Even with one-fourth the bandwidth of the 'normal' x16 PCIe slot, in many cases the NVIDIA option shows its advantage. The eGPU offers an alternative to the "FirePro Only" option in the 'late 2013' Mac Pro cylinder. The single active FirePro D700 edged out the TITAN Xs. Then we played animation in fullscreen mode at 2560x1440 with a maximum frame rate of 500 (to push it to as fast as it would go). We took the Car Chase animation that Cinebench runs in a 1024x768 portal and opened it in Cinema 4D Rev 17. Notice how none of the GPUs have a clear advantage. NVIDIA GPUs excel in many GPU intensive games.Ĭinema 4D R17 - Here is an example of a CPU bound animation that can confuse and frustrate Mac users seeking to boost FPS with a better GPU. (HIGHER number means FASTER in average Frames per Second.) Tomb Raider - We use the built-in benchmark using the HiGH preset. When it comes to GPU intensive games, the NVIDIA GPUs dominate whether in the eGPU or native PCIe x16 slot. (HIGHER number means FASTER in Frames per Second.) Quality was HIGH, no AA, no Vsync Fullscreen. (LARGER number in Frames Per Second means FASTER.)Įven when 'hobbled' by the limited bandwidth of Thunderbolt 2, the eGPU TITAN X 'buries' the AMD FirePro D700 on this OpenGL test.ĭiablo III - We observe FPS while our character was by the New Tristam Waypoint. It tends overheat the GPU making it a perfect stability and stress test tool.
#External display for mac pro 2013 software
The dual FirePro D700s get no score because they are not 'CUDA capable.' This is one of the primary justifications for the eGPU: to run 'CUDA only' pro software like Octane Render.įurMark OpenGL 'GPU burner' uses fur rendering algorithms to measure the performance of the graphics card. OctaneBench allows you to benchmark your CUDA capable GPU using Octane Render code. However, when dual D700s are rendering OpenCL, they, of course, beat the TITAN X. The single GeForce TITAN X beats the single FirePro D700. (LARGER number means FASTER in Thousands of Samples per Second.) For this round we used the default Sala scene (488K Triangles) and rendered using all available GPUs, but GPUs only. LuxMark 2.1 is an OpenCL benchmark that renders scenes of various complexity.
0 kommentar(er)
