Techpowerup First GTX 280 Review is gone

[ownage]

Hi Pal!
The issue with just pulling the first post is that the rest of the thread wouldn't have made sense. I know, I did pull just the first post and then said the rest has to go.

Or the first post could have been edited like Perkam did with the leaked R600 review when he had moderator status.

[perkam]

Or the first post could have been edited like Perkam did with the leaked R600 review when he had moderator status.

Let's not go there

Actually this thread shouldn't be open for long. My apologies again

Perkam

[Errik89]

we do, wait till monday

- also investigating 3870x2 scores now

so you really have two different version of GTX280 at launch date?

anyway the benchmark you have removed have already been posted on other forum around the world...

[W1zzard]

where did all the people complaining about the "fake" crysis benchies go ? i'm waiting for your input now

errik89: i know they have been posted everywhere and i'm wasting my time trying to contain this now. because then i can tell the judge "at least i tried" so he won't send me to nvidia's factory to make graphics cards for the rest of my life for punishment

[RPGWiZaRD]

Seriously why would any1 be ebothered so much about the thread, I'm not the slightest bothered by that it's removed. Let's just look forward instead.

errik89: i know they have been posted everywhere and i'm wasting my time trying to contain this now. because then i can tell the judge "at least i tried" so he won't send me to nvidia's factory to make graphics cards for the rest of my life for punishment

Honestly such punishment wouldn't sound so bad to me. *j/k*

Nah I hope it turns out well, you've contributed a lot to the community and I don't think it's that much fair in acclaiming you to be responsible for this but oh well...

[ownage]

errik89: i know they have been posted everywhere and i'm wasting my time trying to contain this now. because then i can tell the judge "at least i tried" so he won't send me to nvidia's factory to make graphics cards for the rest of my life for punishment

Haha, you make it sound funny, but seriously. How serious are the consequences for you?

[W1zzard]

Haha, you make it sound funny, but seriously. How serious are the consequences for you?

worst case i assume is we wont receive any future nvidia products for review before the launch date.

[Errik89]

where did all the people complaining about the "fake" crysis benchies go? i'm waiting for your input now

errik89: i know they have been posted everywhere and i'm wasting my time trying to contain this now. because then i can tell the judge "at least i tried" so he won't send me to nvidia's factory to make graphics cards for the rest of my life for punishment

...to answer your question: because the scores seem high with all vga!

[W1zzard]

do you see anything wrong with the settings? we are using the benchmark_gpu timedemo with multiple runs because the first run is really low because of hdd accesses.

if you say the scores "are too high" - they are consistently high through all our cards, right? so the relative positioning is correct?

crysis does not support AF at all? then our graph titles are wrong

[dinos22]

wizzard
you don't need to prove anyone here that your scores are legit
no quality issues with TPU here that's for sure

i really see no big deal that you are using an overclocked system. Afterall review sites should go into detailed explanation what CPU bottlenecking is and pros and cons of overclocking and its impact on GPU performance considering the fact we are now entering the age of monster GFX cards

looking forward to your full review when it comes out soon

[MpG]

The scores DO seem a little high.

It wasn't too long ago the [H] put up an article outlining the reasons and methodology behind their "Real" testing, as opposed to using canned benchmarks. As justification, it was shown how drivers can (seemingly) be optimized for specific benchmarks, rather than the game the player actually plays. By way of demonstration, they showed how the Crysis benchmarks were showing considerably higher framerates than could be reasonably expected within the game itself, to the point where a hardware setup that gives an acceptable benchmark may not actually yield a playable experience. Without getting into the gritty debate over whether that's true or not, the bottom line is that they were finding real world performance to fall short of the benchmarks. Whereas you seem to be getting numbers that are actually BETTER than the benchmarks, at the given settings.

So with all that in mind, I'm a little interested in what kind of a demo/situation you use to create you own numbers. Is there something distinguishing that would account for most of the difference? Just curious, now that it's been brought up.

[s1nykuL]

do you see anything wrong with the settings? we are using the benchmark_gpu timedemo with multiple runs because the first run is really low because of hdd accesses.

if you say the scores "are too high" - they are consistently high through all our cards, right? so the relative positioning is correct?

Correct me if I am wrong but could it be that at 1024x768 the scores are CPU limited.

[W1zzard]

you don't need to prove anyone here that your scores are legit

if people see issues i will do my best to explain them. naturally i will investigate if my own testing is flawed first, that's what i'm trying to figure out now

[Morais]

At least I came before the link removal

[dinos22]

yeah i know

[W1zzard]

Correct me if I am wrong but could it be that at 1024x768 the scores are CPU limited.

not completely cpu limited, otherwise there would be no difference between the cards

The scores DO seem a little high.

Whereas you seem to be getting numbers that are actually BETTER than the benchmarks, at the given settings.

as i said we are using the standard gpu test flyby with above settings. anyone got a timedemo i can try?

[RPGWiZaRD]

I think most people got confused due to it said AA4x in the slides although as you showed here no AA was used and in that case the numbers look normal to me (and to my own system comparing 8800GT performance at 1600x1200).

I personally prefer to use Crysis Benchmark Tool v1.05 and "Sphere" timedemo is a good one imo.

[xlink]

Sorry guys, we have to respect NDAs, I requested removal of the thread.

Please do not repost the info, it's not the final data anyway.

If you want to discuss our review methodology, how GTX280 reviews should be done or anything similar without using our NDA info feel free to continue.

no worries. I only saved it to two harddrives... still i get to gloat over knowing 3 days early...

[s1nykuL]

not completely cpu limited, otherwise there would be no difference between the cards

I stand corrected, I should have thought about that

And I am not smart enough to notice anything wrong

[W1zzard]

at 1600x1200 we do use AA. sorry that screenshot was only from 1024x768 where there is no AA indeed.

hrmmm



thats an 1:1 capture from 1600x1200 4xAA .. looks like no AA ?


crysis says AA is on?!!

and thats the result on 3870 X2 .. 1600x1200 4xAA:
===== Crysis =====
Average: 28.6 FPS
Minimum: 20.9 FPS
Maximum: 32.6 FPS

my existing review data says 29.1 so no difference

[RPGWiZaRD]

at 1600x1200 we do use AA. sorry that screenshot was only from 1024x768 where there is no AA indeed.

hrmmm



thats an 1:1 capture from 1600x1200 4xAA .. looks like no AA ?

Difficult to say from that small pic but it really looks like no AA but I highly suspect it didn't use any AA in the tests u ran or otherwise I'll have to agree with the others that FPS just is too high. I get like mostly between 10 - 20 FPS with my 8800GT in 1600x1200 with AA4x. To me numbers looked normal for no AA performance though.

[[XC] hipno650]

it's amazing how many sites Google pops up with complete screen shots i feel for you man once it's out there is no way to contain it. well now i have a taste and i am satisfied for today.

[W1zzard]

omgz it seems that the benchmark script for crysis ended up using the wrong resolutions :banghead:

[Movieman]

I know I know

But he was talking in a legal sense, which I am familiar with, and I just wanted to correct him on the legal obligations on XS, which are to remove TechPowerUp's content only. XS is not liable to remove the comments.

My apologies if I'm stretching the issue

Perkam

Yes, on that point we agree and NP as to stretching the issue.
Debate diseminates information and that is always a plus.

[T_M]

The question then becomes do we serve the interests of a member with a problem that needs to be addressed or ignore that request and toss him to the wolves.

great

My thinking was that when an XL at XS PM's me and asks for help I do what I can and right then.
That green in the name says to me he's earned that position and I do what I can.

Huh? So a regular member wouldn't get the same treatment?