• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Is Haswell Ready for Tablet Duty? Battery Life of Haswell ULT vs Modern ARM Tablets

Sweepr

Diamond Member
Interesting read.

AnandTech said:
The idea for this article struck me as I was in meetings last month. Sitting in that conference room for 8 hours straight each day would’ve killed my rMBP13 without plugging it in. The 2013 MacBook Air on the other hand did just fine. At one point there was some drama around a few power outlets not working. Much like using a tablet, I didn’t care. Even when I had only 50% of my battery charged, I had more than enough juice to get through the day without hunting for a power outlet. Given a very light usage model, Haswell ULT behaved like an ARM tablet platform. The difference being that if/when I needed more performance, it was available.

This whole situation convinced me to run a test that a few AnandTech readers had asked me for a few weeks ago: run our tablet battery life workload on the 2013 MacBook Air. Even our lightest Mac battery life workload is still heavier than what we run on smartphones/tablets, so the light workload battery life numbers aren’t really representative of a tablet usage model with Haswell ULT. Luckily our tablet battery life tests are fairly portable, so I prepped the 2013 13-inch MBA the same way I would one of our tablets: brightness calibrated to 200 nits running the very same workload as what we would on a tablet. You’ll notice two bars for the 2013 MacBook Air, one indicating its result and one with that result scaled down to simulate what would happen if it had 78.7% of its actual battery capacity - putting it on equal footing to the 42.5Wh iPad 4. With workload and performance constant, it’s safe to assume that battery life scales linearly at best with battery capacity. In other words, our MacBook Air numbers at 42.5Wh should be indicative of what we’d expect if the 13-inch MBA actually had a 42.5Wh battery rather than 54Wh unit...

AnandTech said:
56106.png


56107.png

http://www.anandtech.com/show/7117/haswell-ult-investigation
 
I just picked up a Haswell Macbook Air last week. The battery life is insane.

I really don't have to treat it as a laptop anymore where I'm watching the battery percentage, planning for when I need to charge or not, is there going to be an outlet where I'm going. It's more like how I treat my smartphone. Charge it at night and that's good enough.
 
I just picked up a Haswell Macbook Air last week. The battery life is insane.

I really don't have to treat it as a laptop anymore where I'm watching the battery percentage, planning for when I need to charge or not, is there going to be an outlet where I'm going. It's more like how I treat my smartphone. Charge it at night and that's good enough.

mmm, sounds good. I'm just waiting for more windows versions (Sorry, not a mac fan).
 
mmm, sounds good. I'm just waiting for more windows versions (Sorry, not a mac fan).

Yeah if there was a PC version with that battery life I would've gotten it. But the wife needed a laptop NOW and I figure it couldn't hurt to try out a Mac for once since it probably does everything she needs to do.
 
I wonder how much the battery life drops if you put Windows 7 on it. Apple has some kind of driver trick in OSX last time I checked.
 
The Haswell version of the MBA will achieve pretty similar battery life scores on Windows vs. OS X, but idle time will most likely be less on the Windows side.

But I do not recommend Bootcamp for someone who just wants Windows.

I am really looking forward to the Surface Pro 2.
 
The comparison between the tablet web browsing and video playback power consumption between the various platforms is quite interesting. Specifically, if you divide battery capacity by run-time to arrive at power consumption in watts you see that the Haswell MBA goes from 3.78W to 5.18W, the IVB Surface Pro goes from 7W to 7.95W, and the iPad 4 goes from 4.48W to 3.16W. (Some of the other ARM tablets also reduce power with video playback, but not to quite the same extent.)

One interesting point there being that the larger increase in video playback power consumption for Haswell (1.4W versus 0.95W on IVB) certainly makes me think that Apple made use of a PSR enabled display as such would only decrease power consumption on mostly static screen content (not video.) Another point being that you're effectively looking at a 2.72W delta between the Haswell MBA and iPad 4 between the 'active idle' and video playback power consumption. (Haswell MBA goes up 1.4W while iPad4 goes down 1.32W.) Intel's definitely doing something wrong there.
 
Thats a myth. You can install Windows on a Macbook btw. We use that here on them.

He's correct. Any macbook running windows will not have proper battery life, period. You TECHNICALLY can run windows but it is a waste of time. Apple's windows drivers leave a lot to be desired, and when using bootcamp windows you lose a lot of the battery life advantages of the Haswell. Some of those advantages are baked into drivers and firmware, which oddly enough are tailored towards OSX.

See this for more info:

http://www.pcper.com/reviews/General-Tech/MacBook-Air-11-2013-Review-Windows-Perspective


Personally, I would prefer using bootcamp Windows on the MBA but there are too many usability issues - Using windows full time on a macbook is completely not worth it; Using non bootcamp windows is also near impossible. There are something like 50 workarounds you can do to make it work on older MBAs, but on current macbooks - they're using a customized UEFI that will not allow you to boot from windows media for a fresh Windows installation. Older models had workarounds but current ones do not. So if you're using windows, you're using bootcamp. And it's worthless.

But, this is actually okay. OSX isn't all that bad to use really - it grows on you.
 
Last edited:
He's correct. Any macbook running windows will not have proper battery life, period. You TECHNICALLY can run windows but it is a waste of time. Apple's windows drivers leave a lot to be desired, and when using bootcamp windows you lose a lot of the battery life advantages of the Haswell. Some of those advantages are baked into drivers and firmware, which oddly enough are tailored towards OSX.

