Prey demo benches

[n7]

http://www.rage3d.com/articles/preydemoperf/index.php?p=1

Another superb review by Ratchet.

 

[Elfear]

Interesting article but I assume they tested at default video driver settings which isn't apples to apples IMO. I could be wrong about that but I didn't find anything in the article to suggest otherwise.

 

[Wreckage]

Despite the bias you will find in the Rage3D forums (of course), Ratchet's reviews are always well done.

Simple, to the point and all on one page.

I'm not going to even think about this game until I get a new system in a few months, but it seems to run just fine on current video cards.

Which leads me to the next note: try as I might I couldn't get the 7950 GX2 to run in dual-GPU mode on the A8N-SLI (even with the beta BIOS(s) that supposed added support for the card). I ended up benchmarking the GX2 on the A8R32-MVP instead, which worked fine.

:laugh:

 

[Ackmed]

Rathet is one of my fav reviewers. I really like the way the page is setup, and he does top notch work. I do wish he had given a conclusion of some sort though.

Kinda odd to me that the 7800GTX SLI was only a few frames slower at 1920x1200 4xAA/16xAF than the 7900GTX SLI. Guess its getting pretty CPU bound. Every single card from the 7800GTX and up is very playable at very high settings, which is nice. Wish there was a 7900GT for the folks looking to compare that card. From this review, NV has the lead by around 3-6fps at 1920x1200 (my res).

 

[Kyanzes]

I love this review. The 7950 GX2 shines, very nice results indeed.

 

[Wreckage]

Originally posted by: Kyanzes
I love this review. The 7950 GX2 shines, very nice results indeed.

In the single card benchmarks it passes the X1900XTX by 28fps at 1600x1200 with AA/AF. :Q

People who picked up one of these are going to be very happy with it.

 

[cmrmrc]

weird how all cards don't take a large performance hit when 16xAF is activated and how all cards take a pretty huge performance hit when only 4xAA is enabled

 

[fbrdphreak]

Very very nice review. Simple and straight to the point. Awesome bar graphs. :thumbsup: All video reviews take note, that is the shiz

 

[Ackmed]

The benchmarks were done on "Quality", or on the 3rd notch out of 4 in the drivers for NV quality wise. And done on the highest driver setting for ATi. Hardly the same setting. Im doubtful NV would have kept its lead with the drivers set to HQ, and optimizations turned off, but I guess we wont know.

 

[5150Joker]

Originally posted by: Ackmed
The benchmarks were done on "Quality", or on the 3rd notch out of 4 in the drivers for NV quality wise. And done on the highest driver setting for ATi. Hardly the same setting. Im doubtful NV would have kept its lead with the drivers set to HQ, and optimizations turned off, but I guess we wont know.

We'll know soon enough when I review my 7900 GTX vs. XTX at stock speeds.

 

[Cookie Monster]

Any IQ comparisons. To my knowledge, NV cards always looked better on OpenGl or doom3 engine based games.

 

[m21s]

Originally posted by: Wreckage

Originally posted by: Kyanzes
I love this review. The 7950 GX2 shines, very nice results indeed.

In the single card benchmarks it passes the X1900XTX by 28fps at 1600x1200 with AA/AF. :Q

People who picked up one of these are going to be very happy with it.

Already happy

 

[Ackmed]

Originally posted by: Cookie Monster
Any IQ comparisons. To my knowledge, NV cards always looked better on OpenGl or doom3 engine based games.

Where do you get that impression? NV has typically been faster, but they have looked the same from what I recall.

 

[Cookie Monster]

Originally posted by: Ackmed

Originally posted by: Cookie Monster
Any IQ comparisons. To my knowledge, NV cards always looked better on OpenGl or doom3 engine based games.

Where do you get that impression? NV has typically been faster, but they have looked the same from what I recall.

From playing doom3. Im not sure but it looked better on the NV card i was using then my friends ATi card.

 

[Ackmed]

First Ive heard about it. Never say anything in any review suggesting it looked better on NV hardware, than ATi hardware. Screen comparisons didnt show anything either.

We know that many of you are concerned with the image quality provided between these two manufacturers' video cards. In our evaluation, which consisted of swapping many video cards back and forth, we did not notice any differences in image quality between the cards at all. With both brands of cards, DOOM 3 was being presented exactly the way it was meant to.

