• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

DX10 ATI vs NVidia gaming benchmarks

Then read about what Nvidia were unable to do in time for the 8800 release, and WHY they use the hardware AA :) Interesting reading, and shows clearly why AMD get spanked (in general) when using AA on the 2x00 range. Basically Nvidia asked to have the dx10 spec changed as they couldn't/didn't develop good enough for proper spec, which then screwed AMD with their model. I think Microsoft allowed it as they NEEDED dx10 in order to help sell Vista, hence why they changed it for Nvidia. That is just my opinion however.

I suppose it means nothing now as the spec is the way it is, and nothing can be done about it. Nvidia will 'probably' be doing AA the same way with their next gen of 10.1 cards. Just know there is a reason why the 2x00 gets caned with AA on.

This doesn't however, make up in any way for the AMD/ATI failure to bring a product to market that can match the GTX.

EDIT: What I make of that review/test is that it there is still a lot to learn both from ATI and Nvidia on drivers optimistation with DX10, and that the current dx10 mid-range hardware is only just coping. High-end (gtx ulta and uncertain about the 2900 1gb but prob mid or mid-high anyway), stuff does better but will be quickly replaced with the new stuff.

Better review than the lame selective one done with knights of the sea or whatever it was called.
Matthew
 
Last edited:
Meh, big deal, all crap DX10 titles engineered towards nvidia hardware, ill reserve judgment until i see what the performance of crysis is like. ATI will push the boat out for performance in the drivers for crysis big style.
 
Cyber-Mav said:
have to agree with willhub here, those benchmarks don;t look good for ati at all, and they even used the nvidia 162.22 driver which is not the latest one to help ati a bit in the tests. not good at all.
Click to expand...

Obviously AMD have no high-end, but mid range was pretty close from what I see Mav. Company of heroes shows 8800 gts hardware getting a spanking. What are your thoughts on this Mav? Sarcasm: Clearly I should steer clear of Nvidia 8800 gts hardware now.

Raven> Was under the impression that Crysis was biased towards Nvidia hardware, as they haven't demo'd or mentioned on anything other than Nvidia? I may be wrong as I try to stay clear of Crysis hype.

Matthew
 
Scougar said:
Obviously AMD have no high-end, but mid range was pretty close from what I see Mav. Company of heroes shows 8800 gts hardware getting a spanking. What are your thoughts on this Mav? Sarcasm: Clearly I should steer clear of Nvidia 8800 gts hardware now.

Raven> Was under the impression that Crysis was biased towards Nvidia hardware, as they haven't demo'd or mentioned on anything other than Nvidia? I may be wrong as I try to stay clear of Crysis hype.

Matthew
Click to expand...

they not using the latest drivers for the nvidia cards, they using 162 driver not the newer 163 driver.
 
Cyber-Mav said:
they not using the latest drivers for the nvidia cards, they using 162 driver not the newer 163 driver.
Click to expand...

I see. When AMD get a unusually low score, all hell breaks loose and it's the hardware to blame, but when it happens to Nvidia, it's the drivers?

The benches look pretty even to me, just with a severe lack of high end from AMD. Well done Nvidia for raising the bar with the GTX and Ultra :cool:

Matthew
 
Scougar said:
I see. When AMD get a unusually low score, all hell breaks loose and it's the hardware to blame, but when it happens to Nvidia, it's the drivers?

The benches look pretty even to me, just with a severe lack of high end from AMD. Well done Nvidia for raising the bar with the GTX and Ultra :cool:

Matthew
Click to expand...

Well the latest drivers are supposed to provide a performance boost compared to the ones TH used, so it stands to reason that NV would be even further ahead.

I'm just discraced by the performance of the 2900, I'll probs keep it untill christmas, sell it and put it with my parents and get an 8800GTX for christmas.
 
this is not the first time THG has done this, thier last review used the nvidia 162 drivers and catalyst 7.6
dunno what they playing at to be honest.
 
The 2900 seems to suffer most when AA is enabled.

Shame they didn't run all the benchmarks with and without AA as some of use never use it, would have been good to see these results

AD
 
Who knows, I think its fair to say there is work to be done on both sides with the drivers. DX10 drivers really need work on, but I think both Nvidia and AMD know that, with DX10 compatability, features and performance slipping on both sides; however as DX10 becomes more common, I'd imagine it'll shift further and further up thier priority list, given that most of thier userbase is on XP, not Vista right now, Im not entirely surprised DX9 still seems to be thier current main focus. THG I use for graphics comparison charts and thats about it really, they produce some good bumph now and then.

Glad to see that Nvidia is finally starting to pickup the ball on drivers, its a shame the release of the HD2900XT was needed to prompt it though, thier previous driver release record was pretty dismal, whereas at least ATI are holding strong in that arena.

At the moment, given the cash involved for buying these cards, everything around pretty much shows it to be swings and roundabouts, so Im not taking any particular sides, I've seen the 8800GTS and 2900XT in action, and both are good cards, just doing things slightly differently to each other. Drivers may or may not improve for both parties, but given past record, performance for all the current cards should keep rising for now at least :)
 
Last edited:
Scougar said:
Then read about what Nvidia were unable to do in time for the 8800 release, and WHY they use the hardware AA :) Interesting reading, and shows clearly why AMD get spanked (in general) when using AA on the 2x00 range. Basically Nvidia asked to have the dx10 spec changed as they couldn't/didn't develop good enough for proper spec, which then screwed AMD with their model. I think Microsoft allowed it as they NEEDED dx10 in order to help sell Vista, hence why they changed it for Nvidia. That is just my opinion however.
Click to expand...

Perhaps you could provide links for this? As far as I'm aware, nvidia use hardware AA because it's faster, not because they're missing any functionality. I suspect that what you are alluding to is ATI's use of shader based AA in place of hardware AA. This was ATI's choice. It was not mandated by DX10. As far as I know nvidia does shader AA just fine too. It's a standard part of DX10. Nvidia chose to support both software and hardware AA. ATI only do software.

I know that nvidia complained about Call of Juarez using shader AA, but I believe that they actually run it just fine and benchmarks suggest that performance isn't an issue. I think their complaint was simply that it would run even faster with hardware AA (although it's a separate argument whether this would have affected image quality).
 
Tom|Nbk said:
Yawn at another of these threads, yawn at the same arguments that have been going on for months, yawn at your constant moaning. :o
Click to expand...

Dont post then, I was not moaning tbh, I seen the benchmark, I said what I thought, not a problem with that, then someone has to complain about me saying what I want, then as usual you come into the thread and start.
 
i cant believe how can hd2900 be so behind gtx :( thats insane im sad edit: maybe its just software drivers ?
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom