Radeon HD 2900 XTX: Doomed from the Start

I dont really get the point your trying to say here.

You declare X2900XTX total loser based on some 3Dmark tests which nobody can say are legiment and then you go off wondering what could be wrong. I would like to point out that X2900XT seems to be providing very much competition for Nvidia if these benchies hold any water. So its not all dooms day scenario yet, lets not jump into conclusions before we get the reviews at our hands :)

edit:

There seems to be alot of speculation that X2900XTX will not even be released on April 14th.

Are we looking at the same benchmarks?!

I am seeing FPS scores for REAL GAMES....not 3Dmark....where the 8800 GTX is absolutely bitch slapping the 2900 XTX

I happy to be corrected, but thats what it seems I am looking at :confused:
 
It seems that X2900XT will compete with 8800GTS only and X2900XTX has been delayed till Q3.

edit:

Defiant007 dunno how that 3Dmark thing slipped in. I think we were both looking at the same dailytech benchies, my bad :)
 
I dont really get the point your trying to say here.

You declare X2900XTX total loser based on some 3Dmark tests which nobody can say are legiment and then you go off wondering what could be wrong. I would like to point out that X2900XT seems to be providing very much competition for Nvidia if these benchies hold any water. So its not all dooms day scenario yet, lets not jump into conclusions before we get the reviews at our hands :)

edit:

There seems to be alot of speculation that X2900XTX will not even be released on April 14th.

He's just saying that the performance crown, which is what matters in the high-end, when comparing these cards, still is on NVIDIA's side, 6 months after the release of the 8800 GTX. Considering of course, that these benchmarks are true.

Also, I'm totally sure that neither XT and XTX, will be released on, and I quote:

sam0t said:
April 14th

I know it was a typo :)
 
I know it was a typo :)

Haha got me on that one! :D I counted the months with my fingers and still got it wrong, time for more classes :eek:

Iam sure alot of 8800 people are pretty happy about the situation, their cards life time just doubled I would imagine!
 
This is not good. 6Months late and ATI still can't catch GTX. Nvidia has the ULTRA ready to go but there is hardly a point. XTX won't even be released as it is completely uncompetitive.

In all this time NVIDIA is no doubt respinning the silicon for the whole lineup. They are not sleeping waiting for ATI to catch up.

It looks like might first NVIDIA card may be soon after 1 matrox, 1 3dfx, and 4 ATI cards.

This reminds me of the relentless pressure that destroyed 3dfx (I was a stockholder.).
 
Hey! ATI was creme of the crop until your Amerikkkan pig-corporate AMD bought it.

ATI was getting whipped and delaying products long before AMD bought them (X1800)
 
Guys, these are meaningless canned benchmarks and timedemos.

I'm sure Kyle will back me up on this, we should stop discussing these in this thread right now.
 
This is not good. 6Months late and ATI still can't catch GTX. Nvidia has the ULTRA ready to go but there is hardly a point. XTX won't even be released as it is completely uncompetitive.

In all this time NVIDIA is no doubt respinning the silicon for the whole lineup. They are not sleeping waiting for ATI to catch up.

It looks like might first NVIDIA card may be soon after 1 matrox, 1 3dfx, and 4 ATI cards.

This reminds me of the relentless pressure that destroyed 3dfx (I was a stockholder.).

I really hope not. AMD needs to, as I've heard many times, get their "act together", or we'll be back to Intel dominating the CPU market entirely.
I just can't help but feel that they adopted a very poor market solution. They seemed to be putting all their eggs in one basket, wanting both Barcelona and R600, to cream Intel and NVIDIA's offerings. AMD's been cutting prices in their processors, which I believe is a good solution, but to get back on their feet and have larger profit margins, they need a hell of a product. Since "it seems" R600 is not it, let's hope Barcelona is.
 
I really hope not. AMD needs to, as I've heard many times, get their "act together", or we'll be back to Intel dominating the CPU market entirely.
I just can't help but feel that they adopted a very poor market solution. They seemed to be putting all their eggs in one basket, wanting both Barcelona and R600, to cream Intel and NVIDIA's offerings. AMD's been cutting prices in their processors, which I believe is a good solution, but to get back on their feet and have larger profit margins, they need a hell of a product. Since "it seems" R600 is not it, let's hope Barcelona is.

What are you talking about?

These are meaningless canned benchmarks. We know nothing about R600.
 
What are you talking about?

These are meaningless canned benchmarks. We know nothing about R600.

Read more carefully. I said "it seems". Nothing, where rumors or unconfirmed benchmarks are used/shown, is true by default.
 
