The 2018 Apple iPad Pro (11-Inch) Review: Doubling Down On Performance
by Brett Howse & Andrei Frumusanu on December 4, 2018 10:00 AM ESTBattery Life
Apple is generally quite conservative when quoting battery life, and the iPad Pro 11 and 12.9 both are rated at up to 10 hours of web usage. The smaller model offers a 29.37 Wh battery, and the larger model offers 36.71 Wh of capacity. Both of these capacities are much lower than a Surface Pro 6, which has 45 Wh, or a typical Ultrabook, which would be well over 50 Wh.
Our battery tests are performed at 200 nits of brightness.
Web Browsing Battery Life
Our iPad achieved well over the rated ten hours, coming in at 12:13 on our web rundown test. This is a couple of hours longer than you’d get on an iPhone XS Max, and well ahead of the battery life on a Surface Pro 6 on this same test. This is one area where the efficiencies of the SoC, coupled with the operating system, pay big dividends compared to the PC space.
Battery Life Movie Playback
Movie playback is a unique situation where the workload can be offloaded to fixed function hardware in the media block, which is much more efficient than doing the work on the CPU. The iPad Pro achieved just over 15.5 hours of movie playback of a locally stored video. This is a couple of hours longer than you’d get on a Surface Pro with the same workload, despite the much smaller battery capacity.
Normalized Results
One thing we do on our PC reviews is to look at the efficiency of the device by removing the battery capacity from the equation. This shows the current gap between tablets and PCs. The Surface Pro 6 is one of our most efficient devices around, offering over 12 minutes per Wh of battery capacity, and the iPad over doubles that efficiency at almost 25 minutes per Wh. Or put in other terms, the iPad, on average, was drawing 2.4 Watts of power during the web test, and the Surface Pro 6 was drawing about 5 Watts. Considering much of the Surface Pro draw is the display, it shows you how effective Apple has been in driving down all of the power drain.
Charge Time
The other end of the spectrum is the charge time. Apple ships the iPad Pro with a USB-C power adapter with 18 Watts of output. That is quite a bit lower than you’d see on a laptop, and for example the MacBook ships with a 30 Watt AC Adapter. That means that the iPad charge time is quite long, despite the small battery capacity.
In addition, Apple ships an almost comically short USB-C cable with the iPad Pro. At three feet long, it will almost certainly be impossible to charge and use the iPad unless you happen to have an outlet right on your desk. At least with the move to USB-C getting a longer cable is not an issue, but for such an expensive device, this is a bit silly.
Facebook
Twitter
RSS
145 Comments
View All Comments
Spunjji - Wednesday, December 5, 2018 - link
Honestly wondered this too. Apple don't seem to like performance segmentation at the top-end, and I'd bet that 1TB unit uses TLC flash, so extra caching in system RAM would massively help performance and longevity. Just a supposition though.Socius - Wednesday, December 5, 2018 - link
I picked up the 1TB model for the 6GB of ram. Apps/games stay in memory far longer than they do on the iPhone XS Max with 4GB of ram. Of course, that could also be related to the way iOS is designed to perform on iphones vs. ipads. But either way, the 1TB model also has a much faster disk. If you use your iPad for productivity, I would try to find a way to justify the extra expenditure. My justification was the pending release of Adobe Photoshop, as well as trying to do some of my 4K video editing strictly on the iPad.Speedfriend - Tuesday, December 4, 2018 - link
If the single core power consumption is 3.6-4.3, what would it be at most stressed multicore?eastcoast_pete - Tuesday, December 4, 2018 - link
Writing this as somebody who is not an iOS fan or Apple fanboy, this tablet is clearly the reigning king of tablets, with the pricing to match. However, a laptop replacement it is not, despite what Apple keeps saying. The one aspect that I did find disappointing is the rear camera setup; despite statement by some of Apple's head-honchos, plenty of people I know do use their tablet (usually iPads) for photography, and for the current iPad Pro prices, Apple could have thrown in the camera setup from the iPhone XS.KPOM - Tuesday, December 4, 2018 - link
I think what Apple is saying is that it does what most people need a computer to do, not that it would replace everything a notebook does. For me, it does about 90% of what I need. This plus a low-end ultralight notebook with a big screen would be all I need right now. My employer insists on giving me a quad-core notebook that weighs a bulky 3.5 lbs. I’d take a Spectre over my PC setup. On business trips, the iPad is all I need.WasHopingForAnHonestReview - Wednesday, December 5, 2018 - link
So, email. K.crash86 - Tuesday, December 4, 2018 - link
On the GPU test - it looks like there is no change in resolution between the iPAD 1112x834 and 2224x1668 settings. Instead, it looks like the rendered resolution is exactly the same, just scaled by 2X (the @2.0?). Are you sure the rendered resolution actually changed between the tests? It seems unreasonable that the average / frame would be completely unchanged.Ryan Smith - Tuesday, December 4, 2018 - link
We're sure.Take a good look at the photos. The Retina resolution photos are far clearer since they're not being internally rendered at 1/4 resolution.
tipoo - Tuesday, December 4, 2018 - link
That's wild.*ask to render 4x more pixels*
A12X: *Shrugs*
KateH - Wednesday, December 5, 2018 - link
the author said the framerate appeared to be capped at 27fps (presumably for the sake of power consumption and thermals). So we don't know how fast it could render Civ at full tilt, only that it's at least 27fps at both resolutions. I assume this just means the GPU is just working less hard to hit the framerate at the lower resolution, but we don't know how much... Apple may have GPU design chops but even they can't magic their way out of resolution performance scaling limits, only obfuscate said limits ;)