See this for more info:

http://www.pcper.com/reviews/General-Tech/MacBook-Air-11-2013-Review-Windows-Perspective


Personally, I would prefer using bootcamp Windows on the MBA but there are too many usability issues - Using windows full time on a macbook is completely not worth it; Using non bootcamp windows is also near impossible. There are something like 50 workarounds you can do to make it work on older MBAs, but on current macbooks - they're using a customized UEFI that will not allow you to boot from windows media for a fresh Windows installation. Older models had workarounds but current ones do not. So if you're using windows, you're using bootcamp. And it's worthless.

But, this is actually okay. OSX isn't all that bad to use really - it grows on you.

You didnt notice that I said we have these running with both OSX and WIndows? It is a myth.

Also look at the review you posted:
In order to test these claims, we ran the PC Perspective Battery Test under both Windows and OS X. Our battery test is a simple concept, but provides a peek at real world web browsing usage. The script automatically loads a different website every 30 seconds and browses the page.

I doubt they even used the same browser for example.
 
I wonder how much the battery life drops if you put Windows 7 on it. Apple has some kind of driver trick in OSX last time I checked.

Not really a driver trick. One of the major benefits of Haswell is a new power state for idle. Windows 7 doesn't support that (and probably never will), currently only Windows 8.1 is rumored to provide any support for it (from the microsoft side of things).
 
You didnt notice that I said we have these running with both OSX and WIndows? It is a myth.

Also look at the review you posted:


I doubt they even used the same browser for example.

Yes, the Windows test used twice as much power because they used Chrome instead of Safari, my god you have spotted the problem.

Ooooor, maybe OSX is tuned (ie. Apple exploits their control of hardware and software) for increased battery life. And installing Windows on the laptop shows that the advantage rapidly disappears without OSX.
 
Not really a driver trick. One of the major benefits of Haswell is a new power state for idle. Windows 7 doesn't support that (and probably never will), currently only Windows 8.1 is rumored to provide any support for it (from the microsoft side of things).

Supporting things like extra C-states is exactly what driver trick means? I don't mean it as a subterfuge or anything, it's a positive thing that Windows 7 hasn't implemented that's all.
 
The comparison between the tablet web browsing and video playback power consumption between the various platforms is quite interesting. Specifically, if you divide battery capacity by run-time to arrive at power consumption in watts you see that the Haswell MBA goes from 3.78W to 5.18W, the IVB Surface Pro goes from 7W to 7.95W, and the iPad 4 goes from 4.48W to 3.16W. (Some of the other ARM tablets also reduce power with video playback, but not to quite the same extent.)

Many of the ARM chips have dedicated video decoding hardware built in to the SoC in order to save on battery life. Intel's CPUs have been powerful enough to just use the CPU cores to decode and haven't been as worried about battery life since they haven't traditionally been put in phones.

Haswell is definitely going to make me more interested in the next Surface Pro though.
 
Supporting things like extra C-states is exactly what driver trick means? I don't mean it as a subterfuge or anything, it's a positive thing that Windows 7 hasn't implemented that's all.

So maybe we are arguing semantics. To me a driver trick is what nVidia and AMD would do to artificially inflate their performance numbers in different synthetic tests or different games (cheating in computations, etc).

On the other hand, Intels new power state is a documented processor feature that needs OS support to work. It isn't really trying to cheat a benchmark, rather it is offering tangible benefits to all users.

To call it a driver trick, to me, would mean that you think that SSE is also "driver trick" since it too must be supported by the OS before you can take advantage of it.
 
Well, maybe they are right to bring Broadwell mobile first. Maybe if Haswell mobile had come first it wouldnt have gotten such a bad reputation. I do wonder if the temperature issues that affect the desktop carry over into mobile though.
 
Many of the ARM chips have dedicated video decoding hardware built in to the SoC in order to save on battery life. Intel's CPUs have been powerful enough to just use the CPU cores to decode and haven't been as worried about battery life since they haven't traditionally been put in phones.

Haswell is definitely going to make me more interested in the next Surface Pro though.

I'm interested to know where these power usage numbers are coming from. Intel CPUs have dedicated video decoding hardware (as do AMD cpus) So I'm curious to know if it is a bad decoder implementation.
 
I'd actually been curious about this too. Considering Baytrail/Atom or whatever is on the way, I was wondering why they didn't just downclock Haswell and toss it in a Tablet.

I'm waiting on some reviews of some Haswell laptops from Windows that show some great batterylife and good form factor like the Macbook Air. I just can't use OSX. I find it too hard to do simple tasks sometimes.
 
Much more interested in Intel's Bay Trail for tabs instead of Haswell.

It would be perfect if M$ released a 10.6'' Haswell-U/Y Surface Pro and a cheaper 8'' Bay Trail-T Surface Pro tablet (full Windows 8.1 x64 instead of RT). Also:

3DCenter said:
Broadwell Y is a 5W sku that is capable as a fanless detachable form factor. Please take note that it comes in a smaller package and is not footprint compatible with the Broadwell U-Processor. It supports LPDDR3 as well as DDR4RS.

Fun times ahead.
 
Last edited:
Much more interested in Intel's Bay Trail for tabs instead of Haswell.

Well of course I'm VERY interested in that. But Haswell has brought revolutionary battery life improvements. I wonder what a Haswell specced Surface Pro would function like? Surface Pro launching with Haswell may have saw a different kind of reception as well.
 
Broadwell will make "Core" a legitimate tablet contender. Hoping that the Gen8 GPU is beefed up on the decode/fixed function side.

Very exciting times ahead. This is what happens when you light a fire under Intel's butt...we actually get some pretty darn revolutionary things.
 
Back
Top