Why is Apple understating battery life?

Discussion in 'MacBook Pro' started by akramer, Oct 28, 2013.

  1. akramer macrumors member

    May 20, 2010
    So I've owned the following Apple laptops:
    2002 TiBook 800
    2003 17" Powerbook G4
    2006 Macbook
    2008 15" Macbook Pro
    mid-2010 15" Macbook Pro
    2010 Macbook Air

    For work I've had:
    2005 15" Powerbook G4
    early-2006 15" Macbook Pro
    late-2007 15" Macbook pro
    mid-2009 13" Macbook pro
    2012 15" retina Macbook Pro

    I guess you could say I have extensive hands-on experience with a lot of Apple laptops.

    In the past Apple's marketing materials would list an estimated battery life that was optimistic. Sitting in a dark room with the screen on dimmest surfing the web (but not scrolling much!) was what it took to achieve the times they stated. This has been the case with every one of those laptops, except maybe the 2012 15" rMBP.

    Now I've got a new Haswell 15", and it absolutely destroys their 8 hour estimate.
    Sitting with the screen at 25% brightness surfing the web looks like it'll net me 11-12 hours, proof attached.

    Why is Apple doing this? They've never been ones to understate the performance of their products, and battery life is a significant differentiator.

    Are they waiting for Broadwell to announce "now our laptops last 14 hours!" when they really only added 2 hours of runtime? I don't get it. Anyone have more insight into this?

    Attached Files:

  2. UBS28 macrumors 6502a

    Oct 2, 2012
    Doesn't Apple do the test with 50% brightness?
  3. actuallyinaus macrumors regular

    Feb 13, 2013
    actually they run the tests now at 75% now, go look at the small print point 2 http://store.apple.com/au/buy-mac/macbook-pro

    also the 15" with and without dgpu are both listed as 8 hours, i wonder if the dgpu never kicked in or if they are just listing the dgpu time and thus the igpu only version would get even more...
  4. jamets515 macrumors regular

    Mar 23, 2011
    Gloucester, UK
    I thought apple did a 'better battery test'. I vaguely remember them explaining it in a keynote. Because all their competitors quote the maximum possible length of charge without wifi and an unusably dim screen, apple decided to do a test based on real world usage - giving you a battery life spec that's actually achievable through normal use.
  5. akramer thread starter macrumors member

    May 20, 2010

    Small print point 2 links to http://www.apple.com/au/batteries/
    I can't find a description of test methodology there?
    Did they start testing their laptops differently in the last couple years? For sure my fall-2008 and mid-2010 MBPs required lots of babying to achieve the stated battery life, let alone exceed it.
  6. Atomic Walrus, Oct 28, 2013
    Last edited: Oct 28, 2013

    Atomic Walrus macrumors 6502a

    Sep 24, 2012
    I have zero insight into why they do this (I could speculate, but I don't know anything more than you) but I can say that it's definitely true this year.

    I'm doing about the same thing; Sitting here using the machine at about 30% brightness. Just reading some PDFs and wasting more time than I should on forums not unlike this one. I'm at 77% right now with a predicted 7.5 hours to go and approaching 3 hours of on-battery usage. If I maintain this level of usage that puts me in the same range you're seeing (10-11)

    This is basically what Haswell can do in theory. The closer you stay to idle, the less power it uses (this has always been true, but Haswell can reach much lower power stats at near-idle than previous CPU generations).

    To go back to speculation, I think the reason Apple quotes such low figures by comparison (aside from using 50% brightness) is simply because it would be somewhat dishonest to use ultra-light usage as the measurement for performance on a machine like this. With 4 cores and a 3.5Ghz peak (for the 2.3 anyway) the slightest nudge will knock the machine out of its Air-like power efficiency state and into serious battery consumption territory. Even something like spending too much time in one of these post editing boxes will seriously compromise the battery life and bring it much closer to Apple's estimates.

    So while it may be entirely possible to see 11 hours of usage, it's not realistic for most rMBP users.
  7. nbevans macrumors newbie

    Oct 25, 2013
    MS is understating their estimates too with the Surface. A lot of reviews of the Surface 2 are indicating 12 hrs battery life.

    Maybe they don't want to get sued?
  8. AirThis macrumors 6502a

    Mar 6, 2012
    Because they have a new fancy OS or app in the making that requires tons of power. So they now have some margin to work with.

