More ATI Driver News from Derek Smart
- Thread starter antlers
- Start date
- Status
Sabastian said:I am saying it is a possibility.....But for some reason that "possibility" is dismised imeadiately??!! Whatever...
Everything's a possibility Sabastian. But let's face facts, if you and Doomtrooper try and move the conversation in that direction (that there is in fact no problem and DS is lying his ass off) then this thread might as well be locked right now - it will degenerate into the usual VoodooExtreme flame war.
Mr Smart, your various posts probably mirror many developers.... except that they are more diplomatic and know they probably shouldn't post their problems in public forums, and, instead, keep a level head and sort things out via emails and phone calls with various IHVs.
Personally, I'd prefer the other developers's approach. Developer relations problems and driver problems (yes, they always exist) aren't things I'd want to hear of... they are the very things that helps to continually improve drivers, to the state where they can be released.
You don't want to hear about how I convinced Tim Sweeney to bitch about cubemap support in the GF3, do you? There are things best kept private, for the benefit of the audience that you appear to only care about, your fans/gamers.
Personally, I'd prefer the other developers's approach. Developer relations problems and driver problems (yes, they always exist) aren't things I'd want to hear of... they are the very things that helps to continually improve drivers, to the state where they can be released.
You don't want to hear about how I convinced Tim Sweeney to bitch about cubemap support in the GF3, do you? There are things best kept private, for the benefit of the audience that you appear to only care about, your fans/gamers.
Bigus Dickus
Regular
What I'd like to know is how one minor game developer's discovery of driver issues with his unique ways of using new hardware makes the drivers TERRIBLE. Or did he put TERRIBLE in red (I can't remember, everything seems so drama-queen like)?
If the drivers show good stability, compatibility, and performance with the vast majority of game titles, all with the largest hardware deviation from previous generation seen in years, I'd say that makes them pretty damn solid drivers.
If they worked with every obscure code path some egotistical developer decides to use with no fixes required, I'd say that would make them magical.
Am I surprised that ATI's driver developers aren't magicians? No, why should I be?
Apparently, in Mr. Smart's universe (of which is is undoubtedly the center), his code is the only valid criteria for judging drivers.
[edit] I'm still struggling with what W-buffer or 32 bit Z-buffer support has to do with the driver developers... isn't that a hardware design decision? [/edit]
If the drivers show good stability, compatibility, and performance with the vast majority of game titles, all with the largest hardware deviation from previous generation seen in years, I'd say that makes them pretty damn solid drivers.
If they worked with every obscure code path some egotistical developer decides to use with no fixes required, I'd say that would make them magical.
Am I surprised that ATI's driver developers aren't magicians? No, why should I be?
Apparently, in Mr. Smart's universe (of which is is undoubtedly the center), his code is the only valid criteria for judging drivers.
[edit] I'm still struggling with what W-buffer or 32 bit Z-buffer support has to do with the driver developers... isn't that a hardware design decision? [/edit]
Galilee said:
I am not saying ATIs drivers are "perfect" but they certainly are not terrible which is mr smarts conclusion. AFAIK they are not "terrible" nor are they even close to being "terrible". I will make a note Galilee.... "hrm ATIs drivers are not perfect" but IMO there is a large discrepency between "terrible" and "not perfect".
Galilee said:
I don't think there's anyone here who would insist the 9700 drivers are 100% problem free. Heck, any of us who frequent Rage3D's forums know they aren't as there are a few very helpful ATI driver folks over there collecting information and keeping us informed about the fixes they are making. However, that does not invalidate Doomtrooper's point here. I don't think I've seen Doom insist that multi-texturing is working just fine. However, given that there ARE other games out there that are doing multitexturing just fine in Direct3D, this would lead one to believe that Derek is doing -something- different since he's getting a different outcome. But he has so far Mr. Smart has not responded to the question of what he's doing different from other devs out there that's causing this. It may very well be a hard-to-find problem in ATI's D3D code that's at fault, and it just hasn't been found because other devs aren't doing things that way and not exposing the buggy code. It may also be that Mr. Smart has put in a broken implementation of multitexturing and his game is at fault. Whether it works on other cards or not is irrelevant, drivers can compensate for faulty code just like code can compensate for faulty drivers. But without some more information about the nature of his code, how are we to draw any conclusions?
Doomtrooper
Veteran
Gerry said:Joe DeFuria said:
Actually I think you'll find Doomtrooper, perhaps tongue in cheek, was looking to do exactly that.
Do you have a Radeon 9700 ?? IF so test Serious Sam or any other Multitexturing game for that matter and report back any issues...otherwise my point is valid.
He states the drivers are TERRIBLE...well the funny thing is that every game thrown at the 9700 runs PERFECT, the only game that us grief was the buggy DEMO of BF 1942.
If there is a bug in Multexturing it is not affecting current games besides his, so ATI stated they will fix it..what else do you want them to do..write the game code for him too.
