Tweet
Thread summary for your amusement... or not... (LONG)
Well, just to satisfy myself that I wasn't "talking through the wrong end" as Rags put it, I went through the entire thread and summarized it all below.
It all began innocently enought when...
1) Rags stated that Matrox had good driver development and support.
1.1) I then basically asked... well, if that really is the case, then where are Matrox Whistler drivers?
2) Rags then stated "Hehehe, you aren't too bright. Whistler isn't even in a public beta right now."
2.1) I then provided Rags with a <a href="http://msdn.microsoft.com/subscriptions/resources/subdwnld.asp">link</a> to the Microsoft MSDN download site which cleary shows that the Whistler Beta 1 is available to those MSDN subscribers who have the appropriate subscription level.
So I stated that it seemed, in my opinion, that Whistler has in fact been generally available to the Windows development community (or any member of the general public who has the appropriate MSDN subscription) since early November 2000. I'm also pretty sure that the MSND subscribers should have received the CD by now as well.
3) Rags then stated that "the key phrase in what you just said was "WIN2K". I don't see anywhere on the sight where they say these will work under whistler". And that he assumed members of the Matrox developers program would currently have access to Whistler drivers.
3.1) I then agreed, that yes indeed, the Windows 2000 drivers were only stated to work in Windows 2000. But, I also pointed out that it seems to me that other vendors didn't seem to have troubles in this area, in my opinion. I don't have access to Matrox beta driver program so I wouldn't know about that, but I haven't seen any developer drivers for Whistler at this point.
4) DGhost then said "If you are a *legit* microsoft beta tester, then you understand that companies don't release drivers for the platform until it has been released." (NOTE: I never stated anywhere if I was or was not a beta tester)
4.1) I than responded that yes, he was infact correct on that point. No vendor could claim Whislter support until RTM and WHQL have been run against the package. But, I wasn't asking for official released and offically supported drivers. Just any type of drivers at all that would provide more comprehensive suppor than the "in the box" Microsoft divers. I am fairly sure that preliminary drivers can be offered by hardware vendors through Micrsoft using the download site.
4.2) DGhost then accused me of being a warez kiddie or something when he stated "you are obviously not a legit tester." It should also be noted that I never said the 5.20 drivers didn't work with SMP systems, despite DGhost claiming "i have yet to experience any problems on my system with the drivers. And i was affected by the problem in the 5.1x series of drivers." But, one wonders, do they have the same OpenGL issues that affected the Win9x released at the same time? Well, from this posting it woudl seem that I was NOT alone with issues in OpenGL with the 5.20 drivers.
5) Rags then stated "Can you imagine all the tech support calls from Warez kiddies complaining the the latest drivers aren't working under build 21xx when they did under build 20xx?"
5.1) I then simply pointed to the Linux drivers which are offered by Matrox yet are not officially supported. Do these Linux drivers cause support nightmares for Matrox? I wouldn't thing so, becuase they aren't supported by Matrox and they are free to simply ignore the posts or simply filter them out. Again, they could restrict access to be done through Microsoft. My issue was not one of lack of support the Linux drivers, rather that there would most likely not be a techincal support calls arriving from Warez kiddies. Support for Whislter drivers would, supposedly to be done though Microsoft who, if needed will interface with the hardware vendor through the appropriate channels if needed.
6) DGhost stated "I think Matrox is faster to fix its problems than any other companies - fast to release the drivers is a different story. ."
6.1) I then provided SMP support as a counter example to that. I dunno, but is SMP not an high priority issue? Especially since it is essential to WHQL certification, at least the last time I checked it was.
7) DGhost that stated that "I think they test them far more rigidly than nvidia or 3dfx does."
7.1) I pointed to the OpenGL issue in the Nov 2 drivers as a counter example to the calimed lots of testing. How did that (in my opinion) rather serious issue slip through their testing then? It totally hosed BG2 and I don't recall see that as a known issue in the release notes from Matrox.
8) I then said others have different expectations since it should be expected that every one uses theie hardware differently.
8.1) Rags said that "drivers for running whistler isn't a reasonable expectation. It's not a feature, and it definitely is NOT a bug."
