Anyone else wondering if 6800 Ultra is really the 6800?

Dave, have you tried overclocking the card yet to see what's bottlenecking the card at the high end?

One would guess naively that it was bandwidth, but it could be the core gives a good contribution as well.
 
Joe DeFuria said:
I'm personally not expecting any major differences in pixel shaders. Vertex shaders maybe. For ATI this time around, I think they're really going to try and just be balls-out the fastest...which considering the good foundation of their R3xx core, isn't necessarily a bad thing.

I agree and disagree :)

Sure, the R3xx core is a very good foundation to build from. But i personally think it's pretty weak if they don't support SM3.0. After all, it's nothing like PS 1.4 which was pretty much defined by ATI and was also a defined after the hardware was released.

Just as i think it's weak that Nvidia doesn't support higher modes of MSAA, although i consider not supporting SM3.0 to be a much bigger problem.
 
Bjorn said:
Joe DeFuria said:
I'm personally not expecting any major differences in pixel shaders. Vertex shaders maybe. For ATI this time around, I think they're really going to try and just be balls-out the fastest...which considering the good foundation of their R3xx core, isn't necessarily a bad thing.

I agree and disagree :)

Sure, the R3xx core is a very good foundation to build from. But i personally think it's pretty weak if they don't support SM3.0. After all, it's nothing like PS 1.4 which was pretty much defined by ATI and was also a defined after the hardware was released.

Just as i think it's weak that Nvidia doesn't support higher modes of MSAA, although i consider not supporting SM3.0 to be a much bigger problem.

I agree as I said above ATI not supporting SM3.0 will be a huge marketing nightmare after how they bashed NVIDIA last generation on the PS 2.0 support. The situations are very comparable.
 
Bjorn said:
After all, it's nothing like PS 1.4 which was pretty much defined by ATI and was also a defined after the hardware was released.

Or Ps 1.1 which was pretty much defined by nVidia? Or 2.0 which was pretty much defined by ATI?

I honestly don't see ATI's non-support of PS 3.0 as that much of an issue. Obviously, it's better to have it than to not have it. It could be a significant issue if in addition to that, ATI's PS 2.0 performance isn't as good as nVidia's.

Just as i think it's weak that Nvidia doesn't support higher modes of MSAA, although i consider not supporting SM3.0 to be a much bigger problem.

Well, I see them as different problems. Practically speaking, One is a problem of not being better "out of the box." The other is potentially a problem of not being as "future proof" down the road. Each a problem in it's own way. I don't think either one is a bigger problem than the other marketing wise. They both have "AA" checkboxes, and they both have "DX9 support" checked off.
 
Stryyder said:
I agree as I said above ATI not supporting SM3.0 will be a huge marketing nightmare after how they bashed NVIDIA last generation on the PS 2.0 support. The situations are very comparable.

And I have to ask again...where did ATI bash nVidia on DX9 feature support, rather than performance?
 
surfhurleydude said:
digitalwanderer said:
Just wait 'til you see what the R420 can hit with low-k! 8)

After just quickly glancing thru 9600 XT reviews, I'm not very excited.

Remamber that the 5700U has external power and can draw more watts.
The RV360 only has the AGP for its power and it OC much better when you volt mod the core. The problem is you need a better cooler then the stock one to keep up with the heat.
 
Joe DeFuria said:
Stryyder said:
I agree as I said above ATI not supporting SM3.0 will be a huge marketing nightmare after how they bashed NVIDIA last generation on the PS 2.0 support. The situations are very comparable.

And I have to ask again...where did ATI bash nVidia on DX9 feature support, rather than performance?

Don't get me wrong I am an ATI fan but plenty of NVIDIA is not fully DX9 compliant was thrown around either by ATI or by proxy about the last generation. I don't think that splitting the above hair is going to matter much.
 
Stryyder said:
Bjorn said:
Joe DeFuria said:
I'm personally not expecting any major differences in pixel shaders. Vertex shaders maybe. For ATI this time around, I think they're really going to try and just be balls-out the fastest...which considering the good foundation of their R3xx core, isn't necessarily a bad thing.

I agree and disagree :)

Sure, the R3xx core is a very good foundation to build from. But i personally think it's pretty weak if they don't support SM3.0. After all, it's nothing like PS 1.4 which was pretty much defined by ATI and was also a defined after the hardware was released.

Just as i think it's weak that Nvidia doesn't support higher modes of MSAA, although i consider not supporting SM3.0 to be a much bigger problem.

