2017 MacBook Pro: Severely Degraded Performance for More than Short Usage

2017-07-16 • SEND FEEDBACK •
Related: 2015 MacBook Pro, 2016 MacBook Pro, 2017 MacBook Pro, GPU, laptop, MacBook, MacBook Pro, Macs, Other World Computing, Photoshop, post processing, video tech

Mac wish list • all 15" Apple MacBook Pro 2017 models • MPG gets credit if you buy through those links. Suggested accessories include the OWC Thunderbolt 3 Dock and 1TB Envoy Pro EX.

MPG tested a fully-loaded 2017 MacBook Pro with 1TB SSD.

...

While testing the 2017 (and 2016) MacBook Pro, a consistent pattern was noticed in that performance declines with each iteration of a test. For example, with 20 iterations of of the Photoshop sharpening test, the 2017 and 2016 MacBook Pro decline in performance within 5 to 30 seconds, with up to 28% slower speed. No such pattern was seen on the iMac 5K or 2013 Mac Pro.

Update 16 July 2017: it turns out that this is a GPU throttling behavior. Disabling the GPU results in identical performance when tasks use the CPU instead. Apparently the GPU cannot handle sustained loads, and must be severely throttled back after at most about 30 seconds. This dirty little secret is not mention in the “brag sheet” aka specifications.

As well, there are strange delays at times with the 201 (and 2017) MacBook Pro that MPG has never experienced with the 2015 MacBook Pro—multi-second pauses launching a simple application like Script Editor. These do not occur with the 2015 MacBook Pro. Something is muddled in the latest models.

It seems fair to ask: has Apple degraded the 2017 MacBook Pro as a tool that can offer reliable performance? By making it more thin, its ability to shed heat is impaired. But it seems that it runs deeper than that, with a design which is inherently unsuited for computing tasks lasting longer than 10 seconds or so, in terms of reliable performance.

NOTE: performance is task specific, not some hard rule. The 2017 MacBook Pro is very likely to outperform earlier models for some tasks, possibly by a good margin, especially GPU-intensive tasks. The results here do not set a general rule, they merely show that a laptop is not a desktop in terms of reliable performance.

Degraded performance with sustained usage

2017 MacBook Pro, 3.1 GHz

Checked on macOS 10.12.5.

Below are times for iterations of the Photoshop sharpening test.

Observe that at first, the 2017 MBP runs the test in the 3.2 to 3.3 second range. But with later iterations, the time increases to an average of 4.0 seconds, which is a 28% degradation in performance. This is even worse than the 2016 MacBook Pro losses.

It is worthwhile to ask whether the CPU speed upgrade is a waste of money for those whose work involves sustained loads; it might be that the slower CPU runs just as fast under load, since both might degrade to some level that keeps heat dissipation at a steady state level. Or it might not be true, but MPG does not have that other model to test.

Accordingly, all benchmarks with the 2017 MacBook Pro are dubious; the performance in sporadic use may be significantly faster and in sustained use, significantly slower. And with extended use, significantly slower. MPG used 20 iterations for Photoshop filter tests, which offers a more balanced look, but users doing only sporadic “blip” tasks should see the faster speeds.

42 iterations of sharpening show a 28% decline:
2.8, 3.0, 3.0, 3.1, 3.1, 3.4, 3.4, 3.7, 3.4, 3.5, 3.6, 3.6, 3.6, 3.7, 3.7, 3.6, 3.7, 3.6, 3.7, 4.0, 3.7, 3.7, 3.9, 4.1, 3.9, 3.7, 3.7, 3.7, 3.6, 3.6, 3.6, 3.6, 3.6, 3.6, 3.6, 3.6, 3.9, 3.8, 3.7, 3.7, 3.7, 3.6, 3.5, 3.6, = 3.6

2016 MacBook Pro, 2.9 GHz

Checked on macOS 10.12.1.

Below are times for 424 iterations of the Photoshop sharpening test with results for the 2013 MacBook Pro further on.

Observe that at first, the 2016 MBP runs the test in the 3.2 to 3.3 second range. But with later iterations, the time increases to an average of 4.0 seconds, which is a 23% degradation in performance.