8.2) Kruzin chimed in with "You can't expect public drivers for an OS that is not final. We heard the same thing about 2k before it was released."
8.3) True, I guess. But then I guess Matrox will just have to suffer the lack of extensive beta testing of the hardware. Their and their customers loss I guess. Bit it sure doesn't seem to me at least that Matrox is fully behind Whistler.
But then I guess that I will also be told that is isn't realistic to expect "bug free" drivers will complete and equivalent support for what is available in Windows ME the day that Whislter is available through retail channels.
Fine, but that's just your opinion. On the one hand people tout that Matrox "have Linux support, OS/2 support, Win95 support, Win98 support, Win98 SE support, Win Me support, and Win2k support from matrox" but then state that support for Whistle isn't a desirable feature.
9) Rags then seemed to challenge me with the following query "OpenGL problems? Can you be more specific. You know, it sounds more and more like people come here and just shoot out OpenGL as an issue with Matrox drivers without really having any issue themselves, or just talking out of the wrong end."
9.1) I then pointed to a Dec 2 posting (which he even referenced earlier in the thread) on Matrox forums which confirms a know OpenGL bug with the Matrox Nov 02 drivers and shows that I was not in fact "talking out of the wrong end". This bug did in fact affect a lot more OpenGL applications than just Alice.
10) Rags then basically said that "As far as SMP goes, the SMP stability issue was one of the first fixes in driver releases and has been fixed for quite some time now."
10.1) I then stated 5.04 Apr 2000 was first stable SMP driver. And the latest good one was 5.20 Nov 2000 fully fixed SMP but seemed to have some OpenGL issues. I haven't tried the 5.31 Dec 22 drivers, but I would expect that they work fine as the Win9x ones do.
11) Rags then stated that "Sorry, but I believe it was the 5.14 drivers that included the SMP fixes, 5.20 was just a WHQL release that was probably in development and testing before the 5.14 drivers."
11.1) All I have to do is point to technical support posts stating to use the 5.04 drivers for SMP systems instead of 5.14 ones. As well, DGhost's own posting in the thread seeme to indciate that "And i was affected by the problem in the 5.1x series of drivers" in the thread.
Here Rags, just to keep you informed is a <a href="http://forum.matrox.com/mgaforum/Forum1/HTML/002987.html">acknowledgement</a> posted Oct 31 from Matrox of know SMP bugs in the 5.14 drivers.
12) You then said the fact that Matrox fixed the Alice bug (which they introduced in the Nov 02 drivers) in the Dec 22 drivers was proof of quick fixes.
12.1) I point out the fact that this bug shouldn't have been there in the first place, as the previous OpenGL drivers do not have these specific bugs.
13) Greebe then stepped in stating that "Alice, Baldur's Gate II, Oni all have updates coming out for them because the games were released before these issues were corrected."
13.1) I guess you are right, the issues were fixed in the Dec 22 drivers and all those games, and to the best of my knowledge IN EVERY APPLICATION available prior to Dec 21, has been released before the Dec 22 drivers. What's your point with that statement?
To the best of my knowledge, <a href="http://www.interplay.com/bgate2/">Baldur's Gate 2</a> shipped September 26. That is about a month prior to the Nov 02 drivers being released which broke Baldur's Gate 2, right?
The point is the Nov 2 drivers broke this game and several as yet unrelease games. But, going back to the previous drivers available immediately prior to the Dec 22 drivers, does fix the problems generated by the Nov 2 drivers in Alice and Oni. In the Dev 22 drivers these issues are fixed as well. Uninstalling and using a previosu driver is what the Matrox technical support people were telling customers who were reporting issues with OpenGL Nov 2 drivers to do in some cases when the proposed workarounds didn't work. Uninstalling the Nov 2 drivers and go back to the previous ones always seemed to fix the new issues created by the Nov 2 drivers.
Greebe then stated that "at least know what your talking about first. At least that way you won't appear to be a TROLL talking out his arse!"
