[Teddy Gage]

Ian, Quadro cards are not faster than consumer gaming cards. In fact many are more or less the exact same hardware as much older gaming cards. I think these benchmarks, being a raw test of computing power, prove this.

What you are paying for in the quadro cards is:

– better drivers
– greater viewport accuracy
– better binned chips. ie. the quadros are tested much more rigorously for faults
– they typically run cooler, have smaller form factors and draw less power
– 24/7 customer support

but they are definitely not “faster”.

IMO the GTX Titan is actually the sweet spot between CUDA performance, viewport accuracy, raw processing power and gaming performance. If you can find one, the GTX 580 is also a great deal. People believe they are faster because they have paid thousands of dollars for the quadro name.

[Teddy Gage]

You’re insane! Nice results on the 3x gtx 580, that’s the fastest render recorded so far. but I’m surprised to see the gains offered by a third card are pretty modest. Although if it comes to rendering long projects it could be handy, I think 2x GTX 580 SLI is best price point to performance ratio.

[Teddy Gage]

Just because I was curious, with double-precision on, the benchmark renders slower. Just tried it. not by much, maybe 5-10%? Honestly I have no idea what double-precision is useful for. Anyone can fill me in?

I was also wondering whether it was possible to edit a preferences file allowing AE to use more available VRAM. I understand limiting cards with less memory, but I’m only getting 4.5 GB of usable VRAM with the titan in AE. Granted, that is a lot but it would be great to get another GB and I doubt the OS needs more than 500 mb of overhead.

[Teddy Gage]

I just got a GTX Titan and wanted to post my numbers

intel 3930K @ 4.6 ghz, with Titan @ stock speed = 3:57
intel 3930K @ 4.6 ghz, with Titan OC @ 1110 mhz base clock = 3:40

I will be testing this against a 2x GTX 580 in SLI, but still waiting on my second card

right now

GTX 580 3GB I am getting 5 min 5 sec with GPU clock at 925 mhz, or 5:45 at stock speed. testbed is intel 3770k @ 4.7 ghz

[Teddy Gage]

Guys, I have posted a spreadsheet of these results for everyone in .pdf and .ods format in the benchmark folder.

https://www.teddygage.com/AEBENCHCS6/

Hope these results are informative and thanks to everyone who has posted their info so far. If I didn’t include your results I only used one system per GPU unless the results were radically different (ie amd vs intel).

I would love to get more quadro and tesla results if possible to add to the list. Thanks!

[Teddy Gage]

I find that render time extremely hard to believe from a 6xx series card, unless it is watercooled. Make sure your cache was cleared (if the frame are already on disc it will be 10x faster)

If confirmed, I will add it to the spreadsheet thanks

[Teddy Gage]

This is very interesting. Been very curious about the Titan’s performance in AE

So basically, the titan seems to brute-force its way through to times that can compete (although not outlass) a speedy 580 setup, despite having less than optimal CUDA performance like the rest of the 6xx kepler series.

However it seems it may be software-limited performance, given that it’s basically just a consumer version of the k20 used in the tesla.

What I’m curious about is viewport performance in maya, and whether it can outperform the radeon 7970…