It is worthwhile to ask whether the CPU speed upgrade may be a waste of money for those whose work involves sustained loads; it might be that the slower 2.7 GHz CPU runs just as fast under load, since both might degrade to some level that keeps heat dissipation at a steady state level. Or it might not be true, but MPG does not have that other model to test.

Accordingly, all benchmarks with the 2016 MacBook Pro are dubious; the performance in sporadic use may be significantly faster and in sustained use, significantly slower. And with extended use, significantly slower. MPG used 10 iterations for Photoshop filter tests, which offers a more balanced look.

424 iterations of sharpening:
3.2, 3.2, 3.4, 3.2, 3.3, 3.3, 3.2, 3.3, 3.2, 3.6, 4.0, 4.3, 3.9, 3.9, 4.1, 4.0, 3.9, 4.0, 4.0, 4.0, 4.1, 3.9, 4.4, 4.0, 4.1, 4.0, 4.0, 4.1, 3.9, 4.0, 4.0, 4.3, 4.0, 3.9, 4.0, 3.9, 3.9, 4.1, 4.1, 4.1, 4.2, 3.9, 3.9, 4.0, 4.0, 4.2, 3.9, 3.9, 4.1, 4.0, 4.0, 4.1, 4.0, 4.1, 4.0, 4.1, 4.0, 4.0, 3.9, 4.1, 4.0, 4.0, 4.0, 4.0, 4.0, 4.0, 4.1, 4.1, 4.0, 4.0, 4.0, 4.0, 4.0, 4.0, 3.9, 4.0, 4.1, 4.1, 4.0, 4.1, 4.1, 4.2, 4.0, 4.0, 4.1, 4.0, 3.9, 4.1, 4.0, 4.1, 4.1, 4.0, 4.0, 4.0, 4.1, 4.0, 4.1, 4.0, 4.0, 4.1, 4.0, 4.1, 4.1, 4.0, 4.0, 4.0, 4.0, 4.0, 4.1, 4.0, 4.0, 4.0, 4.1, 4.0, 4.0, 4.2, 4.0, 4.0, 4.0, 4.1, 4.1, 4.0, 4.1, 4.0, 4.1, 4.0, 4.1, 4.0, 4.0, 4.1, 4.0, 4.0, 4.1, 4.0, 4.0, 4.0, 4.1, 4.1, 4.0, 4.0, 4.0, 4.0, 4.1, 4.1, 4.1, 4.0, 4.0, 4.1, 4.0, 4.1, 4.2, 4.0, 4.0, 4.1, 4.2, 4.0, 4.1, 4.2, 4.0, 4.0, 4.2, 4.0, 4.0, 4.0, 4.1, 4.0, 4.0, 4.1, 4.0, 4.0, 4.0, 4.2, 4.1, 4.2, 4.0, 4.0, 4.0, 4.2, 4.0, 4.2, 4.1, 4.2, 4.0, 4.0, 3.9, 4.1, 4.1, 4.0, 4.2, 4.0, 3.9, 4.0, 4.3, 4.0, 4.0, 4.0, 4.0, 4.3, 4.0, 4.1, 4.3, 4.2, 4.0, 4.3, 4.1, 4.0, 3.9, 4.0, 4.1, 4.0, 3.9, 4.0, 4.0, 4.2, 3.9, 4.0, 4.0, 4.0, 4.0, 4.0, 4.1, 3.9, 3.9, 4.0, 4.0, 4.0, 4.0, 4.2, 4.0, 4.0, 3.9, 4.0, 4.0, 4.0, 3.9, 4.0, 4.0, 3.9, 3.9, 4.0, 3.9, 4.2, 3.9, 4.0, 4.0, 3.9, 3.9, 4.0, 4.0, 4.0, 4.0, 4.0, 4.0, 3.9, 3.9, 4.1, 4.0, 4.0, 3.9, 4.0, 4.0, 4.0, 3.9, 4.0, 4.0, 3.9, 3.9, 4.0, 4.1, 3.9, 3.9, 4.0, 3.9, 4.0, 3.9, 4.0, 4.0, 3.9, 4.0, 4.0, 4.0, 4.0, 4.0, 4.0, 4.0, 4.0, 3.9, 4.0, 4.2, 3.9, 3.9, 4.0, 4.0, 3.9, 3.9, 4.1, 4.0, 3.9, 4.0, 4.0, 4.1, 3.9, 4.0, 4.0, 3.9, 4.0, 3.9, 4.0, 4.0, 4.0, 3.9, 4.0, 4.0, 4.1, 3.9, 4.0, 4.0, 3.9, 3.9, 4.0, 4.0, 3.9, 3.9, 3.9, 4.0, 3.9, 3.9, 4.0, 4.0, 3.9, 3.9, 4.0, 4.1, 3.9, 3.9, 4.0, 4.1, 4.0, 3.9, 4.0, 3.9, 3.9, 4.0, 4.0, 4.0, 3.9, 3.9, 4.0, 3.9, 3.9, 3.9, 4.0, 4.1, 4.0, 3.9, 3.9, 4.0, 3.9, 3.9, 4.0, 4.0, 4.0, 3.9, 4.0, 4.0, 4.0, 3.9, 4.0, 4.0, 4.0, 3.9, 4.0, 4.1, 4.0, 4.0, 4.0, 4.0, 4.0, 3.9, 4.0, 4.2, 4.0, 3.9, 4.0, 4.0, 4.0, 6.1, 4.1, 4.2, 4.1, 4.5, 4.2, 4.0, 4.1, 4.0, 4.0, 3.9, 4.0, 4.0, 4.1, 4.1, 4.1, 4.0, 4.0, 4.0, 4.0, 4.0, 4.2, 4.2, 3.9, 4.1, 4.0, 4.0, 4.0, 4.1, 4.0, 4.2, 4.0, 4.0, 3.9, 4.0, 4.1, 3.9, 4.1, = 4.0

