HD 2900XTX Benches

[Pugnate]

Originally posted by: Wreckage
NOVEMBER 2006....

Computer Geek1: Should I buy a G80?
Computer Geek2: Nah, just wait a little bit and the R600 will be out.

hahaha... dude that's what happened with me, except I went ahead and got the 8800GTX. I thought,"Screw it. I want to play some games."

 

[5150Joker]

Originally posted by: jiffylube1024


As an aside, it seems that in general the gaming industry is in a lull, as development teams are all working on "next-gen" looking games, which means even longer development times than before...

Not to mention an increase in console exclusive (or cross-platform) titles, such as Gears of War and many more Unreal 3 engine games, which appear to be in development for Xbox360, PS3 (in many cases) as well as PC. Ever since Microsoft introduced the original Xbox, it has stolen a lot of PC gaming's thunder, and the mockery that has become the expensive PC graphics card race has really turned a lot of people off (myself included). [though MS has their own problems with the ****** life expectancy of Xbox360 systems, and PS3 reliability is still to be determined]

The only thing that drives me back to PC games is the kb/mouse interface. If Sony and MS truly embraced kb/mouse for all their FPS games, I wouldn't touch the PC again apart from owning a laptop for Office.

 

[Wreckage]

The 8800GTX can beat a 1950XTX crossfire setup.

Maybe the GTX was just a swift kick to the nuts that no one saw coming.

I think ATI was just side swiped by the GTX release, getting bought out by AMD and the royal pain-in-the-ass it is creating drivers for Vista.

They were probably all set to go, on schedule. Then the above "perfect storm" hit and they had to race back to the drawing board.

Not to mention that the Core 2 Duo was a swift kick to AMD's big green sack. So their focus is split right now.

This year is pretty much going to suck for AMD. Next year? Who knows. The tech industry moves so fast that they could be right back on top again with-in another generation or so.

 

[UnTech]

Originally posted by: Noema

Why are you happy? A strong ATI card is beneficial to the consumers, even if you don't ever buy an ATI card in your life.

No reason for nVidia to drop prices on the GTX, I reckon

[/quote]

I was going to say maybe Bitboys can come to the rescue.

Then I learned ATI bought them and fully integrated them into their company. Heck! I'm surprised the R600 is even coming out after learning that.

 

[vadp]

Originally posted by: jiffylube1024
Sigh, if these performances benchmarks are true

They're not.
The sites are full of BS, Speculation and Disinformation at the moment.
Everyone who says otherwise is full of it.
Wait for a final product to make a conclusion.
AMD is making an awfully big deal with all those NDA's.
I refuse to believe that do all that just to hide a miserably performing product.

 

[swtethan]

Originally posted by: vadp

Originally posted by: jiffylube1024
Sigh, if these performances benchmarks are true

They're not.
The sites are full of BS, Speculation and Disinformation at the moment.
Everyone who says otherwise is full of it.
Wait for a final product to make a conclusion.
AMD is making an awfully big deal with all those NDA's.
I refuse to believe that do all that just to hide a miserably performing product.

could you link me to something that was reviewed in the release then actually performed much better when it was mainstream?

 

[yacoub]

BREAKING NEWS: This just in...

In other news, a little girl's My First Hairdryer, first reported stolen nearly a year ago, turned up today attached to a flagship videocard. No charges have been filed yet, though authorities are investigating Vijay Sharma as a "Person of Interest".

The following image of the device was captured during the police raid:
http://images.dailytech.com/nimage/4558_r600xtx_1.jpg

 

[vadp]

Originally posted by: Cookie Monster
Hmm, wait for the real reviews to decide.

However to add more insult to injury, the 8800 ultra will release (a G80 on steroids) and that would create even a bigger gap in comparison with the DTs leaked benches.

The 8800 is just an insanely overclocked GTX with an even bigger Heatsink, plenty of thermal issues, additional noise and huge price tag with miniscule production numbers.
The 8800 Ultra is just for publicity and bragging claims.

 

[swtethan]

Originally posted by: vadp

They're not.
The sites are full of BS, Speculation and Disinformation at the moment.
Everyone who says otherwise is full of it.
Wait for a final product to make a conclusion.
AMD is making an awfully big deal with all those NDA's.
I refuse to believe that do all that just to hide a miserably performing product.



The 8800 is just an insanely overclocked GTX with an even bigger Heatsink, plenty of thermal issues, additional noise and huge price tag with miniscule production numbers.
The 8800 Ultra is just for publicity and bragging claims.

gee, someone isnt an ati fanboy

 

[vadp]

In addition, all these engineering samples are being presented like it is a real thing!
We don't know that and AMD goes out of its way to restrict any flow of info.

