Battery Life

Apple has always advertised the iPad as achieving all day battery life. I've generally found that to be true based on my usage, although on days where I use the larger iPad Pro with the Apple Pencil heavily I can find myself looking for a charger by the time the work day is over. With our 2016 mobile test suite we've rolled out a new WiFi web browsing test that is more taxing on devices than the previous one, along with moving from GFXBench's T-Rex HD battery test to Manhattan Metal. With the 9.7" iPad Pro having a similar battery capacity to the iPad Air 2 it can hopefully be expected that battery life remains the same, or possibly improves.

Web Browsing Battery Battery Life 2016 (WiFi)

In our 2016 WiFi web browsing test the 9.7" iPad Pro comes in at roughly the same runtime as the iPad Air 2. Apple has always advertised a ten hour battery life for iPads, and that generally held true in our old test which did static page loads. But in our new test the iPads simply can't last that long due to the heavier and more representative workload.

Unfortunately I no longer have the Tab S2 for comparisons to a non-Google Android tablet in this test, but when it comes to battery life the 9.7" iPad Pro lasts a bit longer than the Nexus 9 and significantly shorter than Google's Pixel C. The Pixel C is also significantly thicker and heavier, but its LTPS display also helps it to drive down platform power, and I wouldn't expect to see the iPads approach it any time soon with Apple's continued focus on driving down mass and thickness, and reliance on IGZO displays to achieve their switching refresh rate.

Video Playback Battery Life (720p, 4Mbps HP H.264)

The 9.7" iPad Pro does well in our video playback test. This test hasn't changed from the 2014 one, as my measurements found that the impact of moving to higher resolution and higher bitrate test files has a negligible impact on battery life due to the fact that the decoding for H.264 is all handled by dedicated hardware. You can see that the Pixel C still leads the pack, and the Tab S2 is close behind it even though it's incredibly thin and has a relatively small battery, and this is due to its AMOLED display using much less power in low APL videos than your typical IPS LCD on a tablet.

It's worth noting that we see a regression here from the iPad Air 2. The 9.7" iPad Pro actually has a slightly larger battery than the iPad Air 2, and I'm not sure where the difference here is coming from (possibly the display?), as our video playback test is conducted in airplane mode with no background tasks or location services, so there's not much room for software variance.

GFXBench Manhattan 3.1 / Metal Battery Life

GFXBench Manhattan 3.1 / Metal Final Frame Rate

In the GFXBench 3.1 Manhattan battery test the 9.7" iPad Pro performs incredibly well. Not only does it last slightly longer than the iPad Air 2, but the GPU performance throughout the test is more than double that of the A8X. With our old T-Rex HD test we saw Apple's devices essentially maintain the same frame rate for the entire test, as the GPUs were so fast that they were actually spending time idling. With Manhattan this isn't the case, and we see that the 9.7" iPad Pro does show signs of throttling. In general the performance is still very stable, and the throttling is not near as serious as what I've seen on competing Android tablets running the old T-Rex test with its lighter workload.

Charge Time

Smartphones have gone from taking several hours to charge a few years ago to only taking an hour or two today. Unfortunately we haven't see such improvements with tablets. While charge times have certainly gone down with Apple's 9.7" iPads, that's due to shrinking batteries rather than significantly faster charging. With the 12.9" iPad Pro Josh measured a charge time of over five hours, which is really unacceptable to say the least. With the 9.7" iPad Pro using a battery of similar size to the iPad Air 2, it can hopefully be expected that the charge time will be similar too.

Charge Time

Apple includes a 10W brick with the 9.7" iPad Pro. I actually haven't kept track of when they switch between 10W and 12W, but I can tell you that they need to start shipping these tablets with something closer to 20W or 30W. They happen to already sell something that fits the bill quite nicely, and customers would appreciate it. In the case of the 9.7" iPad Pro the charge time is much shorter than the five hours that the 12.9" model takes, but I think we really need to see improvements in how long it takes to charge these devices. When you use an iPad all the time for writing with Apple Pencil it will most certainly be nearly dead when you get home, and that means you can't use it for the rest of the night unless you keep tethered to a power outlet. Hopefully we see improvements made here with the next generation of iPads, along with the rest of the tablet market in general.

