Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror
Windows Operating Systems Software Hardware

Boot Windows Vista In Four Seconds 326

Posted by timothy
from the says-them dept.
arcticstoat writes "Asus' budget motherboard wing, ASRock, claims that it's found a way to load a clean boot of Windows from a full shut down in just four seconds, using its new Instant Boot technology. The technology takes advantage of the S3 and S4 features of ACPI, which normally enable the Sleep/Standby and Hibernation modes in Windows respectively. However, by calling them at different times in the boot-up and shutdown process, Instant Boot enables you to boot up to your Windows desktop in three to four seconds, even after a proper shut down. Two modes are available; Fast mode, which uses S3 and boots up in around four seconds, and Regular Mode, which uses S4 and apparently takes between 20 and 22 seconds to boot. The advantage of Instant Boot when compared with normal Sleep and Hibernation modes is that you get the advantage of a clean boot of Windows, without what ASRock calls 'accumulated garbage data,' and you also get the security of knowing that you won't lose any data if there's a power cut and you lose AC power. There's also a video of it in action at the link above."
This discussion has been archived. No new comments can be posted.

Boot Windows Vista In Four Seconds

Comments Filter:
  • Video (Score:5, Insightful)

    by AKAImBatman (238306) * <akaimbatman@gma i l . c om> on Wednesday November 12, 2008 @06:04PM (#25740243) Homepage Journal

    Those guys in the video are having WAY too much fun with their jobs. "Why your computer boot so fast? I must get ASRock motherboard!"

    Now I know why ASUS mobos tend to be so good. They encourage a fun workplace. ;-)

    • Re:Video (Score:5, Funny)

      by Anonymous Coward on Wednesday November 12, 2008 @06:21PM (#25740451)
      Really? I was sure I just learned Asus motherboards are such high quality because their engineers have a lame sense of humor, leading to a diminished social life with more time to devote towards motherboard perfection.
      • by geekoid (135745)

        There so good bacause they strive to be as good as gigabyte boards. Always falling short.

        On the plus side.

    • by duguk (589689)
      Honestly? I've had problems with both ASRock and ASUS motherboards, generally to do with BIOS problems - all on Windows XP and Vista. I've never had one myself and can only comment on my customers' machines, this is purely my experience, but often I've had machines that just won't boot, and after a BIOS reset and/or BIOS upgrade, they've been fine. One had to have a lower processor put in before an upgrade could be done; yet it was working fine! Is this just something I've had problems with as a fluke, my
      • I bought an asrock for a Pent D 805 to test out the over clocking I could get on the that CPU. I had stable to 4.3 Ghz on air. Granted the air cooling also used a floor fan, but it was stable. The funny think was the motherboard liked every linux distro I tried. The same cannot be said for the Asus, Gigabyte, and Biostar motherboards that I have tried. Something came up that made ne switch to a different distro at install with the other brands. I figured with the AGP and PCIE slots the asrock would cause mo

        • by duguk (589689)
          So, a no then? iirc it was an AM2 with SLI I had the weird problem with it not booting, yet had been up until then. No new hardware or anything but the update (after putting a lower CPU in first) fixed it.

          I've got a Gigabyte AM2 at the moment, running Gentoo Linux, and Arch in the past. I have had some trouble with it a year back, but not had any problems with newer kernels. Hope that helps!
      • Re: (Score:3, Insightful)

        by Netsplitter (983360)
        My personal experience is the same as yours. The motherboards are great, if you can get your hardware to work on them. I've found compatibility with different kinds of memory to be especially terrible, but they usually work after BIOS updates which, thankfully, Asus seem to release often. Of course, one could argue they should get it right the first time; I do, and it's why I don't buy Asus anymore.
        • by duguk (589689)
          I'm glad its not just me! Agreed, I'm glad they do release their updates often too. Certainly better than some manufacturers, but some models do seem to have some problems, especially with the first revisions. The same with any manufacturer I guess, purely often down to hardware compatibility.
    • by Cylix (55374)

      Asus has an trend of set it and forget it.

      When other motherboard manufacturers patch some holes (sometimes as best can be done) Asus happily moves on to the version 2 of the motherboard. I have purchased at least three motherboards that experienced this same fate. One of the last system builds stability was partially increased with a "beta", but the update can easily bork the board. (Mostly less agressive timing ratios, but the voltage regulators on the memory had some real issues).

