OS Mobility Explored

by Jarred Walton on September 21, 2009 6:00 PM EST

Gateway NV52 (AMD) - Futuremark

Since we were already installing the various operating systems and running battery life tests, we thought it might be interesting to run some of the Futuremark benchmarks (while waiting for the battery to recharge...). Windows XP can't run the latest Vantage versions of PCMark and 3DMark, but we included 3DMark03/05/06 and PCMark05. Here are the results.

Gateway NV52 Futuremark Performance

The results in the 3DMark tests are very close, with the largest gap coming in 3DMark03. XP leads Vista by 3% in that test, which is hardly noticeable, and the results in the other 3DMark suites are within 1%. PCMark shows a much larger difference, with PCMark05 putting Vista in the lead by 7% over XP and 8% over Windows 7. We'll take a closer look at those results below, as the composite score is very deceiving. PCMark Vantage goes the other way, with Windows 7 beating Windows Vista by almost 20%. Let's look at the individual test results in both PCMark benchmarks to get a better idea of what's going on, beginning with PCMark05.

Gateway NV52 PCMark05 Breakdown
  XP SP3 Vista SP2 Win7 RTM
PCMark05 Score 3590 3875 3623
HDD XP Startup (MB/s) 6.664 5.948 6.168
Physics and 3D (FPS) 111.1 97.4 103.7
2D Transparency (Windows/s) 214 2730 478
3D Pixel Shaders (FPS) 55.02 54.81 52.86
Web Page Rendering (Pages/s) 2.450 1.644 1.962
File Decryption (MB/s) 35.91 34.02 36.05
2D 64 Line Redraw (FPS) 331.8 362.7 392.1
HDD General Usage (MB/s) 4.233 4.038 3.941
Multitasking 1 1000 930 948
Audio Compression (KB/s) 1936 1313 1163
Video Encoding (KB/s) 310.4 366.8 402.2
Multitasking 2 1000 889 934
Text Editing (Pages/s) 113.0 88.4 100.7
Image Decompression (MPixels/s) 23.64 23.54 23.09
Multitasking 3 1000 956 1047
File Compression (MB/s) 4.224 3.86 3.283
File Encryption (MB/s) 21.04 20.72 29.27
HDD Virus Scan (MB/s) 68.37 59.38 52.88
Memory Latency (MAccesses/s) 6.73 7.103 8.382

When you look at the composite score, Windows Vista looks very attractive in PCMark05. The individual results tell a completely different story! (Note that we calculated results for the multitasking tests relative to the XP score, which is why it scores 1000 on all three tests.) The high composite score of Vista is a result of the 2D Transparency test, where it is nearly 13 times as fast as XP and almost 6 times as fast as Windows 7. Exactly how important is 2D transparency? It probably helps in Vista when you're using Aero Glass, but it shouldn't matter much at all in Windows XP.

Obviously 2D transparency is a weak point of XP - or at least the XP drivers - so we went through and calculated the relative performance in the PCMark05 tests with and without 2D Transparency. We used the XP result as the baseline metric. Including 2D Transparency, Vista's average performance is 200% of XP and Windows 7 is 108%. Remove that one result and XP ends up being 8.6% faster than Vista and 3.5% faster than Windows 7. The composite PCMark score is weighted, and we don't have exact details on their formula. It's clear that 2D Transparency does not have the same weight as the other tests, but it's still enough to skew the results.

Gateway NV52 PCMark Vantage Breakdown
  Vista SP2 Win7 RTM
PCMark Vantage 2566 3055
Memory 1467 1529
TV and Movies 1541 1835
Gaming 2121 2126
Music 2170 3347
Communications 2971 3652
Productivity 2499 2558
HDD Test 2445 2372

The results for the individual test suites in PCMark Vantage are a lot closer than the 2D Transparency result from PCMark05, and Windows 7 leads in most of the tests. Gaming performance is essentially a tie, Vista leads by 3% in the HDD test suite, but everything else favors Windows 7 - sometimes by a large margin. We don't know exactly why Windows 7 scores so much higher in the TV and Movies, Music, and Communications test suites. It could be that driver differences play a part, or it may be that Windows 7 is simply better optimized for some of these tests. We do know that most users think Windows 7 performs better than Windows Vista, and the PCMark Vantage results clearly support that impression.

Gateway NV52 (AMD) - Battery Life Gateway NV52 (AMD) - OS Benchmarks
Comments Locked

106 Comments

View All Comments

  • JarredWalton - Tuesday, September 22, 2009 - link

    I'll look into powertop in a while. The issues with Windows really have a lot more to do with users than with the platform as a whole. My work PC has been running without any problems and without reinstalling the OS for over three years. My gaming system is in a similar state, and both have been through a few hardware upgrades, plus various driver and software installations.

    I don't run any "internet security suite" - be it McAfee, Norton, AVG, or anyone else. No AV, not even anti-malware (though I have scanned with Spybot S&D, Ad-Aware, and HiJackThis on occasion just to make sure). Why am I problem free? Because I know what I'm doing.

    I think the same could be said of Linux users: they're mostly problem free because they know what they're doing, and they could be problem free in Windows if they wanted to put in a small amount of effort.

