![]() ![]() We used the DEMO version with a test scene called octane_benchmark.ocs. However, it only runs only on CUDA capable NVIDIA graphics cards. and does so in a fraction of the time it takes with a CPU based renderer. This is a "GPU only" standalone renderer that can process scenes created in Maya, ArchiCAD, Cinema 4D, etc. LONGEST graph bar means FASTEST in thousands of samples per second. We rendered the Sala scene (488K triangles) using GPUs only. ![]() LONGEST graph bar means FASTEST in frames per second. We applied one noise reduction node to the Standard Candle project and the Standard_Flashlight project. The GPU's core utilization was at maximum. We found that Noise Reduction rendering during playback was pure GPU. NVIDIA GPUs used web driver 343.02.01f01 and CUDA 6.5.37. Rad HD 5870 = AMD Radeon HD 5870 (1GB, 850MHz)Īll GPUs were tested in a 2010 Mac Pro 3.33GHz 6-core with 24G of RAM running OS X 10.10.1 Yosemite. GT 750 Ti = NVIDIA GeForce GT 750 Ti (2GB, 1254MHz) GTX 680 = NVIDIA GeForce GTX 680 (2GB, 1058MHz) GTX 780 = NVIDIA GeForce GTX 780 (3GB, 900MHz) ![]() GTX 980 = NVIDIA GeForce GTX 980 GPU (4GB, 1215MHz) That makes it a prime option for owners of Mac Pro towers - especially when they run pro apps that lean on the GPU for help in rendering. This model of GPU offers high performance while using the normal 6-pin power feeds of the Mac Pro tower. We have continued to test the NVIDIA GeForce GTX 980 after our November 2014 report. Posted Thursday, January 29th, 2015 by rob-ART morgan, mad scientist ![]()
0 Comments
Leave a Reply. |