Really close, its Refractive Software buying AutodeskSo is autodesk buying out refractive software?
/me slaps self with a trout for even thinking that thought

Really close, its Refractive Software buying AutodeskSo is autodesk buying out refractive software?
/me slaps self with a trout for even thinking that thought
you'll read this in the release notes of beta2.andrian wrote:I have a question, what is the benefit of Cuda 3.0 over 2.0? Some examples what you can't do in cuda 2.0 that you can do in 3.0 will be great.
Quite enough.. I'll wait.radiance wrote:you'll read this in the release notes of beta2.andrian wrote:I have a question, what is the benefit of Cuda 3.0 over 2.0? Some examples what you can't do in cuda 2.0 that you can do in 3.0 will be great.
saves me the time to type it up here, which i don't have...
Radiance
it took about 900mb.metalliandy wrote:Awesome job with the physical sun and sky system guys!
I cant wait
How much GPU memory does a 7million poly model actually take btw?
I imagine it must pretty much max out any of the current cards.
with instances you can go as high as you want.andrian wrote:So if a person have GTX 260 with 1792 (or better with more memory) can get some decent renders with a LOTS of polygons - 14+ millions.
What about instancing, are they gonna improve that number ?
Example : GTX 260 896MB ram, I have building set up approximately 1,5 millions polygons, and I add greenery, all greenery pieces (trees. grass. flowers, etc.) will take like 5 million more, so we are on the threshold of 900 MB, but if I scatter instances lets say equivalent to 15 million more polygons what will happens then? We will still need only 900MB to render or ?