AMD Radeon HD 6870 and HD 6850 confirmed to be launched on 22.10.2010

[kaktus1907]

take your pick

6870 = 239-279$

6850 = 179-209$

http://www.newegg.com/Product/Produc...D=3332167&SID=

[tbone8ty]

Finally at Newegg

6870 = 239-279$

6850 = 179-209$

http://www.newegg.com/Product/Produc...D=3332167&SID=

the XFX 6870 is $279... fail

looks like the sapphire is the one to get it comes with a hdmi cable and display port adapter

[Metroid]

6870 = 239-279$
6850 = 179-209$

6850 price / performance / overclocking ratio can't be beaten for $179.

Nvidia will have to drop the 460 1GB prices even further.

[kuroikenshi]

wow unless nvidia drops its prices even further, the'd be fighting a losing battle

[kaktus1907]

http://www.legitreviews.com/article/1445/1/
http://www.hardwarecanucks.com/forum...50-review.html

[Metroid]

http://www.legitreviews.com/article/1445/1/

Thu, Oct 21, 2010 - 12:00 AM

[kaktus1907]

http://www.tweaktown.com/reviews/360...ard/index.html
http://www.hardocp.com/article/2010/...eo_card_review
http://www.tweaktown.com/reviews/360...ard/index.html

[kaktus1907]

http://www.techpowerup.com/reviews/HIS/Radeon_HD_6870/
http://www.techpowerup.com/reviews/A...850_Direct_Cu/
http://www.techpowerup.com/reviews/HIS/Radeon_HD_6850/
http://www.techpowerup.com/reviews/A...870_CrossFire/
http://www.techpowerup.com/reviews/A...850_CrossFire/

[ethomaz]

So Barts = Cypress+? Same shader structure.

[Loque]

from http://www.legitreviews.com/article/1445/14/



my score:



so much for twice the tessellation if the 6870 is at a 5850 level..

[zerazax]

from [url]

so much for twice the tessellation if the 6870 is at a 5850 level..

Dude you can't
1) compare numbers directly across different setups
2) 2x tess doesn't mean other things aren't going on. The fact it can beat a 5870 despite having less units in the test should tell you things are being boosted somewhere

[kaktus1907]

http://www.hexus.net/content/item.php?item=27053&page=1

[Sn0wm@n]

I would really consider the 6850, but with it's 150w load draw, and the fact that it only has 1 aux power connector I'm affraid it may limit it's OC potential...

Does that make sense to anyone else or am I wrong...

and those are only refference board for now ... we might get better stuff for non refference boards

btw will we get an official thread listing all the 6800 reviews instead of this thread???

[Jowy Atreides]

lol

[OCX600RR]

lol

Wtf lol!

[kadozer]

Mutli_GPU configurations should be tested on the latest drivers for all Graphics Card not only HD68xx, for that you can not compare scaling with HD58xx,GTX4xx

Indeed, good thing I was comparing it with sli. How much fps difference are we talking bout here if they did latest drivers. Care to quantify? No? ok good just take it as face value for what it is and know that crossfire looks good enough looking at the techpowerup review added to the mix.

2nd edit Here ya go Maxforces AT crossfire: http://www.anandtech.com/show/3987/a...range-market/6
http://www.anandtech.com/show/3987/a...ange-market/12



I love that Hardocp included Civ V. That game really taxes any current gen Nvidia or AMD in DX11.

[SKYMTL]

That game really taxes any current gen Nvidia or AMD in DX11.

I would love to know how sites are testing Civ. I have yet to experience a way to objectively benchmark it without running into a processor bottleneck on high-end systems.

[kadozer]

I would love to know how sites are testing Civ. I have yet to experience a way to objectively benchmark it without running into a processor bottleneck on high-end systems.

Yeah I know exactly what you mean as a Civ V player. Hardocp states the simulated a small world environment with army battles. Just shows how really taxing this game is, as the larger maps and ai actions in between turns are exponentially bigger as you go from 8 , 16+ players.
They state they ran FRAPS for 5 min runs. Not sure how reliable that is.

[SKYMTL]

Yeah I know exactly what you mean as a Civ V player. Hardocp states the simulated a small world environment with army battles. Just shows how really taxing this game is, as the larger maps and ai actions in between turns are exponentially bigger as you go from 8 , 16+ players.
They state they ran FRAPS for 5 min runs. Not sure how reliable that is.

I'm a big Civ player as well (when I have the time) but I have yet to see a situation where a "battle" results in lower FPS than just panning the map. Even a nuke being used later in the game doesn't really have all that much of an impact.

Heck, it isn't even true 3D; it uses an updated version of 2D sprites with some DX11 features thrown in.

From my experience, the game is seriously CPU bottlenecked once the AI kicks in which means the lowest performance will be when changing turns. Not when the GPU is loaded.

[SamHughe]

Guru3d review is up again + Crossfire review

[zerazax]

Strategy games are notoriously hard to benchmark since they're so CPU intensive... and my experiences with Civ 5 are almost always CPU related it seems

[cegras]

I wanted to chime in and say good review too, SKYMTL. As usual techpowerup is great, and techreport is another I'm waiting for.

[SKYMTL]

I wanted to chime in and say good review too, SKYMTL. .

I take it you read the conclusion this time too?

[cegras]

I take it you read the conclusion this time too?

Some I agree with ... some I don't. At least I gave you my reasons!

Again, I find it almost ridiculous how nvidia is trying to do a price war when the performance isn't on the same level.

[zerazax]

Some I agree with ... some I don't. At least I gave you my reasons!

Again, I find it almost ridiculous how nvidia is trying to do a price war when the performance isn't on the same level.

Well, ATI did the same thing with the 3800s (to a lesser extent) and the 4800's as well. Although in the case of the 4800s, the pricing more or less made the GT200's look absurd if nothing else.