This message has been edited by slyfox (edited 15 December 2000).]

I installed a fresh copy of win2K how much clearer can I get. He's just one beta tester, hell if every company went by just what one beta tester said then their would be tons of bugs in software. But since you probably didn't read the complete thread i'll sum it up.

1. I'm not the only person that has these problems.
2. I did install win2K on a seperate drive and installed just the bare essentials to run quake3 with the latest matrox drivers.

In final what is pissing me off the most if that Rags, and Gurm are ignoring the fact that I am not the only person with this problem which proves it to be a bug and not something special to just my machine.

Rags or Gurm you ever think that the matrox drivers may have issues with certain chipsets ? For matrox to truely test the problem they need to get a CUSL2 board like mine, install the same hardware like mine and look for the bug.

So that's why I am getting so pissed off.

[dZeus]

Slyfox: Matrox even confirmed the 'texture corruption' as being a driver bug: http://forum.matrox.com/mgaforum/For...ML/000463.html

It's the second picture in Galvins post. Now I am not denying that you also get this problem, I am just saying that I don't believe his software config/arragements of PCI cards can be changed to fix this. I think it is indeed a driver issue (I get the thing myself on a totally different config).

And I am sure that this particular bug did not show up in the earlier drivers (pre 5.10).

However, I agree, I also run in ACPI mode with all the PCI/AGP devices on one IRQ (IRQ 11 in my case). However, I seem to remember that Gurm said millions of times that this is perfectly normal behaviour, and that everything should work just fine. And everything on my pc does work fine, except for the texture bug in the post 5.10 drivers.

However, tomorrow I will test this on a Win98 pc with NO ACPI and NO IRQ sharing... if that pc does show up this problem, than the probability of this being a driver bug is 99%

[This message has been edited by dZeus (edited 15 December 2000).]

[Rags]

Galvin,

Your summation was very poor, and you have a tough time reading posts. It appears that mr Dzeus has the same affliction as you.

I am NOT trying to solve texture corruptions on your machine. I acknowledge there are some bugs in the drivers, but what you fail to realize is that a system locking up is NOT one of the damned bugs. How hard is that to understand? Also, if you have your system set correctly, and your Q3 settings set properly, you won't get near as much glitches graphically as you do using a GeForce or Radeon. Believe me, I have used both. I also have used SEVERAL i815 chipset based motherboards, so don't go telling me that it's a problem with the drivers and your motherboard, because that is PURE BS. I repeat, your screenies don't show corruption. Corruption is an entirely different error, and once texture corruption occurs, it doesn't just magically disappear, the texture(s) that corrupt will stay corrupted until it is dumped from memory, then re-entered it may be okay.

I suggested to you to install win2k in a standard PC HAL because it is VERY clear, without a doubt in my mind (especially since I have that very motherboard in my box right now as I type), there are issues with ACPI and this board. This would hopefully stop the lockups, and if not, it would at the very least eliminate one cause.

But you know what? I could care less if you ever get it set up. That's up to you, but don't come here saying it's Matrox's or Microsoft's fault for something that is clearly the fault of the user.

Have a nice day.

Rags

[slyfox]

Actually Galvin I think there are about 8 or 10 external testers as well as those internally. Also if I am not mistaken most of them are members of this board. So I would be willing to bet you that this problem(?) that you are having has been discussed and researched. And maybe this has been mentioned before but is there anyway that this problem could be heat related. I don't know, I not the expert here. I defer to true experts. I was mainly commenting on your arrogant attitude and how it will get you no where on these boards.

slyox

[Rags]

dZeus,

That link you posted just confirmed what I said earlier. What is your point? I know there are lightmap errors. This has been known for some time. Like I posted earlier, other graphics cards experience lightmap errors in Q3A, and with more frequency under win2k.

Rags

[Galvin]

Originally posted by Rags:
Galvin,

Your summation was very poor, and you have a tough time reading posts. It appears that mr Dzeus has the same affliction as you.