 

[vadp]

That someone runs three Nvidia cards at the moment (and non ATIs)

 

[Creig]

The more I think about it, the more I'm convinced that something just isn't right with numbers DT is providing. Saying that ATI is releasing a new card that can't keep up with an 8800GTX is one thing. Saying that ATI is releasing a new card that is barely faster than their previous generation champ (X1950XTX), even when using a quad core processor, is another.

Company of heroes 1280*1024:
1950 XTX 99 (tweaktown) E6600 (dual core)
2900 XTX 97 (dailytech) QX6800 (quad core)

Company of heroes 1600*1200:
1950 XTX 70 (tweaktown) E6600 (dual core)
2900 XTX 73 (dailytech) QX6800 (quad core)


FEAR 1280*1024:
1950 XTX 80 (tomshardware) No softshadow 4AA 8AF , X6800 (dual core)
2900 XTX 84 (dailytech) with softshadow 4AA 16AF , QX6800 (quad core)

FEAR 1600*1200:
1950 XTX 57 (tomshardware) No softshadow 4AA 8AF , X6800 (dual core)
2900 XTX 58 (dailytech) with softshadow 4AA 16AF , QX6800 (quad core)


Gibbo over at OcUK had the following to say after benching a 2900XT and comparing it to his 8800GTX OC:

Hi there

I cannot say too much, but the card is well built, good quality, is quiet even when underload and is cooler running then previous ATI cards.

Performance wise everyone will have to wait but I think everyone will be impressed with its performance, especially if suggested pricing from other threads and articles from other websites are to be believed. I really cannot say anything or I will get my ass kicked from AMD. As soon as I can you guys will know more.

Something


just


doesn't


add


up.

 

[NoStateofMind]

From the looks of it, AMD has not only screwed up with the GPU, but also with people in the market to buy a new GPU. They double ****** themselves. R600 was something I recommended others to wait for, not anymore. G80 is what I'll recommend now.


I do think there is something to think about though. AMD has had 6 months to play with their competitors GPU. They must have known before hand that the R600 was only "comparable" to nVidia's DX10 solution. My hope is that they were smart enough to see that, and have prepared a "refresh" to offset things. Altho going by the last few months, it doesn't look too promising.

 

[Aberforth]

Originally posted by: vadp

Originally posted by: jiffylube1024
Sigh, if these performances benchmarks are true

They're not.
The sites are full of BS, Speculation and Disinformation at the moment.
Everyone who says otherwise is full of it.
Wait for a final product to make a conclusion.
AMD is making an awfully big deal with all those NDA's.
I refuse to believe that do all that just to hide a miserably performing product.

You are a typical example of a *FANBOY*...anyway why would these sites discredit R600 if it's actually good, why didn't they do it for Nvidia GTX and G80 drivers?.....*think*...got it?...I think not

 

[NoStateofMind]

Originally posted by: Creig
The more I think about it, the more I'm convinced that something just isn't right with numbers DT is providing. Saying that ATI is releasing a new card that can't keep up with an 8800GTX is one thing. Saying that ATI is releasing a new card that is barely faster than their previous generation champ (X1950XTX), even when using a quad core processor, is another.

Company of heroes 1280*1024:
1950 XTX 99 (tweaktown) E6600 (dual core)
2900 XTX 97 (dailytech) QX6800 (quad core)

Company of heroes 1600*1200:
1950 XTX 70 (tweaktown) E6600 (dual core)
2900 XTX 73 (dailytech) QX6800 (quad core)


FEAR 1280*1024:
1950 XTX 80 (tomshardware) No softshadow 4AA 8AF , X6800 (dual core)
2900 XTX 84 (dailytech) with softshadow 4AA 16AF , QX6800 (quad core)

FEAR 1600*1200:
1950 XTX 57 (tomshardware) No softshadow 4AA 8AF , X6800 (dual core)
2900 XTX 58 (dailytech) with softshadow 4AA 16AF , QX6800 (quad core)


Gibbo over at OcUK had the following to say after benching a 2900XT and comparing it to his 8800GTX OC:

Hi there

I cannot say too much, but the card is well built, good quality, is quiet even when underload and is cooler running then previous ATI cards.

Performance wise everyone will have to wait but I think everyone will be impressed with its performance, especially if suggested pricing from other threads and articles from other websites are to be believed. I really cannot say anything or I will get my ass kicked from AMD. As soon as I can you guys will know more.

Something


just


doesn't


add


up.

Good point, I hope you are right.

