The Apple iPad Pro Review
by Ryan Smith, Joshua Ho & Brandon Chester on January 22, 2016 8:10 AM ESTSystem Performance Cont'd
Continuing on with our more game-like benchmarks, tests like 3DMark and GFXBench are supposed to replicate gaming workloads to help determine relative performance in most common 3D games. In the case of the iPad Pro, the GPU is a 12 cluster variant of the PowerVR Series7XT GPU architecture. This is double the number of clusters relative to the A9’s GPU, which should prove to be quite impressive judging by the GPU performance that we saw in the iPhone 6s.
The iPad Pro manages to maintain superiority in 3DMark, but we're really starting to see the limitations of this test. The physics test generates non-sequential data structures with memory dependencies, which can penalize devices with lower core count and clock speed, but the workload is able to be spread across multiple cores to exploit TLP, which benefits devices with more real cores, or virtual ones (hyperthreading). We also see that the graphics test isn't really scaling well at this point as it's just too light to take advantage of the full potential of the A9X GPU. This likely also explains why the iPad Pro isn't closer to the Surface Pro 4 in performance on this benchmark, given what we know about A9X's GPU.
In GFXBench we can see the major benefits that really come with the larger GPU. It's pretty obvious here that clock speeds are basically identical when comparing the A9 GPU and A9X GPU as the scaling is almost perfectly double. In this benchmark the iPad Pro quite handily beats the Surface Pro 4, but it's important to keep in mind that the Surface Pro 4 is running a higher level of precision and that the iPad Pro is running OpenGL ES rather than OpenGL in this test, so it isn't strictly apples-to-apples (nor is such a thing truly possible at this time). Overall though the GPU of the iPad Pro is incredibly impressive, and I doubt that anyone will really have issues with gaming performance on this device.
NAND Performance
At this point it’s pretty well understood that storage performance can often be a gating factor in performance. Although caching is an amazingly effective method of hiding memory latency, for the first hit it’s mandatory to miss the cache unless you’ve managed to prefetch the data in question. The other issue where storage performance becomes obvious are cases where it’s necessary to commit data to storage first. Some cases where this is going to be obvious is app installation or iCloud restores, especially when network performance is at the point where installation can actually be gated by writing to disk rather than downloading from the network.
In the case of the iPad Pro, Apple claims that they’ve implemented a storage controller comparable to some desktop SSDs. It turns out that this controller is a familiar one, as the storage controller identifies itself as the APPLE SSD AP0128K in the case of this review unit. It turns out that everything about this SSD is identical to what we saw in the iPhone 6s as well, down the use of Hynix for at least one of the NAND vendors and the hybrid SLC/TLC architecture discussed in previous articles. In order to test how this storage solution performs we once again use Eric Patno’s StorageBench, which provides a rough analogue to AndroBench 3.6.
It turns out that in this test, performance is basically identical to the iPhone 6s. This isn’t quite the equal of something like the Surface Pro 4’s PM951 SSD, which has the advantage of more NAND dies working in parallel, but given that the iPad Pro PCB size isn’t going to be anywhere near that of the Surface Pro 4 it’s likely that this is a concession to gain better battery life. I definitely wonder what performance would be like relative to a Surface Pro 4 if the iPad Pro had a 512GB SKU, but given that the iPad Pro tops out at 128GB this isn’t really a question with a relevant answer.
Facebook
Twitter
RSS
408 Comments
View All Comments
willis936 - Monday, January 25, 2016 - link
I am very interested in energy per calculation comparisons between the A9X and the Core M. Yes Core M will beat out the A9X from a power perspective but are both within the same power budget? If so then Intel has done some impressive work.Constructor - Monday, January 25, 2016 - link
That's not even cut and dried. The Anandtech performance comparison leaves quite a number of question marks. It looks a lot as if some of the tests were written originally so autovectorization would work with known desktop compilers but LLVM for iOS just didn't catch on to it.The drastic swings between the various tests are not very plausible otherwise.
Which makes that comparison utterly useless if that's the case. And that the testers didn't even bother to check the generated code is highly disappointing.
