More details about Vista's startup times

More details about Vista's startup times

Summary: Is Windows Vista really slow to start up? Over the weekend, I played lab rat again, clicking a stopwatch over and over agin to measure startup times on a room full of Windows PCs. Is Vista really faster than XP? The results surprised even me.

SHARE:
TOPICS: Windows
102

Last week, when I published my preliminary test results documenting start-up times for Windows Vista, the most common comment, by far, was that my test was unfair because I used hardware that was too powerful. Personally, I think this particular Dell system (which costs around $1200 and is far from their top of the line) is pretty mainstream, but more data never hurts. So I went back to the lab and performed similar tests on two more systems.

First, a recap and an explanation. The original story in ComputerWorld was based on quotes from end users on the Windows Vista Performance and Maintenance Forum, complaining about dreadfully slow start-up times for machines with Windows Vista installed. Now, do I think those people were lying? Of course not. I’ve seen systems that take an exorbitantly long time to boot up, and I’ve usually been able to resolve the issue by finding the configuration problem or third-party application that was dragging down performance.

Do those randomly selected posts represent a trend? Not necessarily. Computerworld could just as easily have searched the Windows XP Support and Maintenance forum and found dozens of similar complaints about Windows XP. Or they could have gone to various Mac support resources, where they would have been overwhelmed with recent examples of people struggling with slow startups on their OS X machines.

Unfortunately, ComputerWorld didn’t include any testing to try to duplicate those results under controlled circumstances. Which is why I’m playing lab rat.

In my earlier controlled tests on a Dell XPS 410, I found that it took 61 seconds to load Windows XP and 72 seconds to load Windows Vista. Those results were consistent. But can they be extrapolated? What happens if I run a similar test on less robust hardware?

To answer that question, I added two more test systems to the one I tested last week. In addition, I performed startup tests on several other systems running Windows Vista without installing XP as a control. (For details of the methodology I used, see the end of this post.)

These are not sterilized, tweaked test systems. They’re machines I use all the time, including the ones on which I do daily work. Two of the three systems were using older, single-core CPUs and had a total of 1.5GB of RAM. (For detailed system configurations, see the end of this post.)

The results prove to my satisfaction that Windows Vista is not inherently slow. I was consistently able to achieve startup times on Windows Vista machines ranging from 44 seconds to 1:39 (the longest startup time I recorded was 2:04). In one of the three cases, Vista was much faster than Windows XP running on the same hardware. Here’s the data (all times in mm:ss, rounded to nearest second):

System #1

XP: 1:01

Vista: 1:12 (15% slower)

System #2

XP: 1:47

Vista: 1:20 (34% faster)

System #3

Windows XP Professional: 0:58

Windows Vista Home Premium: 1:14 (22% slower)

Ubuntu Linux 6.10: 1:49 (47% slower)

Additional systems (Vista only)

Dell XPS 210, Intel E6700, 2GB RAM, Vista Ultimate: 0:44

Dell Inspiron 6400, Intel T2050 (dual-core), 1GB RAM, Vista Business: 1:39

HP TX1000, AMD Turion TL-60, 2GB RAM, Vista Home Premium: 1:05

I’ll have a final post drawing some conclusions from this data and from my testing tomorrow.

SYSTEM SPECIFICATIONS

Here’s a summary of how the three systems were configured:

System #1 is the Dell PC I used for last week’s tests. It has an Intel Core 2 Duo E6600, 4GB of RAM, and a 160GB 10,000 RPM SATA drive. I used Dell-supplied disks to install clean copies of Windows XP Professional and Windows Vista Business Edition. This system was built in 2006 and it’s one of two desktop systems I use for daily work.

System #2 is an Acer C310 series Tablet PC, designed and built in 2005. It includes a 2GHz Pentium M (a single-core CPU), 1.5GB of RAM, and a 100GB ATA hard drive. It’s exceptionally well built and is my primary traveling computer. For my tests I used the original OEM installation of Windows XP Tablet PC Edition as supplied by Acer, and upgraded that installation to Windows Vista Business using Acer’s Upgrade Assistant.