14) I then stated that, it seemed logical to expect that Microsoft wasn't going to have their Windows 2000 OS support DualHead(tm) without Matrox making the effort to give them drivers EARLY in the OS development cycle.
14.1) Rags then said that "MS knew very well what was needed to allow dualhead/twinview/etc." and that Microsoft "chose to take the easier way out and stick to the more NT4 like setup in that regard."
14.2) Sorry, but unless you work or worked for Matrox and/or Microsoft I don't really know how you could be so well versed in this issue to know with absolute certainty that this is actually the case.
Well, how can you realistically expect Microsoft to support something if there are no drivers for them to attempt to test and develop their operating system with?
How would Microsoft even know that the limitation existed unless Matrox told them about it? Yet you assume that Microsoft well versed and toally familiar with the G400 card at that point?
Didn't Mircosoft state at one point that Windows 2000 SP1 would add support for DualHead? Didn't they then later change their position on that stating that it would not be all that feasible to do so? Obviously, this would seem to point to the fact that Microsoft did not in fact know the Matrox hardware all that well, in my opinion at least. I can't see why Microsoft would state that publically and then reverse their decisions stating that it would be too complicated to add? Perhaps, they overlooked something since they didn't fully understand who DualHead worked, or maybe they felt the effort to add it to Whistler was more important. Of course, considering when they started working on Whistler this woudln't be the least bit surprising to me.
You need to remember that (I think) Matrox is using a unique method on their card. It is always very important (in my opinion) that IHVs work closely with Micrsoft to push for support of their products. One of the keys however, is a commitment to making stable drivers in addition to providing the appropriate amount of hardware and providing the needed engineering support to the right groups within Microsoft.
The whole point of beta testing is to test as much hardware with as much software to discover issues as soon as possible, which is why I the love Plugfests a lot. Right now, I think that it may be difficult to do using the Matrox cards, due to lack of drivers, in my opinion. Of course, I don't know as I am not in the Matrox beta program so who knows whats is going on there.
15) Rags then stated that "Most of the OpenGL problems I have seen in games on the Matrox forums were due to misconfiguration, or game patchese a utility that is not part of the driver package to configure the OpenGL drivers to get them to work as expected."
Well, I would assume from this statement that you are very familiar or work with Matrox technical support in order to have such detailed knowledge of all the OpenGL problems being reported to them. The posts you tend to see on these forums, more likely than not, do not reflect the majority of the issues reported directly to Matrox. Of course I as well have nothing to back that opinion up.
Unfortunaltely, I have yet to see the OpenGL compliance test results for the Matrox cards and I doubt I ever will. I also don't have access to the Windows OpenGL DDK, so unfortunately I can't run them myself to see how good or bad they really are.
I would assume that you would have access to such information to be making such a statement. Or is that simply your opinion of the quality of the drivers based on postings in some news groups and such?
Most people seem to have trouble distinguishing between opinions and informed opinions, because you have no idea of who you are talking to outside of the veracity of their postings.
Which is why people seem to get attacked for simply stating what seem to be rather trivial opinions.
To Geebee who stated "at least know what your talking about first. At least that way you won't appear to be a TROLL talking out his arse!"
Sdly, after examining this entire rather long thread, I failed to see where I was the one not knowing what I was talking about... but then, maybe I just missed it or something... huh...
Basically, I am left with DGhost accusing me be of being a warze kiddie or some sort, Rags pretty much accusing me of lying about the real OpenGL problems in the Nov 2 drivers and Greebe stated I did "know what I was talking about".
Well, it sure is nice to see that the "regular" posters around here are so friendly to new posters!
Anyways, thanks again for that tip ROM... the one on needing to use the Tweak Utility to get the 32 bit textures to work properly in OpenGL. Saved me from spending more time on this issue. BTW, what else does using the most accurate setting do? There doesn't seem to be any documentation that I could find on that option.
I guess I'll be sticking to the official Matrox forums from now on, seeing as how most of the people claim to "who know what they are talking about" don't seem to be all that high here... to me at least and most certainly only in my own opinion, of course.
Kruzin, sorry if this clutters up your board, but I really felt that I couldn't let those various accusations go unchallenged.