 

[ShadowOfMyself]

Originally posted by: Creig
The more I think about it, the more I'm convinced that something just isn't right with numbers DT is providing. Saying that ATI is releasing a new card that can't keep up with an 8800GTX is one thing. Saying that ATI is releasing a new card that is barely faster than their previous generation champ (X1950XTX), even when using a quad core processor, is another.

Exactly... As I said, even the x1950 catches up with the GTS lots of time... This is just ridiculous... Sounds like the x1800 benchmarks that came out from that Sander guy, where the card was getting owned by the 7800GTX, and it turned out to be a pile of FUD... Ill wait for my trusty review sites such as Firingsquad, Guru3d, Legitreviews etc

 

[vadp]

Originally posted by: Aberforth
You are a typical example of a *FANBOY*...anyway why would these sites discredit R600 if it's actually good, why didn't they do it for Nvidia GTX and G80 drivers?.....*think*...got it?...I think not

A fanboy?
By saying to wait for the real thing to make conclusions?
Ahhh... Got it.

 

[ShadowOfMyself]

Originally posted by: Aberforth

Originally posted by: vadp

Originally posted by: jiffylube1024
Sigh, if these performances benchmarks are true

They're not.
The sites are full of BS, Speculation and Disinformation at the moment.
Everyone who says otherwise is full of it.
Wait for a final product to make a conclusion.
AMD is making an awfully big deal with all those NDA's.
I refuse to believe that do all that just to hide a miserably performing product.

You are a typical example of a *FANBOY*...anyway why would these sites discredit R600 if it's actually good, why didn't they do it for Nvidia GTX and G80 drivers?.....*think*...got it?...I think not

Because.... Nvidia didnt leak any info until the very last moment? Because.. maybe Nvidia paid them? Well they sure pay Kyle over at HardOCP

 

[swtethan]

Back when c2d benches were released, everyone was like THAT CANNOT BE POSSIBLE OMGGGGGG WAIT FOR REAL BENCHES INTEL MESSED WITH THEM!!!!!! Well what happened? AMD had control of the testing in the first benches, so WHY would they intentionally mess it up? The xt and xtx are the same core just with different memory (with different clocks also). Remember 8800gts 320 & 640?

 

[TecHNooB]

Originally posted by: Creig
The more I think about it, the more I'm convinced that something just isn't right with numbers DT is providing. Saying that ATI is releasing a new card that can't keep up with an 8800GTX is one thing. Saying that ATI is releasing a new card that is barely faster than their previous generation champ (X1950XTX), even when using a quad core processor, is another.

Company of heroes 1280*1024:
1950 XTX 99 (tweaktown) E6600 (dual core)
2900 XTX 97 (dailytech) QX6800 (quad core)

Company of heroes 1600*1200:
1950 XTX 70 (tweaktown) E6600 (dual core)
2900 XTX 73 (dailytech) QX6800 (quad core)


FEAR 1280*1024:
1950 XTX 80 (tomshardware) No softshadow 4AA 8AF , X6800 (dual core)
2900 XTX 84 (dailytech) with softshadow 4AA 16AF , QX6800 (quad core)

FEAR 1600*1200:
1950 XTX 57 (tomshardware) No softshadow 4AA 8AF , X6800 (dual core)
2900 XTX 58 (dailytech) with softshadow 4AA 16AF , QX6800 (quad core)


Gibbo over at OcUK had the following to say after benching a 2900XT and comparing it to his 8800GTX OC:

Hi there

I cannot say too much, but the card is well built, good quality, is quiet even when underload and is cooler running then previous ATI cards.

Performance wise everyone will have to wait but I think everyone will be impressed with its performance, especially if suggested pricing from other threads and articles from other websites are to be believed. I really cannot say anything or I will get my ass kicked from AMD. As soon as I can you guys will know more.

Something


just


doesn't


add


up.

Hm... in that case... Back aboard the S.S. ATI! Might have to bring along a bucket.

 

[jiffylube1024]

Originally posted by: 5150Joker

Originally posted by: jiffylube1024


As an aside, it seems that in general the gaming industry is in a lull, as development teams are all working on "next-gen" looking games, which means even longer development times than before...

Not to mention an increase in console exclusive (or cross-platform) titles, such as Gears of War and many more Unreal 3 engine games, which appear to be in development for Xbox360, PS3 (in many cases) as well as PC. Ever since Microsoft introduced the original Xbox, it has stolen a lot of PC gaming's thunder, and the mockery that has become the expensive PC graphics card race has really turned a lot of people off (myself included). [though MS has their own problems with the ****** life expectancy of Xbox360 systems, and PS3 reliability is still to be determined]

The only thing that drives me back to PC games is the kb/mouse interface. If Sony and MS truly embraced kb/mouse for all their FPS games, I wouldn't touch the PC again apart from owning a laptop for Office.