I agree as I said above ATI not supporting SM3.0 will be a huge marketing nightmare after how they bashed NVIDIA last generation on the PS 2.0 support. The situations are very comparable.

Couldn't agree more. ATIs failure to support SM3.0 is IMHO really going to hurt. Now would be the ideal time to take avantage of CPU bound conditions and where as NVidia concentrated on bringing their architecture up to date albeit with perhaps slightly antiquated FSAA and AF I don't see ATI improving either FSAA, AF or bringing new features?
A speed bumped R360 just doesn't cut it to me as a gamer nor developer and with OEMs being fickle I don't see them amassing either. Surely the whole concept of 12-18 month product cycles was to ensure you could get both new features and performance?
Of course this all could be a moot point if ATI have an ace up their sleeves ;)
 
Joe DeFuria said:
Stryyder said:
I agree as I said above ATI not supporting SM3.0 will be a huge marketing nightmare after how they bashed NVIDIA last generation on the PS 2.0 support. The situations are very comparable.

And I have to ask again...where did ATI bash nVidia on DX9 feature support, rather than performance?

I seem to remember that the problem with no MRT support was brought up quite a lot with regards to Half Life 2.
 
Bjorn said:
I seem to remember that the problem with no MRT support was brought up quite a lot with regards to Half Life 2.

By ATI, or the developer? And isn't that particular feature now being touted by nVidia as well?

It appears that this more a vindication of Valve's stance on MRT, than it is ATI bashing nVidia for a feature ATI "claims" is significant.
 
I'm not going to make any conclusions on how PS3.x compliant the R42x will be until it arrives. Some rumors say it will be 32bit floating point, others say 24, some say one feature will be suppported, others say different ones are. Right now pretty much all we have is a bunch of unrealiable rumors.
 
Joe DeFuria said:
Or Ps 1.1 which was pretty much defined by nVidia? Or 2.0 which was pretty much defined by ATI?

But PS1.4 was added after the basic DX 8 (PS 1.1) standard was released. SM3.0 was added at the release of DX9 which was quite some time ago. The IHV's had a clear target for once.
 
I thought 2.0 was defined by the API with ATI input because NVIDIA tried to force their viewpoints and pulled out when no one wanted to play ball.
 
The list of games that will have PS 3.0 support this year is about 3 times longer than the current crop of PS 2.0 games.

From a developers point of view it seems like PS 3.0 is the way to go if you really want to get the best performance out of your shaders. The ability to simply disregard parts of the shader can save lots of resources and increase performance.

It will be interesting to see if these games that come out will show any actual increase in performance over PS 2.0 versions. If they do show a real increase in performance and developers pick up on it, ATI could be in a world of hurt on this.

It will be interesting to see the performance differences in the newest games. We got through the development cards(9800,5900) and now are onto the meat of the DX9 game.
 
Joe DeFuria said:
By ATI, or the developer? And isn't that particular feature now being touted by nVidia as well?

It appears that this more a vindication of Valve's stance on MRT, than it is ATI bashing nVidia for a feature ATI "claims" is significant.

That might be correct. But Valve and ATI has pretty strong bonds though.
 
Maintank said:
From a developers point of view it seems like PS 3.0 is the way to go if you really want to get the best performance out of your shaders. The ability to simply disregard parts of the shader can save lots of resources and increase performance.

I got the feeling that the branching in PS3.0 wouldn't necessarily be faster on the NV4X. More of a convenient thing for developers but care has to be taken to avoid slowing things down. VS3.0 branching should be faster though.
 
The real question is whether or not PS 2.0 will become a deflection point with game developers. I think nVidia certainly wants it to be, what with their NV3x's poor performance with 2.0 shaders and their support of the 3.0 model with their NV4x parts. But I think Intel's new integrated graphics and the market share of ATI's R3xx chips won't allow it. Regardless, it's a shame ATI doesn't support 3.0, just as it was a shame nVidia's FP performance has been so poor, pre-NV4x.
 
Bjorn said:
Maintank said:
From a developers point of view it seems like PS 3.0 is the way to go if you really want to get the best performance out of your shaders. The ability to simply disregard parts of the shader can save lots of resources and increase performance.

I got the feeling that the branching in PS3.0 wouldn't necessarily be faster on the NV4X. More of a convenient thing for developers but care has to be taken to avoid slowing things down. VS3.0 branching should be faster though.

Wouldn't PS3.0 have to be faster on NVIDIA if ATI is not supporting it?? Or am I totally confused.
 
Back
Top