What was that about Cg *Not* favoring Nvidia Hardware?
- Thread starter Hellbinder
- Start date
SpellSinger
Newcomer
It would be nice if Nvidia actual provided a schedule for PS1.4 support. Just saying they plan to add support is not enough. Plans change. Without 1.4 support Cg is not useful for development with ATI products. Are they just pulling another fast one? Why wouldn't Nvidia garner support from the other graphics vendors before announcing Cg? Don't answer this, cause I know the reason already.
Is there a specific issue you're referring to? I have not seen anymore issues with ATI's drivers not following specs than I have with Nvidia's. Why wouldn't ATI refer developer's to use a 1.4 shader, especially if it helped performance? Why do you keep making this comment without any data to back it up?
Is there a specific issue you're referring to? I have not seen anymore issues with ATI's drivers not following specs than I have with Nvidia's. Why wouldn't ATI refer developer's to use a 1.4 shader, especially if it helped performance? Why do you keep making this comment without any data to back it up?
SpellSinger said:
Just pulling another fast one? It's not like it was an official announcement. It looked very informal to me, which may mean that it was never supposed to be stated in the first place.
Well, there are a few. One is John Carmack's statements a while back (where he stated that nVidia's drivers are his "gold standard") where he stated that ATI's drivers don't always adhere properly to the spec. Derek Smart has also had a number of complaints, and there have been a few posts on these forums and others about other specific ideas (one is where the texm3x3vspec, if I remember the instruction correctly, doesn't work in the same way that nVidia's does, though it was never solved in that thread what the reference rasterizer did).
Valid points but how old now is JC's comment? A lot has changed over the past 8 months with respect to ATIs drivers. Then in one of these threads we had some Microsoft DX guys saying that ATI's drivers were on par with nV. Not sure what he ment but it shows that things are/have been changing....
Chalnoth said:
Hey, you almost got me worried there ealier on, Chalnoth!
The things you're mentioning is - as you know - a bit old news, so I would give the benefit of the doubt to ATI on the judgement whether they have improved drivers the last half year or not. Why are you sure that developers are still bitching a lot more over the ATI drivers of today compared to nVidia's (just asking)?
Anyway: I wouldn't be surprised if nVidia still have the edge here, but the question is whether the gap is still big enough to keep on making this 'point'.
It'll be interesting to see what version of Nvidia's card can emulate ps1.4. In d3d gf3/4 have four texture stages or samplers, perhaps internally the cards can have more, don't know. Ps1.4 can sample 6 textures so how they emulate this is unknown. If I were to guess I would say gffx could emulate it but gf3/4 won't.
Doomtrooper
Veteran
Chalnoth never includes the other part of JC's comment from the the plan...
ATI had been patiently pestering me about support for a few months, so last
month I finally took another stab at it. The standard OpenGL path worked
flawlessly, so I set about taking advantage of all the 8500 specific features.
As expected, I did run into more driver bugs, but ATI got me fixes rapidly,
The chip architecture has not really changed in two years including the Geforce 3 and 4 cards, its no wonder there is no driver bugs..will the 9700 have driver bugs in 6 more months?? Probably not.
ATI had been patiently pestering me about support for a few months, so last
month I finally took another stab at it. The standard OpenGL path worked
flawlessly, so I set about taking advantage of all the 8500 specific features.
As expected, I did run into more driver bugs, but ATI got me fixes rapidly,
The chip architecture has not really changed in two years including the Geforce 3 and 4 cards, its no wonder there is no driver bugs..will the 9700 have driver bugs in 6 more months?? Probably not.
SpellSinger
Newcomer
I have had the same experience with ATI. A few driver issues (a few bugs of my own though, which ATI helped to point out). ATI Developer Relations provided quick fixes for me and continues to stay in contact.
ATI has also helped to test my code with many cards that I don't have. Hell they even provided feedback on their testing with a GF3.
ATI has also helped to test my code with many cards that I don't have. Hell they even provided feedback on their testing with a GF3.
Doomtrooper
Veteran
That would be a hardware limitation IMO..otherwise it would have been fixed.
Doomtrooper said:
But I don't see why that's relevant. In particular, JC is a big-name in the gaming industry. Any company that doesn't fix those things quickly is going to have problems. Some others, such as Derek Smart, have complained that they have found blatant driver bugs where ATI has put fixes on the "back burner," so to speak (side note: yes, Derek Smart is arrogant and abrasive, but that doesn't necessarily make him wrong).
Doomtrooper said:
Which doesn't help things any. The problem is still there, regardless of cause. For example, the amount of time that the Morrowind z-buffer issues have been around make one thing that it's a hardware issue. But that doesn't matter to me. I like the game, so I want to play it on a video card that doesn't have such problems. Regardless of the cause, it's still incorrect rendering (apparently, in this case, a failure to have backwards-compatibility).
