Windows Drains MacBook's Battery; Who's To Blame?
ericatcw writes "Users hoping that Windows 7's arrival will mean less power drain on their MacBook laptops may be disappointed, writes Computerworld's Eric Lai. Running Windows 7 in Boot Camp caused one CNET reviewer's battery life to fall by more than two-thirds. But virtualization software such as VMware Fusion suffer from the same complaints. Some blame Apple's Boot Camp drivers (the last ones were released in April 2008); others lay the blame at Windows' bloated codebase. With Apple and Microsoft both trying to avoid responsibility for improving the experience, Windows 7's reported improvements in power management will be moot for MacBook users for a while."
Windows Drains MacBook's Battery; Who's to blame?
I blame Microsoft. Much like the title, I was expecting Windows 7 to actually recharge my laptop battery, not drain it.
This is a whole new and special kind of whining.
/. has reached a new level.
Waaaaahhhh!!!
I have a new MBP and use Fusion. I have an XP image and a Vista image loaded up. I have not noticed any unusual power drain, but that's kind of to be expected, IMO. Also, I have to question the wisdom of using a VM session for more than an hour or so on just the battery.
I can see some instances where this would be an issue for some, but this seems like senseless "hating" to me. No, I'm not trying to troll or anything else, I'm just having a hard time figuring out why someone would spend a long-ish amount of time in Fusion running a guest OS on battery power. It seems obvious to me that there are issues running a non-native OS on a laptop designed for a specific OS...
Sent from your iPad.
Macbooks are essentially the same hardware as Windows machines, down to battery capacity. It is unlikely that a "bloated codebase" would chew through the battery like nobody's business on one x86 machine and suddenly become perfectly benign on a practically identical x86 machine. Bloat doesn't magically appear when you put an Apple logo on something.
No kidding!!! What do you say at this point?
I have a MBP 5.1, one with both the on-board and discrete Nvidia cards. OS X switches between them depending on whether it is going for power savings or performance.
The drivers for Windows XP and Linux do not seem to have this ability. When I'm doing nothing but surfing, I get about 4.5 hours of battery life in OS X, but only about 2.1 hours in Linux (Ubuntu Jaunty) and Windows XP.
I always assumed it was the inability of XP and Linux to switch to the on-board graphics card.
Just another proletarian malcontent.
No matter how bloated Windows is, battery life is only a function of ACPI drivers --- bootcamp's fault
FTFA: Other than that, Windows 7 has been working great on my MacBook Pro... It looks good, too, even prettier than when it is installed on PC hardware.
This reminds me of the iPod Nano review here at Slashdot that claimed that the Nano sounded great, even in a moving convertible with the top down. (http://apple.slashdot.org/article.pl?sid=05/09/08/1439244)
Yes, it's the Apple magic that makes the software look better.
How can we know that the battery isn't simply returning strange battery level information to the OS that OSX knows how to parse but Windows doesn't? What a strange review.
Boot Camp just resizes the hard drive so it can accomodate a Windows install and then you are able to dual-boot your system. It's also possible to install Linux on the other side for example. So it seems like Windows has an issue with the Intel or NVidia chipset, the processor or just plainly consumes more resources than Mac OS.
A good comparison would be to install Linux on the other side and see what it's battery life is then. Mac OS X offloads a lot (all) of the desktop rendering to the GPU while the Windows XP desktop doesn't and although Vista's top-end version does, it is offset by the amount of graphics that need to be rendered and the low-end version still doesn't.
There is a reason that the battery dies quicker and since there is no layer of Mac OS X between Windows and the hardware I doubt it's because Apple did something wrong. It's either Windows or the Intel or NVidia drivers. You can't really compare VMWare or Parallels performance because it's running Windows on top of Mac OS X, it is of course going to consume more resources.
Custom electronics and digital signage for your business: www.evcircuits.com
I have a slightly dated Macbook with an integrated Intel graphics chips. Has anyone with similar specs tried to run Windows 7 on it? If so, how does it stack up against XP in terms of performance and responsiveness, and how does Windows 7 fare in a VMware session?
