$10.000 MacPro vs $10.000 PC

I have just created a video comparing two of these workstations.


I have checked GPU-Z and GPU was not involved in any Adobe rendering so that was disappointing. Please let me know if there are some way to improve performance (other than overclocking).

What exactly did you find disappointing?

I expected Adobe software to use $1800 Quadro K5000 CUDA to speed up the rendering

I don't know if it supports CUDA, but before buying a 10k workstation you might wanna have Googled whether or not the software you use actually supports the cards you're buying. These are the things you check before buying when getting a 10k workstation. 

Adobe has been in bed with Nvidia for years, hell Nvidia GAVE them hardware out the nose so that Adobe software would always run better on their cards. It was CUDA or nothing until CC started making the switch to OpenCL.


Seeing two monitors with noticeably different calibrations is one of my triggers so watching this had me tweaking a bit in my chair. Aside from that I'm not surprised. Apple has done some optimizations that could never be done in an open environment like Windows. Sure those specs aren't the absolute best but everything running on it knows how to squeeze every last drop out. I'm particularly curious as to why you'd side with more cores over a more balanced clock/core Xeon.

Filters aren't going to break the bank unless you're processing medium format raws, huge composites, or absurdly high res drum scans. I'd love to see you grab some PhaseOne IIQ files and put them through the ringer.

I wouldn't know, but what I do know is that a simple search would have revealed it- and that that is what the TS should have done. No offence though.

The workstations are bought for 3D and number of cores does make a high difference. CUDA seems to benefit Adobe for previews but not for final rendering.

The systems are cloned from one so calibration is yet to be done. 

More cores because of 3D as you can see the difference in speed is double. Also, 3ds max is not available for OSX.

What kind of test did you have in mind with these images?

Just pass them through ACR and run the same radial test. A small 4mb jpeg is nothing, of course they'll come out like dragsters. Why not load them up with something beefy and see how they pull as semis.