GPU benchmarks

[matanov]

Hi there, we have run some tests in the office.

Scene used - http://ftp.chaosgroup.com/support/benchmark_scene.zip with max paths per pixel = 512, so we are looking at the time of completion .
V-Ray version official 3.10.02 for 3ds Max
Note the different times with different driver versions.

Soooo here are the results:
Quadro K2200 / 341.05 - 9m 43,7s
Quadro K6000 / 341.12 - 2m 13s
Quadro K5200 / 341.05 - 3m 9,7s
Quadro K4200 / 341.21 - 4m 54,7s
GTX 550Ti / 340.52 - 11m 25s
GTX 560 / 340.52 - 6m 49,7s
GTX 560 / 335.23 - 6m 48s
GTX 770 / 344.75 - 4m 22,6s
GTX 770 / 340.52 - 5m 17,5s
GTX 770 / 344.60 - 4m
GTX 750 / 340.52 - 8m 55s
GTX 570 / 334.89 - 4m 22s
GTX 980 / 344.16 - 2m 57,7s
GTX 980 / 347.09 - 2m 30,9s
GTX 980M / 344.45 - 4m 32s
GTX 480 / 340.52 - 4m 9.5s
GTX 660Ti / 331.82 - 6m
GTX 660Ti / 344.48 - 5m 45,7s
GTX 690 / 340.52 - 2m 11s
GTX 590 / 334.67 - 5m 12s
Tesla K40 / 332.76 - 3m 10,6s

[urbanite]

Thanks for the results Peter. I just run the test with my K6000 and the time is 2.41... I wonder what could be the reason of such difference, vray and driver version the same.

[3LP]

GTX 780 Ti / 340.52 - 2m14
GTX 780 Ti / 340.52 - 1m52 (OC +220Mhz)

I do have one of the best GTX 780 Ti though, the "ZOTAC GeForce GTX 780 Ti OC"
I thought newer cards would actually be better than mine, doesn't seems so, makes me think again about buying those new GTX 9xx.

Stan

[3LP]

I quickly jumped in gDocs and did a spreadsheet with the data.
It should be shared and editable by anyone here :
https://docs.google.com/spreadsheets...it?usp=sharing

I think the image should be updated automatically every-time we add info, so this could be a good way to have users input :

Stan

[Arjan Jacques]

GTX 580 / nvidea drivers 347.25 - 3m22,8 sec
Max 2014, vray 3.10.02

I was not able to update the spreadsheet

[3LP]

ah ha, just saw I gave the wrong credantials

Try now :

https://docs.google.com/spreadsheets...it?usp=sharing

Cheers

[Arjan Jacques]

Yes, added my score and the graph below was auto updated. Nice!

[3LP]

That's the spirit

[walee]

my score with my GeForce GTX TITAN

nvidia 347.25
V-RAY RT 3.10.03
1503.897 KPaths/s
2m18.2s elapsed

nvidia 347.52
V-RAY RT 3.10.03
1461.614 KPaths/s
2m22.1s elapsed

[3LP]

Just updated your results to the chart, but someone just added a Titan Z!
It's beating my Tri-GTX 970!

I should probably have gone for that one. Price is about the same but uses only 1 pci and has 6Gb in staid of 4.
Next time

[walee]

i tested again under the latest drivers 347.52 and i never get the same results:

2210.219 KPaths/s - 2m13.9s
1450.108 KPaths/s - 2m15.1s
1484.116 KPaths/s - 2m13.3s

does anyone knows what the differences are coming from ?

[3LP]

Not sure,
I also got different result each time I render again, it's always quite close though.

[savage309]

i tested again under the latest drivers 347.52 and i never get the same results:

2210.219 KPaths/s - 2m13.9s
1450.108 KPaths/s - 2m15.1s
1484.116 KPaths/s - 2m13.3s

does anyone knows what the differences are coming from ?

The KPaths per second we show is not an average value (and some fluctuations there can occure). You have seen its current value of the moment the render has ended. The total render time is what is important here and that deviation is small and completely normal (and is caused by the fact that V-Ray runs on top of OS & GPU driver).

[MattClark]

Updated the chart

4 x 980
Drivers 347.52

40.8 seconds!

[3LP]

Oh nice well done!

I wonder how the new titan gonna be!

Stan