Chalnoth said:
Kinda like certain texture compression modes on a certain series of chips that's been present for years! now.
I just got back into Morrowind and I have no complaints on how it looks on my 9700 Pro (4xAA/8xAF at 1280x960). There is still a slight problem with water boundaries, but it's been greatly lessened compared to the first driver set. In fact, I have to actively look for it to notice it and, regardless, the game looks awesome with the above settings.
Doomtrooper
Veteran
Please don't refer to Morrowind issues on a 9700 Chalnoth, when you refuse to even take people sugestions by starting with a clean install...or trying newer drivers.
On your own home base forum...
Or Better yet..
http://www.motherboards.org/articlesd/hardware-reviews/1220_2.html
Video Card: All In Wonder 9700 Pro Review
Benjamin Sun
Mini-rant snipped out to avoid thread plunging into p*ssing match. JRR
On your own home base forum...
Or Better yet..
http://www.motherboards.org/articlesd/hardware-reviews/1220_2.html
Video Card: All In Wonder 9700 Pro Review
Benjamin Sun
Mini-rant snipped out to avoid thread plunging into p*ssing match. JRR
Doomtrooper said:
I have tried to check with each new driver release. Just checked with the CAT 3.0's, and the z-buffer issue still exists. Most of all, I have heard over at Rage3D semi-official confirmation that this is an issue known by ATI.
As a side note, the z-buffer errors do not appear everywhere, and it was actually a little while of playing before I spotted the problem. But, as it so turns out, one place that I tend to frequent in the game (Just outside the Imperial Shrine in Wolverine Hall, Sadrith Mora), the z-buffer issues show up especially well in the water (while walking down the stairs facing the water, and, perhaps even moreso, when leaping down the stairs facing the water), and the issues also show up when walking directly towards or away from one of the nearby guards (wearing basic imperial armor...).
There are also far-plane clipping problems, where on the GeForce4 the clipping plane appears to move smoothly as the character moves, on the Radeon 9700 pieces of objects sort of leap into and out of view (Usually trees and similar objects).
And I really do not believe this can possibly be an issue related to not doing a fresh installation. Stuttering, possibly. Instability, possibly. But this? I don't think so.
As for Rogo, I'll have to talk to him to make sure he's not seeing the problems I am.
Lastly, I'll reiterate what I've stated previously. I'm not willing to do a complete reinstall at the juncture. I have work to do in Linux, and I currently need to use my GeForce4 for that. I'm not going to do a fresh installation of Windows only to have to switch back to my GeForce4 not long afterwards.
John Reynolds said:
Yes, it's exactly like that problem, with one crucial difference. It can be fixed with user-side workarounds. Granted, this shouldn't have to happen either, but it's still better than an issue that just has to be dealt with by the user. Anyway, you're right in a way. I couldn't have anything to complain about on the Radeon 9700's "correctness in rendering" if this was the only problem. The thing is, it isn't. There is still the problem of anisotropic forcing bilinear/trilinear on all textures (blurring text in some games), and improper rendering in the Tenebrae mod for Quake.
Well, I think the most noticeable is the z-buffer errors on the imperial guards (and some other characters). As a side note, these problems don't seem apparent in the Tribunal expansion pack (disclaimer: I only stayed in Mournhold for a few minutes). This may be because the expansion pack doesn't appear to exist "outdoors." That is, there is no outside terrain rendered. The z-buffer errors in Morrowind only seem to appear when outdoors.
Chalnoth said:
And Nvidia shouldn't have 'broken' AF + MT with their GF4s, making it (AF) basically unusable in a lot of newer games. Things happen, no product is perfect. But reaching for a user-coded mod of a game that's now 6.5 years old is a bit ridiculous, don't you think?
I've been playing outdoors and haven't noticed any problems with models yet. Play NOLF1 on an Nvidia card lately? The menu has some nasty corruption in it. Try HoMM 4 when it first came out last spring? Crashed to the desktop with the then-latest WHQL Detonators. Ever notice the LOD changing in various Detonator releases? I sure have, and the increased texture aliasing (NWN really showed this). I could go right on nit-picking, but for the most part my GF3/GF4s worked with most games and gave good performance and pretty good IQ. The 9700 of course obliterates them, but that's only to be expected since it is a next-gen. product. It's not perfect, but when viewed fairly and in the large picture it's a damn fine piece of gaming hardware.
DT, that the r300 has z-buffer issues with Morrowind is a known problem that ATI, through sireric, has stated they are working on. In reality, Morrowind uses w-buffering, and the r300 does not have hw support for it. Hence the flickering textures, funky water, and overlapping paperdoll textures.