"The best opinions tend to be offered by those who are the best informed." -- Ono Tadaki
It all began innocently enought when...
1) Rags stated that Matrox had good driver development and support.
1.1) I then basically asked... well, if that really is the case, then where are Matrox Whistler drivers?
2) Rags then stated "Hehehe, you aren't too bright. Whistler isn't even in a public beta right now."
2.1) I then provided Rags with a <a href="http://msdn.microsoft.com/subscriptions/resources/subdwnld.asp">link</a> to the Microsoft MSDN download site which cleary shows that the Whistler Beta 1 is available to those MSDN subscribers who have the appropriate subscription level.
So I stated that it seemed, in my opinion, that Whistler has in fact been generally available to the Windows development community (or any member of the general public who has the appropriate MSDN subscription) since early November 2000. I'm also pretty sure that the MSND subscribers should have received the CD by now as well.
3) Rags then stated that "the key phrase in what you just said was "WIN2K". I don't see anywhere on the sight where they say these will work under whistler". And that he assumed members of the Matrox developers program would currently have access to Whistler drivers.
3.1) I then agreed, that yes indeed, the Windows 2000 drivers were only stated to work in Windows 2000. But, I also pointed out that it seems to me that other vendors didn't seem to have troubles in this area, in my opinion. I don't have access to Matrox beta driver program so I wouldn't know about that, but I haven't seen any developer drivers for Whistler at this point.
4) DGhost then said "If you are a *legit* microsoft beta tester, then you understand that companies don't release drivers for the platform until it has been released." (NOTE: I never stated anywhere if I was or was not a beta tester)
4.1) I than responded that yes, he was infact correct on that point. No vendor could claim Whislter support until RTM and WHQL have been run against the package. But, I wasn't asking for official released and offically supported drivers. Just any type of drivers at all that would provide more comprehensive suppor than the "in the box" Microsoft divers. I am fairly sure that preliminary drivers can be offered by hardware vendors through Micrsoft using the download site.
4.2) DGhost then accused me of being a warez kiddie or something when he stated "you are obviously not a legit tester." It should also be noted that I never said the 5.20 drivers didn't work with SMP systems, despite DGhost claiming "i have yet to experience any problems on my system with the drivers. And i was affected by the problem in the 5.1x series of drivers." But, one wonders, do they have the same OpenGL issues that affected the Win9x released at the same time? Well, from this posting it woudl seem that I was NOT alone with issues in OpenGL with the 5.20 drivers.
5) Rags then stated "Can you imagine all the tech support calls from Warez kiddies complaining the the latest drivers aren't working under build 21xx when they did under build 20xx?"
5.1) I then simply pointed to the Linux drivers which are offered by Matrox yet are not officially supported. Do these Linux drivers cause support nightmares for Matrox? I wouldn't thing so, becuase they aren't supported by Matrox and they are free to simply ignore the posts or simply filter them out. Again, they could restrict access to be done through Microsoft. My issue was not one of lack of support the Linux drivers, rather that there would most likely not be a techincal support calls arriving from Warez kiddies. Support for Whislter drivers would, supposedly to be done though Microsoft who, if needed will interface with the hardware vendor through the appropriate channels if needed.
6) DGhost stated "I think Matrox is faster to fix its problems than any other companies - fast to release the drivers is a different story. ."
6.1) I then provided SMP support as a counter example to that. I dunno, but is SMP not an high priority issue? Especially since it is essential to WHQL certification, at least the last time I checked it was.
7) DGhost that stated that "I think they test them far more rigidly than nvidia or 3dfx does."
7.1) I pointed to the OpenGL issue in the Nov 2 drivers as a counter example to the calimed lots of testing. How did that (in my opinion) rather serious issue slip through their testing then? It totally hosed BG2 and I don't recall see that as a known issue in the release notes from Matrox.
8) I then said others have different expectations since it should be expected that every one uses theie hardware differently.
8.1) Rags said that "drivers for running whistler isn't a reasonable expectation. It's not a feature, and it definitely is NOT a bug."
