About to run this test with my 680 FTW+ 4GB but which test do you use ? I think there are 3 variations on it here .

 

Yeah, may bad. Sorry Paul

 

Here's the one

 

https://dl.dropbox.com/u/19353176/GPUMark2.zip

Thanks Dazz

OK guys something seriously wrong with this card, tested with i5 2500k @4.5 Ghz, 1600 RAM. 680 FTW+ 4GB

 

 

 

 

You can see GPU is being used to its max, fluctuates between 90-100% in GPUz

 

Scores are min 8, max 11. Average 8.863. Same scores over 3 x 5 mins test. This cannot be right, surely ?????????? Could it be a problem with the driver ? Ban cap made a post a while ago about how to completely wipe the driver but I'm dammed if I can find it.

OK guys something seriously wrong with this card, tested with i5 2500k @4.5 Ghz, 1600 RAM. 680 FTW+ 4GB

 

 

 

 

You can see GPU is being used to its max, fluctuates between 90-100% in GPUz

 

Scores are min 8, max 11. Average 8.863. Same scores over 3 x 5 mins test. This cannot be right, surely ?????????? Could it be a problem with the driver ? Ban cap made a post a while ago about how to completely wipe the driver but I'm dammed if I can find it.

 

Did you load the Inspector profile (.nip file) from the provided zip?

AA should be 8xS instead of 16xS.

 

Looks like I screwed up again ouch!

 

At 16xS, 4xSGSS and 16xAF with those nasty clouds all over the place it's perfectly normal to see FPS in the single digits

OK redone the test with your inspector file Dazz, sorry missed that. Scores now are similar to those who posted with a 580

 

Frames Time Min Max Avg

4878 300000 14 21 16.26

 

Same for all 3 tests. I can only conclude that the 680 is no better than the 580 in this test.

OK redone the test with your inspector file Dazz, sorry missed that. Scores now are similar to those who posted with a 580

 

Frames Time Min Max Avg

4878 300000 14 21 16.26

 

Same for all 3 tests. I can only conclude that the 680 is no better than the 580 in this test.

 

Well, I don't think any of the GPUMark runs above where done with that test suite apart from yours. I would say that's quite a great score for such high IQ. Wish I had my 580 to compare, but I don't

Now if you set AF to app controlled, and if you were to overclock, you would probably be averaging some 25 - 30 FPS with mins above 20. And that's with 4xSGSS.

 

Pretty amazing if you ask me. Would you be so kind to at least try the App controlled AF tweak please?

 

was that at your native 2500x1440 resolution? If it is, it's still more impressive man.

 

Well, this test is probably not going to be useful to compare cards in different systems since there are far too many factors to take into account (display resolution, cloud types, cloud resolution, coverage...) but you could run this test with your 560Ti for a real comparison with the 680 everything else being the same.

 

Then it's up to you to decide if the difference in visual / shimmering control is worth the price of that 680 or not

Will do Dazz, is this correct , all other settings the same ?

 

 

Will do Dazz, is this correct , all other settings the same ?

 

 

 

That's correct. You also need to set Filtering to Anisotropic in the FSX options (Graphics tab)

Here are my scores for an i5 [email protected] Ghz, 1600 999 24 RAM, 680 FTW+ 4 GB. All tests ran in fullscreen mode using GPUmark2.

 

1920x1080x32 resolution

Frames Time Min Max Average

4696 300000 23 30 26.117

4690 300000 23 30 25.77

 

2560x 1440x32 resolution

4696 300000 14 18 15.653

4692 300000 14 18 15.64

 

Using APP controlled AF tweak, filtering set to Anisotropic in FSX options, graphics tab

1920x1080x32 resolution

4719 300000 23 30 26.123

4729 300000 23 30 26.208

 

2560x1440x32 resolution

4729 300000 14 18 15.763

4692 300000 14 18 15.64

Just tried 2x SGSS using APP controlled AF tweak, filtering set to Anisotropic in FSX options, graphics tab

 

2560x1440x32 resolution

 

Frames Time Min Max Avg

6481 300000 19 25 21.603

6502 300000 20 25 21.673

 

Will revert back to my 560ti tonight and compare the 2.

Just tried 2x SGSS using APP controlled AF tweak, filtering set to Anisotropic in FSX options, graphics tab

 

2560x1440x32 resolution

 

Frames Time Min Max Avg

6481 300000 19 25 21.603

6502 300000 20 25 21.673

 

Will revert back to my 560ti tonight and compare the 2.

 

Looks like the AF tweak is not doing much for your frame rates here, oh well.

Anyhow, 20 - 25 FPS @ 2560x1440x32 + 2xSGSS... and on stock clocks... wow, just wow.

 

Anticipate single digits with the 560Ti in this same conditions.

 

Another thing you could try if you want to further enhance performance (apart from overclocking the GPU of course) is to lower the cloud resolution in REX, assuming you use REX textures, that is

Cloud test with 560 ti GPUMark2 using AF tweak from my previous post

 

4xSGSS Min 6 Max 9 Av 7.3

 

2xSGSS Min 11 Max 15 Av 12.66

 

So the 680 FTW+ 4Gb is around 100% faster than my current 560ti. Would I keep it for FSX? no, I can get by with my 560ti OK just by lowering settings a bit. But the 680 is a different story with X-plane 10 demo. With the same settings frames increased from 9 FPS with the 560 to 30 FPS with the 680 . X-plane does seem to make more use of a more powerful graphics card.

Paul