Gateway NV52 (AMD) - OS Benchmarks
The final aspect of operating system performance we want to look at involves common OS tasks. We measure how long it takes to perform the routine operations of booting, shutting down, hibernating, resuming, suspending, and waking your computer using the four test operating systems. Unlike other test results, there was a huge degree of variability in some of the tests. We ran each test at least five times, and we took the best result - the reason being that some of the results were completely out of line with the average result, so including one or two results where the system took four times as long it didn't seem logical. Ubuntu and Windows 7 didn't show as much variation as XP and Vista, particularly in the sleep/wake testing, but the hibernate/resume test once again varies quite a bit between test runs.
The variability remained even after following a relatively straightforward test procedure. First we booted each laptop using the test OS (timing how long it took). Then we let the system stabilize for at least 15 minutes, with no extra applications running and the system sitting idle at the desktop. Finally, we performed one of the other tests - shut down, hibernate/resume, or sleep/wake. Obviously, you can't test resume time without first hibernating, and you can't test wake time without first putting the system to sleep, but otherwise we made sure to shut down the system between test runs.
We also need to mention that Windows XP has an inherent advantage in the hibernate/resume testing because we are using a 32-bit OS instead of a 64-bit OS. That means the system doesn't make use of all 4GB of memory, and the hibernation file is 3GB. Writing 3GB of data will obviously be easier than writing 4GB of data, especially when you're dealing with a 5400 RPM laptop hard drive.
The results are rather scattered this time, with certain functions favoring one OS and others looking better on a different OS. Boot times favor Windows XP, followed by Windows 7 which takes 11% longer. Vista trails Windows 7 by 10%, and Ubuntu brings up the rear at 5% longer than Vista. Shutdown times are a different matter, where Ubuntu takes the lead. Vista and Windows 7 are essentially tied, taking 55-60% longer to shut down than Ubuntu. Windows XP brings up the rear this time, at over 2.5 times as long as Ubuntu, although we have to say that this appears to be a driver issue as the result for the NV58 is radically different. When you shut down Windows XP on the NV52, there's a delay of about 15 seconds where there is no apparent activity, after which the hard drive activity light begins blinking. It would seem that without that handicap, Windows XP would be roughly the same as Vista and Windows 7.
As already stated, Windows XP 32-bit has an advantage for the hibernate/resume testing because it doesn't have as much addressable RAM. It ends up walking away with the hibernate/resume performance crown, followed by Windows 7. That said, the time to hibernate in Windows 7 is 34% longer than Windows XP and resume time is 32% longer; Win7 has 33% more memory, so a 32-bit version of Windows 7 ought to be as fast as Windows XP (or a 64-bit XP as slow as Win7). The gap between Windows 7 and Windows Vista is extremely large for hibernation time, with Vista taking 65% longer to hibernate and 46% longer to resume. Ubuntu is almost tied with Vista on the hibernation time, but it takes 50% longer to resume - about twice as long as Win7 - making it by far the slowest OS for this particular test.
We need to take a moment to discuss boot times and resume times, however, as there's more going on than meets the eye. When booting or resuming a system, the computer needs to perform a Power On Self Test (POST). That takes around nine seconds on these Gateway systems, and once that's complete the boot loader is activated which then loads the OS. Ubuntu uses GRUB and that appears to add an extra ~10 seconds relative to Windows. Without that time penalty, Ubuntu would be far more competitive in boot time (about equal to Win7), although it would still be in last place in the resume test. There are also laptops that have a fast boot feature so that they don't perform a lengthy POST, which can shave another 6 or 7 seconds off the boot/resume times. This is definitely a feature we would like to see implemented on more systems going forward, particularly laptops where users routinely boot or resume several times per day.
Wrapping up the OS performance analysis, the sleep/wake times were highly variable on the Windows XP and Vista operating systems. This is especially true if you put the system to sleep right after waking it up, booting, resuming, or any other complex task; in those cases, it might take over a minute to go back to sleep. Ouch! It may also take over 20 seconds for Windows to go to sleep if you have a bunch of applications open - that's about six times as long to put your laptop to sleep, which can be very annoying if you just want to pack up and go. You might think that you can just put your system to sleep and throw it in its bag and walk off, but sometimes a glitch will prevent a laptop from going to sleep so we like to make sure our laptops actually shut off before putting them in an enclosed space. We have heard of a few instances where a laptop has apparently overheated because it didn't go to sleep and a user put it in a laptop bag, resulting in a dead system. That's some food for thought anyway. Ubuntu and Windows 7 did a lot better, with the worst-case result being around twice as long as the best time.
