ATI whoops the GX2

dchrsf

Gawd
Joined
Oct 12, 2005
Messages
842
Looks like we have new reigning champ(s)! ATI FTW!!

http://www.dailytech.com/article.aspx?newsid=3727

Radeon X1950 CrossFire puts Quad SLI in its place

DailyTech has received early benchmarks of ATI’s upcoming Radeon X1950XTX and X1950 CrossFire graphics cards. ATI’s upcoming Radeon X1950XTX and X1950 CrossFire is expected to make its debut on August 23rd. Specifications for the Radeon X1950XTX and X1950 CrossFire are finalized with a 650 MHz core clock and 2 GHz effective memory clock. The core clock is unchanged from the previous Radeon X1900XTX while memory clock receives a hefty 450 MHz boost. This time around the Radeon X1950XTX and X1950 CrossFire are equipped with 90nm Samsung GDDR4 memory.

The benchmarks compare ATI’s Radeon X1950XT in CrossFire against NVIDIA’s Quad SLI. The test setup used for NVIDIA’s Quad SLI is a Dell XPS 700 system equipped with an Intel Core 2 Extreme X6800, 1GB of DDR2-800 memory and two GeForce 7950GX2 graphics cards for four total GPUs. The ATI CrossFire test system is identical with the Quad SLI system except the nForce 590 SLI Intel Edition motherboard was swapped out for an early Radeon Xpress 3200 (RD600) motherboard and two ATI Radeon X1950XTX/CrossFire graphics cards.

These early benchmarks are favorable to ATI. Call of Duty 2, Half-Life 2: Episode 1 and Serious Sam II heavily favor ATI, most likely due to better multi-GPU scaling on ATI’s side. FarCry, Quake 4 and Doom 3 performance shows the Quad SLI system creeping up to the performance of ATI’s X1950XTX/CrossFire system. However, the Quad SLI system still falls behind, close but no cigar. F.E.A.R. is the only game that can take advantage of Quad SLI and shows the Quad SLI system beating out the similarly configured X1950XTX/CrossFire system; though at 2560x1600 with 4xAA and 8xAF the X1950XTX/CrossFire takes the lead once again. With Elder Scrolls IV: Oblivion the X1950XTX/CrossFire system manages to take a nice lead over Quad SLI at 1600x1200 with 8xAF. The lead narrows when the resolution is raised to 2560x1600 with 8xAF.

Could Nvidia's GX2 video drivers not be optimized (or the othe way around, the games not being optimized) for quad yet? I wonder if these benches are with the new forcewares..
 
The Benches say X1950XTX vs 7950GX2 I think the article is jsut wrong. I think their is some misinformation. Even with GDDR4 I dont think ATis Crossfire could beat 4 cards in Sli. Thats a tough task.
 
Where did the benches come from? It says:
Benchmarks provided to vendors for the Radeon X1950XTX CrossFire versus GeForce 7950GX2 Quad SLI

Which means they must have come from ATI. Which means they're more than likely FUD.
 
Its not really a case of who is faster in my opinion, the price is a big factor. One of the main reasons i got my 7950GX2 was because it was only about £50 more than a single 7900 GTX

If the ATI card is faster then great, but if its 2x more expensive then it's unlikely to get a lot of punters.

FYI A sapphire x1900xtx was exactly the same price as my Leadtek 7950GX at the time I got it (a few weeks ago)
 
Frosteh said:
Its not really a case of who is faster in my opinion, the price is a big factor. One of the main reasons i got my 7950GX2 was because it was only about £50 more than a single 7900 GTX

If the ATI card is faster then great, but if its 2x more expensive then it's unlikely to get a lot of punters.

FYI A sapphire x1900xtx was exactly the same price as my Leadtek 7950GX at the time I got it (a few weeks ago)

Why you but a 1900XT thats just a WEE bit slower, and can o/c to XTX speeds easily for about 50-90 bucks Cheaper.
 
search the forums, this has been posted 3 times.

in every time the validity of those tests has been called into question. but wth ill restate my points:

1: if the momory Bandwith was such a huge bottleneck a whole lor more R&D woulda been put into it, and we would be on something a whole lot better then GDDR4

2: the Quad SLI drivers are trickey. i bet that they are only using two of the available 4 cores.

3 its been stated in this thread that its FUD from ATI. anyone remember the 6600GT vs X700PRO benchmarks ATI put out? yeah.... i can tell you from first hand experiance. they lie. they lie bad.

btw anyone seen the heat sink on that thing? coolist (both terms implyed) biggest bestest air cooling heat sink ever. no aftermarket cooling required. (fyi the close up in the title of that artical shows the prototype heatsink. they actually went with a differant design..... :rolleyes: )
 
This is most likely just another preview thing. Remember the Conroe review ages ago? :rolleyes:
 
My point is that for the price the GX2 is a very fast card, if you have a budget of £375 for a video card you can eigther get a GX2 or a X1900XTX and the GX2 is just a lot faster, it can also overclock by quite a lot even with just stock cooling.