Have you guys even looked at those numbers?

Look again. There is clearly something odd about these numbers. When you compare them with their previous numbers in the 2900xt department, then it doesn't add up. The "review" is borked. Some scores are even only half of what they should be, even for the XT.
 
Huh?
Do you not read Kyle's posts?
Canned benchmarks are meaningless!

OK Kyle, calm down. You should stop posting with the username "Sharky974". We recognize your posts no matter which username you use. :D
 
um hello, they used an overclocked 8800GTX

That doesn't really matter. There is not much to be gained from a factory overclocked GTX, when compared to a normal GTX. It's faster no doubt, but not by much. Take roughly 8-10 fps max, off those GTX numbers and that's your "normal" GTX.
 
So ATI lost out on the ultra high-end market..... the true test will be to see how the midrange cards do (and even how the 2900XT is priced). These cards are where the real money is made. :)

(but yes, it does affect ATI's reputation for sure -- though they can bounce back, as nVidia has proven since the old leafblower FX5800 Ultra).
 
What are you talking about?

These are meaningless canned benchmarks. We know nothing about R600.

Meaningless canned benchmarks? There are at least 3 games tested there at different resolutions and the difference is staggering!

There are a lot of wait-and-see R600 guys like myself who are now going to buy 8800GTX's. If ATI wants to save a lot of sales, they need to release some benchmarks of their own that counter what we have seen from DailyTech.

Gamers spread rumors like WILDFIRE in the days of Digg. When these specs get out (they already have) the damage to ATI will be IRREPARABLE.
 
So ATI lost out on the ultra high-end market..... the true test will be to see how the midrange cards do (and even how the 2900XT is priced). These cards are where the real money is made. :)

And indications so far is that the HD2600, HD2400 will be a paper launch. Meaning ATI is 6 months late with any new generation part and won't actually have a midrange for a while yet. The will only have the the part they aimed at the high end and missed with. A $400 second place card, that will neither gain mindshare nor volumes.

This is a disastrous launch for a company that is bleeding money big time. If Barcelona is not a home run, it will be very grim indeed for AMD.
 
I sincerely doubt that those are the final clock speeds. I'm going to wait for the final [H] review before I judge.

Oh, and Fudzilla and theInq do not serve as sources.
 
Benchmark is Bull

Compare with 1950XTX on same settings on Xbits lab:

Company of Heroes - 1600x1200 - 69.2fps - R600: 73,7
Company of Heroes - 1920x1200 - 53.2fps - R600: 53,2

F.E.A.R. - 1280x1024 - 70fps - R600: 84
F.E.A.R. - 1600x1200 - 53fps - R600: 58
F.E.A.R. - 1920x1200 - 47fps - R600: 53,7

Half-Life 2: Episode One - 1280x1024 - 107fps - R600: 117,9
Half-Life 2: Episode One - 1600x1200 - 82fps - R600: 91,5
Half-Life 2: Episode One - 1920x1200 - 69fps - R600: 68,2

Source: http://gathering.tweakers.net/forum/list_message/27923277#27923277

that is strange. maybe ATI screwed up by keeping 16 ROPs? I always thought that was a bad idea.
 
Something doesn't seem right.:confused:

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

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

http://www.tweaktown.com/articles/10...oes/index.html


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

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

http://www.tomshardware.com/2007/02/12/the_amd_squ...

Posted here. http://rage3d.com/board/showpost.php?p=1334912940&postcount=5212
 
i didnt notice it say anywhere what each games in games setting was set to. max everything? aa? af? hdr? i didnt see it say aa or af was turned on in any of those games. if aa and af arnt on than those benchmarks are kinda useless.
 
I posted this elsewhere on the same topic and thought it might add a bit to the discussion here too.

_________________________________________________________________________
Their investment parts are actually fine. The XT (and presumably below) compete decently and seem to beat the 8XXX series. Combine that with the purported benefits of having onboard sound and that's a good card.

The problem lies in the high end/flagship area (XTX and GTX). I personally believe the XTX is poorly clocked as is evidenced by DailyTech's OC results of the "simple" XT. Hopefully this is a mere case where ATI can simply go "let's up the clockspeed" as manufacturers often do in the last few weeks with their cards and is not a yield problem with cards capable of reaching those speeds. I guess what I'm saying is that let's hope the XT that DailyTech got is indicative of a normal output card and was not a "golden sample" one. This would mean that ATI could boost the clockspeeds which (according to DailyTech) puts the XT performance on par with the GTX. Combined with better and more memory and the sound option the XTX could end up being a decent card/alternative.... just not 6 months late decent I guess.