Doomtrooper said:If there is a bug in Multexturing it is not affecting current games besides his, so ATI stated they will fix it..what else do you want them to do..write the game code for him too.
LOL, mayhap that is what he wants... then he can turn around and sell the patch..
hehe Oh dats bad... fitting though.
Joe said:
sure, that's how adults are meant to communicate (i was tempted to ask you to clarify your 'Yup. And what does that have to do with the price of tea in China?' comment but i'll pass on it)
You are telling me that consumers make it difficult for hardware and software IHVs alike. Big news there.
no. i'm telling you that you're paying extra for something which could have costed you $n, but due to unforseen obstacles (broken api features fall exactly in this category as seen from a developer's perspective) eventually costed you $(n+m), where m > 0.
now, if you want to consider 'a broken api feature' as being just as much of an act of god as is, say, the accident the lead coder of this title had where he broke his leg and had to spend 1 month in hospital, which eventually stacked up to the overall title costs, or you wouldn't consider it as such, is beyond my scope.
i was not telling you anything, i was just asking you if you were aware of that fact that an unforseen game-project obstacle such as a broken api feature costed you. see, i even didn't ask you if you were fine with that.
YeuEmMaiMai
Regular
Mr Smart,
Hey are you talking our of your a$$ again? I have a 8500 and I can tell you that I WORKS AS ADVERTISED for what games I play. No body cares about your constant complaining and no one really cares to play your games because they
a. suck graphics wise
b. no interesting story line
c. game play...do you even KNOW what that is?
DN manhattan project is much more fun than any game you wrote...wait a minute, jazz jack rabbit is much more fun as well.
Who do you think you are John Carmack? hahahhah dream on sucker...
Hey are you talking our of your a$$ again? I have a 8500 and I can tell you that I WORKS AS ADVERTISED for what games I play. No body cares about your constant complaining and no one really cares to play your games because they
a. suck graphics wise
b. no interesting story line
c. game play...do you even KNOW what that is?
DN manhattan project is much more fun than any game you wrote...wait a minute, jazz jack rabbit is much more fun as well.
Who do you think you are John Carmack? hahahhah dream on sucker...
Doomtrooper said:Do you have a Radeon 9700 ?? IF so test Serious Sam or any other Multitexturing game for that matter and report back any issues...otherwise my point is valid.
He states the drivers are TERRIBLE...well the funny thing is that every game thrown at the 9700 runs PERFECT, the only game that us grief was the buggy DEMO of BF 1942.
If there is a bug in Multexturing it is not affecting current games besides his, so ATI stated they will fix it..what else do you want them to do..write the game code for him too.
Hmmm. Are you delibarately missunderstanding me? I've no problem with people taking issue with DS calling the drivers terrible. As other people have stated from a consumer point of view they clearly aren't, and from the looks of it the majority of developers are happy as well. Clearly however DS does raise (in his own abrasive way) issues why cannot be denied. You were beginning to go down the road of claiming this is all a figment of his imagination, which isn't exactly helpful.
Yes, he's probably doing something other developers don't do - I doubt whether he cares. Yes, ATI sounds like they're fixing things for him - so you agree there was something to be fixed?
From HIS standpoint it way well be that the drivers ARE terrible - if you're not looking to play any BC games then you're not likely to give a frig. I don't understand why you're getting so defensive about this all.
Joe DeFuria
Legend
Yes, I am aware of that. And though you didn't ask, I'm not "fine" with it, but I do accept it as a reality. Just like death and taxes.
Are you also aware that consumers pay also pay extra when one IHV writes "loose" driver code that is "forgiving" of certain coding errors?
darkblu said:Sabastian said:
heh, you can what?
AFAIK comes from reading dozens of reviews. Granted some more indepth then others. None mention MT problems....none. There really is little evidance to substaintiate his claims. So tell me how AFAIK disqualifies my post? All I am saying is that show me where else others are having difficulties with MT...what is so hard to grasp about that darkblu? Countless others have no problems but when some halfass coder says he has a problem then all of a sudden it is undisputable evidance in some strange way that ATI has "terrible" drivers. Get a grip man.
Doomtrooper
Veteran
Gerry said:
Nobody is getting defensive, you insinuated that I think Derek Smart is mistaken or defending ATI's drivers.
Well I'm giving you a consumer level response...all current games that support Multitexturing are not exhibiting any issues...if its a driver bug it belongs with the DEV team..not whining all over the damn internet.
Joe DeFuria said:
yes, joe, i am. but that IHV has all rights to say 'up yours' to the respective developer who made those coding errors. and the rest of the devs would only be happy about such a response on this account. and trust me, IHV do tend to give such responses
point being, when an IHV says 'yes, we'll fix that' it's usually that:a) they actually did screw it up
b) for some reason they considered that particular developer of special value to them, to the point they'd please him at the cost of the creating troubles for the rest devs.
now, to get more on topic, do you believe ati acted in the b) manner in derek's case?
- Status