I think we can safely say that this new ATI card is going to be faster than a signle X1900XTX and will be a fair bit more expensive which puts in the price range that far exceeds the GX, unless they decide a price cut across the whole range is in order.

If its pure speed you want no matter the cost, then Quad SLI is still likely to Nvdia in firm posetion of the speed crown.
 
Hawk said:
This is most likely just another preview thing. Remember the Conroe review ages ago? :rolleyes:

"shows an average of 40% faster then the Fx62!!!"

a week later:
"shows an average of 30% faster then the FX62!!"

a week after that:
"shows an average of 20% faster then the FX62!"

the week after that
"shows an average of 10% faster then the FX62!"

now:
"doesnt really make a differance in most games at 1600X1200, bar some flight simulators"

yeah i remember that well. who wants to bet the same thing happens here?
 
You have to be running uber settings to stress these cards.

Mine overclocked can chew through Oblivion in 1600 1200 with all settings maxed and 8xaf and gets about 30 fps outside, for a comfortable 45 or so 1280 960 maxed with 8xaf is a bit more playable (i like high frame rates)

If you only play games like CSS then it will make zero difference, I started a server and was running in 2048 x 1536 with 4xAA and 16xAF with max settings and getting a minimum of 90FPS and a peak of about 220, average was about 160

Online with players its a bit less, not by much.
 
This post is a repost and the information in that article is assumed false.

http://www.hardforum.com/showthread.php?t=1085944&page=1&pp=20


This article writer is an ATI supported site, so it will cary pro-ATI bias. Also, a few members including myself refuted the benchmarks with our quad-sli setups (scores are much higher for the quad sli setups than they show there).
 
MrWizard6600 said:
"shows an average of 40% faster then the Fx62!!!"

a week later:
"shows an average of 30% faster then the FX62!!"

a week after that:
"shows an average of 20% faster then the FX62!"

the week after that
"shows an average of 10% faster then the FX62!"

now:
"doesnt really make a differance in most games at 1600X1200, bar some flight simulators"

yeah i remember that well. who wants to bet the same thing happens here?


HAHAHA
So so so true!
 
One way or the other...it makes no difference to me. Until the day that I can afford and justify the money for something like the 3007FPW then my current 7900GT SLI is a slight overkill for my 1280x1204 lcd. I care about the bang for the buck in my pricerange..
 
Heh, I dont post much news, but thought this would be interesting. Sorry I looked before I posted, but didnt reconize any topics relating to mine. Plus, I had to run and catch the bus after I posted (I missed it anyways0 :D ) so didnt have much time,

If people are really doing this, canning benchmarks; why would they though because it will only hurt their rep, and not sell cards..
 
dchrsf said:
If people are really doing this, canning benchmarks; why would they though because it will only hurt their rep, and not sell cards..


See but when you are being paid by ATI it doesn't matter much, also, how many people do you think right now is telling their buddies "OMG, the 1950XT RAWKS all other cards!!!1!1!!"? I can bet you a ton are taking place with the viral marketing.
 
dchrsf said:
Heh, I dont post much news, but thought this would be interesting. Sorry I looked before I posted, but didnt reconize any topics relating to mine. Plus, I had to run and catch the bus after I posted (I missed it anyways0 :D ) so didnt have much time,

If people are really doing this, canning benchmarks; why would they though because it will only hurt their rep, and not sell cards..
Not really...once the seed is planted that it whoops everything out there then even later benchmarks showing a smaller gap won't do much to damage reputations. They can always claim that Nvidia optimized their drivers at a later date or something..
 
If them benchmarks (hardwarezone) are anything to go by it seems the GX2 will be the faster of the 2, not supprising really since all they appear to have done is increase the memory clock speed.

They'll also have to drop their prices a LOT to match the same with Nvidia, currently the X1900XTX is priced the same as the GX2.
 
{NG}Fidel said:
The Benches say X1950XTX vs 7950GX2 I think the article is jsut wrong. I think their is some misinformation. Even with GDDR4 I dont think ATis Crossfire could beat 4 cards in Sli. Thats a tough task.


7950GX2 is not 4 cards in sli....?

it is a dual GPU card all tied in as 1 card.
 
MrGuvernment said:
7950GX2 is not 4 cards in sli....?

it is a dual GPU card all tied in as 1 card.


It is still four physical boards but I still agree with you as they are counted as a single card.
 
Thats with 1 core enabled. try running them with both GX2 cores enabled. this is a review here with the 7900GTX/ X1900XT/ and 7950GX2 a run with 1 core alone and then both. it seems to murder even 2 X1900XT's in some stuff. which clearly shows that the before posted benchmarks of the DUAl X1950's is fake. seein is its barley faster than a regular X1900XT.. i wouldnt pay any attention to those benches before. no way 2 video cards can beat 4. lol....if so then obviously theres some driver issues.

http://www.driverheaven.net/reviews/GX2/HL2.htm
 
Ockie said:
It is still four physical boards but I still agree with you as they are counted as a single card.

They used single cards in the review as far as I can tell.
The GX2 is a 2 GPU solution with 2 boards.
Am I missing something?
 
Back
Top