The Mac Pro Review (Late 2013)
by Anand Lal Shimpi on December 31, 2013 3:18 PM ESTCPU Performance
I like to have large historical databases of performance so I can put new products in perspective. The Mac Pro and its funny lineage make this a little difficult. For starters, I simply haven't reviewed all of the Mac Pro CPU combinations that have existed over the years. Then there's also the fact that not all of my Mac suite applies well to a 12-core/24-thread Mac Pro. I'm going to try my best to put the new Mac Pro's performance in perspective, but it's going to require a couple of subsections.
Let's first start with a look at the historical performance of the Mac Pro. I really have to thank @elfurberino and @tapbot_paul for lending their time (and their Mac Pros) to help flesh out this comparison. With their help I managed to put together performance data for almost every single generation of Mac Pro.
We'll start with Cinebench R10's single threaded test. Unfortunately the benchmark crashes on Macs with 16+ threads so single threaded performance is all we'll be able to look at:
That's right, I still have my old PowerMac G5 Dual 2.5GHz (upgrade from my original 2.0 model). It's interesting to note that single threaded performance has only improved by 2.8x over that 2.5GHz dual G5 machine from around a decade ago. If we were able to also look at multithreaded performance we'd see a much larger gap. The dual G5's multithreaded performance is actually lower than the single threaded performance of the new Mac Pro's IVB-EP (3346 vs 5187). And the new Mac Pro has 12 of those cores.
Here you can see a very healthy increase in single threaded performance over the 2010/2012 12-core system. The 34% increase in performance is because the Mac Pro never got the Sandy Bridge bump. All previous Mac Pros topped out at Nehalem/Westmere. Couple all of the Sandy Bridge improvements with the much higher peak clock speeds (3.5GHz vs. 3.0GHz) and the performance gains make sense.
The 8C system from early 2009 gives us an example of how it's very possible to have a newer Mac Pro actually perform worse than its predecessor. Apple has done a relatively good job this round of keeping the core count/frequency tradeoffs sensible, but you still have to align your silicon choices to your workload.
Moving on to Cinebench 11.5, we lose the PowerMac G5 comparison but we gain a more modern benchmark. Once again we'll start with the single threaded numbers:
There's that healthy single threaded performance bump again. It is pretty incredible to me just how far we've come in single threaded performance since the mid-2000s. What's even crazier is that 2.0GHz Mac Pro from 2006 is only about 40% faster than a Bay Trail tablet with an Atom Z3770.
The multithreaded story is more evolutionary for sure, especially compared to the previous generation 12-core model. Here we're showing a 20% gain over the previous 12-core design. If you're migrating from a machine with fewer cores you can expect a corresponding increase in multithreaded performance. What is most surprising here is that a 2.3GHz 15-inch MacBook Pro with Retina Display (Late 2013) actually offers better multithreaded performance than the 8-core 2.66GHz Mac Pro from early 2008 in the chart above. The new 15-inch rMBP scores a 6.62 here compared to the 6.58 of that old Mac Pro.
I also shared my Firefox build test with Adam and Paul, who helped me fill out the chart below:
There isn't much of an advantage to having 12 cores here, but the new Mac Pro does deliver an amazingly quick build time compared to anything else. The new Mac Pro is good for around a 24% improvement in build performance compared to the outgoing 12-core model.
267 Comments
View All Comments
Ppietra - Friday, January 3, 2014 - link
An object with black color only implies that it absorbs visible light. Thermal radiation is mostly infrared not visible light, so being black has no consequence since there is nothing emitting visible radiation internally. Externally the surface is very reflective so no problem there either - not that there would be one it wasn’t reflectivecosmotic - Tuesday, December 31, 2013 - link
It would be nice to see storage performance of the Mac Pro SSD against RAID on mechanical disks and SSD disks from a previous Mac Pro model.cosmotic - Tuesday, December 31, 2013 - link
Including IOPSacrown - Tuesday, December 31, 2013 - link
The early 2008 Mac Pro does not support hyperthreadimg as your charts indicate. Of course I could just be doing something wrong with mine...acrown - Tuesday, December 31, 2013 - link
Stupid onscreen keyboard. I meant hyperthreading of course.Anand Lal Shimpi - Tuesday, December 31, 2013 - link
Whoops, you're right! Fixed :)acrown - Tuesday, December 31, 2013 - link
Great article by the way. I'm so on the fence about whether to get one to replace my current Mac Pro.The read is tempting me more and more though...
ananduser - Tuesday, December 31, 2013 - link
It's actually simple, it's the best OSX workstation for seemingly only Apple software that actually makes full use of the GPU setup.If your workflow revolves exclusively around FCX, it is the only workstation you'll need. If you're an average consumer wanting a powerful OSX machine you'd better get a consumer oriented imac.
PS: No need for me to mention that if you'll need CUDA and Windows then it's a bad buy.
akdj - Wednesday, January 1, 2014 - link
I'd expect OpenCL to become more and more and MORE ubiquitous as time marches on and Moore's law in relation to CPU slows...and more computing can be taken care of via screaming fast GPUs. Again. Early adopters. But CUDA/Windows options are aplenty. Just more expensive and without twin GPUs. Without PCIe storage. And.....oh yeah, their Windows boxes. At least with the MP you can run Windows...and perhaps, as we saw Adobe so quickly do with HiDPI support post rMBP release (along with hundreds of other apps and software companies)---hopefully Windows 8.2/9.x realizes the more significant 'all around' gains utilizing OpenCL (nVidia too?) than the very, VERY select software titles that take advantage of CUDA....and when they do, it's primal in comparison to what OpenCL opens the doors for. Literally. Everythingmoppop - Wednesday, January 1, 2014 - link
Considering CUDA is a GPGPU API there's no door that OpenCL opens that CUDA can't do...in fact, you could say that CUDA opens more doors on an Nvidia GPU. Nvidia also supports OpenCL since it was among the parties to help expand the api spec, but make no mistake, their flagship is CUDA.Aside from shunning Nvidia to market-segments whose software will be at least CUDA-accelerated (if there's GPU accelaration at all), my main beef with the Mac Pro, however, is that there are artificial limits placed by the design. Namely 1 CPU socket and only 4 DIMM slots.
For VFX/3DCG pros, the reality is that GPU rendering simply isn't there yet. Your PRMan, Mental Ray, V-Ray (not real-time V-Ray), Arnold, and Mantra renderers are still very much in the CPU world. When professionals buy a machine they need it to work now, and not 5 years from now. While the Mac Pro certainly appeals to portions of the pro market-segment, it was a simply foolish reason to castrate the Mac Pro.