Tested: The truth behind the MacBook Pro's 'terrible' battery life | Hacker News Reader

Tested: The truth behind the MacBook Pro's 'terrible' battery life

macworld.com

So far none of the reviews (including this one) have been that relevant for makers/hackers.

I have the new MacBook Pro and yes the battery isn't great. On average you're pressed to get 3-4 hours of realistic development use. I say "realistic" because all the test I've seen have been very artificial and don't mirror a typical hacker's environment (using Chrome/Firefox with a zillion tabs open) multiple electron based apps (hear: Slack, Frontapp etc), with sublime and terminals running.

This contrasts the 7-9 hours of realistic use I'd get with a 13" MacBook Air (which I still think is Apple's best machine ever made, in spite of it's non-retina screen).

Just a heads up the changes to the keyboard are terrible (especially the arrow keys layout). Surprisingly the virtual escape key hasn't bothered me as much as I thought it would, even in vim.

The other thing you should be aware is that this is the first MacBook where I get unintentional input on the touch pad when typing. It's really annoying to be writing and have the mouse jump the cursor position to another part of the doc. This is something that I experience all the time on non-apple HW (and scoff at), and it's really sad to see Apple's once perfect touch pad drop to that level.

If you're thinking of buying the new MacBook Pro, my short answer is "don't", see what you do about hanging out until the next revision or possibly switch platforms.

As a counterpoint, I am still annoyed by the touch bar. Whenever you're idle for a minute the touch bar powers down, which means the keys stop working. So when you're watching video first the touch bar lights up, reflecting the light onto the screen which is distracting, especially at night. After a minute the touch bar turns dark, even though the touch bar contains the brightness slider and media keys, the only keys you're interested in! So you first have to touch the touch bar, wait for it to turn on, hunt & peck for the icon you need and press it. It's remarkably bad.

The app-specific touch bar features are gimmicky and useless. So I've turned them off and I'm now using the touch bar with fixed icons, because that way at least I regain some touch-typing ability.

I expected the touch bar would be useful to input emojis, but it turns out that scrubbing through 300+ emojis on a tiny horizontal bar takes way too long. Most apps can pull up a list of favorite or recently used emojis, and that works just fine.

When in an intense debugging session I like to rest my fingers on the step-over, step-into, step-out function keys, while I compare mentally what should happen with what happens in reality. Ruining a debugging session and losing your concentration because of the damn touch bar is infuriating.

When using Windows or Linux on a MBP the touch bar is worse than conventional function keys in every single way.

The touch bar is worse still when rebooting the MBP after a system update, because the touch bar doesn't turn on until after you log in. Last time this happened the screen was at minimum brightness, but without the touch bar I couldn't turn the brightness up. And when you accidentally trigger voice control with the power button you can't hit escape to shut it up, because escape isn't powered on either!

The touch bar is after a few months of use still a miserable experience.

There is enough physical space on a 15" MBP for a full size keyboard, but instead we get minuscule arrow keys. Ugh.

If you're interested in quickly inserting emojis on Mac, try `Cmd+Shift+Space`[1]. You'll get a searchable input, no need to hunt and peck through the touch bar nor remember how to do them in various apps.

[1] http://osxdaily.com/2015/05/27/quick-type-emoji-mac-keyboard...

It feels like they should have added that touch bar as a bar in addition to the function keys if anything. That is likely off the table because of space and aesthetics though.

I think I would have preferred the touch bar as a strip on the top of the trackpad. Could use it with your thumb while typing, or as a fixed space for shortcuts when you're mousing.

Touchbar needs more customisability. The fact that now I have to make multiple presses and look into it to change the volume is making it a time waste.

I used to have a laptop with a built-in tiny hardware volume dial. It was hooked directly into the power to the amp and speakers so it worked instantly even if the machine was bogged down with loud busy autoplay video web sites or other awfulness.

I'd pay extra for another one like that.

If you don't care about the new App Controls, you can go into Keyboard preferences and set the Touch Bar to show the Expanded Control Strip.

BetterTouchTool

And?

This is exactly why I picked up the older 2015 13" MBP for work a couple of weeks back. Just trying out the keyboard and the annoyingly large trackpad for a few minutes at the Apple store was a big NO.

Is the older MBP still value for money? Probably no. But its still a better choice for most.

I have a mid-2014 MBP and have been seriously considering purchasing a lightly used one and putting it in a closet until my current one dies.