System Performance Display Analysis: Color Accuracy in DCI-P3 and sRGB
Comments Locked

144 Comments

View All Comments

  • ragingfighter - Wednesday, July 13, 2016 - link

    If $1000 for a tablet is something you could afford that's fine but keep in mind $1000 can get you a good laptop as well I don't not your choice and this is not a criticism to what you chose for your workflow so I don't want you to take it as such it's just a general opinion.
  • syxbit - Wednesday, June 1, 2016 - link

    Google should be ashamed of themselves with those internal NAND (sequential write) results. The iPad is 10x the performance of the Pixel C...
  • Wolfpup - Wednesday, June 1, 2016 - link

    I wonder if that's with encryption on or off? I sure hope that's with it on :-O

    I love that iOS just comes with it on now, and users don't need to know anything, manage it, etc., NOR worry about performance.

    I know my Android tablet was seriously dirt slow...well, dirt slow in general, and in storage performance specifically. (What blew me away was my equivalently speced Surface running close-to-full-Windows ran circles around it...which is like WHAAAAAT, the limited mobile OS is supposed to at least be faster LOL)
  • grayson_carr - Thursday, June 2, 2016 - link

    New Android devices all come with encryption on as well. And while Apple does hold a huge lead in NAND performance, I haven't noticed the storage performance of Android devices negatively impacting day to day use since the Nexus 6.
  • ragingfighter - Wednesday, July 13, 2016 - link

    I think it's the cost of having a filesystem a more Advanced interaction with it all and having the full on options of tweaking and placement and so on that also can bog down an operating system. A lot of how we perceive a ui or user interface or operating system is in which the speed and workflow comes across such as finding what you're trying to work on how fast it comes on screen how much time you save. With iOS, either not have a file system as you just have apps apps that do specific things. Depending on what type of workflow you have this can be easier or harder you do not have to look through a lot of submenus to find a file or something you're working on. Is your work something that you can transferred to the cloud and just so you have access to other devices after it's done or do you work on a tablet at home and have a come up on your computer at work?
    When you start getting into the sub areas of an operating system terminal command lines and all types of other advanced features and code and stuff that is one in a way I believe the operating system gets perceived as cluttered. Even something as simple as removing programs and just drag them to the garbage can for being deleted is something common sense and I was amazed that is not implemented across all operating systems. The complexity of having to go to your ad and remove programs list on windows to delete something still is amazing to me that people would have to do such a thing for such a basic function Especially during the year 2016. The thing that makes iOS really good is also the same thing that makes it really bad for some and that is its basic simplicity iOS is at its greatest for young children for people that have a writing type of workflow for consumption, but especially for the elderly that do not have as much technical know-how as others have in life I know I have relatives that if it wasn't for the iPad they would still be thinking of 56K not know how to check their emails and truly enjoy the web. They are not great typists not nearly there not know what the Home row is. In this regard voice dictation gives them another way to interact with their tablets which are computers and allows them to use them much better than without.
  • Hemlocke - Thursday, June 2, 2016 - link

    It shows in daily usage, too. My Pixel C went out the door to get my Pro 9.7, and after a few hours, I know I made the right choice. The Pixel was irritating, with hiccups and stutters when there is no reason for it. My Air 2 was a better tablet.
  • lucam - Wednesday, June 1, 2016 - link

    Hi Brandon,
    There is the new GFX Bench 3.1 available for IOS too; especially the new one 1440 resolution.
    It's worth trying it.
  • Oyeve - Wednesday, June 1, 2016 - link

    Would be nice if you actually FINISHED your S7 review to see those pic samples against this tablet. I own the 9.7 pro and the S7 and the S7 just blows the pro away. Hmmm, thats probably why you guys never finished the S7 review. It would crush your apple love.
  • Brandon Chester - Wednesday, June 1, 2016 - link

    I stated in the article that the sensor used in the 6s/SE/9.7" is not the best sensor in a smartphone. I would argue that the Galaxy S6 is superior, let alone 2016's Android devices. Your perception of bias is unfounded.
  • KPOM - Wednesday, June 1, 2016 - link

    Apple haters gonna hate. For whatever reason they can't accept a decent review of an Apple product, even one that points out its flaws.

    I like AT and am glad the new ownership doesn't seem to have affected its editorial independence. Kudos to Ryan and the whole team.

Log in

Don't have an account? Sign up now