Windows 8 Consumer Preview vs. Windows 7: Benchmarked

Windows 8 Consumer Preview vs. Windows 7: Benchmarked

Summary: Can Microsoft's upcoming operating system keep up with -- or even beat -- Windows 7, or does Microsoft still have work to do?


It's time to see how Microsoft's Windows 8 Consumer Preview stacks up against Windows 7. Can the upcoming operating system keep up with -- or even beat -- Windows 7, or does Microsoft still have work to do?

This is my second attempt at benchmarking the Windows 8 Consumer Preview. I attempted to benchmark the operating system soon after it was released back in February, but ran into troubles with graphics card drivers issues, and problems getting consistent results from a couple of the benchmark tools I was using. It seems that these issues have been ironed out, finally allowing me to complete the testing.

The hardware

The following hardware platform was used for benchmarking the two operating systems. The system was purpose-built for the job of benchmarking:

  • Intel Core i7-2600K processor
  • Crucial 4GB DDR3 1600 (PC3 12800) RAM
  • EVGA 01G-P3-1460-KR GeForce GTX 560
  • GIGABYTE GA-Z77MX-D3H motherboard
  • Western Digital Caviar Black WD1002FAEX 1TB hard drive
  • CORSAIR Enthusiast Series TX650 V2 650W power supply unit

Everything on the system was set to stock speeds, with no component overclocked.

For the tests I used two Western Digital Caviar Black WD1002FAEX 1TB hard drives from the same batch to eliminate the possibility of an update causing one drive to be faster than the other. One drive had installed Windows 7 Ultimate (SP1) 64-bit, and on the other I installed the Windows 8 Consumer Preview 64-bit. All drivers and updates were installed, along with all the software that would be needed for the tests. The drives were then defragmented using the Windows tool before the benchmarking was carried out.

The benchmark tests

Here's a rundown of the tests that were run on the two operating systems. I've chosen a mixture of real world and synthetic benchmark tests.

Each test was run three times and the results averaged.

  • Boot time Measured using a handy tool called BootRacer. This measures both the time it takes to get to the logon screen and the time to boot to the desktop.
  • Audio transcode time Transcoding an audio test file from WAV to MP3 format using iTunes. A measure of the operating system's ability to handle multimedia.
  • Video transcode time Transcoding video test file from DVD to MP4 format using Handbrake. A measure of the operating system's ability to handle multimedia.
  • PCMark 7 A benchmark run with PCMark 7. The industry standard PC test for CPU, HDD, SSD, memory, and graphics performance.
  • 3DMark 11 A benchmark run with 3DMark 11. This is a set of six demanding benchmark test measuring the graphics performance of gaming PCs.
  • FurMark A benchmark run with FurMark. This is a VGA stress test, GPU burn-in test and an excellent OpenGL benchmark. This is a very stressful benchmark and can damage or even destroy hardware if used incorrectly, and therefore I do not recommend running this tool on a system unless you know exactly what you are doing and fully understand the risks associated with it.
  • Cinebench 11.5 A benchmark run with Cinebench 11.5. This is a real world cross, platform test suite that evaluates a computer's CPU and GPU performance capabilities.
  • Heaven 3.0 A benchmark run with Heaven 3.0. This is a DirectX 11 GPU benchmark based on the advanced UNIGINE engine. Not only does this tool give us the maximum frames per second (FPS), it also records minimum frames per second, which is handy observing dips in performance during heavy load.
  • Alien vs. Predator A benchmark run using the in-built benchmark tool available in Alien vs. Predator. The benchmark is run at 1920x1080 screen resolution with DirectX 11 enabled. This is a real world gaming test.

Topics: Windows, Hardware, Microsoft, Operating Systems, Software

Kick off your day with ZDNet's daily email newsletter. It's the freshest tech news and opinion, served hot. Get it.


Log in or register to join the discussion
  • Minimum FPS

    I'm not sure if you marked your graphs wrong or if you read them wrong but the Windows 8 DX11 min FPS is higher, albeit only by 1, than the Windows 7 DX11 min FPS in Heaven 3.0. Since Heaven is really a DX11 benchmark I'm not surprised that their DX9 test would show some odd variations.

    Other than that this is line with what most people have found. Also, you should compare the Windows 8 hybrid shutdown to normal Windows 7 shutdown on a mechanical drive. From end of post to login on my Core i5 laptop it's about 11 seconds on 8 and it was about 20 or so on Windows 7.
  • Windows 8 Consumer Preview vs. Windows 7: Benchmarked

    Microsoft Windows 8 is better in almost every way, that's a good sign of things to come. The most impressive part is that this is still beta code so its not optimized yet meaning when its release time you will have a much more improved OS.
    Loverock Davidson-
    • Disagree completely

      Windows 8 is Windows 7 with an added interface! What is improved for desktop, and laptop users??? Give me 10 reasons why people should upgrade and spend 200 on windows 8!!! Perhaps the reason it is better is you do not have anything installed on win 8 compared to windows 7. Windows 8 is a major step back, in windows designand evolution!!! You are probably a newbie to windows os, and never used windows XP!!!
      Pollo Pazzo
      • In FACT ...