I am NOT trying to solve texture corruptions on your machine. I acknowledge there are some bugs in the drivers, but what you fail to realize is that a system locking up is NOT one of the damned bugs. How hard is that to understand? Also, if you have your system set correctly, and your Q3 settings set properly, you won't get near as much glitches graphically as you do using a GeForce or Radeon. Believe me, I have used both. I also have used SEVERAL i815 chipset based motherboards, so don't go telling me that it's a problem with the drivers and your motherboard, because that is PURE BS. I repeat, your screenies don't show corruption. Corruption is an entirely different error, and once texture corruption occurs, it doesn't just magically disappear, the texture(s) that corrupt will stay corrupted until it is dumped from memory, then re-entered it may be okay.

I suggested to you to install win2k in a standard PC HAL because it is VERY clear, without a doubt in my mind (especially since I have that very motherboard in my box right now as I type), there are issues with ACPI and this board. This would hopefully stop the lockups, and if not, it would at the very least eliminate one cause.

But you know what? I could care less if you ever get it set up. That's up to you, but don't come here saying it's Matrox's or Microsoft's fault for something that is clearly the fault of the user.

Have a nice day.

Rags

How do you know there are ACPI issues with the board ?

[dZeus]

Rags:

I agree that his locking up problems proably don't have anything to do with the G400 drivers, but I never said so in the first place.

However, YOU were the one stating his 'texture corruption' problems were not in the drivers, but probably related to his configuration, because according to you, nobody said 'yah' when he mentioned his texture corruption problem. I was merely contradicting that. I see you are saying that ACPI putting all the PCI devices on one IRQ may cause trouble, but I doubt that. (and btw, I own a BX mobo, not a i815). Tomorrow I will look into that (as I already said)

Oh, and I am not a graphics expert, and I don't know what these 'bugs' should be named. If you give me a better description, I will take that. I do have screenshot listed for these kind of problems, so what else do you want to have?

And please, still tell me what bugs on my page are only applicable to 'previous drivers': I am anxious to know!

[Rags]

Glavin,

Because I do, and your IRC buddies know this as well. It has interupt problems when allowing win2k's ACPI to handle the interrupts, it just doesn't want to release full control for some reason.

I fought with this board every way possible to get ACPI to work stable for me, and it was a no go. The board works great using the standard PC HAL, and is rock solid.

Rags

PS-Again, I hate to repeat myself. Please read my posts, and I wouldn't have to. It really gets old. It seems like you just skim over the words and form a general idea of what is said (I HOPE you aren't illiterate)



[This message has been edited by Rags (edited 15 December 2000).]

[dZeus]

Rags: what exactly do you mean with 'your link'... I've linked to about 3 different OpenGL problems in this thread.

I repeat, your screenies don't show corruption. Corruption is an entirely different error, and once texture corruption occurs, it doesn't just magically disappear, the texture(s) that corrupt will stay corrupted until it is dumped from memory, then re-entered it may be okay.

My point is that other people are getting the texture corruption problems Galvin is talking about too, and those people use a config that is not even the slightest bit the same as his (well.... except for ACPI, and I won't say anything about this anymore, untill I check this out tomorrow on a non-ACPI pc).

What is your point however (in your quoted part of my post)? That the problems ARE in the driver, but don't have the names that Galvin and I give them? or that those problems are NOT in the drivers, but cause by the pc config? That's what I am trying to figure out

[This message has been edited by dZeus (edited 15 December 2000).]

[2Whyzzi]

Interesting. Matrox says bug in driver. Perhaps they didn't write the driver with IRQ sharing in mind. VAR rule of thumb: IRQ sharing = usuability complaints from customers = bad idea. Thats why I suggested what I did. Glad I'm out of the business. 'nuff said.

BTW - don't knock off advice until you try it. Just don't come crying to us if it still doesn't work - Go to Matrox and file a bug report/complaint. If S3 can fix come back and fix a problem (Savage4/A90 on MVP3) so can Matrox.

dZues: If you can, I would appriciate it if you could test W2K + 5.14 with and without ACPI but NO IRQ sharing -- to see if it ACPI or IRQ sharing could be the cause. Thanx.