8.2) Kruzin chimed in with "You can't expect public drivers for an OS that is not final. We heard the same thing about 2k before it was released."
8.3) True, I guess. But then I guess Matrox will just have to suffer the lack of extensive beta testing of the hardware. Their and their customers loss I guess. Bit it sure doesn't seem to me at least that Matrox is fully behind Whistler.
But then I guess that I will also be told that is isn't realistic to expect "bug free" drivers will complete and equivalent support for what is available in Windows ME the day that Whislter is available through retail channels.
Fine, but that's just your opinion. On the one hand people tout that Matrox "have Linux support, OS/2 support, Win95 support, Win98 support, Win98 SE support, Win Me support, and Win2k support from matrox" but then state that support for Whistle isn't a desirable feature.
9) Rags then seemed to challenge me with the following query "OpenGL problems? Can you be more specific. You know, it sounds more and more like people come here and just shoot out OpenGL as an issue with Matrox drivers without really having any issue themselves, or just talking out of the wrong end."
9.1) I then pointed to a Dec 2 posting (which he even referenced earlier in the thread) on Matrox forums which confirms a know OpenGL bug with the Matrox Nov 02 drivers and shows that I was not in fact "talking out of the wrong end". This bug did in fact affect a lot more OpenGL applications than just Alice.
10) Rags then basically said that "As far as SMP goes, the SMP stability issue was one of the first fixes in driver releases and has been fixed for quite some time now."
10.1) I then stated 5.04 Apr 2000 was first stable SMP driver. And the latest good one was 5.20 Nov 2000 fully fixed SMP but seemed to have some OpenGL issues. I haven't tried the 5.31 Dec 22 drivers, but I would expect that they work fine as the Win9x ones do.
11) Rags then stated that "Sorry, but I believe it was the 5.14 drivers that included the SMP fixes, 5.20 was just a WHQL release that was probably in development and testing before the 5.14 drivers."
11.1) All I have to do is point to technical support posts stating to use the 5.04 drivers for SMP systems instead of 5.14 ones. As well, DGhost's own posting in the thread seeme to indciate that "And i was affected by the problem in the 5.1x series of drivers" in the thread.
Here Rags, just to keep you informed is a <a href="http://forum.matrox.com/mgaforum/Forum1/HTML/002987.html">acknowledgement</a> posted Oct 31 from Matrox of know SMP bugs in the 5.14 drivers.
12) You then said the fact that Matrox fixed the Alice bug (which they introduced in the Nov 02 drivers) in the Dec 22 drivers was proof of quick fixes.
12.1) I point out the fact that this bug shouldn't have been there in the first place, as the previous OpenGL drivers do not have these specific bugs.
13) Greebe then stepped in stating that "Alice, Baldur's Gate II, Oni all have updates coming out for them because the games were released before these issues were corrected."
13.1) I guess you are right, the issues were fixed in the Dec 22 drivers and all those games, and to the best of my knowledge IN EVERY APPLICATION available prior to Dec 21, has been released before the Dec 22 drivers. What's your point with that statement?
To the best of my knowledge, <a href="http://www.interplay.com/bgate2/">Baldur's Gate 2</a> shipped September 26. That is about a month prior to the Nov 02 drivers being released which broke Baldur's Gate 2, right?
The point is the Nov 2 drivers broke this game and several as yet unrelease games. But, going back to the previous drivers available immediately prior to the Dec 22 drivers, does fix the problems generated by the Nov 2 drivers in Alice and Oni. In the Dev 22 drivers these issues are fixed as well. Uninstalling and using a previosu driver is what the Matrox technical support people were telling customers who were reporting issues with OpenGL Nov 2 drivers to do in some cases when the proposed workarounds didn't work. Uninstalling the Nov 2 drivers and go back to the previous ones always seemed to fix the new issues created by the Nov 2 drivers.
Greebe then stated that "at least know what your talking about first. At least that way you won't appear to be a TROLL talking out his arse!"
14) I then stated that, it seemed logical to expect that Microsoft wasn't going to have their Windows 2000 OS support DualHead(tm) without Matrox making the effort to give them drivers EARLY in the OS development cycle.