Can you expect "power savings" when VMware is running? You are basically running two computers at once.
One word : games. Unless you have a very high spec machine, VMs are next to useless for any modern games.
upon the advice of my lawyer, i have no sig at this time
Running Windows XP dual-boot on a MacBook Pro (what you people call "boot camp") also drains the battery a lot faster than OSX. I'm pretty sure Apple didn't put much effort into making sure all the hardware drivers worked anywhere near as well under Windows as they do in OSX. (additionally, I've seen display driver quirks and more iffy trackpad operation)
That's actually a designed feature. Maximum CPU usage is a terrible drain on environmental resources, and the extra heat generated by the CPU contributes to global climate change. So to encourage users to use less CPU power (or to focus on using it in bursts rather than constantly keeping the CPU pegged, Apple used a slightly underpowered power supply.
I thought it was a joke too when I first heard it from my friend at Foxconn (who worked on the HW design with Apple). He told them they would have this problem (and could solve it with a very simply HW change), and they explained that this was a deliberate design decision. So sometimes what looks like a bug really isn't, especially when it comes to Apple.
1: have your OSX drivers switch between on board and dedicated GPU as needed.
2: Make the drivers for every other OS use the dedicated GPU constantly even if there's no real need.
3: Claim the sucky battery life is MS' fault and that their OS is poorly programmed
Apple and it's customers are the only losers if something doesn't work on the Macbook. Microsoft never claimed it would. This situation is very similar to the Palm Pre / Itunes fiasco. If you're a Palm Pre owner, just STFU if Itunes doesn't behave the way it should.
First, a quibble with your argument. You do not pay a "massive premium". Depending on what product you bought, you paid a slight premium or slight discount vs. a similarly-spec'd Dell.
Second, it should be pretty clear why one would occasionally need to run Windows in native mode. Aside from saving the additional cost of virtualization software, Boot Camp simply runs some programs faster. Also, if I hand my IT department my notebook to configure and it is in Mac mode, they'd have absolutely no ability whatsoever to load it up with the VPN software since they don't do Mac. Put it in Windows mode, and they give no complaints.
W..w..W - Willy Waterloo washes Warren Wiggins who is washing Waldo Woo.
The problem is not specific to Windows or MacBooks. Many developers code as if the only machines that will run their software are permanently el-grid-connected servers or workstations. Polling loops with insane timers (like 1000hz), and they also take the advice "don't optimize prematurely" to mean "don't optimize unless you are payed for it". Re-drawing the display even if it is not needed at all, copying data structures all over, etc. No wonder batteries drain.
In this case I believe all three are to blame - neither alone is the culprit - I mean Windows usually is compatible with real hardware enough to last couple-three hours on an average laptop battery doing average desktop stuff, MacBook is about the same. Probably BootCamp taking battery awareness too lightheartedly and/or unable to optimize for specific cases like virtualized Windows code running.
This is /. ain't no ladies round here.
All the BootCamp tool really does is partition the drive. You can just live partition the disk in Disk Utility and then install Windows, Linux, Whatever. The BIOS fakeout is something that is part of the EFI to begin with, not part of BootCamp.
We have an explaination!
No kidding!!! What do you say at this point?
Seeing as both Windows 7 and VMWare are affecting things, maybe virtualization and power saving just aren't all that compatible.
I blame intel for this one.
Almost correct. They buy it, then complain when power-saving features won't work in an unreleased OS with driver support in beta (if any).
Comment removed based on user account deletion
On latest gen (nv9300 based) Mac Mini, I have installed Win7 64bit. It installed all the drivers and even clever to figure mainboard driver giving direct link to nvidia driver exe which is absolutely a very serious risk but anyway...
The ATA chipset driver is missing from Win7 since Apple didn't really put nv9300 chipset in exact way. So, it falls down to non DMA generic MS driver. Every single byte transferred to/from disk is guaranteed to use massive CPU along with horrible (down to 15MB/sec from 70MB/sec under OS X) slowness.
So, if Macbooks have similar issue with Windows 7, it could be same issue. As they are battery powered, it would be visible in battery life too.
