no 6800ultra Extreme, still the same 6800ultra

[BRiT]

Uhm, can I find that just a tad hypocritical sounding considering that nVidia is just now giving SM 2.0 any form of realistic support this generation? :?

Well you do know that SM2 was never ever the target that Nvidia was going for; it was always targetting SM3. At least that's what the Nvidia appologists are saying. It seems it took them 4 hardware cycles (NV30, NV35, NV38, NV40) to get it right.

And even then we're not sure how well it turned out. They still have significant issues running SM3 technology demos. Perhaps one day they'll have decent enough drivers to run them.

 

[KimB]

Uhm, can I find that just a tad hypocritical sounding considering that nVidia is just now giving SM 2.0 any form of realistic support this generation? :?

No, digi, I don't. They had better than SM2 support. It was just a poor implementation. ATI decided to not support SM3. nVidia failed to support SM2 at high speed. I think there's a huge difference. At least nVidia has been working to fix that mistake.

 

[BRiT]

No, digi, I don't. They had better than SM2 support. It was just a poor implementation. ATI decided to not support SM3. nVidia failed to support SM2 at high speed. I think there's a huge difference. At least nVidia has been working to fix that mistake.

So you're saying you would have prefered that ATI should have made the same mistake with the R420 as Nvidia did with the NV3x?

 

[anaqer]

They had better than SM2 support. It was just a poor implementation.

That's crazy. Every feature is only as good as its actual implementation.

 

[Dave Baumann]

Personally I think the argument is somewhat pointless - we're not at a point where PS2.0 is the widespread development starting point yet, so I don't think it going to be the case that ATI not supporting it at this point in time is going to have any effect what-so-ever as developer have the option to develop SM3.0 features / effects into their titles now and by the time there is a widespread uptake of SM3.0 or greater the hardware will have evolved through several iterations. It does appear to be the case, though, that NVIDIA's performance has hurt the uptake of SM2.0, which is a shame because there is actually a widespread consensus of hardware support for it.

However, the biggest driver in game feature useage / effects will be the next generation consoles, not whether ATI or NVIDIA are trading places with SM2.0/3.0/4.0 etc.

 

[AlphaWolf]

However, the biggest driver in game feature useage / effects will be the next generation consoles, not whether ATI or NVIDIA are trading places with SM2.0/3.0/4.0 etc.

Unfortunate, but true.

 

[KimB]

Personally I think the argument is somewhat pointless - we're not at a point where PS2.0 is the widespread development starting point yet, so I don't think it going to be the case that ATI not supporting it at this point in time is going to have any effect what-so-ever as developer have the option to develop SM3.0 features / effects into their titles now and by the time there is a widespread uptake of SM3.0 or greater the hardware will have evolved through several iterations.

That's not so much the point as this idea:
If ATI succeeds by allowing nVidia to be the first to put out new technology, and thereby allowing themselves to have the highest-performing graphics card, why would ATI ever want to come first in the technology race again? And furthermore, why would nVidia want to continue to increase the technology of their products? It's performance in current games that matters anyway, right?

I just think it's dangerous and underhanded of ATI to attempt this.

 

[BRiT]

I just think it's dangerous and underhanded of ATI to attempt this.

Did you also think it was dangerous and underhanded for Nvidia to be doing exactly this up to the NV40?

 

[Dave Baumann]

Chalnoth, you appear to be being petulently stupid in your comments here. There is nothing dangerous or underhanded - would would be dangerous is to produce a part that you believe to be borderline profitable across the entire range, thus jeoparising your business.

ATI succeeded in being the first to put out shader 2.0 parts in the last generation, just because they haven't done so now doesn't mean they won't do it in the future. The drivers for SM3.0 right now are not as compelling seeing as there is still some considerable discorvery still to be done on the development side of things with SM2.0 yet -- I mean, be realisitc, we've not even scratched the surface yet -- and there were no new silicon options available to do it.

The situations will change as we go down the line and different requirements will drive different decisions - as I mentioned, ATI are going to be behind one of the the key next generation consoles that will drive technology ahead and they are going to want to make damned sure they are going to be able to tap into that with the PC products as soon as they can.

 

[KimB]

Chalnoth, you appear to be being petulently stupid in your comments here. There is nothing dangerous or underhanded - would would be dangerous is to produce a part that you believe to be borderline profitable across the entire range, thus jeoparising your business.