    Put your average user in front of a Ubuntu installation -- or any Linux installation -- and they're going to be lost as soon as they want to do more than run email, OpenOffice, and surf the Internet. "But that's all you need!" Exactly: all you need and all you want are not the same. Windows makes it perhaps too easy for people that haven't got a clue to install programs and screw things up. LOL
  • Gamingphreek - Wednesday, September 23, 2009 - link

    Honestly, all the configuration I suggested takes a few minutes at most. There is no compiling from source and no generating makefiles.

    As someone said earlier, Linux and Windows have different philosophies in terms of setup. Linux is an OS that depends on customization, at least SOME time needs to be put in configuring it.

    Running powertop and writing a bash script takes a mere 5 minutes and can save >30 minutes of battery life in some cases. Enabling laptop mode is literally as simple as opening a file and changing the value from 0 to 1.

    Updating the kernel, mesa, and graphics is literally 6 (2 for each -- 1 to add the repository and 1 to add the key) commands long and takes <5 minutes.

    I understand this isn't an article on Linux, but these are not in depth tweaks and are things that anyone running Linux on a laptop will typically do.

    Finally, you state that an average user would be lost in Linux. Honestly this is Anandtech - a very well known and reputable Tech based website. "Average users" typically do not venture here. Should you guys stop talking about Processor Architecture since people wont know what it means?
  • strikeback03 - Wednesday, September 23, 2009 - link

    My father needed to move suddenly due to a job change, so I hooked him up with a computer I scavenged out of the trash at work. It had a XP license key on it, but thanks to their wonderful restrictions on what what disc can install what version I had no working installer for that key. So I threw Ubuntu on there, which does indeed do everything he needs (allow him to play with his investments and watch Hulu). He called last weekend asking where to find a driver for a printer that was not included by default, as the solution involved several lines of code he decided to just wait until I can walk him through setting up Remote Desktop and do it from here.
  • Mattus27 - Monday, September 21, 2009 - link

    I can't believe you tested using live dynamic websites, instead of just downloading a page and all its resources and running the test from disk. The drawbacks of that should have been fairly obvious.
  • JarredWalton - Monday, September 21, 2009 - link

    The problem is, there is a ton of JavaScript involved with any current website, and getting all of the content for an offline version isn't quite that simple. Go try it: download everything for www.AnandTech.com to your PC and then look at that file and compare it to how the site actually looks. Try that with Facebook, YouTube, Yahoo, MSN, CNN, HardOCP, FiringSquad, SlashDot, TechReport, etc. and you will find they all need tweaking to look anything approaching correct. (Well, maybe not every single one of them, but most will still end up with JavaScript files that load content from dynamic web servers.)

    Anyway, I wanted to test with "real" content and not some bogus artificial test that doesn't have anything to do with what real Internet sites are like (i.e. some of the MobileMark stuff has very questionable testing procedures). I wanted something I had control of that would still tax PCs like a real website. I've got some downloaded sites and I've been going through the HTML and modifying it to hopefully create a "static" page that I can host on our server and still have it work more or less correctly. That of course means another batch of testing, but so far it's looking good.

    I debated scrapping the current article, but figured some would enjoy the read and the joys of testing multiple different OSes. Besides, this way I can get feedback on what other tests you might like to see.
  • gstrickler - Friday, September 25, 2009 - link

    How about setting up a local caching web proxy and having all the machines connect via the proxy? It won't completely eliminate the variances of internet routing/throughput and connecting to live sites, but it should minimize them.

    As for Wi-Fi (and interference from your 2.4GHz phone), leave the Wi-Fi enabled and connected, but go ahead and connect the machines using Ethernet. The difference is drain of a modern Wi-Fi transceiver when transmitting/receiving vs "idle" is fairly small, so simply having it on and connected to the Wi-Fi access point (which will occasionally "talk" to each connected client") should be sufficient from a power draw standpoint while using Ethernet for the active network connection will be more reliable and won't be subject to the interference. You can put the Wi-Fi on a subnet that does not route to your proxy/firewall and/or give it no default gateway, which will ensure that all the real traffic in on Ethernet. You could add in an occasional Ping from each client to the WAP to ensure that the Wi-Fi card/connection stays active.


  • andrewaggb - Tuesday, September 22, 2009 - link

    Well I certainly agree that testing dynamic websites at different times is not the way to go.

    I can also understand what you are saying about trying to download a site and all the related resources, I've tried in the past and the browser method doesn't work.

    You'll probably need an external tool to download the site and all it's resources. A quick google search came up with this tool, http://www.surfoffline.com/">http://www.surfoffline.com/, I've never used it but it has a trial and it sounds like you can download entire websites and export them. Worth a shot anyways, there's probably a bunch of tools like this that "hopefully" work as advertised :-)

  • Lowly Worm - Monday, September 21, 2009 - link

    ".. we ran the same test under Windows Vista and a cheat significantly better battery life. "

    Heh.. DragonSpeak "a-cheating" on you? Makes for interesting phonetic typos.
  • JarredWalton - Monday, September 21, 2009 - link

    Yup. Thanks for the correction - was supposed to be "achieved", naturally. :)
  • lordmetroid - Monday, September 21, 2009 - link

    The Linux distribution you tested while mainstream, maybe not as tailored to mobility as say the Moblin distribution by intel, though still in Beta, I would loved to have seen that distribution tested considering it is specially built with focus on mobile platforms.

Log in

Don't have an account? Sign up now