BTW, there is no point testing Windows 7 until Apple releases boot camp for Windows 7. Apple computers aren't really PCs. If MS was really clever and wanted Windows 7 to be _really_ tested, they should have printed a very clear privacy policy on screen and actually make machine report all kinds of anonymous stats. That way, they could really figure what is going on. For example, a core duo powered 2009 machine shouldn't really max to 15mb/sec with a SATA 2 drive.
I couldn't even find something similar to bugreporter.apple.com when I wanted to report issues. All I saw is a stupid forum which beginner level MS engineers are monkeying with templates. They even made their own wrong answer as 'answer to the issue' while it would create massive compatibility problems in one occasion.
My wife uses a MacBook Pro and switches between Vista and OS X with Boot Camp (mostly using Vista). When the computer is idle in Vista, I've noticed it quite often thrashing the hard disk for many, many minutes and repeats this at intervals, like every half hour or so. On OS X, it never does this - sleep is sleep and the thing is always quiet. I wonder is this behaviour (whatever it's for) is the cause of the power drain?
I write driver level embedded code for a living. Everything from bootstrapping embedded linux to SoC level power management.
Power management is usually the last thing to get done (if at all)... why? Because management usually sees it as icing on the cake. Attitudes are typically just make it work and we'll ship a bigger battery to make it last. Or we'll ship an upgrade in 6 months, if the product starts to take off and we decide to fund further development.
Time to market is everything.
Power management is also really hard to get right 100% of the time. It's really hard to debug code/hardware where stuff is shutting itself off, or worse, a controller uP is shutting you off unexpectedly.
It has NOTHING to do with 'bad code' or 'shitty programmers'. It's just management grinding down on the engineers to do it: better, faster, cheaper, pick two. Usually faster and cheaper win.
http://nvidia.custhelp.com/cgi-bin/nvidia.cfg/php/enduser/std_adp.php?p_faqid=2243 Pretty straight forward. Regular PC laptops with the dual/triple gpu's can use Nvidia's Hybrid SLI.
Even funnier, did the author install Vista/7 to a 1.42 Ghz machine with 133Mhz system bus and 32MB ATI card and gained massive performance compared to XP?
That is what Mac Mini G4 users experienced when they upgraded to Leopard...
CBS should arrange a meeting with all the editors, authors of CNET and simply remind them they aren't the failed TV channel who is alive with MS money anymore, they don't have to be MS fans in absurd degree. If they don't fix this attitude soon, very soon, their cool domains will be sold to some porn site and they will be laid off. This is really getting beyond funny.
You would be absolutely correct if Windows 7 was SUPPORTED with Boot Camp. Damn thing can't even get rid of "boot menu" as MS is fixated to partition 1 for booting.
What suggests you that Apple is a generic PC anyway?
Keep in mind this is Apple we're dealing with and we know how vicious they can get when someone tries to step in and actually compete on one of their platforms. My bets on a line of code in the bios that says something like (in pseudocode of course):
I won't necessarily argue with the fact that Apple can get vicious, when dealing with iPhone/iPod developers via the AppStore they can often take that extra step to being downright stupid as well. Even some of the developers who are assigned bug reports form their bug tracking software could do with some lessons in manners when dealing with customers who go out of their way to report bugs in Apple software. But BootCamp and the associated drivers were made by Apple. If the fact that Windows runs on OS X really bothered Apple that much they would have killed BootCamp long ago. Not that it would do them much good. If they killed BootCamp you would probably have half a dozen software companies and FOSS projects stepping in to provide replacements in a matter of months necessitating a followup campaign of strongly worded nastygrams from Apples legal weasels. Personally I rather doubt this will ever happen. The most likely explanation for these BootCamp performance problems is not a grand conspiracy but rather the simple fact that the Windows BootCamp drivers get a lot less attention development effort from Apple's developers than their OS X equivalents do. I doubt that if the situation was reversed, that Microsoft would be in any hurry to make sure their OS X drivers were as good and highly optimized as their Windows drivers.
Only to idiots, are orders laws.
-- Henning von Tresckow
To each their own.
