GTX 480 and GTX 470 Reviews

[ajaidev]

I wanted them on xbitlabs

[Chumbucket843]

I find this very weird


http://en.hardspell.com/doc/enshowco...23&pageid=6891

So with increase in resolution the ATI's card performance actually increases instead of decreasing??

its possible with tessellation. on average 32 pixels cover a triangle and thats where evergreen runs most efficiently. if its less than that then you cant use the shading power of the gpu. at the lower res with tessellation a triangle might cover <10 pixels which kills performance. if we double the res we can increase the amount of work the pixel shaders can do instead of bogging the gpu down with culling/setting up triangles and whatnot. this is why the 480 has less advantage over 5870 at higher resolutions. that my theory at least. the data doesnt lie.

[Chickenfeed]

Their 1280x800 numbers are flawed. A 5870 will do 50fps on the preview build of the stone giant demo as I mentioned before.

[Sadasius]

its possible with tessellation. on average 32 pixels cover a triangle and thats where evergreen runs most efficiently. if its less than that then you cant use the shading power of the gpu. at the lower res with tessellation a triangle might cover <10 pixels which kills performance. if we double the res we can increase the amount of work the pixel shaders can do instead of bogging the gpu down with culling/setting up triangles and whatnot. this is why the 480 has less advantage over 5870 at higher resolutions. that my theory at least. the data doesnt lie.

Thanks. I was trying to figure that out myself but what you said does make sense.

[Russian]

I'm wondering why there are no fermi reviews for tech report, anyone know?

Probably politics...

[SKYMTL]

Probably politics...

This one I would love to hear the reasoning behind. I was actually hoping for a review from them and XBit.

[Sadasius]

This one I would love to hear the reasoning behind. I was actually hoping for a review from them and XBit.

It was Homeland Security that got them for hashing a terror plot using an explosives vest filled with Fermis and running into ATI headquarters and setting them off. Who knows, it might be true! Did I mention Charles is my hero?!?

Okay, okay kidding aside.....It was aliens!

[kaktus1907]

new review GTX 470 runs @ 775/1550/1000mhz

http://forum.donanimhaber.com/m_38810174/tm.htm

[Russian]

This one I would love to hear the reasoning behind. I was actually hoping for a review from them and XBit.

I mean, you and i both know that If nvidia's PR doesn't like you or what you're doing or thinks that you'll definitely without a doubt write a bad review they probably won't sample you. That kinda politics. I don't know what the situation was for TR, but it probably had something to do with their attitude towards Nvidia or Fermi. Pure speculation.

[Loque]

new review GTX 470 runs @ 775/1550/1000mhz

http://forum.donanimhaber.com/m_38810174/tm.htm

so a 28% oc on a 470 equals a stock 480 that you can barely oc, maybe two oc'ed 470's in sli shouldn't be that bad :p

[Sadasius]

I mean, you and i both know that If nvidia's PR doesn't like you or what you're doing or thinks that you'll definitely without a doubt write a bad review they probably won't sample you. That kinda politics. I don't know what the situation was for TR, but it probably had something to do with their attitude towards Nvidia or Fermi. Pure speculation.

Pretty much common knowledge. It's the reason why I came up with fantastical stories. Seriously though....It's aliens!

[SKYMTL]

I mean, you and i both know that If nvidia's PR doesn't like you or what you're doing or thinks that you'll definitely without a doubt write a bad review they probably won't sample you. That kinda politics. I don't know what the situation was for TR, but it probably had something to do with their attitude towards Nvidia or Fermi. Pure speculation.

That is unfortunately the same with every PR person I talk to in this industry. They all have the ability to cut you off on a whim.

However, I don't think that's what happened since TR and XBit were front and center for the Deep Dive briefings.

[Loque]

Updated Hitler review http://www.youtube.com/watch?v=If0Bkfnifi4

[jmke]

I'm wondering why there are no fermi reviews for tech report, anyone know?

you guys should actually read TR before beginning with the crazy theories
http://techreport.com/discussions.x/18681

[Cybercat]

That is unfortunately the same with every PR person I talk to in this industry. They all have the ability to cut you off on a whim.

However, I don't think that's what happened since TR and XBit were front and center for the Deep Dive briefings.

I don't know what guys are talking about, but TR confirmed some days ago that they have cards and have done testing, it's just a matter of getting the article done.

http://www.techreport.com/discussions.x/18681

[onethreehill]

Techreport Review
http://techreport.com/articles.x/18682

[zalbard]

Techreport Review
http://techreport.com/articles.x/18682

Great, they have FPS over time graphs (for CoD:MW2 @ 2560x1600 only, though)! And 2GB 5870!
And the extra GB of VRAM only seems to improve the min FPS by 1-2, that's it...
And check the load noise levels of 2GB 5870!

[Russian]

That is unfortunately the same with every PR person I talk to in this industry. They all have the ability to cut you off on a whim.

However, I don't think that's what happened since TR and XBit were front and center for the Deep Dive briefings.

Yeah, unfortunately i wasn't there... even though Nvidia said it was Chief Editors only there were still some sites that had more than one person go.

As for the reviews...
Looks like a few sites were late to the Party.

I still find it funny that our review wasn't posted even though it was posted on Launch Day.

[damha]

Techreport Review
http://techreport.com/articles.x/18682

Another reviewer who doesn't know how to review.

http://techreport.com/articles.x/18682/9

Look at the frametimes for the 5870 2GB vs the 5870. Clearly the 5870 2GB is at least 3-4FPS faster on average than the 1GB model, and I am being pessimistic. How they screwed this one up this badly I will never know.

Review gets

Ah well, good thing we have no limits on garbage bins.

[insurgent]

you guys should actually read TR before beginning with the crazy theories
http://techreport.com/discussions.x/18681

I only asked a question, no conspiracy theories, just a straight question.

I just check the RSS feeds for reviews, I don't read every article, but thanks for the link

edit:

I still find it funny that our review wasn't posted even though it was posted on Launch Day.

Which site?

[cal_guy]

Another reviewer who doesn't know how to review.

http://techreport.com/articles.x/18682/9

Look at the frametimes for the 5870 2GB vs the 5870. Clearly the 5870 2GB is at least 3-4FPS faster on average than the 1GB model, and I am being pessimistic. How they screwed this one up this badly I will never know.

Review gets

Ah well, good thing we have no limits on garbage bins.

"To test, we played through this scene for 60 seconds while recording frame rates with FRAPS. This firefight is chaotic enough that there's really no hope of playing through it exactly the same way each time, although we did try the best we could. We conducted five play-throughs on each card and then reported the median of the average and minimum frame rate values from all five runs. The frame-by-frame results come from a single, representative test session."

Consider reading the reviews.

[Russian]

I only asked a question, no conspiracy theories, just a straight question.

I just check the RSS feeds for reviews, I don't read every article, but thanks for the link

edit:


Which site?

BSN*

I just don't want to site pimp so i refrain from posting any links in here.

[rusty]

so what 480 suck ?

[Helloworld_98]

so what 480 suck ?

depends on your definition of 'suck'

[insurgent]

BSN*

I just don't want to site pimp so i refrain from posting any links in here.

Oh, will check that later