14.1) Rags then said that "MS knew very well what was needed to allow dualhead/twinview/etc." and that Microsoft "chose to take the easier way out and stick to the more NT4 like setup in that regard."
14.2) Sorry, but unless you work or worked for Matrox and/or Microsoft I don't really know how you could be so well versed in this issue to know with absolute certainty that this is actually the case.
Well, how can you realistically expect Microsoft to support something if there are no drivers for them to attempt to test and develop their operating system with?
How would Microsoft even know that the limitation existed unless Matrox told them about it? Yet you assume that Microsoft well versed and toally familiar with the G400 card at that point?
Didn't Mircosoft state at one point that Windows 2000 SP1 would add support for DualHead? Didn't they then later change their position on that stating that it would not be all that feasible to do so? Obviously, this would seem to point to the fact that Microsoft did not in fact know the Matrox hardware all that well, in my opinion at least. I can't see why Microsoft would state that publically and then reverse their decisions stating that it would be too complicated to add? Perhaps, they overlooked something since they didn't fully understand who DualHead worked, or maybe they felt the effort to add it to Whistler was more important. Of course, considering when they started working on Whistler this woudln't be the least bit surprising to me.
You need to remember that (I think) Matrox is using a unique method on their card. It is always very important (in my opinion) that IHVs work closely with Micrsoft to push for support of their products. One of the keys however, is a commitment to making stable drivers in addition to providing the appropriate amount of hardware and providing the needed engineering support to the right groups within Microsoft.
The whole point of beta testing is to test as much hardware with as much software to discover issues as soon as possible, which is why I the love Plugfests a lot. Right now, I think that it may be difficult to do using the Matrox cards, due to lack of drivers, in my opinion. Of course, I don't know as I am not in the Matrox beta program so who knows whats is going on there.
15) Rags then stated that "Most of the OpenGL problems I have seen in games on the Matrox forums were due to misconfiguration, or game patchese a utility that is not part of the driver package to configure the OpenGL drivers to get them to work as expected."
Well, I would assume from this statement that you are very familiar or work with Matrox technical support in order to have such detailed knowledge of all the OpenGL problems being reported to them. The posts you tend to see on these forums, more likely than not, do not reflect the majority of the issues reported directly to Matrox. Of course I as well have nothing to back that opinion up.
Unfortunaltely, I have yet to see the OpenGL compliance test results for the Matrox cards and I doubt I ever will. I also don't have access to the Windows OpenGL DDK, so unfortunately I can't run them myself to see how good or bad they really are.
I would assume that you would have access to such information to be making such a statement. Or is that simply your opinion of the quality of the drivers based on postings in some news groups and such?
Most people seem to have trouble distinguishing between opinions and informed opinions, because you have no idea of who you are talking to outside of the veracity of their postings.
Which is why people seem to get attacked for simply stating what seem to be rather trivial opinions.
To Geebee who stated "at least know what your talking about first. At least that way you won't appear to be a TROLL talking out his arse!"
Sdly, after examining this entire rather long thread, I failed to see where I was the one not knowing what I was talking about... but then, maybe I just missed it or something... huh...
Basically, I am left with DGhost accusing me be of being a warze kiddie or some sort, Rags pretty much accusing me of lying about the real OpenGL problems in the Nov 2 drivers and Greebe stated I did "know what I was talking about".
Well, it sure is nice to see that the "regular" posters around here are so friendly to new posters!
Anyways, thanks again for that tip ROM... the one on needing to use the Tweak Utility to get the 32 bit textures to work properly in OpenGL. Saved me from spending more time on this issue. BTW, what else does using the most accurate setting do? There doesn't seem to be any documentation that I could find on that option.
I guess I'll be sticking to the official Matrox forums from now on, seeing as how most of the people claim to "who know what they are talking about" don't seem to be all that high here... to me at least and most certainly only in my own opinion, of course.
Kruzin, sorry if this clutters up your board, but I really felt that I couldn't let those various accusations go unchallenged.
"The best opinions tend to be offered by those who are the best informed." -- Ono Tadaki
Comment