OTOY Forums

Hi!
I wonder if there is a mathematical relationship between number of CUDA cores and render times.
For example 100 CUDA/60 min, 200 CUDA/30 min, 400 CUDA/15 min, ...

using the same gpu at the same clock rates and having enough memory bandwidth it is just like that. but usually gpus with different cores have also different specs in terms of vram speed and bandwidth, so the picture is a bit more complex; if you look at the benchmark chart, you can get an impression: http://www.refractivesoftware.com/forum ... f=9&t=7383

Interesting info T_3.
Do you know If you can get more speed having 2 cards instead of 1? (same total CUDA cores, VRAM and bandwidth in both options).
I mean if the bandwidth affects to the speed, if in the first case gpu's can operate faster because each one has i.e. 100 GB/s for 200 cuda's and the second one 100 GB/s for 400 cuda's.

octane scales perfectly with the number and power of the cards used.

e.g. one card: rendering speed n, two cards of the same type: rendering speed n*2
or one card: speed a, another card: speed b, both cards together: speed c = a+b
...

So, you mean that it doesn't matter to have the same power in 1 card or divided in 2, right?
In that case would be better to have all the power condensed in 1 card to have more free space.

stinger wrote:So, you mean that it doesn't matter to have the same power in 1 card or divided in 2, right?
In that case would be better to have all the power condensed in 1 card to have more free space.

exactly.
just in case you are looking for a gtx 590, be aware that this one doesn't have 3gb vram at once - as most retailers suggest - but only 1.5gb per gpu, because it's a dual-gpu card...

Thanks for the answers t_3!

Is there a way a newb like me can view that information?

I get "You are not authorised to read this forum."


I have a GeForce GTX 560 Ti 2GB in my PC, and I'm keen to know about how fast it would be, and also, how it deals with RAM requirements. Many of my renders require >4GB of RAM in Windows.

UHF wrote:Is there a way a newb like me can view that information?

I get "You are not authorised to read this forum."

I have a GeForce GTX 560 Ti 2GB in my PC, and I'm keen to know about how fast it would be, and also, how it deals with RAM requirements. Many of my renders require >4GB of RAM in Windows.

yes there is: http://dl.dropbox.com/u/10375307/octane_bench1.pdf

quoting from another post in the licensed users forum, this might help regarding ram useage:

t_3 wrote:some numbers:

• for 5mio resulting tris, octane needs around 800mb vram, ~2gb system ram, and additional ~2 to 8+(!) gigabytes of system ram while parsing and voxelizing (depending on the complexity of the scene)
• a 4096x4096 "image" map needs 64mb vram, and -apparently- ~2 times the original raw image size (~100mb for a rgb image);
• a 4096x4096 film (=output size) would need 320mb vram (20 byte per pixel) and the same amount of system ram

note: system ram usage not counted to the exact byte, only taken from the changes in the taskmanager...

t_3 wrote:using the same gpu at the same clock rates and having enough memory bandwidth it is just like that. but usually gpus with different cores have also different specs in terms of vram speed and bandwidth, so the picture is a bit more complex; if you look at the benchmark chart, you can get an impression: http://www.refractivesoftware.com/forum ... f=9&t=7383

This relates to the question I'm posting in the forum. How much improvement in terms of time/% do I get going from 3x580 GTX 3Gb to 4x580 GTX 3gb?

Who are allowed to see these numbers? Perhaps I just have to joint another forum, or is it for already established customers?

Thanks!