At this point I'm not worried about specs-dollars value but the development efficiency-dollars value of purchasing an inferior product (new MBP) or switching platforms (XPS with Ubuntu).

Apple sure is trying _really hard_ to get me to switch.

Apple sure is trying _really hard_ to get me to switch.

Yes. It sure doesn't look like Apple is going to relent on the silly touchbars, the $300 price bump you pay for the touchbars, the shrunken batteries, and ongoing trials they subject us to.

The only saving grace for them is that Dell XPS machines don't really cost much less. Similarly outfitted 15" units cost a couple hundred dollars less than Apple's. Pre-touchbar they'd actually cost more.

I'm considering doing the same for my personal device. I have a 2012 15" rMPB and its showing its age. I want to get a beefed up refurbished replacement before Apple stops selling them.

Surprisingly the virtual escape key hasn't bothered me as much as I thought it would, even in vim.

Put this

    ino kj <Esc>

in your .vimrc and instead of escape, just hit the j and k keys quickly to leave insert mode. It's a life changer.

Give Safari a shot. For me, it increases battery by at least 25% vs. Firefox or Chrome with many tabs open, probably more like 50%+ sometimes.

Cutting the Electron apps might also help a lot.

I can’t imagine having many terminals open takes any appreciable CPU, unless you’re using them to continuously run compilers or something.

You can also try reducing the display brightness, turning off bluetooth, and poking around in the Activity monitor to see what other processes are using lots of CPU time.

I actually do/did! I much prefer Safari for the iCloud Keychain syncing with the iPhone. (Apple actually stuffed thr keychain app up too recently where on the macs with touch bar keychain takes a few seconds to do a query). I have been a huge proponent of Safari but currently there is a real annoying bug where the URL/Search input won't register the "return/enter" key press for a URL or query. This happens once every 20 or so times and to get round it I need to close the window, reopen it and paste my query.

I've nuked the caches and removed extensions to no avail.

That annoys me too :/

Changing focus to another app and then back to Safari fixes the bug too. Still a pain.

Safari dev tools don't seem to be up to par with Chrome. Especially on the react.js front.

I just can't bring myself to use one browser for work and another for personal.

It’s hardly fair to blame Apple when third-party browsers and other applications are battery hogs though.

I do not blame them for this.

I blame them for building an inferior web browser with substandard feature support just because they want to wall people into the App Store ecosystem.

It does seem like they're changing and incorporating more JS features into the browser. Let's hope they keep up.

I want to use Safari, but its lack of developer tools (e.g. the React debugger console) make it a no-go except for light browsing.

I've actually really liked the keyboard and mouse-pad, but I came from a windows laptop. Keys give a firm click which is very satisfying for me despite low travel. Track pad is expansive and glorious, always used a mouse before on a laptop, don't see the need now except if I try gaming on it. Weirdly no problems with it picking up my wrist and jumping around.

I definitely find the up/down arrows to be annoyingly small, though.

My thoughts exactly, especially arrow keys and trackpad.

I used to rely a lot on tiptap gesture to switch terminal, browser, ide tabs but now it too sensitive as my palms always touch corners of trackpad.

Someone needs to write an app to setup ignore areas of the trackpad. I would pay for that.

> this is the first MacBook where I get unintentional input on the touch pad when typing

I don't think your hands or wrists are supposed to touch the laptop at all while you are typing though.

This reminds me of the "you're holding your iphone wrong" explanation for the signal problems.

I find that I need my wrists to provide stability so it does not slide off my lap.

Really? Why would I not want them to?

Well I'm not a doctor but it seems to be what many keyboard manufacturers say, and I guess they've thought about it. Das Keyboard for example say your wrists should be well above the level of the top of the keys, which is the level the trackpad is at.

http://www.daskeyboard.com/blog/wordpress/wp-content/uploads...

Why not?

I don't know. But all those guides to correct posture using a computer say it.

http://media.istockphoto.com/vectors/guide-to-correct-hand-p...

Interesting. I'll have to pay attention to where my hands go next time I'm at a keyboard.

This sounds about right. They improved the efficiency of the machine overall and they felt that they could decrease the size of the battery and retain the same unplugged runtime under "normal use."

Add a "pro" workload with more frequent CPU or GPU intensive tasks, then those efficiency gains count for less and you start to notice the smaller battery.

Did they make the right choice? The smaller battery does make the machine lighter and probably contributes to the additional thinness they were able to achieve, but you can also make a good argument that for pro users, heavy workload battery life matters more.

You would think they would've tested the battery life of a "Macbook Pro" for "pro" work loads, wouldn't you?