Sleep times are a tie between Ubuntu, XP, and 7 at just under 5 seconds. Vista took noticeably longer even in our best-case result, trailing XP by 59%. Wake times mix things up quite a bit and show a lot less variation between runs, with Ubuntu falling to dead last and Vista and 7 essentially tied at ~3.5 seconds. XP takes 46% longer than Vista to wake and Ubuntu is 55% longer than XP (over twice as long as Win7). Again, though, variability in sleep times makes us prefer Windows 7 and Ubuntu followed by XP, with Vista being the worst at consistently going to sleep without a large delay.
Facebook
Twitter
RSS
106 Comments
View All Comments
ascl - Tuesday, September 22, 2009 - link
I was reading this thinking that it was an unusually bad review from anand.... then I reached the conclusion 2 + 3 and my complaint was answered! Using randomised web testing is terrible if you want repeatable results. Use an internal server with a fixed set of pages (and ads).Kudos!
vailr - Tuesday, September 22, 2009 - link
Maybe compare battery life on a MacBook Pro running OSX Snow leopard vs. Windows 7 64-bit/Win XP 32-bit (in Boot Camp) vs. Ubuntu(?). Using VLC Media Player sequenced to play a series of several DIVX movies, for finding the running time under each OS.gstrickler - Saturday, September 26, 2009 - link
Anand has already done that with a MBP and OS X vs XP (and maybe Vista). Anyway, it would be good to see it repeated with Win7 in a couple months when they've had time to produce some reasonably power efficient Win7 drivers.In any case, a MBP is a great platform for the task, it will run all those OSs, the Nvidia chipset is well supported in all of them and it's pretty power efficient. Of course, with it's battery life, the tests might take a while.
Gamingphreek - Tuesday, September 22, 2009 - link
Battery life in Linux, as it is a *nix based OS, needs configuring. You claimed that you couldn't find it, but honestly you need to find someone who has *nix experience then because it is honestly the most customizable of the OS's.Additionally, while you did a good job in following that website for advice, staying with the 'Safe' configuration is a grave error. All that merely does is use an old version of MESA (OpenGL driver) and an old Intel driver. I would honestly like to see you try out the 2.6.30.5 or the 2.6.31rc kernel along with the most up to date drivers from the XorgEdgers repository. Performance with those optimizations is honestly quite remarkable.
Furthermore, I would suggest looking over lesswatts.org as well as running the PowerTOP application to see what is unnecessary. For instance, I have a script that disables my PCMCIA slot given that I do not use it. I also have my RJ11 based modem disabled. I have LaptopMode enable automatically when I unplug A/C power and disable when I plug in A/C power.
Additionally, why did you turn off auto dimming?? That is a great feature that severely crippled the performance of Ubuntu yet again.
As for Firefox/Shiretoko (Shiretoko is the codename for Firefox 3.5) that is a known issue with lazy programming. Downloading the Noscript or Adblock extension helps immensely with performance. Additionally Shiretoko/FF3.5 has a vastly improved engine when compared to FF3.0.
Additionally, I don't believe you stated what file system you were using. EXT4 is vastly superior to EXT3 (While it isn't the default, among Linux users, it is rare for someone to choose EXT3 over EXT4) - especially when boot times are involved. Even still, it sounds like there was a broken script or something - Ubuntu 9.04 has the fastest startup/shutdown I have ever experienced.
Honestly, Ubuntu seemed to draw the short end of the stick here. It takes time to configure the OS - I honestly expect more time to be given to configuring it like the other ones.
smitty3268 - Tuesday, September 22, 2009 - link
While I agree with most of your points (you can definitely tweak linux down to the bare bones much more than something like XP to save power), I think it is perfectly acceptable to use a default distro that is commonly used. After all, he didn't go through the Windows registry, disabling services and hacking stuff there either.But in the end, let's face it. Firefox and Flash are horribly optimized for Linux. It's not exactly a surprise that they suck down more juice, given that they usually take about 5 times more CPU power than under windows.
Gamingphreek - Tuesday, September 22, 2009 - link
Well that is the inherent difference between Linux and Windows.Thats like saying you aren't going to download and install drivers for Windows - Linux comes with all of them...
And as a slight correction, Firefox isn't really the problem and Flash is horrible for any/everything BUT Windows.
