Here's what AMD didn't want us to see - HAWX 2 benchmark

[n0x1ous]

Interesting....

http://www.hardware.fr/articles/804-17/dossier-amd-radeon-hd-6870-6850.html
http://www.hardware.fr/articles/804-1/dossier-amd-radeon-hd-6870-6850.html

 

[jinsaotomex4]

Delicious green bars. That's a hell of a performance difference.

 

[Stoneburner]

WHy the wafers look like chexmex?

 

[Elfear]

Looks like Hardware France took it down. Your link doesn't work for me and the HAWX 2 benchmark page also shows no graphs.

 

[Madcatatlas]

Ye, you dont think theres something wrong with that benchmarks? and that is what and why AMD asked reviewers to not test it?

...

edit:
Elfear, the benchmark is still up and showing graphs on IE8

 

[n0x1ous]

It was there fine with just the picture in my post and then it disappeared. I edited to just add the link to the page on the review which is still working for me. Give it a try again. Sorry about that.

I have the image saved, but I can't add attachments apparently....

 

[crislevin]

well, AMD said they showed to the Ubi programmer that 6800 can do much better, but Ubi refused to implement the codes.

I heard nVidia sponsored Ubisoft...does that matter?

 

[Bateluer]

Meh, Barts losing in a single mediocre game means little. Probably be addressed with some driver releases regardless.

The Barts cards offer fantastic bang for the buck right now.

 

[crislevin]

Meh, Barts losing in a single mediocre game means little. Probably be addressed with some driver releases regardless.

The Barts cards offer fantastic bang for the buck right now.

actually, the game is not final, and the benchmark is not final neither.

 

[n0x1ous]

the picture shows up fine, and then i hit submit and its nothing......

 

[Madcatatlas]

its there noxius

 

[n0x1ous]

its there noxius

in my post or on the site itself? maybe my browser has gone crazy....

 

[Lonyo]

They don't show because the site probably doesn't allow hotlinking. They showed for you because you've been to the original page.

You would need to save them and then upload to Imageshack or something and then use those links and they should show up fine.

 

[n0x1ous]

Thanks for the explanation Lonyo, that makes sense. Apparently the link to the page is working in my original post so I guess I'll leave it as is, but thank you.

 

[Genx87]

This sholdnt surprise anybody. AMDs own graph shows their tesselation performance above factor 10ish starts to arrive near the 5000 series with these chips. In other words the tesselator is still weak compared to Nvidia's 400 series. It should be interesting to see if the 6900 series fixes this issue. But considering Tessealation was one of the 3 big additions in DX11 it is disappointing AMD didnt address this issue with these cards.

 

[Soleron]

This sholdnt surprise anybody. AMDs own graph shows their tesselation performance above factor 10ish starts to arrive near the 5000 series with these chips. In other words the tesselator is still weak compared to Nvidia's 400 series. It should be interesting to see if the 6900 series fixes this issue. But considering Tessealation was one of the 3 big additions in DX11 it is disappointing AMD didnt address this issue with these cards.

If Nvidia was using the same die area (255mm^2), their tesselator power would be reduced accordingly because it scales with shaders for them.

So in a heavily tessellation limited benchmark like this, in order to make comparisons on the tesselator being weak/strong you should actually be testing a GTS 450 against the 6870. Similar die sizes means a fair fight.

I'm sure if AMD was allowed a 366mm^2 die like the GTX 460 where the extra area was just tesselator units it would own that benchmark.

 

[OCGuy]

well, AMD said they showed to the Ubi programmer that 6800 can do much better, but Ubi refused to implement the codes.

I heard nVidia sponsored Ubisoft...does that matter?

Nope, doesn't matter. AMD wants to water down DX11 so it "appears" faster.

 

[Skurge]

This sholdnt surprise anybody. AMDs own graph shows their tesselation performance above factor 10ish starts to arrive near the 5000 series with these chips. In other words the tesselator is still weak compared to Nvidia's 400 series. It should be interesting to see if the 6900 series fixes this issue. But considering Tessealation was one of the 3 big additions in DX11 it is disappointing AMD didnt address this issue with these cards.

I thinks its a case of AMD going for a more balanced chip. Yes the 400series has much better tesselation, but at almost twice the size (GTX470) It had better. It ties the 460 in tess heavy games and beats in everything else, including other less tess heavy games. Do I wish they had more tess power? Hell Yes, do I think they should have compromised other parts of the chip to achieve that? No.

 

[formulav8]

Is that a nvidia sponsored game? Was Batman and a couple other games that wouldn't do intelligent AA on Ati cards, twimtbp games as well? Detect a pattern?

 

[crislevin]

Nope, doesn't matter. AMD wants to water down DX11 so it "appears" faster.

I would like to hear your theory about how AMD's implementation is bad for gamers.

 

[Genx87]

If Nvidia was using the same die area (255mm^2), their tesselator power would be reduced accordingly because it scales with shaders for them.

So in a heavily tessellation limited benchmark like this, in order to make comparisons on the tesselator being weak/strong you should actually be testing a GTS 450 against the 6870. Similar die sizes means a fair fight.

I'm sure if AMD was allowed a 366mm^2 die like the GTX 460 where the extra area was just tesselator units it would own that benchmark.

The 450 sells for what, 100-120 bucks, the 6870 240? Not in the same league. AMD can make the thing as big or small as they like. Nothing stopped them from making the die larger and add a more robust tesselator.

 

[OCGuy]

I would like to hear your theory about how AMD's implementation is bad for gamers.

It's bad for an "apples to apples" comparison between competing discreet graphics solutions.

 

[Skurge]

Nope, doesn't matter. AMD wants to water down DX11 so it "appears" faster.

No, you think they want to. You don't know that they do. Who knows that the bench even follows DX spec accordingly. Not any of us I'm sure.

You don't see them complaining about the LP2 bench do you? So there must be something wrong with this one.

 

[DaveSimmons]

Meh. nV bought nV-specific optimizations. It's the same marketing game as MS paying for timed exclusives on DLC, or Sony paying for an exclusive in-game character.

 

[crislevin]

It's bad for an "apples to apples" comparison between competing discreet graphics solutions.

lets be fair here, what is apple to apple anyway?

AMD thinks apple to apple is 6870 vs 4601G, and who is comparing that?
Batmen disable AA on ATi card, where is the apple?
6870 consumes half of the power than 470, what is the apple?

the standards are made for the end-users and developers, thats whats important.