That's what he said.
the battery manufacturer and apple.
their battery doesn't last long enough under load.
Apple's mac book doesn't keep the battery cool enough under load.
They're using their grammar skills there.
of course the users who buy a MacBook, get a great OS shipped on it and want to install - agh - windows.
For all the Windows hating that Apple fanboys do, I can't believe they are really angry that they can't run Windows for long periods of time. In a VM. On battery. As a BlackBook owner running XP and 7 in Fusion on the occasions I need it, I understand the need to run Windows occasionally. But if you really must use Windows for so long that you notice battery depletion, you should just go buy a Dell.
"Consuming Internet bandwidth since 1991."
Apple does not do power management in the main processor. They use a separate little processor to do this. This little processor will tell the CPU to slow down. If the CPU does not slow down -- say because it is not running an OS that understands the commands, the little processor may halt the clock of the CPU for a while.
It makes sense that running Windows under VMware could give better results.
Religion is the main cause of atheism.
Why use BootCamp? Because sometimes you need to have Windows.
I have a Buffalo NAS - yeah, I know, it seemed like a good idea at the time - which crashed after a power outage. To get it into a state that would permit me to remove everything from its drives I needed to reinstall the OS. The installer only works from within Windows. Not Leopard. Not Linux. Windows. Without BootCamp, it would have taken me a whole lot longer to it working.
I also have other equipment that has installers that work only from Windows. It sucks, yes, but that's the way it is...
Hate to disagree with an expert but I have seen several cases where badly designed code directly affected battery. I am on Linux, and have got a habit of peeking into its source code, after checking with Intels "powertop" utility. Recently, I found out for example that ffmpeg uses a 1000hz timer to play a HTTP-delivered MP3 stream with bitrate of 56Kbps. Stuff like that. Of course not always it is due to bad code, but often it actually is.
Also, I do application software developing for a living, and in my case there has been minimal occurences where said manager actually came back and revised the software to include better powermanagement. Granted, they do exactly as you described - "let us first make it work". But after it is shipped, nobody cares for the code. Of course, many may say that I have been employed by really shitty managers/employers, and it would partially be true.
So, it is NOT JUST management. It is a multitude of factors, involving management AND coders. Both affect each other as well.
Faster and cheaper is the curse of our times. I hate that shit :-) I acknowledge its necessity in practical living, but it should be a temporary step on the way to "faster, cheaper, better" but it usually stops after the first two qualities are achieved.
I've used both Macs and Windows machines for quite a while now, though I always used a Mac in my private life. I'm in IT, so not using a Windows machine is simply not an option in many cases.
However, at home, ever since I started using iTunes and Mac OS X I noticed that when no sound is playing from my laptop (first a Powerbook, then a Macbook) it started zooming. This was probably because my stereo wasn't properly grounded at the time. As soon as the computer made noise however, it stopped. If I turned on iTunes it stopped! When the album I was playing was finished, it started up again.
I've always guessed that this is because of the way Mac OS X will manage memory. If something isn't used, why bother supplying power to it? Since, interestingly, this phenomenon doesn't occur when I boot up Windows XP or Vista, meaning power is always supplied to a jack not used.
If they turn off many other things 'not used' atm and switch them on as soon as a cable is detected or something similar, imagine how much power you could save ...
Just a thought.
Life is great! (as told by Lady Susan)
In all likelihood, there are some power-management features of the hardware that are better controlled in OS X than in Windows. The ACPI and CPU-based power management should be pretty standard, but how good are the Windows drivers for the 9400M? There might be updated ones on Windows Update or nVidia's site.
The MBP has a big battery (60 Wh), but still, I see similar things in Linux vs. Windows. Even my cheapo Dell Inspiron with a 48Wh battery will last 4 hrs in Windows 7, but only about 2.5 in Ubuntu.
1. Write Windows 7
2. Drain MacBook battery
3. Profit!
Perhaps NOTHING was the wrong choice of words.
If its not open source, blame everyone.
If that doesn't work blame the user (for not using open source).