If Apple has seen that most people buy their Macbook Pros as Facebook machines, then maybe launch a new line that actually targets the pros, and give it another name for differentiation purposes.

At max CPU it's 15min less battery life. That's really not that much different for any real purpose.

MacBook Safari?

SafariBook Pro?

I got a non-touch bar 13" MBP and the tests back up what I've experienced over the past couple weeks. Normal usage the battery is stellar, do something high CPU bound for a long time and you'll suffer.

Coding I've been doing in my spare time (Euler problems) the battery life has been 10 - 12 hours with medium brightness. Stellar compared to anything I've had before,

Using the coconut battery app though I can see when my CPU is taxed (churning out truncated harsahd numbers or w/e) I see the estimated battery life go down to about half that.

I think that's pretty fair for the machine, and I doubt any other small notebooks would fair much better under intensive tasks.

Looks like what really ate them was the 15" model could get the discrete GPU stuck on without cause, which kills the battery.

I have been using gfxCardStatus https://gfx.io/ for years. It offers a finer level of visibility and control over what's going on with my GPUs.

Yep, gfxCardStatus comes highly recommended. I still don't understand why Apple has a "Keep the dGPU on always" option in power settings but not "Never use the dGPU".

There are also a lot of people with bad VRAM or a busted GPU that use it to keep their machines from kernel panicing.

I've been using this on my 2012 Macbook Pro for years, always set on the integrated graphics and have never looked back.

Thanks - this is amazing. I can't believe I hadn't seen it before.

How are you finding only 2 ports affecting your workflow, if at all?

I've mainly used it for personal python projects and have found two ports fine. When it gets dicey is if I'm using an external monitor and backing up on a thumb drive, a third port would be nice for charging. That was just one situation, been fine besides that, I just don't use ports all that much.

When I switch over to it for work (web dev / project management) I am going to get a hub like this, https://www.bourgedesign.com/archub/ so I can have external storage, a monitor, and anything else all gloriously connected through one cord.

I have a new 15" MacBook Pro with touch bar and can confirm that this OS X update (10.12.3) made a very noticeable improvement on my battery life. It went from 4 hours on moderate usage to at least 8... This relieves the majority of my anxiety with the new MBP and I'm now really happy with my purchase.

Yup, I see 8-10 hours of real use since 10.12.3

Seems crazy that a software update can make that big of a difference. It almost makes me think they held this back on purpose.

Extremely doubtful, considering how much negative PR they've gotten on their new line of MBPs because of this... The costs must far outweigh the benefits.

Do many people do "pro" workloads on a battery? I always have a charger with mine and frequently lots of peripherals; USB keyboard, mouse, monitor, external HD, etc. and if I'm running 4 docker containers and a full stack of software, it's getting pushed pretty hard. Not sure I'd really expect much of the battery when you're whipping the machine like that.

Do you commute? I often run unittests and re-compile whilst on the train.

Depending on the tech stack, these things can be heavy, or unexpectedly heavy.

I hadn't considered that. Commuting by bike there's not a whole lot of unit testing going on for me

I have actually seen someone using a laptop while biking. Not that I'd advise it.

Yes, very much so. Editing images on a train after a photo shooting is a great way to gain some time and do something productive while on commute. Besides, I take advantage of things still being fresh in my head, instead of trying to remember them on the next day

What, you don't have electrical sockets on trains? In normal 1st world countries we have sockets in all train seats and even on inter city buses.

I don't know, and honestly don't want to, what a pretentious sentence like “normal 1st world countries” mean to you, but many countries with world leading HDI have only two sockets per cabin. Usually the two seats by the window.

My Google-foo isn't working so I can't provide the link, but I read somewhere that USB-C does't provide enough power so it will actually discharge your battery while plugged in if you run it under heavy CPU/GPU load for an extended period of time.

Can anyone confirm this?

I don't think that's likely to be the case. It comes with an 87 watt charger, but USB-C is specced to provide 100W. Surely if it needed more power, they just would've included a higher powered charger?

However, I just saw that Twitter thread from MKBHD linked below, so... maybe it's true. Perhaps Apple simply failed to include a high powered enough adapter and should've gone to the full 100w.

[Edited with more relevant info].

USB-C can provide as much as 100W, which is more than the MacBook needs. However, that does not mean the cable and charger you're using will.

Same thing with other power standards.

I can't confirm from my own experience, but popular tech reviewer MKBHD tweeted something similar[0].