In the rest of the screenshots shown above, we did not see any differences in image quality. We even took a close look at the lighting, as you can see in the fifth picture, and at High Quality the lightmaps and specular are compressed, yet we still see no difference in image quality between the video cards. Whether you have a NVIDIA or ATI based video card, you are going to be receiving the full gamut of video quality that DOOM 3 has to offer. Shadows, specular lighting, bump mapping, heat haze, the quality is all the same between the latest NVIDIA and ATI video cards.

http://enthusiast.hardocp.com/article.html?art=NjQ0LDQsLGhlbnRodXNpYXN0

 

[Cookie Monster]

I guess my eyes were playign tricks on me

 

[Polish3d]

It is stupefying to me that nvidia can win benchmarks simply by defaulting their drivers to a lower quality setting, and 4/5 benchmarkers just let it go. :disgust:

Maybe ATI can suddenly release drivers that default to "High Performance/Lowest Quality" settings and suddenly 0MFG they have a faster card than the 7950GX2!!11

 

[Elfear]

Originally posted by: Frackal
It is stupefying to me that nvidia can win benchmarks simply by defaulting their drivers to a lower quality setting, and 4/5 benchmarkers just let it go. :disgust:

Maybe ATI can suddenly release drivers that default to "High Performance/Lowest Quality" settings and suddenly 0MFG they have a faster card than the 7950GX2!!11

Lol. But if ATI did that they would be cheating....

 

[BFG10K]

weird how all cards don't take a large performance hit when 16xAF is activated

This has been the case for a long time as 16xAF is basically free on cards that support it.

Im doubtful NV would have kept its lead with the drivers set to HQ, and optimizations turned off, but I guess we wont know.

The benchmarks I linked to at EliteBastards have nVidia running HQ and you can see ATi pull ahead.

From playing doom3. Im not sure but it looked better on the NV card i was using then my friends ATi card.

At default driver settings Doom 3 visibly shimmers on modern nVidia cards but it doesn't on ATi cards.

 

[martinez]

The other thing is you can't tell IQ just from Screens alone. On my lowly crappy(these days)6600GT I noticed massive IQ performance increase in AoE3 with the advent of the more recent drivers. But when I took screens to try to compare the two they looked the same. I guess the smoothness of the rendering has a large impact on how the image is perceived, as I was getting 5-6 frames more in the same scenes.

 

[Shamrock]

Originally posted by: Ackmed
The benchmarks were done on "Quality", or on the 3rd notch out of 4 in the drivers for NV quality wise. And done on the highest driver setting for ATi. Hardly the same setting. Im doubtful NV would have kept its lead with the drivers set to HQ, and optimizations turned off, but I guess we wont know.

I cant find it anywhere. Where does it say he used "Quality" settings?

**EDIT** nevermind, I found it in the forums, it IS Quality

Here is something noteworthy...From Ratchet on the forums

Originally posted by: RatchetI would think that the lead for NVIDIA has something to do with it being based on the Doom 3 engine, yes. Note, however, that it has an ATI logo on the spalsh screen and is not an NVIDIA TWIMTBP game.

 

[theprodigalrebel]

I seriously think that is close to the BEST way to publish graphs/charts for a Video-Comparision. Only nitpick is that the 0xAA/0xAF data isn't required at all because 16xAF seems to have a <1% performance hit (barely a fraction of a frame in most cases).

Graphs make it very easy to see how AA affects performance for various cards. Also, nice to see one chart for ATi, one for nVidia, one with both head-to-head and another for MultiGPU. Good job Ratchet!

 

[josh6079]

Its also good to see an anticipated game launch and actually more of a game than a benchmark (i.e. Oblivion, GRAW, TRL). Don't get me wrong, Oblivion certainly is nice, but hardly as flexible between as many different video cards as Prey. Really, a lot of people will be able to play this at some pretty optimal settings.

 

[Cookie Monster]

Originally posted by: josh6079
Its also good to see an anticipated game launch and actually more of a game than a benchmark (i.e. Oblivion, GRAW, TRL). Don't get me wrong, Oblivion certainly is nice, but hardly as flexible between as many different video cards as Prey. Really, a lot of people will be able to play this at some pretty optimal settings.

Well the doom3 engine was made to run on a geforce3.. runs great on all hardware. Even on a voodoo3!!!

 

[Powermoloch]

I downloaded prey, my god it's crazy !!!!! portals? (awesome) gravity? (awesome!!!!) ? spirit mode? (duddeeeeee...)

my 7600GT KO really shines @ 1600x1200 with 2AA. It's a wowzer.