PeterV

------------------
Abit BH6 r1.01
Celeron2-566 o/c to 850MHz+Slotket!!! 1.8v
256 Megs PC-133 Cas3
Matrox G400 SH OEM (not oc'd) rev=03h
Diamond Monster Sound MX400

[Rags]

dZeus,

However, YOU were the one stating his 'texture corruption' problems were not in the drivers, but probably related to his configuration, because according to you, nobody said 'yah' when he mentioned his texture corruption problem. I was merely contradicting that.

And I am totally correct when I say that there aren't texture corruption in his drivers. If he is getting texture corruption, he has a misconfiguration or faulty hardware. It's easier and more sensible to fix KNOWN configuration problems rather than pull hardware in a guessing game.
And if you are contradicting that fact, then I am sorry for you.

And please, still tell me what bugs on my page are only applicable to 'previous drivers': I am anxious to know!

Right off the top of my head, mind you I am not looking at your page this very moment, the shadow error is fixed, the clipping errors are fixed, I will look at the page tonight to see if there are others. Remember, I am speaking from my OWN experience, and not speaking for Matrox.

. I see you are saying that ACPI putting all the PCI devices on one IRQ may cause trouble, but I doubt that. (and btw, I own a BX mobo, not a i815).

This means absolutely nothing to our discussion. The chipset itself isn't the problem, it's the motherboard's very own implementation that causes problems. My Abit BF6 has zero problems with ACPI using the latest BIOS. If a system properly supports ACPI, then I suggest using ACPI, I am not saying ACPI is bad and should never be used.

Rags

[dZeus]

My Abit BX6 r.2 is almost identical to your BF6. And yes, I am using the latest BIOS.

Right off the top of my head, mind you I am not looking at your page this very moment, the shadow error is fixed, the clipping errors are fixed, I will look at the page tonight to see if there are others. Remember, I am speaking from my OWN experience, and not speaking for Matrox.

HUH? I don't see any 'clipping errors' or 'shadow errors' listed on my page????? Maybe 2 moths ago???? not today, not yesterday, not the week before yesterday, not the week before that either, etc.

update: I never contradicted ACPI causing this issue... I was just saying it was very unlikely, and that I would check into this issue tomorrow

update2: as a matter of fact, the shadow bug was listed as 'fixed' when the 5.14 driver was released... go see my '5.11 -> 5.14' driver changes list

[This message has been edited by dZeus (edited 15 December 2000).]

[Rags]

dZeus,

Rags: what exactly do you mean with 'your link'... I've linked to about 3 different OpenGL problems in this thread.

The link to the Matrox forum where you said they admitted this bug. If you read through this thread (again, you really should if you are going to argue), you will see I said there were lightmap errors, and the link you posted said the same thing I did, so what are you trying to prove? I already agreed it was there. Are you just trying to argue for the heck of it, or did you just not read this thread?

What is your point however (in your quoted part of my post)? That the problems ARE in the driver, but don't have the names that Galvin and I give them? or that those problems are NOT in the drivers, but cause by the pc config? That's what I am trying to figure out

My point is that when someone is claiming to have all this texture corruption, and is really meaning lightmap errors, then there is a lack of communication. The lightmap errors are there, I have never denied that. As far as the misplaced and missing textures on respawn, I have not seen this in win2k yet.

Rags

[dZeus]

ok, thanks for giving me the right name for the 'lightmap errors' in the driver. I will change the name of this problem on my website to 'lightmap errors' too, if you give me some time (1 - 2 days, not more).

And yes, I think most of the arguing in this thread has been going on because of mis-communication. And I am as much to blame for that as others... here's a sorry for that

As for the problem that there is screen corruption (or whatever is the right name) when you spawn, if you follow the directions on either my website or here in one of the posts I made (with the three links), than you can see if it is possible to reproduce it on your system. On my system, those steps always lead to a 100% chance of reproduction.