      So the last board was ret

  • ASRock is not ASUS (Score:5, Informative)

    by vwpau227 (462957) * on Wednesday November 12, 2008 @06:06PM (#25740265) Homepage

    ASRock is not ASUS. Hua Ching, the subsidiary that was spun off from ASUS is not any longer a part of the ASUS organization. See http://www.theinquirer.net/en/inquirer/news/2002/11/05/asus-distances-itself-from-asrock-subsidiary [theinquirer.net] for details. There are a number of companies locally and elsewhere that have been pushing cheap ASRock mainboards as being the same quality as ASUS mainboards. We have seen many issues with the ASRock mainboards, both in premature failure and incompatibilities, that we have not seen at all in ASUS mainboards. ASUS has its own low-end set of mainboards and they are much better than the ASRock, from my experience. The sooner this sort of misinformation gets sorted out, the better informed the consumer will be.

  • by Anonymous Coward on Wednesday November 12, 2008 @06:07PM (#25740271)
    Unfortunately a proper shutdown takes 2 minutes ....
    • by Dan541 (1032000) on Wednesday November 12, 2008 @08:18PM (#25741609) Homepage

      Really I can shut my laptop down in less than 10 seconds, leaving the battery cover off enables me to improve that to 2-3 seconds.

      • by Khyber (864651) <techkitsune@gmail.com> on Wednesday November 12, 2008 @08:39PM (#25741793) Homepage Journal

        Hp laptops only take half a second to shut down - at least the DV 2000, 6000, and 9000 models, because the slightest bump to the battery release latch most often causes the battery to drop out, contacts first, causing immediate shutdown if you're not plugged in.

        Side note and totally off-topic - WHAT THE FUCK SLASHDOT? Your site has been making my pop-up blocker and Firefox go fucking NUTS today!

        • by binaryspiral (784263) on Wednesday November 12, 2008 @11:43PM (#25742991)

          Side note and totally off-topic - WHAT THE FUCK SLASHDOT? Your site has been making my pop-up blocker and Firefox go fucking NUTS today!

          Ditto... popups are probably the least effective on sites that attract technology minded folks - no, they don't see the ads - it just pisses them off everytime the blocker has to prevent them.

          Screen ads much, Taco?

        • Re: (Score:3, Informative)

          by PJ1216 (1063738) *
          Thats odd. I don't have any ad-blockers (only popup blockers) but I haven't gotten any popup notifications saying any popups have been blocked at all yesterday or today.
    • Re: (Score:3, Interesting)

      by rtb61 (674572)

      In this case it is likely true. The only way to accelerate clean boots would be to create it during shutdown, so a clean boot for standby or a clean boot for hibernation. Basically clearing memory of it's current running state and replacing it with a clean boot state and using that as the stored standby or hibernation state. Configuration changes would get messy. Far cleaner just to embed core elements of the Linux kernal straight into the CPU, you can only really do it with Linux because of course closed

  • Cool. (Score:3, Interesting)

    by Anonymous Coward on Wednesday November 12, 2008 @06:08PM (#25740285)

    I wonder if something like this could be done with Linux now that 2.6.27.5 has been out for a few days and that situation with the RESET_REG_SUP bit has been resolved. This certainly is great news for Vista users looking for a new board.

    • Re: (Score:3, Insightful)

      by GroundBounce (20126)

      It shouldn't be hard, as long as you have any kernel/motherboard combination where suspend or hibernate work reliably. Just create a shutdown level that actually does a reboot, then modify the startup scripts to immediately do a suspend or hibernate as soon as the machine is booted if that shutdown level was used. The effective shutdown time will be longer (because it's actually a reboot), but the effective "boot" time will be very quick.

      While probably do-able, this actually seems like overkill. Why not

    • by k1e0x (1040314) on Wednesday November 12, 2008 @07:31PM (#25741205) Homepage

      I wonder if something like this could be done with Linux now that 2.6.27.5 has been out for a few days and that situation with the RESET_REG_SUP bit has been resolved. This certainly is great news for Vista users looking for a new board.

      It's been done in 5 seconds..

      Doesn't even require a special motherboard, they did it by modifying Fedora on a EEE pc (something not known for it's speed)
      http://lwn.net/Articles/299483/ [lwn.net]

      Video: http://www.youtube.com/user/arjanintel [youtube.com]

      • Re: (Score:3, Interesting)

        by k1e0x (1040314)

        After reading how the implementation in windows is done by doing pre-hibernation, it makes the Linux method look even more impressive. They start from a clean cold boot.