You're apparently totally misunderstanding my argument. It's obviously not dangerous for ATI's business in the short-term. It's dangerous for technological advancement in the future. It's underhanded of ATI to support short-term gain for to our (potential) long-term detriment.

 

[KimB]

Did you also think it was dangerous and underhanded for Nvidia to be doing exactly this up to the NV40?

Um, they weren't. Every single one of nVidia's new architectures (since the TNT) has been a technology leader upon release. Most of them have also been performance leaders.

 

[Eronarn]

Um, they weren't. Every single one of nVidia's new architectures (since the TNT) has been a technology leader upon release. Most of them have also been performance leaders.

Well, there's that whole FX thing, you know.

 

[KimB]

Um, they weren't. Every single one of nVidia's new architectures (since the TNT) has been a technology leader upon release. Most of them have also been performance leaders.

Well, there's that whole FX thing, you know.

Which was a technology leader upon release. It had problems, unfortunately, that prevented it from performing as well as nVidia wanted, or as well as we hoped, but those problems have been fixed with the NV40.

 

[Dave Baumann]

You're apparently totally misunderstanding my argument. It's obviously not dangerous for ATI's business in the short-term. It's dangerous for technological advancement in the future. It's underhanded of ATI to support short-term gain for to our (potential) long-term detriment.

No, I understand exactly what you are saying, and I think you are being very stupid about it all.

Long term, its going to make very little difference whether one doesn't support SM3.0 right now - it will have a devasting effect on the advancement of 3D graphics if one or the other goes under though.

You are also taking this one case as some kind of marker that this is always going to happen, when there is no basis for assuming this will be the case - they have both leapfrogged each other in features and capabilities throughout various generations and there is zero reason to assume this will not continue to be the case now.

Again, its also not the case that offering more performance won't advance games anyway since it will encourage developers to take up a greater useage of features/capabilites within larger sections of the game.

 

[KimB]

You are also taking this one case as some kind of marker that this is always going to happen, when there is no basis for assuming this will be the case - they have both leapfrogged each other in features and capabilities throughout various generations and there is zero reason to assume this will not continue to be the case now.

That's just the point. If the leapfrogging shifts to performance at the expense of capabilities, which could well happen if ATI "wins" this generation, then we have a problem.

 

[Dave Baumann]

There is no precident for that though - GF3 to GF4 was only a speed enhancment, while ATI has intersperced that with a more advanced shader model; was that a precident for it happening then?

 

[AlphaWolf]

You are also taking this one case as some kind of marker that this is always going to happen, when there is no basis for assuming this will be the case - they have both leapfrogged each other in features and capabilities throughout various generations and there is zero reason to assume this will not continue to be the case now.

That's just the point. If the leapfrogging shifts to performance at the expense of capabilities, which could well happen if ATI "wins" this generation, then we have a problem.

The gf4 created a problem?

 

[Tahir2]

For the longest time NVIDIA only supported SM1.3 and ATI supported SM1.4.

In the long term it did not matter - however even in the short term having the SM1.4 did not help ATI to surpass NVIDIA...

Now we have a situation that is analagous to the above mentioned... and again in the long term it does not matter.

ATI is planning on a SM3.0 featureset from what I have heard. Unfortunately I was not able to confirm whether this was SM3.0 on its own or combined with SM4.0 for their next generation technology.

In the end though as long as both NVIDIA and ATI continue their advancement in featureset and are basically inline with MS DX plans it doesn't matter - just as long as people keep buying these new cards.

 

[Tahir2]

Oh and by initial indications from where I work we have 10's of X800 Pro orders and IIRC zero Geforce 6800 orders...which means Chalnoth that your worst nightmares are likely to become reality... a lesser featured card will dominate the market (again).

 

[KimB]

For the longest time NVIDIA only supported SM1.3 and ATI supported SM1.4.

I think that was a completely different situation. ATI has released a new architecture without (many) new features. This is really the first time this has happened since 3dfx and the Voodoo3.

For every generation until the R300, nVidia was the first-adopter of a new featureset. They were also the performance leader (starting with the GF 256). Now ATI is trying to get ahead by skimping on features, by rehashing an architecture that is nearly two years old.

And SM3 is not an intermediate step between generations. It's the basis for the current generation of graphics programming interface, and will be essentially the standard for the next 1.5-2 years.