OWC Thunderblade Thunderbolt SSD

Blazing fast, up to 32TB.

YEE HAH!

MPG review of OWC Thunderblade

√ No more slow and noisy hard drives!

Reader comments

Andrew A writes:

This is a sad sad day of recognition for me. I really wanted to keep on loving Apple, have their phones (7 and 7+) laptops (2011 15” MacBook Pro and a 13” 2015 MacBook Pro), 2013 Mac Pro, their Airport Express (use it with my DAC and HiFi) and Airport Extreme router, and 24” Cinema display plus several iPods. Most of all I used to really loved their software. All since 2003 when I first starting buying their equipment and stock I looked forward to their next innovation and improvement eagerly. Not anymore. Seems like everything need several updates to repair problems newly created.

Now, no more display or router, high priced crappy performance for new machines, everything is soldered and or requires a series of connectors and a dock to use all for the sake of small and light. What happened to useful and reliable??? They may have had a vision of where we should be going (at least with Jobs), but I think their vision has turned to greed and is very much shortsighted (cannot see the tree before the forest with each amputation of their product lines).

My 2013 machines may be the last Mac products I ever buy at least for quite a while. Hopefully they will last 10 years (I have a 2006 iMac that still works well; gave it to my daughter with an upgraded SSD). Most of the software I use for my visual and musical entertainment and work should be good enough for my needs for the next 5 years maybe more. Even the iPhone will soon be no different than the rest.

Eventually the only thing they will be selling is the phone at over $1,000 (my current 7+ cost over $800 with the 128GB up from $600+ for the iPhone 6+) and service subscriptions for things that used to be free (TV and music); don’t forget the fee for storing your stuff in the Cloud. Microsoft and Adobe are not too far behind and getting all of us hook on a perpetual fee service.

Perhaps that is where the world is going, everything is rented or leased with a few owning all the capital assets. Hopefully no Soylent Green in my lifetime, please…

MPG: Apple Core Rot moves to hardware? But in fact all of the prior laptops also slow down, so the 2016 15" MacBook Pro is just continuing that behavior, albeit in a more pronounced way.

Apple should be very glad that Macs are only 5% of their business because I receive emails from former Apple fans who still want to love Apple products, but are tired of being kicked in the groin.