        Still pre-hibernation is pretty interesting.. I never shut my systems off anyhow though, I always believed that systems left on have a longer life span because the chips never cool down, but that can't be true for some of the computers moving parts..

      • by DrYak (748999) on Wednesday November 12, 2008 @10:54PM (#25742711) Homepage

        It's been done in 5 seconds..
        Doesn't even require a special motherboard, they did it by modifying Fedora on a EEE pc (something not known for it's speed)

        And above all the strategy behind the speedy boots are completely different :

        - Linux basically Just boots in 5 seconds. In a plain normal fashion. Everytime you push the button, no matter what, the system is up in 5 seconds (well except if you trashed your machine and disk have to be rechecked).
        But it's a plain standard boot.

        - Whereas, for Windows, ASRock has to resort to abusing the sleep/hibernate system. With the subtle difference is that they are not actually suspending the system to RAM/disk (in order to avoid accumulating garbage, as they say).
        They are resuming a special suspended "freshly booted" state.
        i.e.: when in fast boot mode, you are not actually booting Windows. You are resuming an image of a "Windows-that-just-got-started" suspend on RAM.

        The main implication is that the first time you boot, and after each system update (and you know that, given microsoft's track of security, you're still going to have patches coming often) or any other change that might render the pre-suspended image obsolete, you can't do this. You have to go through a slow boot, rebuild a pre-suspended state, and only after that it'll work.

        It's not a standard boot. It's not as robust as a real boot, and frequently it won't work. (not to mention that these pre-suspended image will be the perfect place to inject a vm-based rootkit).
        In short : not as useful as you would hope.

  • by Anonymous Coward on Wednesday November 12, 2008 @06:09PM (#25740287)

    I'd boot Vista off my work PC in a millisecond if I could

  • by Svartalf (2997) on Wednesday November 12, 2008 @06:09PM (#25740295) Homepage

    It sounds like all they did was allow you to store a Hibernate to Disk snapshot of your system at startup before anything else gets done- which is technically cheating. ANYTHING can boot up in about 4 or seconds that way. :-D

    • by Shin-LaC (1333529) on Wednesday November 12, 2008 @06:14PM (#25740375)
      That was my first thought too. Do they make any attempts at detecting whether the OS was updated, or new software was installed that requires a reboot, so they can perform a full boot and update the startup snapshot when needed?
      • by erikina (1112587) <eri.kina@gmail.com> on Wednesday November 12, 2008 @06:22PM (#25740463) Homepage
        My guess, is that it wouldn't matter. As when you turn off their computer, they probably behind the scenes turn it back on again. Then hibernate. So a normal "reboot" would be a little slower than usual, and to a user every power on is like opening a fresh copy.
        • by Tacvek (948259) on Wednesday November 12, 2008 @06:34PM (#25740611) Journal

          Correct. That is exactly what they do. So they just shift the boot time to be part of the shutdown time, so when you arrive at your computer again, and turn it on, you are just unsuspending it, or are loading an unusually clean hibernation file. This is a very interesting idea, but it is one that doe not need motherboard support. This can be done by the OS alone.

          • by roc97007 (608802) on Wednesday November 12, 2008 @07:49PM (#25741383) Journal

            This used to be called "cutting part off the string on this end and tying it to the other end". The technique is sometimes useful, but in this case, what if you're also concerned about shutdown time? For instance, I sometimes shut down my laptop at the end of a meeting for various reasons. Shutdown time is important because I have to wait until it shuts down completely before closing it, else it'll suspend and then resume shutting down when I'm trying to boot it up.

            Windows *already* takes too long to shut down -- I'm not sure I want to wait even longer so it can also do prework for the next boot. Instead of tricks like this, why not load less cruft at boot?

            • Re: (Score:3, Insightful)

              by AmberBlackCat (829689)

              I think this is an excellent time for a car analogy. In the coldest of winter, I sit in my tiny subcompact sedan for 10 minutes waiting for it to reach a normal operating temperature. Would it be so bad if they found a way to make the car warm up in 4 seconds, with the side effect being it takes an extra 15 minutes go cold again after I turn it off?

