Low End iMac Pro versus two Mac Pros and one iMac 5K
December 26th, 2017, by rob-ART morgan, mad scientist
December 27th - Added 10-core iMac Pro results
This is the first of multiple iMac Pro test pages. In this article we highlight CPU and GPU performance of the 'low end' iMac Pro (with optional GPU) compared to two popular Mac Pro configurations and the fastest iMac 5K.
CPU TESTS
Geekbench 4 Multi-Core CPU Test
Each CPU workload models a real-world task or application, ensuring meaningful results. These tests are complex, avoiding simple problems with straightforward memory-access patterns, and push the limits of your system.(HIGHER Score = FASTER)
Cinebench Multi-Core CPU test
Fascinating to watch as each core (real and virtual) rendering its part of the 3D model. (HIGHER Score = FASTER)
Luxmark 3 Multi-Core CPU test
In this two minute OpenCL benchmark we used all CPU cores to render the LuxBall Scene. (HIGHER KSamples per Second = FASTER)
GRAPH LEGEND
iMac Pro 10c Vega64 - AMD Radeon Pro Vega 64 (16GB) GPU inside the 2017 iMac Pro 3.0GHz 10-Core Xeon W-2150B CPU
iMac Pro 8c Vega64 - AMD Radeon Pro Vega 64 (16GB) GPU inside the 2017 iMac Pro 3.2GHz 8-Core Xeon W-2140B CPU
nMP 8c D700s - AMD FirePro D700s (6GB each) GPUs installed in the 2013 Mac Pro 3.0GHz 8-Core Xeon E5-1680 v2 CPU; TB2 port connected to the LG UltraFine 5K display
cMP 12c Vega64 - AMD Radeon RX Vega 64 (8GB) GPU installed in 2010 Mac Pro 3.33GHz 12-Core Xeon X5680 with DisplayPort of GPU connected to a Dell 5K display.
iMac 5K Pro580 - AMD Radeon Pro 580 (8GB) inside the 2017 iMac 5K 4.2GHz Quad Core i7 (7700K).
GPU TESTS
DaVinci Resolve 14
Using the Candle project, we measured how fast it could playback the 117 frame video clip while applying one node of noise reduction. GPU Processing mode was set to OpenCL for all GPUs. In the case of the 2013 Mac Pro, we used both FirePro D700s since that's the way it was designed to work. (HIGHER Frames per Second = FASTER)
Luxmark 3
This two minute OpenCL Benchmark uses the selected GPU to render the LuxBall Scene. In the case of the 2013 Mac Pro, we used both FirePro D700s since that's the way it was designed to work. (HIGHER KSamples per Second = FASTER)
Geekbench 4 METAL Compute Test
Measures the compute performance of your GPU from image processing to computer vision to number crunching, Geekbench 4 tests your GPU using relevant and complex challenges. Though the 2013 Mac Pro has two D700s, Geekbench only uses one to run this test. (HIGHER Score = FASTER)
Geekbench 4 OPENCL Compute Test
Measures the compute performance of your GPU from image processing to computer vision to number crunching, Geekbench 4 tests your GPU using relevant and complex challenges. Though the 2013 Mac Pro has two D700s, Geekbench only uses one to run this test. (HIGHER Score = FASTER)
WHAT DID WE LEARN?
Apple did its homework when planning the iMac Pro. As you can see from the results above, it beat both beefy Mac Pros in CPU performance. And if we had not jury-rigged an RX Vega 64 in the 2010 Mac Pro tower, it would have won 3 out of the 4 GPU contests.
We tried two different eGPU boxes with an RX Vega 64 connected to the iMac Pro with no success. Word on the street is that the next rev of High Sierra may include support for eGPUs. Meanwhile, at least one person on eGPU.io reported having a GTX 1080 Ti running in the AKiTiO Node connected to an iMac Pro.
Comments? Suggestions? Feel free to email me,
Follow me on Twitter @barefeats