OMG... proprietary hardware, a proprietary OS, and a proprietary BIOS from 3 different vendors don't play nice together? Who would have thought?
I read an article recently that benchmarked a netbook with both Windows 7 and XP. The differences were minimal, nearly identical in all aspects. XP came out slightly ahead on every test though, by 1 or 2 points. Windows 7 has more features than XP and managed to give the user the same experience in responsiveness and battery life. I'd say that's an achievement but they could really further improve battery life further.
My Acer Centrino-based laptop running Vista used to get 1:45 max at any given time under normal operation (wi-fi on, screen dimmed) while my Macbook Pro would get nearly 5 hours under the same conditions. I did, however notice a significant drop if watching video. This leads me to believe the languages used to write these particular applications on the OS's are more efficient with their use of resources such as redrawing. Video obviously needs to reraw a lot more than my browser window so that drop was expected but I should see that parallel in Windows as well. This was not the case.
Your IT dept. is incompetent.
cmd-q.co.uk - some sort of stupid fucking internet bullshit
yeah I'm with you there, the touch pad is so much better in OSX (you can actually tap to click, but in windows you have to use the button....)
No they aren't, really. They just made a business decision to support Windows only. They do a reasonable job with the Windows machines.
W..w..W - Willy Waterloo washes Warren Wiggins who is washing Waldo Woo.
At least on the Mac Book Pro, Windows is stuck using the more powerful and power consuming Geforce 9600 card. The boot camp drivers doesn't contain the option to change to the 9400. In OS X, using the 9600 all the time also gives quite poor battery life.
A lot of commenters are saying that the difference in battery life is down to OS X using the integrated graphics, and Windows using the discrete graphics. I disagree. My MacBook pro is in discrete graphics mode in OS X, but I still see much better battery life than running Windows.
My old work laptop dual-booted Windows XP and Ubuntu. Doing the same stuff in both (browsing in Firefox 3.5, playing music in VLC), on the precise same piece of hardware, I'd get 3.5 hours battery on Ubuntu and 2 hours on XP. I blame the corporate antivirus - McAfee to be precise. It's a goddamn power suck.
Windows can get out the way and let you run stuff as well as Linux ... until you put that damn AV on. Which you'd better do. (I'm sure many people will comment they do OK without an AV, but anyone who isn't a sufficiently advanced geek better be running an AV on Windows.)
http://rocknerd.co.uk
Considering the latest story on Bing's biased search results, I wonder if it's possible that M$ would hardwire Windows to drain Mac batteries faster. The ultimate Malware -- a MalOS?
Maximum CPU usage is a terrible drain on environmental resources, and the extra heat generated by the CPU contributes to global climate change. So to encourage users to use less CPU power...Apple used a slightly underpowered power supply.
Tell me how - in the long run - spending more time on CPU-intensive tasks saves energy and reduces heat.
Mind you, the most logical reason for the user to be running a CPU intensive application is because he needs an answer quickly.
That also has implications for the environment: "The numbers look good. We can shut down the run."
Yet another reason to dislike Apple. I have no idea why people buy things from them. They don't care about their users, only their image.
My blog. Good stuff (when I remember to update it). Read it.
I guess MacBooks have to fall into the camp of Not Exactly Designed for Windows 7.
"It's the height of ridiculousness to say for those 9 lines you get hundreds of millions."
This was supposed to be a funny suggestion, but then I thought it might be even funnier if I binged it and it really did turn up curious results, so I tried it, and...
Microsoft, you so disappoint me. Boring results. Nothing funny at all. You suck.
Sigh.
That would make a MacBook utterly useless for Folding@Home... seems ridiculous.
Do what thou wilt shall be the whole of the Law
Bootcamp is not a virtual machine. It is a set of tools and drivers that allow you to run Windows on an Apple computer.
To do something like that, it must be supported by the drivers. As an example a coworker got a new Thinkpad with that feature, may have been the same one you got not sure. The switching works fine in XP. However he wanted to run the Windows 7 RC on it. There, we couldn't get it to work, I had to go in to the BIOS and shut down the Intel card. Why? No Windows 7 drivers for it. In fact at the time, Lenovo had no 7 drivers at all. All drivers had to be obtained from manufacturers of the various parts.
