Widescreen Gaming Forum

I have added some updated benchmarks of the 6800 and 6900 series cards, focusing on more modern and/or DX11 titles such as:
* AvP
* F1 2010
* Heaven v2
* Just Cause 2
* Mafia II

http://www.widescreengamingforum.com/wiki/AMD_Radeon_6800_6900_-_Updated_Benchmarks

Nice benchmarks, thanks. Unless I missed it somewhere in the article, in the future could you also post the settings used when benchmarking each game? Ex. Texture quality, AA, AF, vsync, etc.

This would make it more useful for new people who are researching eyefinity and want to see what cards they need to purchase to play games at certain frames with certain settings.

Thanks for the compliment. Sorry for not noting it in this last article. Unless otherwise noted, all testing on the WSGF is dome at max quality settings with 4x/16x.

I have posted my review of the Radeon HD 6990.

I have posted my review of the Radeon HD 6990.

Dirt2 in Eyefinity and Dirt2 in widescreen have the same graph it looks like. i may be wrong tho.

Very nice review. WSGF and [H] are pretty much where it's at for multimonitor benches.

Interesting though that the 6990 was slightly faster than the 6970 xfire on here, but it's the opposite on [H].

Thanks. I fixed that. I'm also loading tables with the major values. Will help to show the performance difference between the HD 6990 and comparable SKUs.

I have added performance tables to each of the benchmark pages.

I have added performance tables to each of the benchmark pages.

So unless i misunderstood those graphs there is really only a minimal FPS Difference in 5870 E6's and the 6990 in 5x1... if so i am depressed now. lol

[quote]I have added performance tables to each of the benchmark pages.

So unless i misunderstood those graphs there is really only a minimal FPS Difference in 5870 E6's and the 6990 in 5x1... if so i am depressed now. lol
Pushing 5x1-P is an intense undertaking. The raw number of fps increase is small in some cases, but the % increase is often substantial. You've got to remember that this is all done at max settings.

The 6990 is great for 5x1. What kind of performance were you expecting? What games were you looking at?

[quote][quote]I have added performance tables to each of the benchmark pages.

So unless i misunderstood those graphs there is really only a minimal FPS Difference in 5870 E6's and the 6990 in 5x1... if so i am depressed now. lol
Pushing 5x1-P is an intense undertaking. The raw number of fps increase is small in some cases, but the % increase is often substantial. You've got to remember that this is all done at max settings.

The 6990 is great for 5x1. What kind of performance were you expecting? What games were you looking at?

IDK i guess it seems like the 5870 was so long ago when it was only 2 versions ago really. I mainly look at the games i play mostly but your review i went all the way through since you have the only 5x1 reviews ive seen and my monitors are on there way for my 5x1p setup. Im just debating on whether it would be worth the Price tag or not to get the 6990 Then get another next year, or just get 2 5870s now and upgrade next year when another E6 or 5x1p supporting card comes out...

IDK i guess it seems like the 5870 was so long ago when it was only 2 versions ago really.

By version, you mean series right? You mean one series ago... :wink:

[quote]

IDK i guess it seems like the 5870 was so long ago when it was only 2 versions ago really.

By version, you mean series right? You mean one series ago... :wink:
:doh Silly me :D

i was just going by 6870/50 and then to 5870/50 which would be 1 and 2 :P but either way it works out haha

Personally, I'd go for the 6990. There is a performance improvement, and I found the cards noticeably quieter and cooler in my office. Looking at NewEgg, the 6990 would set you back about $150 more.

It is amazing to see the Black Edition cards from last year still out there for $1k...

Personally, I'd go for the 6990. There is a performance improvement, and I found the cards noticeably quieter and cooler in my office. Looking at NewEgg, the 6990 would set you back about $150 more.

It is amazing to see the Black Edition cards from last year still out there for $1k...

Yea i bet they will go down eventually along with the 6990 but i think i am going to jump on the next xfx or sapphire one that comes up.

It's purely a supply and demand thing right now. Newegg can't keep them in stock. They're selling out instantly. I just wish they'd have more sales like they did with the 5870 E6 a short while ago.

It's purely a supply and demand thing right now. Newegg can't keep them in stock. They're selling out instantly. I just wish they'd have more sales like they did with the 5870 E6 a short while ago.

I bet that was predicated on the fact they knew the 6990's were coming...

Thanks for the nice review. There is not may reviews that have gotten multi screen up yet. There was one other that had xfire 6990's and 3 x 30" and a few others at eyefinity and 1 6990.

One thing you really really really need to oc your cpu. A basic example on my system with bc2.

At stock 2.66 i get around 45 - 65 gpu usage and 40 - 60 fps.

OC'd to 4 ghz I get 85 - 99 gpu usage and 70 -100 fps!!

Not all games respond so well but these dual GPU set-ups really need cpu power to get them going. Dual 5970's needed a similar amount of power to the dual 480's

Cheers

I would argue that stock and overclocked results are necessary; we should be putting together an article on the impact of both CPU and GPU overclocking on Surround/EyeFinity fairly soon - as there are going to be more people out there who want to look at triplewide without overclocking because, for whatever reason, they don't want to.

Until recently, I tended to leave my i7 rig at stock unless I was encoding video - on a single screen I didn't notice huge differences in framerate. Surround makes it quite different, though.

Some games react to the increase in CPU speed enormously. Others don't even notice it. It's completely engine dependent, even in what should be GPU-limited scenarios like Surround.

Great review!

Why do the majority of the other sites review such a board but then bench nothing that makes use of the multiple outputs?

With regards the CPU and OC'ing, I was thinking that perhaps the reason that the differences between the e6 card and this is not so wide is possibly because you are hitting that CPU wall.

You may find with more grunt behind it, like with a 2600K for instance, that it may open up or they both may increase.