              At the risk of sounding self-centered, I think a lot of computer users are like me. They approach their computer and want it to start working as quickly as possi

    • by vux984 (928602)

      It sounds like all they did was allow you to store a Hibernate to Disk snapshot of your system at startup before anything else gets done- which is technically cheating. ANYTHING can boot up in about 4 or seconds that way. :-D

      Strange... my system takes MUCH LONGER to wake from hibernate than it does from a cold boot. I get some black/green screen that says 'resuming windows' for a few seconds, and then black for up to a minute, then the login screen where I enter my password, then my desktop comes up but the

    • It sounds like all they did was allow you to store a Hibernate to Disk ...

      I don't think so. I routinely use hibernate and it takes around 20 seconds to boot back up. They're doing something beyond hibernate. In my case it doesn't really matter because I use a Sony CRT that takes 15 seconds to power on. So this feature would only save me around 5 seconds on boot.

  • Did they automate the logon? I wouldn't want something on my system that just drops to desktop. ESPECIALLY if I have to go travelling and can't have the laptop in my possession at all times.

  • by crt (44106) on Wednesday November 12, 2008 @06:11PM (#25740323)

    They never show the shutdown process - my guess is that when you shutdown, it actually reboots, then right after Windows boots it puts it to sleep or hibernate (S3/S4). When you turn it back on, it wakes it and looks like you "just" booted up.
    Not really a bad idea I suppose - moves the boot time from boot to shutdown, when you are less likely to care.
    Of course you can get the same effect yourself by rebooting then just putting your machine to sleep when you want to shutdown. Someone could probably even write a simple software solution for this rather than requiring a whole new motherboard.

    • Re: (Score:3, Interesting)

      by rsborg (111459)

      Of course you can get the same effect yourself by rebooting then just putting your machine to sleep when you want to shutdown. Someone could probably even write a simple software solution for this rather than requiring a whole new motherboard.

      I fail to see how what they've done is a bad thing. And I don't think it's quite as easy to script, since you'd have to reconfigure how windows works. If you have an inkling of how to do this in a smooth automated fashion, please do tell me.

      Big question is, if you pull

      • by Bill, Shooter of Bul (629286) on Wednesday November 12, 2008 @06:25PM (#25740505) Journal

        Big question is, if you pull the cord, does it maintain state? Or will that require a "cold boot"?

        They answered that "big question" in the article. There are two different options the "super fast" boot mode that does the " boot" in four seconds. And a regular fast boot that takes 22 seconds. The four second super fast one, needs to stay plugged in to maintain state. The slower fast one does not.

        I wouldn't recommend swapping out hardware in either case.

        • by Gordo_1 (256312)

          Super fast = reboot at shutdown, then boot OS and go into S3/standby before login; S3 requires constant trickle of power to maintain RAM state.
          Regular fast = reboot at shutdown, then boot OS and go into S4/hibernate before login; S4 writes memory contents to disk, so no power required. Takes longer to "boot" back up as it has to read each bit back into RAM.

          Neither scenario would generally permit swapping out hardware that needs to be enumerated by the BIOS, such as RAM, video cards and SATA disks. USB devic

      • by bluefoxlucid (723572) on Wednesday November 12, 2008 @06:55PM (#25740843) Journal

        And I don't think it's quite as easy to script, since you'd have to reconfigure how windows works. If you have an inkling of how to do this in a smooth automated fashion, please do tell me.

        A shutdown script that places C:\Windows\_reboot_quick_boot as an empty file on the file system when you shutdown, and cancels the shutdown in favor of a reboot ('shutdown -a' on Windows, 'shutdown -c' on Unix).

        A boot-time script that runs last and waits about 10 seconds at the log-in screen for keyboard input (and to give the start-up applications a chance to settle, so they don't have to do a lot of work and thrash disk on resume). If it gets keyboard input, it exits. Otherwise, it initiates a hibernate action and then exits. When you resume this hibernate, the program either has exited or is exiting.

        Do I get a blow job now? (I'm a Unix kid, these kinds of problems are obvious to me)

        • Oh, and the boot-time app exits early if that file doesn't exist (rather than hibernating), and deletes the file otherwise. My bad, but this should be obvious.
    • by Egotistical Rant (42993) on Wednesday November 12, 2008 @06:22PM (#25740465)

      This is EXACTLY what it does. The "more images for this article" section at the given link has a flowchart of the process...it's just a reboot and suspend.