System #3 is a homemade PC using hardware that was common in 2003/2004. This is the same system I’m using for my experiments with Linux, with one crucial difference: I removed the add-in SATA controller and disabled the HPT370 RAID controller, reverting to the onboard IDE controllers. The system has a 2.8GHz Pentium 4 processor, 1.5GB of RAM, and two functionally identical 200GB ATA100 drives with very similar performance characteristics (one drive is for Linux, the other for Windows). I used retail copies of Windows XP Professional and Windows Vista Home Premium.

METHODOLOGY

My methodology was consistent on all machines. I used a stopwatch, which I started as soon as the first BIOS text appeared on the display [1]. If the system stopped at a logon dialog box, I paused the timing, entered the password or clicked the logon icon, and resumed timing as soon as I pressed Enter [2]. After the desktop appeared, I immediately clicked the default browser icon (Internet Explorer 7 on all Windows machines, Firefox 2.0.0.3 on the one Linux machine) and stopped timing when the start page was fully loaded. I repeated this test for each system until I had three consecutive consistent results and then took the average of those results; I ran the test a minimum of 6 times per machine.

[1] Timing from the point of pressing the power button would have added no more than 2 seconds to each test.

[2] Results were similar on systems where the primary user account had no password or where auto-logon was enabled.

Topic: Windows

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

Talkback

102 comments
Log in or register to join the discussion
  • Did you do this multiple times as Vista

    Gets faster with each re-boot.
    No_Ax_to_Grind
    • Yes, I rebooted multiple times

      On each system I rebooted at least 6 times. On systems where I had installed clean images I rebooted at least 10 times over a 24-hour period.
      Ed Bott
      • I ask because...

        Doing as you did just didn't produce the results I have seen after using Vista for awhile. (Week or so) I believe defragging and allowing Vista to build the index makes a good deal of difference.

        I also wonder about your three systems, are they all BIOS booted or perhaps one of them is using the newer programable systems. (Which Vista supports)
        No_Ax_to_Grind
        • Not sure I understand

          The Tablet PC is a 2005 design, BIOS-booted. It's been running Vista for a good month, so it's well broken in. I used image files to swap XP in and then restore Vista. The image contains any performance enhancements from superfetch etc.

          The older machine is of course BIOS-based.

          The XPS 410 has an AHCI BIOS and has been running steadily here for a couple of months with predictable performance. This new installation is from April 2, so it has had plenty of time to burn in (performance-wise) as well.

          In all cases I saw a steady reduction in boot times as I went through multiple reboots. Booting the system 10-12 times does a pretty good job of simulating a week's worth of activity.
          Ed Bott
    • Reading comprehension

      From Methodology paragraph:

      I repeated this test for each system until I had three consecutive consistent results and then took the average of those results; I ran the test a minimum of 6 times per machine.
      donniebnyc666
    • Read much?

      From Methodology paragraph:

      "I repeated this test for each system until I had three consecutive consistent results and then took the average of those results; I ran the test a minimum of 6 times per machine."
      donniebnyc666
      • Post much?

        Duh.
        donniebnyc666
    • Why?

      Don't tell me just because Vista has advanced auto-tunning features that OS X lacks they shouldn't be taken into account.
      Fil0403
  • re: boot times

    [b]Many Linux distros and even OSX can take as long or longer then Vista to boot[/b]

    have any links to back up your claims? btw, can I be a abmer today? ;-)

    gnu/linux...giving choice to the neX(11)t generation.
    Arm A. Geddon
    • Did you look at the results for System #3?

      That's my only data point here, but I was surprised.
      Ed Bott
      • re: System #3

        of course I looked. I do happen to read the whole article. :-) in fact, I'm always ready to give a thumbs up or down too.

        any computer at home, is one I've built and every time I build a new one, the specs get better so I hope vista's startup is quicker.

        gnu/linux...giving choice to the neX(11)t generation.
        Arm A. Geddon
        • wrong on the computers.

          guess I'm wrong when it comes to the macs, I can only upgrade them. ;-)

          gnu/linux...giving choice to the neX(11)t generation.
          Arm A. Geddon
      • Surprise

        Ed,

        I'm surprised by the Ubuntu boot time. This morning I started up an old test machine with SuSE 10.1 installed. It is a Compaq SFF Deskpro with a Pentium Coppermine 1GHz processor, 512Mb of RAM. Autologin is not enabled and there are a number of accounts depending what testing is being done. As a result logging in on a cold boot is a two stage process.