[0] https://www.twitter.com/MKBHD/status/827182930954973185

If you read his followup tweets, he wasn't charging with the MBP charger, but off of his monitor.

Ah I hadn't read that far down. Thanks!

That was it, thanks.

edit: seems to be the 5K monitor power output though. Another reason why Apple should still make monitors. Besides, you know, being able to use them next to Wi-Fi signal.

Mimics what I've seen on a maxed out 15" MBP. Author killed the battery in 83 minutes running a CPU benchmark and in 81 minutes with the GPU on Tomb Raider.

Browsing the Internet, I easily get 4-6+ hours of battery life. If I switch to coding (RStudio IDE seems to be very intensive) or similar, battery life drops precipitously.

I still dislike the keyboard after a few weeks of using it; I briefly switched back to my old MacBook Air to move some files a few days ago, and it was really much nicer.

Migrating from a MacBook Air, the biggest improvement (by far eclipsing anything else) was the screen. It's incredible, although I can't compare it to the previous generation MBP. The touch bar feels like a gimmick; I rarely use it. At least it doesn't get in the way, and the virtualized escape key is totally fine. I've also had palm rejection issues with the enormous touchpad that I never had with my MacBook Air.

The weirdest issue I am having is that programs crash after resuming from sleep. RStudio crashes almost 100% of the time. It's hard to interpret the error messages, but I believe it might be due to a bad RAM chip ("memory not mapped"). No Mac I've ever owned has done this before. I'd really appreciate it if anyone had advice for this! It's getting to be problematic.

The RStudio crashing bug should be fixed in daily builds. https://dailies.rstudio.com

I'm a heavy rstudio and Python user and have been wondering how the new MBP would perform with those apps rather than just video, thanks for posting

Python seems fine generally, but I don't use a Python IDE

I get at least 8-10 hours on my tMBP 15". I don't mind the keyboard. I this point it seems like hating on the MBP is 'free karma' and the 'in-thing' to do on HN. My experience has been otherwise.

I wish I could get those numbers! I only ever achieved those on a 13" MacBook Air.

>> 'free karma' and the 'in-thing'

I hate to admit it but I have an "Apple Problem" purchasing pretty much everything they've made. I'm event wearing right this moment a WWDC Apple cap! What I'm trying to say I'm naturally much more biased towards Apple. It's safe to say that this is literally one of the worst Macs I've had (stability/battery/experience wise), and man I've had many! The device as a whole seems really retrograde, and I'm hoping they can get it sorted out via firmware/software.

Here's an example of the bugs I've seen:

- Kernel-panicking when unplugging it from the LG 5K display. - Not waking up from sleep after unplugging it from the LG 5K display. - Randomly restarting when in sleep mode

I haven't seen the above issues "overly reported" like the battery issue is.

After going through a few of these articles, this is my very general takeaway summary:

"The new MacBook Pro's battery life is not as good as the previous generation but is still decent for its class. The decrease in battery life, compared with the older model, is mostly due to a brighter screen, better cpu/gpu, and slightly smaller battery. Ultimately, it is up to the user and the user's use case to gauge the actual battery life and tradeoffs of upgrading to the new MBP."

My only caveat is that the improvements to cpu/gpu typically result in lower power consumption requirements. The screen is also lower resolution (compared to 4k) and with brightness user-configurable, not sure how much it is to blame.

I think the bulk of the issue is the almost 25% smaller battery (as noted in the article). So the real trade off is thinness and weight for battery life, not performance for battery life which is more typical of competing laptops.

I don't think the new MBPs are bad machines, but I probably won't be upgrading my current one to one of the new ones.

But sales figures seem to indicate the thinness trade off appealed to a broader range of buyers...so ultimately it was probably the right business decision for Apple.

Video playback is no longer a benchmark with Hardware video decoding units that let CPU practically shut down 80% of die space. People actually expect to USE the laptop like a computer, not like a movie watching on a toiled tablet. As the last test shows using CPU shows new laptops to be worse than the 2015 models.

Btw Did this macworld.com writer run out of bytes on the page? was that the reason he had to split this article in two? or was it just for additional ad impressions?

Yeah, I'd say video playback is no longer a benchmark because if I'm going to watch video, I'll be doing it on my iPad, not on my laptop.

Or if they're going to make video watching a benchmark, at least make it use modern streaming services like YouTube and Netflix. Who downloads a movie to their drive and watches it from there anymore?