    • by Tacvek (948259)

      I'm pretty sure you are correct. That is what the diagram on the page (at the bottom, the left of the two thumbnails). A very interesting concept, since few people will care about the extended shutdown time, but will enjoy the faster "startup" time.

    • by swillden (191260) <shawn-ds@willden.org> on Wednesday November 12, 2008 @07:06PM (#25740939) Homepage Journal

      Of course you can get the same effect yourself by rebooting then just putting your machine to sleep when you want to shutdown. Someone could probably even write a simple software solution for this rather than requiring a whole new motherboard.

      Hmm. It seems like it'd be really easy to do this with an open source OS. I think I may have just found a nifty little project for this weekend. All it should take is:

      • Add an inittab runlevel (7?) for "shutdown to instant boot".
      • Add an /etc/rc7.d with a script that writes a file that records the fact that we're in "shutdown to instant boot" state, then switches to runlevel 6.
      • Add an init script in late in the normal startup sequence that checks for "shutdown to instant boot" state. If it finds that state, it removes the file and then initiates suspend or hibernate, depending on a configuration option.

      At that point "sudo init 7" should cause your machine to shut down to "instant boot" state. Hitting the power button will then "instant boot" it.

      "sudo init 0" or "sudo init 6" will do a normal shutdown or a normal reboot.

      The final step would be to modify the "shutdown" command to go to runlevel 7 when given some new option, and then to modify the GUI-based shutdown tools to provide the instant-boot option as well, and maybe make it the default. Oh, and maybe modify the ACPI script that's executed when the power button is hit so that the power button does a "shutdown to instant boot" by default.

      Pretty easy. Of course, in Linux I don't ever see any reason to shut the machine down anyway. My laptop pretty much only gets rebooted when there's a kernel update to install. Other than that, it just gets suspended. So, kind of pointless in Linux, but easy. The same would apply to *BSD.

    • by HTH NE1 (675604)

      Not really a bad idea I suppose - moves the boot time from boot to shutdown, when you are less likely to care.

      That would depend on when you want the updates that require restart to be applied.

      And what happens after this "shutdown" and you, say, install new drives? Will it see them when the BIOS (or whatever) has already decided they aren't there? What if you want to boot off a different device like a CD?

      This reminds me of the episode of Turbo Teen where, if he transforms back into the car, the timer on the bomb planted inside him will resume counting down. (Or, if it wasn't an episode, it was a vivid dream I had as

      • And what implications for security like disk encryption?

        If you do WDE the right way, here's what happens:

        1) In your OS, select "shutdown".
        1a) User removes boot USB key from safe.
        2) System restarts, then POSTs.
        2a) User inserts boot USB key into computer.
        3) System boots from USB key and prompts for passphrase.
        4) User enters passphrase.
        5) OS boots, then hibernates.

        To wake the system, follow the obvious steps.

        Now, if you suspend to RAM, you won't need to boot from the USB key, but they keys to your WDE will still be in RAM.

        And what happens after this "shutdown" and you, say, install new drives?

        If it's a hotpluggable device on a system l

  • by mgemmons (972332) on Wednesday November 12, 2008 @06:13PM (#25740343) Homepage
    FTFA: Instant Boot will also only work on Windows systems (XP or Vista) with a single-user account and no password protection.
    • Re: (Score:2, Interesting)

      by Anonymous Coward

      Not quite as useful as it seems at first glance...

      You're absolutely right: it only works for about 95% of home users out there (a conservative estimate). Now, if it booted Linux as well, then it would be useful!

  • BUT... (Score:4, Funny)

    by Anthony_Cargile (1336739) on Wednesday November 12, 2008 @06:17PM (#25740399) Homepage
    Does it come prepackaged with winrar serials, cracks, M$ documents and source code?
  • What about shutdown? (Score:3, Informative)

    by Plocmstart (718110) on Wednesday November 12, 2008 @06:18PM (#25740401)

    Given the flowcharts (not the shiny video that catches your attention first) it appears that instead of shutting down, they simply reboot the system and once it reaches the state where the OS has finished loaded it then goes to sleep or hibernate. Once you power it back on it just returns to the freshly loaded OS.
    So it appears that while it starts up faster, you should end up spending more time shutting down (actually rebooting and reloading the OS). You could also do this manually by rebooting Windows and once it gets to the desktop/login screen go into hibernate/sleep.

