Re: BdB's not so simple interior with area lights
Posted: Thu Jun 24, 2010 11:10 am
At this moment I do a benchmark, with the new Tesla device C2050 and compare this device with the older generation Tesla C1060. The joke is the possibility to load big files sizes in the allocated device memory, the Geforce product line has a maximum of 1,5GB allocated device memory. Tesla devices have for example device memory of 4GB for the Tesla C1060 and 3GB for the Tesla C2060 when ECC modus is not activated and 6GB for the Tesla C2070. So the Octane Render file that’s can be loaded can be larger on a Tesla devices instead of a Geforce devices.
For example I made a object file, which has file format of 1,03 GB and like to render it on a Mesh Preview Resolution of 4096 x 3072, full aperture 4K. This will take a 2149,1 MB of the device allocated memory in my Octane Render scene, this file size will not fit on the Geforce allocated broad memory. I hope to finish my benchmarks this week for publishing it on the Octane forum.
Look at this video http://www.mentalimages.com/index.php?id=634
If you listen to his voice his is real scary
, about Octane Render solution 
, hahaha. If this IRAY solution was a big thing hi will be more confineable 
.
For example I made a object file, which has file format of 1,03 GB and like to render it on a Mesh Preview Resolution of 4096 x 3072, full aperture 4K. This will take a 2149,1 MB of the device allocated memory in my Octane Render scene, this file size will not fit on the Geforce allocated broad memory. I hope to finish my benchmarks this week for publishing it on the Octane forum.
Look at this video http://www.mentalimages.com/index.php?id=634
If you listen to his voice his is real scary
, about Octane Render solution , hahaha. If this IRAY solution was a big thing hi will be more confineable .