Any feature like this that deals with hardware must be supported in drivers. That is the responsibility of the OEM that puts the computer together. They don't necessarily write the drivers, but they work with the companies that make the hardware to get drivers for the OSes they wish to support. If Apple elects not to release drivers for Windows that support that, well then it won't work.
Sure you used google and not bing? ;) http://search.slashdot.org/story/09/08/06/1334225/Bing-Search-Tainted-By-Pro-Microsoft-Results?art_pos=3&art_pos=6
You are all a bunch of idots.
This submission is silly. Why would Apple have updated Boot Camp drivers available for something that isn't even out yet? How about waiting until Windows 7 is actually available for sale in October before looking for people to blame?
... But all those little features OS X knows about and uses properly. Boot Camp Drivers Cover most of them, Windows handles other ones. I know for an instance Windows Vista with boot camp keeps the lights on the keyboard while OS X is a bit smarter then that ... Now comes to the question. Is the Mac made Drivers for Vista keeping those lights on. Or Vista is telling the driver to keep it on. I am betting it is both.
I am also suspicious about Apple supplied drivers degrading battery life, bugs seem too obvious and fortuitous for Apple to pass them off as chance. I've noticed that when my MacBook is on battery Vista is in performance mode by default rather than battery life mode. Updating the configuration for a should-have-been-the-default setting of performance on AC and battery life on battery greatly improved the battery. Remembering to adjust the screen backlighting also makes a dramatic improve, both under Vista and Mac OS X.
There is a reason that the battery dies quicker and since there is no layer of Mac OS X between Windows and the hardware I doubt it's because Apple did something wrong. It's either Windows or the Intel or NVidia drivers.
One of the big factors that makes Boot Camp such a successful Windows on a Mac story is that *Apple* provides various Windows drivers for XP and Vista. So you are a little premature to claim that Apple can not be at fault. Apple drivers *may* be failing to put the CPU into battery life mode when on a battery, fail to dim the screen backlighting, fail to turn off keyboard lighting, etc.
If they are releasing a test dvd paying millions of dollars to Akamai and taking the risk of millions of "non upgrading" beta testers, they should have put that prompt as first thing after the install, right in "check for updates automatically". Defaulting to off of course.
Also communication doesn't need to be re-invented, Apple's method of "use documented compression format(bz2), use http put under https protocol" is more than enough. It will satisfy the paranoid. Apple does it in "System Profiler" application, a complete opt-in which has no inviting prompts. I use it every time when I buy some weird USB thing. Let them know how their OS is used.
What are the battery times for you?
Lars T.
To the guy who modded me down from perfect to terrible Karma - Apple haters still suck
After getting used to Apple laptops and buying an Acer instead recently, this is the one thing that most aggravates me. Apple does power management very well - both windows and linux fall very flat by comparison. I wouldnt be surprised if windows kills batteries faster than OSX on the same hardware - in fact I'd say something was very very wrong if it did not.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Friends don't let friends enable ecmascript.
Any piece of code running on the system can be written in such a way that it forces the CPU to enter a high power consumption mode too often.
Check out the research related to the development of powertop - it has clearly demonstrated that many user-mode programs have a negative impact on power consumption; the use of powertop allowed them to find the "offenders".
The more processes you have, the more code there is - the greater is the probability that a part of it is not written with power efficiency in mind.
The saddest poem
Comment removed based on user account deletion
Thanks for correction. I never used Bootcamp, but I did assume (and obviously it does) it presents some sort of incompatibility for Windows running on a Mac. Probably it is Windows not having a good driver to interface Mac internals.
True. The few times I used a MacBook I was pleasantly suprised how well OSX idles.
The OS uses the drivers for the laptop's logic board to do the power savings. If the drivers do not provide the OS with the right information, or the drivers are missing features necessary for ideal power savings. Then the OS is not going to be able to do much. Apple is the vendor for the Windows drivers for their laptops, and therefor it is probably Apple's fault.
“Common sense is not so common.” — Voltaire