        After the POST completed (this was on a cold boot) and using the office clock as a timer the login prompt appeared in 50 seconds. After entering a user account and password, the time from pressing ENTER to fully loaded and ready to go was 15 seconds.

        So in total that's 65 seconds on a "steam-driven" piece of kit. I don't know what Ubuntu was doing. My only suggestion is it if was not connected to a network and was trying to use the NTP client then that would cause a 30 second pause while it times out. I've seen this happen with SuSE.

        Maybe a quick look in /var/log/messages would be informative?
        bportlock
        • Fully connected to the network

          I've had no other issues with this machine and Ubuntu, and once it starts up it's fine.

          I'm trying to get ready for a trip right now and won't get to check the log files till I get back next week.
          Ed Bott
        • From your own numbers ...

          ... your Linux boot times are comparable and in some cases longer then Vista boot times. It seems that you have confirmed the authors point nicely!
          ShadeTree
          • Please stop trying to mislead....

            ... the boot times for Vista published by Ed were generally over 1 minute. One time was 44 seconds, but the average is well over a minute.

            The time I quoted was for [i]a 1GHz Compaq[/i] used for compatibility testing. Ed's machines were typically, multicores, or 2+ GHz.

            In other words, the antique machine running at 1GHz was comparable to boot times on multicore processors running Vista. So to revisit your "claim" I have indeed proved that SuSE 10.1 has a comparable boot time to Vista when running on a machine about 1/3rd as powerful as that needed for Vista.

            That's why I found Ed's Ubuntu boot time rather strange.
            bportlock
          • Nothing misleading about it!

            CPU spead has very little to do with boot times. The BIOS and hard drives have a much more significant impact. A good 10 to 20 seconds is tied up in post. Cards with their own BIOS extensions also contribute to the overall time. The truth is that most Linux distros even your favorite when not tweeked take as much or more time then most Vista installations. The author's tests bare that out and yours does nothing to refute it.
            ShadeTree
          • Actually, RAM has a lot more to do with improved times...

            ...on comparable systems.

            Take a system with a 1GHZ CPU & 512MB RAM and check boot times of the OS. I.e., from completion of BIOS boot and the start of reading the MBR to the OS being ready to log you in and the HDD being idle.

            Upgrade the CPU to to a 2GHz model of the same CPU family and test again. Not much difference.

            Downgrade the CPU to the 1GHz model and up the RAM to 1.5GB. Test again.

            Big difference!

            I think that Ed should reduce the RAM in all systems to 512MB and test again.

            Looking at the [URL http://www.dell.com/content/products/features.aspx/dimen_versatile?c=us&cs=19&l=en&s=dhs ] Dell site [/URL] two of these mid-level [Versatile PCs] home/home office PCs come standard with 512MB RAM and Vista Home Basic while two come with 1GB RAM and Vista Home Premium.

            Looking at the [URL http://www.dell.com/content/products/features.aspx/dimen_essential?c=us&cs=19&l=en&s=dhs ] entry level [Essential PCs] [/URL] three come standard as 512 MB RAM and Vista HB, one as 1GB RAM and Vista HP.

            Looking at the [URL http://www.dell.com/content/products/features.aspx/dimen_advanced?c=us&cs=19&l=en&s=dhs ] high level [Advanced PCs] [/url], all four come with 1GB RAM and Vista HP.

            I think Ed's 1.5-4GB RAM configurations are far from typical.

            I think BPortLock's 512MB configuration more typical.
            Logics
          • The minimum recommended system memory ....

            ... for Vista is 1 GB. 512MB systems will boot slow. Ed is right however that after the 1GB point additional RAM has little impact on boot times.
            ShadeTree
        • Nothing unusual in that log

          I just upgraded to Ubuntu 7.04 on this machine and repeated the test. Results are identical, perhaps even a couple seconds slower. Nothing in /var/log/messages seems to indicate any sort of problem.
          Ed Bott