ATI R700 Preview

[JohnJohn]

DriverHeaven
PCPer
Anandtech
HardOCP
Techreport
Extremetech
Hexus
Muropaketti
Guru3D
PCGamesHardware

Super short mega analisys

Beyond3D

Ok, I found a new game where 3870 X2 shows very bad microstuttering and it is Race Driver: GRID.
I was checking my benchmark results for R700 preview I'm working on and noticed avg FPS was very bad with 3870 X2. So I immediately applied my previous testing methods using Fraps to log frametimes and analyzed results with Excel: 3870 X2 shows similar behavior in Race Driver: GRID than I noticed in Crysis but it seems to be even worse. In Crysis 3870 X2 rendered every other frame after ~21,5 ms and every other after ~49,5 ms. In Race Driver: GRID 3870 X2 rendered every other frame after ~24,9..27 ms and every other frame after ~40,2..42,4 ms.

I repeated the test with R700 and it showed again very stable rendering times between every frame: ~15,3 - 17,6 ms.

[Knight]

Sad to see not high resolution runs with Crysis.

I know it is to show max playable for very high, but just for fun.

[dunn2953]

"Driver Heaven does not use benchmark scripts. We play the games for long periods of time on various levels and report any unusual findings we see. Then we record with FRAPS across several in game levels recording the averages. This is real world testing and is just how you guys will experience the game. Occasionally we might throw in a time demo as a further comparison, but we will note it. "

this is a very very bad way to benchmark and the results they give for the GTX280 seem higher than normal, albeit only slightly, but the HD4870 CFX results seem very low, and the R700 only slightly above them.

[m0da]

pcper: http://www.pcper.com/article.php?typ...&aid=590&pid=2
anand (from other thread): http://www.anandtech.com/video/showdoc.aspx?i=3354

they show some dif results

[eBoy0]

pcper: http://www.pcper.com/article.php?typ...&aid=590&pid=2
anand (from other thread): http://www.anandtech.com/video/showdoc.aspx?i=3354

they show some dif results

Those previews are the 1gb cards.... where are the 2gb cards?

[AriciU]

http://enthusiast.hardocp.com/articl...50aHVzaWFzdA==

They have 2 of them in CF Can't really see how it scales since they use different settings and res for CF and non CF.

[m0da]

Those previews are the 1gb cards.... where are the 2gb cards?

you're right! thanks. I didn't know they were going to make a 512x2.

first post has the 2gb though.

[eBoy0]

you're right! thanks. I didn't know they were going to make a 512x2.

first post has the 2gb though.

nope, look @ the gpu-z reads 1gb..

[m0da]

From the article:

In terms of memory we have 1 GB of GDDR5 on each side of the card and each RV770 GPU has access to its own gigabyte.

think gpu-z is looking at each core as a card. possible?

[Glow9]

This would be better if we could see it against 9800GX2s in SLI now that be interesting

[AriciU]

Another one -> http://techreport.com/articles.x/15105

[m0da]

http://enthusiast.hardocp.com/articl...50aHVzaWFzdA==

[Bo_Fox]

http://enthusiast.hardocp.com/articl...50aHVzaWFzdA==

Alrite, but wanted to mention on this thread (since it's now moved here) that HardOCP site omitted the GTX 280's CSAA modes without even mentioning it. We all know that those Nvidia cards can do 8x and 16x CSAA modes at little to no performance cost compared to 4x AA. Yet they praise ATI's CFAA modes that are often criticized on other sites (due to edge blurriness in games that heavily utilize post-processing like Unreal3 games and extreme performance hit in many games on single 4870 cards).

While it is truly to be praised that the R700 architecture can do 8x AA very well compared to Nvidia's true 8x MSAA, 8x CSAA and 16x CSAA are done with even less performance hit--usually none at all. And many sites accept that 16x CSAA is better than straight 8x MSAA.

Do you guys agree about 16x CSAA > 8x MSAA?

[strange|ife]

forget crysis, the game wont be playable to next year at near max at any decent reso. Farcry 2 hopefully wont be poop

[hurleybird]

Do you guys agree about 16x CSAA > 8x MSAA?

No way in hell. CSAA sucks serious ass.

[annihilat0r]

PCPer's article looks inaccurate

[annihilat0r]

Extremetech article

http://www.extremetech.com/article2/...2325444,00.asp

[eBoy0]

Extremetech article

http://www.extremetech.com/article2/...2325444,00.asp

3DMark Vantage shows enormous potential for the new ATI card. Scaling over the Radeon HD 4870 is about 80% in High mode, and a whopping 95% in Extreme mode. It's also quite a bit faster than the $650 GeForce GTX 280. So far so good, right? Well, there's a reason the card isn't out yet, as some of our other benchmarks demonstrate.

[ghost101]

3DMark Vantage shows enormous potential for the new ATI card. Scaling over the Radeon HD 4870 is about 80% in High mode, and a whopping 95% in Extreme mode. It's also quite a bit faster than the $650 GeForce GTX 280. So far so good, right? Well, there's a reason the card isn't out yet, as some of our other benchmarks demonstrate.

Dont they know how the 3dmark vantage scores are calculated? Of course the scaling on different presets will be different. The weights given to the gpu and cpu scores for each one is different.

[cakofony]

If i buy this card, will my pcie1 mobo hold me back at all?

[tiro_uspsss]

If i buy this card, will my pcie1 mobo hold me back at all?

nope!

[annihilat0r]

I don't understand. On some sites 4870 CF and x2 scale well (not great, but well) in Crysis. On other sites, they don't at all. I don't think they could be using different drivers, so what's that all about?

[bill_d]

I don't understand. On some sites 4870 CF and x2 scale well (not great, but well) in Crysis. On other sites, they don't at all. I don't think they could be using different drivers, so what's that all about?

i think a lot of the reviews are not retesting all cards on same drivers just the x2 on newer drivers and using scores from past reviews for other cards

[Isaac MM]

I didn't know they were launching a 512mbx2 version of this card, i thought it would be 1gbx2 only

[Warboy]

From What I see, compares to what I said before. It's really going to be a mix bag of chips.

so far from all the articles more or less, The GTX 280 gets better avg FPS, but the 4870 gets a better high FPS.
Typically you wanna go for avg more then high.

I'm wondering why in the Enthusiast article, they said they used a BFG GTX 280 OC, but it's clocked at 615mhz. I thought the standard clock is 620mhz? wtf guys?

But this round is still up in the air, and Who said it was 70% faster then a GTX 280? lol, Because from what I see, it's about 10%-35% Max. Considering it's 2 GPUs vs 1 GPU. That's not too impressive and Now let the price war begin. Of course revisions to the cards too.