    • Given the flowcharts (not the shiny video that catches your attention first) it appears that instead of shutting down, they simply reboot the system and once it reaches the state where the OS has finished loaded it then goes to sleep or hibernate. Once you power it back on it just returns to the freshly loaded OS.
      So it appears that while it starts up faster, you should end up spending more time shutting down (actually rebooting and reloading the OS). You could also do this manually by rebooting Windows and once it gets to the desktop/login screen go into hibernate/sleep.

      It would rock if they could load all this information up in a virtualized environment during idle time while the computer is booted, and then save that image to the hard drive before you even try to reboot. It would need to coordinate itself with windows/program updates to keep that image correct, but it would be the ideal way i think. This could actually really catch on if it's done that way.
      -Taylor

  • Home test (Score:5, Funny)

    by konohitowa (220547) on Wednesday November 12, 2008 @06:23PM (#25740479) Journal

    I pulled on a pair of boots and managed to beat their time by more than 3 seconds.

  • Cheating... (Score:5, Interesting)

    by sofar (317980) on Wednesday November 12, 2008 @06:31PM (#25740575) Homepage

    This is still cheating - it's first of all not actually booting but suspending/resuming (albeit smartly).

    Most importantly the system is not actually shut down, so it still draws power to refresh the memory. This will likely suck on high-performance laptops where the large amounts of ram with high voltages will suck the battery dry in a substantially short time.

    And worse, this technology will take a _long_ time to shutdown. It's sacrificing a lot. We can (really) boot+shutdown a linux box in +- 10 seconds. Would you want a 3 second boot if your shutdown time becomes one minute?

    For people who are on the go a lot and tend to open/close their laptops a lot, this may actually reduce their effective work time a lot.

    • Re:Cheating... (Score:4, Insightful)

      by Colonel Korn (1258968) on Wednesday November 12, 2008 @07:30PM (#25741197)

      This is still cheating - it's first of all not actually booting but suspending/resuming (albeit smartly).

      Most importantly the system is not actually shut down, so it still draws power to refresh the memory. This will likely suck on high-performance laptops where the large amounts of ram with high voltages will suck the battery dry in a substantially short time.

      And worse, this technology will take a _long_ time to shutdown. It's sacrificing a lot. We can (really) boot+shutdown a linux box in +- 10 seconds. Would you want a 3 second boot if your shutdown time becomes one minute?

      For people who are on the go a lot and tend to open/close their laptops a lot, this may actually reduce their effective work time a lot.

      1) It only draws current in "Fast Mode." The "Regular Mode" still allows for 22 second booting with no power draw.

      2) Most people won't mind the slow shut down, even in a scenario where the computer is being turned on and off frequently. Why? You shut down because you want to do something else, so the computer can painlessly finish the process in the background where you don't need to notice it. Yes, it will hurt battery life by extending the powered up state for a little longer than your actual usage time, but it then shaves most or all of that off of boot time, which is subjectively more critical.

      3) How do you boot a Linux box in negative 10 seconds? "+- 10" looks like a statement of error bounds to me. I think you meant "~10."

      • Re: (Score:3, Insightful)

        by Tony Hoyle (11698) *

        2) Most people won't mind the slow shut down, even in a scenario where the computer is being turned on and off frequently. Why? You shut down because you want to do something else, so the computer can painlessly finish the process in the background where you don't need to notice it. Yes, it will hurt battery life by extending the powered up state for a little longer than your actual usage time, but it then shaves most or all of that off of boot time, which is subjectively more critical.

        Actually this matters

    • This will likely suck on high-performance laptops where the large amounts of ram with high voltages will suck the battery dry in a substantially short time.

      For people who are on the go a lot and tend to open/close their laptops a lot, this may actually reduce their effective work time a lot.

      It's a good thing ASRock only makes desktop motherboards, then.

      I mean, theoretically you could install their utility on any machine that supports S3 or S4, but they've most likely tied it to their hardware somehow.

  • How long until a Linux Distro uses that type of Motherboard for doing that? ALso... I want to see how fast Gentoo boots with that... What kind of implications would this have for servers / hypervisors?
    • Re: (Score:3, Insightful)

      by Tacvek (948259)

      The motherboard is not required. Just have your OS automatically hibernate (or for the 3 second boot, automatically suspend) on each clean boot. Then to "shutdown" your computer, just reboot it. Your OS will shut down, and then start again, and enter sleep or hibernation just after it finishes it's restart. So when you approach your computer again, and push the power button, the computer "boots" with very high speed.

