The Mac Pro Review (Late 2013)
by Anand Lal Shimpi on December 31, 2013 3:18 PM ESTThunderbolt 2
The new Mac Pro integrates three Intel Falcon Ridge Thunderbolt 2 controllers. These are the fully configured controllers, each supporting and driving two Thunderbolt 2 connectors on the back of the Pro for a total of 6 ports.
Pairing Thunderbolt 2 with Ivy Bridge EP is a bit tricky as Apple uses Thunderbolt 2 for display output as well as data. Typically you’d route all display through processor graphics, but in the case of IVB-EP there is no integrated graphics core. On a DIY PC you enable display output over Thunderbolt 2 by running an extra cable out of the discrete GPU and into a separate input that muxes the signal with PCIe and ships it out via another port as Thunderbolt. Here’s where Apple’s custom PCB work comes in handy as all of this is done internal to the Mac Pro. The FirePro’s display outputs are available via any two of the six Thunderbolt 2 ports, as well as the lone HDMI port on the back of the Mac Pro.
How does Thunderbolt 2 differ from the original? For starters, it really would’ve been more accurate to call it Thunderbolt 4K. The interface is fully backwards compatible with Thunderbolt 1.0. You can use all previous Thunderbolt peripherals with the Mac Pro. What’s new in TB2 is its support for channel bonding. The original Thunderbolt spec called for 4 independent 10Gbps channels (2 send/2 receive). That meant no individual device could get access to more than 10Gbps of bandwidth, which isn’t enough to send 4K video.
Thunderbolt 2 bonds these channels together to enable 20Gbps in each direction. The total bi-directional bandwidth remains at 40Gbps, but a single device can now use the full 20Gbps. Storage performance should go up if you have enough drives/SSDs to saturate the interface, but more importantly you can now send 4K video over Thunderbolt. Given how big of a focus 4K support is for Apple this round, Thunderbolt 2 mates up nicely with the new Mac Pro.
So far I’ve been able to sustain 1.38GB/s of transfers (11Gbps) over Thunderbolt 2 on the Mac Pro. Due to overhead and PCIe 2.0 limits (16Gbps) you won’t be able to get much closer to the peak rates of Thunderbolt 2.
The impact of chaining a 4K display on Thunderbolt 2 downstream bandwidth
Here’s where the six Thunderbolt 2 and three TB2 controllers come into play. Although you can daisy chain a 4K display onto the back of a Thunderbolt 2 storage device, doing so will severely impact available write bandwidth to that device. Remember that there’s only 20Gbps available in each direction, and running a 3840 x 2160 24bpp display at 60Hz already uses over 14Gbps of bandwidth just for display. I measured less than 4Gbps of bandwidth (~480MB/s) available for writes to a Thunderbolt 2 device downstream from the Mac Pro if it had a 4K display plugged in to it. Read performance remained untouched since display data only flows from host to display, leaving a full 20Gbps available for reads. If you’re going to connect Thunderbolt 2 devices to the Mac Pro as well as a 4K display, you’ll want to make sure that they aren’t on the same chain.
If we start numbering in the top left corner of the 2 x 3 array of Thunderbolt ports and go left to right down the stack, you'll want to first populate ports 1, 2 and 5 before filling in the rest. The diagram below should help simplify:
Facebook
Twitter
RSS
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.