If you "correct" the XTX benchmarks using these overclocks, does it seem completely implausible that the R600 is where it is? Does it seem completely unlikely that Nvidia managed to get a surprisingly good product out much faster that took ATI by surprise and took them a bit longer to match? Do I like the idea? No, but I also don't think that it's a completely unreasonable conclusion to draw.
 
I posted this elsewhere on the same topic and thought it might add a bit to the discussion here too.

_________________________________________________________________________
Their investment parts are actually fine. The XT (and presumably below) compete decently and seem to beat the 8XXX series. Combine that with the purported benefits of having onboard sound and that's a good card.

The problem lies in the high end/flagship area (XTX and GTX). I personally believe the XTX is poorly clocked as is evidenced by DailyTech's OC results of the "simple" XT. Hopefully this is a mere case where ATI can simply go "let's up the clockspeed" as manufacturers often do in the last few weeks with their cards and is not a yield problem with cards capable of reaching those speeds. I guess what I'm saying is that let's hope the XT that DailyTech got is indicative of a normal output card and was not a "golden sample" one. This would mean that ATI could boost the clockspeeds which (according to DailyTech) puts the XT performance on par with the GTX. Combined with better and more memory and the sound option the XTX could end up being a decent card.... just not 6 months late decent.

If you "correct" the XTX benchmarks using these overclocks, does it seem completely implausible that the R600 is where it is? Does it seem completely unlikely that Nvidia managed to get a surprisingly good product out much faster that took ATI by surprise and took them a bit longer to match? Do I like the idea? No, but I also don't think that it's a completely unreasonable conclusion to draw.

But there is more to the GTX than just a clock speed difference over the GTS. So with the XT and apparently XTX only slightly beating the 8800GTS. I wonder if they'll even be able to crank up the clock speeds on the XTX enough to compensate for it being identical to the XT?

But I'll try to wait until the final retail card is tested and not the unfinished OEM product, before declaring ATI a bunch of screw ups. :)
 
As long as the X2900XT makes the 8800GTS 640MB drop to the $300-$325 range, I'll be happy.

ATI has to make some pretty damn good X2600 cards.
 
I still want to see how these ATi cards do in DX10 games and tests. Maybe that will be their saving grace since DX9 is old hat.
I would be willing to bet that DirectX 9 will be more relevant at least until 2009.

I would rather have a card that performs well in most games, instead of some games.

DirectX 10 will be mostly a marketing term and maybe an unsupported patch to a couple games for another year or so.

Look how long it took SM3.0 to really take hold. It came out with the NVIDIA 6 series but now real games fully supported it until the 7 series.

It could be by the time R700\G90 cards are out that DX10 truly matters.
 
I can tell you this much as a huge AMD\ATi fan.

If AMD does not deliver Agena\Barcelona and the RD790 on time in the middle of this year, they are in deep...deep trouble with their perception in the high end segment. So much so that if this Cowboy gets one wiff that if the rest of AMD offering's are late I will jump ship to an Intel\Nvidia solution the very next day. For someone like me to say that should speak volumes to AMD.

They might make all the money in the mid range but it's the high end that holds the bragging rights. *If these numbers turn out to be true ATi (not AMD) has really screwed this cycle.

The question begs to be asked, if they knew the R600 was going to get its ass handed to them by the 8800GTX then in God's name what was the delay for?
 
^^ If true it's called "stay the course"

But I thought I read something about this before and the card was re-designed or something like that.
 
AMD has decided to push forward the launch of the Radeon HD 2900 series to May 2nd instead of the original date on May 14th. Products demonstrations and reviews are allowed to appear on that date so it is considered a soft launch. However, AMD is still keeping Radeon HD 2600 and 2400 under wraps until the big day on May 14th. Radeon HD 2900 XT cards will be available from that day onwards for the price of US$399 to be positioned against the GeForce 8800 GTS. The final clocks for Radeon HD 2900 XT stood at 740MHz core and 825MHz for memories.
.

LINK

Looks like damage control to bring forward the NDA. Yikes this looks terrible for AMD.
 
I still can't believe these benchmarks. Either the benchmark is bunk, or the guys at Rage3D have been waaay out to lunch in regards to their so called "inside information" and performance estimates. If these are indication of what to expect from the final cards, then I am glad that I picked up a GTX this fall for my machine build to keep me amused playing games through the long winter.
 
Back
Top