Icecreamfarmer - Monday, January 25, 2016 - link
I just registered to post this but I have a question?How so cant you draw diagonal lines with a surface 3?
I just tried it several times with and without ruler but they are flawless?
Could you explain?
VictorBd - Tuesday, January 26, 2016 - link
Surface diagonals: The MS N-Trig pen tech manifests a subtle but distinct anomaly when drawing slow diagonal lines in that the lines waver a bit. If you search on this you can see it demonstrated. It is a genuine defect in the current tech. For my use case it is not a concern. I use the pen extensively for interview and meeting note taking (and for light sketching for fun).For my own purposes, the SP4 provides the most compelling overall device available on the market at this time: the power, form factor, desktop docking, OS and apps, ports, and pen when taken all together cannot be matched. It is my primary device every day all day.
At night when I just want to consume web, video, or music, I use an iPad Air 2. Perfect for that. I bought and returned an iPad Pro. I could never try to do production work on it. And it's price and bulk are not worth the beauty of its screen. So I'm keeping the Air for casual consumption. But for work its the SP4 (with a Toshiba dynaPad as a light backup).
Constructor - Tuesday, January 26, 2016 - link
There also seem to be problems properly following the pen near the edges of the screen, even requiring calibration by the user, apparently.The Apple Pencil has neither of those problems. It works very precisely and consistently in any direction and right up to the edges.
VictorBd - Tuesday, January 26, 2016 - link
I initially had pen issues at the edge with the SP4, but it was completely resolved for me by a pen calibration reset. The only thing left is the subtle diagonal - which does not impact me.I also note that the iPad's palm rejection isn't perfect. It allowed my palm to make marks on the screen in OneNote, and it will register finger input as drawing from your "non pen hand" as well in some apps. And right now there's no way to switch off touch input while using the pen so you can grab it however you want. (Another IOS "protected garden" limitation.)
Constructor - Tuesday, January 26, 2016 - link
Nope. The Procreate app, for instance, ignores my fingers completely for any drawing tools but I can still simultaneously draw with the Pencil and operate the UI with my fingers (such as the opacity and size sliders, or the two- and three-finger undo/redogestures.That bit about the "protected garden" is pure rubbish – iOS provides separate APIs for the Pencil and apps already make use of that.
By the way: Palm rejection (in apps where you can't disable finger touch drawing on the canvas) can be trained to some degree. if you're setting your palm clearly on the glass with a larger area touching the surface, it works best. Avoid just light touches with a knuckle of your pinkie finger, for instance (which is when palm rejection can't distinguish it from an intended finger touch), but actually fully rest your hand on the glass for drawing and trust palm rejection to filter that out.
VictorBd - Tuesday, January 26, 2016 - link
Glad to hear that Procreate has done it right. Users will benefit greatly if other apps follow suit. Until they do (and many likely will not)On my other point, I think it unlikely that Apple will either provide or allow others to provide (in the controlled garden of the app store) a utility that toggles the touch input off and on while using the pen. If you haven't used a pen tablet with this feature it may not be obvious at first. But many of those who do discover and use it find it to be a "game changer." All of a sudden your tablet can be handled like a physical piece of paper without any concern for unintended touch inputs. It is the first thing I install on a pen tablet. If iPad Pro had it the experience for me would greatly improve. But I predict that Apple won't allow it. But there is much to the iPad Pro to love no doubt.
VictorBd - Tuesday, January 26, 2016 - link
EDIT: "Until they do.... I don't think Apple will allow the touch toggle ability....."Constructor - Tuesday, January 26, 2016 - link
Your theory is completely wrong. Apple doesn't "disallow" anything!Any app can distinguish between passive finger touch and active Pencil dtection at their own discretion. The APIs already provide that distinction, and I have no idea where you get that idea from that Apple would have any interest to interfere with that.
Again: In Procreate I can simultaneously draw with the Pencil and during the same time move the size slider with a finger while the Pencil keeps drawing – there is no "toggling" of any kind.
It is purely on the application to decide how to treat fingers on the one hand (ahem) and the Pencil on the other – and both are clearly distinguishable at the same time!