    Just a random thought, but who knows? :)
  9. dusk007 macrumors 68040


    Dec 5, 2009
    Personally I think they still overstate or not state enough. Everybody should release 3 numbers for different usage.
    Also there is a big difference in battery life between doing literally nothing and scrolling click around. Most web browsing tests just load a new page every 40 seconds. In between the notebook can go into the deepest sleep states. If you constantly scroll around while reading or open multiple pages in the background, switch around close some again. Battery life is different. The only time I get what Apple advertises is while reading pdfs.

    Also when I am on the go that usually means outside or in more birghtly lit environments than at home at night, I want every bit of brightness I can get out of my display. So 75% is at least somewhat more reasonable. Half brightness is imo a pain at any play that isn't like a dark library. And low 25% brightness I would only accept at home at night with dimmed lights. Usually a place that I don't need too much battery life anyway.
    If I have to, I dimm. It is still better working with a dimmed notebook over with one that is dead, but this really shouldn't be the default for battery life tests.

    Afaik at 100% brightness the MBP last about 7-8h even if it does absolutely nothing.

    They should be standardized 300 nits outdoor test. 200 nits long test. <100 nits dark library max battery life.
  10. thundersteele macrumors 68030

    Oct 19, 2011
    Maybe this is to adapt to different consumer expectations. In the past maybe it was more clear that a laptop with "7 hours battery life" would only last 3-4 hours under anything but web browsing.
    Now people have their iPads that last the advertised 10 hours in most cases, and then they are disappointed if the Macs they buy can't reach the advertised battery life.

    I think they try to model light but realistic use - light web browsing and video playback. By lowballing they give some room for facebook flash games or flash videos that suck away battery life like nothing... so maybe the 8 hours are finally realistic.

    As others have mentioned, scrolling a lot instead of just looking at one site already makes a difference in power consumption.
  11. hallux macrumors 68030


    Apr 25, 2012
    I thought I was getting GREAT life out of my 2012 non-Retina, then I realized that's a cumulative counter since AC power was removed, INCLUDING SLEEP TIME. So if you take it off charge at 7 AM, put it in your bag to go to work and use it during lunch then don't put it on charge until 7 PM it'll show you 12 hours on battery.
  12. Michael Goff macrumors G5

    Michael Goff

    Jul 5, 2012
  13. iosuser macrumors 6502a

    Mar 12, 2012
    Totally. Much better to under-promise and over-deliver; it's how I like to do things in life.
  14. SimonTheSoundMa macrumors 6502a

    Aug 6, 2006
    Birmingham, UK

    They give two main usuage figures. One is a 1080p movie on iTunes, the other is web browsing using 100 top Alexa sites set to load every 20 seconds (or some other time).
  15. actuallyinaus macrumors regular

    Feb 13, 2013
    "Testing conducted by Apple in October 2013 using preproduction 2.3GHz quad-core Intel Core i7–based 15-inch MacBook Pro units, preproduction 2.0GHz quad-core Intel Core i7–based 15-inch MacBook Pro units and preproduction 2.6GHz dual-core Intel Core i5–based 13-inch MacBook Pro units. The wireless web test measures battery life by wirelessly browsing 25 popular websites with display brightness set to 12 clicks from bottom or 75%. The HD movie playback test measures battery life by playing back HD 720p content with display brightness set to 12 clicks from bottom or 75%. The standby test measures battery life by allowing a system, connected to a wireless network, to enter standby mode with Safari and Mail applications launched and all system settings left at default. Battery life varies by use and configuration. See ..."
  16. nealh macrumors 6502

    Jul 20, 2010
    what app are you using to tell how time its been on the battery?

  17. sabbyp macrumors regular

    Oct 26, 2013
    well my battery is getting nothing like that, i'm sat at 60% brightness with no apps using significant energy (just safari open) and i have 4:20 left at 95%

    something isn't right there

    haswell 2.6ghz i7, 16gb ram, 1tb ssd btw
  18. KUguardgrl13 macrumors 68020


    May 16, 2013
    Kansas, USA
    My battery life with my new 13" rMBP tends to fluctuate a lot, but I use Chrome as my browser which seems to suck battery life.
  19. Starfyre macrumors 68030


    Nov 7, 2010
    Sounds like 2.3ghz might be the way to go. That is sooo low...
  20. akramer thread starter macrumors member

    May 20, 2010
    That's Activity Monitor. Click the Energy tab and look at the bottom.
  21. akramer thread starter macrumors member

    May 20, 2010
    That is really interesting. I think that page serves up different content based on IP geolocation. My footnote looks like:

    Attached Files:

  22. nealh macrumors 6502

    Jul 20, 2010
  23. hallux macrumors 68030


    Apr 25, 2012
    Correct, and as I noted in a previous post, that's TOTAL time on battery, including sleep time.

Share This Page