        ??? Windows 8 is a substantial re-working and re-factoring of the entire OS from the bootloader up.

        Whilst, yes, the core underpinnings of the OS remain substantially the same, the entire OS has been completely deconstructed and reassembled in a FAR more modular fashion, with clean layering replacing the ball of string that the Windows code-base used to be.

        Not only this, but thanks to the effort MS put into re-factoring the OS, it now consumes less memory, requires less CPU cycles and IO can now be built for different platforms (i.e. ARM) and many of the OS' previous limitations and issues have been eliminated, resulting in Windows being FAR more scalable, agile and efficient across the board.

        All this whilst not breaking any existing software or apps.

        That's quite the feat in my book.
    • Actually...

      This whole exercise is rather pointless in trying to gauge Windows 8's performance. We have no idea how much debug baggage Windows 8 is carrying, these results might have VERY little in common with the true performance.

      In short we have little to no idea how much Windows 8 was handicapped in these tests. It could easily be that just recompiling the OS would yield better results.

      Additionally PC hardware continues to evolve, and not at an even pace. Windows 8 is, in all likely hood, optimised for tomorrows' PC, not todays'.

      The trends I'm thinking about are: More CPU cores, and multi-GPU setups, moving from spinning disks to SSDs.
      • I expect that the W8CP is doing a lot of telemetry logging

        MS probably will be tweaking the OS a lot, especially the Metro stuff, as a result of all the telemetry they have gathered from the CP.
      • Debug baggage

        Usually Microsoft releases dedicates "checked builds" with debug information in the kernel. The Consumer Preview is not such a build, so it carries very little "debug baggage". True, it does report telemetrics back to MS and it may also log a little more.

        But such overhead is negligible compared to kernel optimizations, parallelized startup processes and - above all - driver optimizations. I think this benchmark is a good indication of the final perf. of Windows 8. It may improve a little due to optimization efforts still going on, but don't expect it to jump.
  • After seeing these benchmarks... we need Windows 8 why?

    • Great Reply Bitcrazed.

      Hopefully, my company will take and spend some money in the 1st quarter of 2013, and upgrade to Win 8 from XP...It really is getting to be old...

      Now, that said, I don't believe on a personal note that I'll upgrade from Win7, on my desktop to Win8. The two laptops we have are still running good, so we'll have to wait and see.

      We still have Win Vista on these, and IMHO, they've been working very well and we're happy with them. So it might come down to how much for the upgrade, and if I can locate all the drivers for them...

  • What about resume from sleep ?

    Adrian, you say :

    We don???t reboot out PCs anywhere near as often as we once did, but a fast boot up time is still appreciated.

    That' OK. But we do resume from sleep mode a lot more than we reboot - my Win7 laptop takes quite a while to come back from sleep zzzzz !

    Have you measured anything related to coming back to life after sleep mode (the computer that is ;)
    Alain in Québec
    • Mine wakes very quickly...

      I basically only reboot when necessary; normally after a MS update. I use sleep mode everyday and mine wakes up almost instantly. Have you checked to make sure your problem isn't related to a particular program or device driver?
    • My resume from sleep is almost instantaneous... 0.5 to 1.0 sec.

      The Danger is Apple
      • Ditto

        FWIW, both my Sony Vaio Z Series (v1) and MacBook Pro running Win8 CP both resume from sleep within 1s - usually before I even get to touch a key on the keyboard.

        If I've left the machine asleep for an extended period, the computer will hibernate (allowing it to stay asleep for weeks). Before Win8, resuming from hibernation usually took 60-90s. Resuming from hibernation in Win8 takes about 15s - a VERY significant improvement over previous versions of Windows.
    • Wakes up fast

      On my Acer Aspire with SSD it takes under 7 seconds for Win 8 to be back. It may take another 10 seconds or so before wi-fi is live
      • Connected Standby compliant hardware will make WiFi and 3/4G instant

        CS is what will make W8 more like a phone/tablet OS as far as connectivity responsiveness goes.
  • Am I dreaming

    A positive AKH article related to MS ? Maybe someone is feeling a tad ill ? I expected my on shedule Windows bashing article! ;)
    • Perhaps it's a question of finding something positive

      Here Adrian found a marginal improvement in OS speed and was able to write an article about it. Enjoy;-)
      Richard Flude
    • And he had to put his iPad down!

      Perhaps it was in for repair or he realised he had to regain his tech cred.
  • Just wait...

    until all the debugging and error reporting code is removed. There should be an even bigger difference between the two.

    Also, my bootup experience is MUCH faster than Win7, as someone else mentioned. :-)
  • Boot Time

    Just curious if you tested boot time with WIN 8 connected to an AD Domain and then compared boot times in the presence of a domain controller and then by relying on cached domain credentials.

    I find that boot times are great until you connect windows to a domain and then they start going up.

    Since a lot of people connect to domains from work, that would be a useful boot time test.

    Also, are you testing boot times logging in locally? Or with the Windows LiveID that Win8 prefers?
    Freddy McGriff