  • by argent (18001) <peter@NOspaM.slashdot.2006.taronga.com> on Wednesday November 12, 2008 @06:43PM (#25740719) Homepage Journal

    What it looks like they're doing:

    1. They're taking a snapshot of the system that was made when you hibernated, and restoring that snapshot.
    2. Next time you shut down it restores the same snapshot you made the first time, after your original clean boot.

    The only real issue I see is that your file system cache (and any other file system state) now contains garbage, and will need to be invalidated (NOT FLUSHED). If the cache was left out of the original snapshot then just remounting the file system from scratch should solve that. This isn't really booting (you'll need to repeat the whole process after just about any time you modify system state, including a lot of things like registry changes), but it's also not specific to Windows and should DTRT with Linux, etc...

  • by holywarrior21c (933929) on Wednesday November 12, 2008 @07:15PM (#25741029)
    it won't boot vista in 4 sec, never. at startup 3 expired anti virus programs, 4 messengers, 3 weather programs, 2 ad block...
  • by istartedi (132515) on Wednesday November 12, 2008 @07:46PM (#25741351) Journal

    The video suffers from YouTube syndrome. Symptoms of this condition include one or more minutes of stuff we don't care about. The condition is most extreme when the event we care about is of short duration. Although it is not entirely curable, it can be treated by posting the time of the interesting event in the comments section. In this case, the event in question occurs at 2:38. Remember, this is a treatment not a cure. The treatment still consumes bandwidth and time. In the future, we may have a Flash plug-in where the annotation feature can be used to denote points of interest, with the ability to skip to a keyframe just before said point. Until then, only video posters can prevent YouTube syndrome. Remember, if the event of interest in your video is of short duration, the video should not be any more than twice as long as the actual event. Introduction, at most, should identify you and/or your company. Anything that can be explained more efficiently as text should be put in the little text section that appears in the upper right.

  • http://www.gigabyte.com.tw/Products/Storage/Products_Overview.aspx?ProductID=2678&ProductName=i-RAM%20BOX [gigabyte.com.tw]

    Just get an I-Ram and there you go. True, you'll only have a 4GB partition for your boot drive, but it's the real deal if you have to boot in 4 seconds. Yes, it's closer to 10-12 given the typical POST process, but since that varies from board to board, it is about 4 seconds total from the time Windows itself starts until it gets to the desktop.

    When the 2.0 version someday comes out, it'll do 16GB,

  • Exciting (Score:2, Interesting)

    by bill_kress (99356)

    I've virtually NEVER had windows work reliably after a hibernate/resume. I acknowledge that there is probably a hardware/software combination out there that can do it, but I've never encountered one.

    For me, things just get unstable, and after 2 or 3 hibernates without a full shutdown, the whole system fails (if it didn't fail on the first resume).

    So although this may work on a pristine install, the thought of owning one of these after a few patches from Microsoft, installing an anti-virus and a few drivers

  • by PPH (736903) on Wednesday November 12, 2008 @08:28PM (#25741709)

    clean boot of Windows, without what ASRock calls 'accumulated garbage data,'

    you also get the security of knowing that you won't lose any data

    You insensitive clod! All my data IS accumulated garbage!

  • by Fantastic Lad (198284) on Wednesday November 12, 2008 @09:45PM (#25742205)

    I'm really beginning to like ASUS. Not only do I like their approach to hardware a lot, but that was the most straight forward, unintentionally entertaining, non-bullshit video I've EVER seen from a corporation. I mean, yeah, it was dippy YouTube stuff, but seriously; Can you imagine any other main stream computer company allowing their tech monkeys to represent the mothership without the message first passing through a legion of marketing directors, lawyers and various haircuts in suits? Hell no!

    Imagine the plasticy, dumbed down crappola video with gawdawful elevator music and bad overlay graphic effects you would have gotten in nearly every other situation.

    Damn. ASUS is run by humans!

    -FL

  • by RogueSeven (965183) on Thursday November 13, 2008 @12:14AM (#25743147)
    Four seconds? That's nothing. I booted Windows Vista the very second I got my new machine to make way for Ubuntu.

    What's that, you didn't mean literally kick? I see.

We can defeat gravity. The problem is the paperwork involved.

Working...