JarredWalton - Tuesday, September 22, 2009 - link
As stated, it wasn't intended to be a Linux review. This is a well-known Linux distro that is supposed to be "easy". I don't have the time nor inclination -- just like 99.9% of users -- to go into detailed steps for hacking and modding Linux. I fully understand that it is highly customizable, but so is a car if you're inclined to go that route. I drive a stock vehicle, and I use a stock OS.Downloading drivers isn't the same thing as downloading the latest kernels, creating your own conf files, and manually entering all sorts of settings that help enable/disable items to provide better power saving. My conclusion pretty much sums up my feelings: the out-of-box experience for Ubuntu is nothing special for a laptop, and if you are expecting it to "just work" you'll be disappointed.
Given how much is available for tweaking in the Linux community, I'm frankly surprised that no one has apparently spent the time to make the default configuration far more sensible and easier to live with. I know how much fun it is to download and compile programs and edit configuration files, but I'd rather just have an easy interface that works without a ton of effort.
I also fully recognize the inherent problems with Flash, so I put in numbers with Firefox and FlashBlock. It helps, but it doesn't help enough to equal the default Windows setup. There are other browser options of course, and if/when I get time I'll see about looking at some of them.
ekul - Tuesday, September 22, 2009 - link
Ubuntu is a poor distro for battery life out of the box as many of its default settings belong on a desktop or even server system. With a bit of tweaking though it can easily get better battery life then windows. My netbook struggled to get 2 hours in XP, in ubuntu I can easily see 2:45 or more.As has been mentioned by at least one other commenter use the powertop tool (sudo apt-get install powertop && sudo powertop). This was written by intel to help find applications and drivers that were waking up the cpu too much and hurting battery life even if they didn't appear to be using very many system resources. The tool itself looks for many settings that are not optimal for battery life and offers to correct them for you so you don't have to go on a treasure hunt at all. I'd love to see what kind of improvement could be made with that tool alone.
As an aside, a major oversight is idle battery life. All of your tests feature the OS as the minor player in the war against power consumption. In each test you have an application eating the majority of the resources. You should fully charge the battery and let the laptop idle at the desktop until it dies, testing each OS's ability to sleep long running processes and services. Perhaps leave an office suite, browser idling on a page and wifi connected to stop the runtime for taking too long.
Finally using flash heavy websites heavily skews the results for both OS and browser battery life tests since you are sending both into battle missing limbs. It is well known flash is poorly written software at the best of times, causing well over 60% of all firefox and IE crashes. It has rudimentary 64 bit support, doesn't support hardware acceleration on anything but windows and if it isn't playing video (covered by the dvd test) is almost certainly on the page to serve an ad. No browser has any sort of control over flash (though chrome does its best to reign it in) so your browser tests amount to little more then a test of flash. Flash sucks even in windows but in linux it is truly awful (as is almost all closed source code for linux).
JarredWalton - Tuesday, September 22, 2009 - link
The second batch of sites was hardly Flash-heavy. Yahoo and MSN have one Flash ad, YouTube has none, and the Facebook login page is just text and images.FWIW, I did run an "idle at desktop" test on Ubuntu on the NV52 and got a time of 204 minutes. That compares to 242 minutes under Vista, or a 18.6%. There are a LOT of other things I still need to look at, however -- including different power schemes, tweaked profiles, etc.
If I'm going to try to improve the Linux results in every way possible, it's only fair to do the same for Windows.
ekul - Tuesday, September 22, 2009 - link
Because the linux and windows philosophy is so different a different approach has to be taken to setting them up and running them. With windows initial setup is very simple and has vendor support for things like drivers. Over time problems begin to appear and cleanup/formatting becomes necessary. Most linux distros integrate all possible hardware support and target lowest common denominator hardware to ensure broad compatibility at the expensive of performance. Once they have been tweaked they will continue to run indefinitely at that level.With windows you have ease now for pain later. With linux you have pain now for ease later. This means running things like powertop and changing the init options to run in parallel would be the same as cleaning a registry in windows rather than something like disabling services in windows. I spent one afternoon tweaking my netbook and now it runs much smoother and faster then it ever did with windows plus the battery life is longer. Changing config options and customizing for your hardware is the reality with linux the same way random problems cropping over time is the reality with windows. If linux distros are to be punished for ease of setup issues then windows must be punished for altered performance 6 months from now. Linux is getting better however, and different distros ship with very different defaults. Ubuntu is really debian unstable repackaged for the server/desktop which means at its very heart it is a server distro. For future tests opensuse may be the best choice to represent laptop/desktop defaults.
FWIW if you want to find the real options in linux for power management you should look in /sys/devices/system/cpu/cpu0/cpufreq/. The options in the gnome power management panel only really deal with monitors. But again 3 minutes with powertop and I'm certain you will see an improvement.