I'm not a Mac user, but bravo to the author for doing a really good analysis. I run a skylake-based Asus running Ubuntu with Gnome and it performs well, but the battery life does vary more than my Lenovo/haswell did. I'm able to get minimum 5h of dev, max 10h, depending on a variety of factors, without using powertop tuning. This post reminded me that my suspicions about Gmaps are worth investigating. Try to always close maps... But now I will be more careful!

I'm going to throw something out there that I think might not be agreeable to many, but I stand behind it either way.

Old Mac line

MacBook Air - "light" workload

MacBook Pro - "pro" workload

---

New Mac line

MacBook 12" - "light" workload

2016 MacBook Pro - "pro" workload

If you think of it in this light, the "Pro" moniker is very fitting. The 2016 MBP is very much the Pro version of the 12" MB.

Yes, it's not as good as older MBP for many, but it still very much earns its name vs the 12".

I'm surprised nobody has yet brought up the update that just came through which fixes issues in graphics card switching in new MBP's. If it preferentially uses the discrete graphics more than it needs to then, well, you know, battery drain problems.

Nobody other than page 2 of the article you mean?

I've been using one of the new 15" with the touch bar for about 2 weeks now.

I agree there's a lot more hate in user forums than on professional review sites. I don't know that either are 100% honest... but I think it's because for so long Apple set the bar so high. My biggest gripe with my 2016 MBP is that it doesn't feel like there's any room to grow. Every MBP I bought to date felt like it was way ahead of the curve. The new ones... meh, they are fine. Just nothing special. (I wouldn't have bought it for myself, but since work was paying...)

1) I don't notice the battery life being any better (or worse). When I'm just surfing the web / doing emails... it lasts a good 4-5 hours. I always plug in when I'm at my desk and I can certainly go between meetings fine. When I play a video, fire up a game, or do a site crawl (for a 250k+ page site) the battery doesn't last. The old ones didn't either.

2) Keyboard took some getting used to... but I do like it now. I really didn't at first. And I didn't like it until I went back to use my old computer and the old keyboard just felt soft and loose -- the old-style keys had more give to them, now typing on that feels like trying to drive a car with huge tires on.

3) Dongles... I miss the MagSafe power plug... I really wish Apple would have left that. But I bought a little hub for all my USB products and it's fine. I have another dongle for HDMI cords. I don't love the new ports, I see no benefit to them... but, I get it -- Apple wants to push the envelope where it can. I carry a lot more dongles with me now that I have at any point in history. Not a fan of this situation at all. I don't want to carry a bag full of wires with me.

4) My old MBP had an upgraded graphics card, and the new one is just stock. I think my 2013 with the upgraded card handles games better than the stock 2016... but that's not fair since I boot to Windows on my old computer. Macs aren't for gaming... I know. But like... sometimes shooting a few aliens in XCOM is a good way to kill time when a client bumps the meeting back. =P The fact it can't run games from 2012 at max settings... this does bug me because of how expensive this computer was. It handles "medium" settings about as well as my 2013 with the upgraded graphics card did...

5) I don't use the touch bar. I just have it set to show function keys most of the time. I miss a tactile "esc" key. I don't know, it's not really something that I think about or have a gripe with. It's just different, but I can't say it's better or worse than the old physical function key row.

6) The new touch pads are way too big. I'm still bumping it when I don't want to. I have no idea how I'm supposed to use a touch pad this large... the old ones were perfect, this was a clear case of "fixing what ain't broke."

>1) I don't notice the battery life being any better (or worse). When I'm just surfing the web / doing emails... it lasts a good 4-5 hours. I always plug in when I'm at my desk and I can certainly go between meetings fine. When I play a video, fire up a game, or do a site crawl (for a 250k+ page site) the battery doesn't last. The old ones didn't either.

That's actually sort of surprising to me, I get like 15-16 hours just web browsing.

Sorry I wasn't clear -- I always plug it in when I'm at my desk, and I'm hardly ever away from my desk for more than 4-5 hours. So I know it works that long at least. Past that... I mean I haven't deliberately not plugged it in just to test...

I got through a movie and a handful of Rick and Morty episodes while on a flight... Drove home from the airport late so I just went to sleep... hit the "Lower Power Mode" warning as soon as I opened it the next day. It wasn't plugged in that night... Battery life seems reasonable. It doesn't seem amazingly better than my old MBP.

I've seen MagSafe-like USB C third party cables.

I wish Apple went this route. They could place the magnet between the cable and the connector. It'd be a win-win, safe & standard.

Reply on news.ycombinator.com