I like a game controller for most of my console games, but it would be amazing if we could use a keyboard and mouse for FPS's, in particular games like Gears of War and the Halo games on Xbox... Especially with cheap, easily available (wireless) USB devices now, it's a no brainer if they would only include the support...

 

[RussianSensation]

Someone help me out,

The Xenos in 360 has a 48-way parallel floating-point dynamically-scheduled shader pipelines unified shader architecture with 2 shader ALU operations per pipeline per cycle (1 vector4 and 1 scalar, co-issued).

And R600 has 64 pipes with 5 shader ALU operations > to get 320?

I am just having a hard time understanding how R600 at 750mhz GPU core is so severely crippled compared to a 128 shader 8800GTX (although its shaders run at 2x the speed). So that would give 256 shaders say equal to ATI (I know the complexity is not comparable so this is again misleading).

Wasn't the whole point of unified shader design to increase performance? IF ATI isnt even able to double the performance of X1950XTX, then why wouldn't they just make a 700 million transistor X1950XTX Part 2 with double the shaders, etc. If unified shader tech cant take advantage of the added 'efficiency' today, perhaps this design is a bit premature and should have been reserved for R700 when DX10 games shift further towards shader intensity/complexity.

We have seen ATI gamble with X1600Pro/XT and X1650XT series by crippling it severely. Sure in some games with lots of shaders it actually matches 7600GT, but both are so slow, it's pointless. I hope ATI isn't making the card "futureproof" so that in 2 year it suddenly outperforms 8800GTX but when we are comparing 28 to say 17 frames per second, what difference does this future proofing make?

 

[Extelleron]

Originally posted by: RussianSensation
Someone help me out,

The Xenos in 360 has a 48-way parallel floating-point dynamically-scheduled shader pipelines unified shader architecture with 2 shader ALU operations per pipeline per cycle (1 vector4 and 1 scalar, co-issued).

And R600 has 64 pipes with 5 shader ALU operations > to get 320?

I am just having a hard time understanding how R600 at 750mhz GPU core is so severely crippled compared to a 128 shader 8800GTX (although its shaders run at 2x the speed). So that would give 256 shaders say equal to ATI (I know the complexity is not comparable so this is again misleading).

Wasn't the whole point of unified shader design to increase performance? IF ATI isnt even able to double the performance of X1950XTX, then why wouldn't they just make a 700 million transistor X1950XTX Part 2 with double the shaders, etc. If unified shader tech cant take advantage of the added 'efficiency' today, perhaps this design is a bit premature.

I wouldn't trust these numbers at all. These performance numbers are clearly BS... I have absolutely no doubt the performance of the HD 2900 parts is better than this. No matter how anti-AMD you are, they wouldn't release a GPU that is barely faster than the X1950XTX and that's what the HD 2900XT/X would be if these numbers were true. Clearly there's either some serious driver problems, or perhaps, which is likely, these numbers are just plain BS.

I'll believe it when Anandtech comes out with the R600XTX barely faster than a X1950XTX.

 

[PingSpike]

I'm really hoping Creig is right about this and there's something weird going on, because this is really the last thing AMD needs right now.

 

[TecHNooB]

Originally posted by: Extelleron

Originally posted by: RussianSensation
Someone help me out,

The Xenos in 360 has a 48-way parallel floating-point dynamically-scheduled shader pipelines unified shader architecture with 2 shader ALU operations per pipeline per cycle (1 vector4 and 1 scalar, co-issued).

And R600 has 64 pipes with 5 shader ALU operations > to get 320?

I am just having a hard time understanding how R600 at 750mhz GPU core is so severely crippled compared to a 128 shader 8800GTX (although its shaders run at 2x the speed). So that would give 256 shaders say equal to ATI (I know the complexity is not comparable so this is again misleading).

Wasn't the whole point of unified shader design to increase performance? IF ATI isnt even able to double the performance of X1950XTX, then why wouldn't they just make a 700 million transistor X1950XTX Part 2 with double the shaders, etc. If unified shader tech cant take advantage of the added 'efficiency' today, perhaps this design is a bit premature.

I wouldn't trust these numbers at all. These performance numbers are clearly BS... I have absolutely no doubt the performance of the HD 2900 parts is better than this. No matter how anti-AMD you are, they wouldn't release a GPU that is barely faster than the X1950XTX and that's what the HD 2900XT/X would be if these numbers were true. Clearly there's either some serious driver problems, or perhaps, which is likely, these numbers are just plain BS.

I'll believe it when Anandtech comes out with the R600XTX barely faster than a X1950XTX.

QFT!