Ubuntu 9.04 Daily Build Boots In 21.4 Seconds

Pizzutz writes "Softpedia reports that Ubuntu 9.04 Boots in 21.4 Seconds using the current daily build and the newly supported EXT4 file system. From the article: 'There are only two days left until the third Alpha version of the upcoming Ubuntu 9.04 (Jaunty Jackalope) will be available (for testing), and... we couldn't resist the temptation to take the current daily build for a test drive, before our usual screenshot tour, and taste the "sweetness" of that evolutionary EXT4 Linux filesystem. Announced on Christmas Eve, the EXT4 filesystem is now declared stable and it is distributed with version 2.6.28 of the Linux kernel and later. However, the good news is that the EXT4 filesystem was implemented in the upcoming Ubuntu 9.04 Alpha 3 a couple of days ago and it will be available in the Ubuntu Installer, if you choose manual partitioning.' I guess it's finally time to reformat my /home partition..."

Comments

Your Goal: One Second or Less

[alain94040]

This is one of my pet peeves: why can't computers boot in a second or less?

Imagine a visionary like Steve Jobs (by the way, enjoy your leave of absence and please come back). He goes to his team and says "I don't care what it takes, build me a computer which boots in one second".

Ignore the past, the legacy of tens of years of layer after layer of OS software. Can it be done?

A 3 GHz dual-core processor can process 6 billion instructions in that first second. I know the disk is a problem. I'm not asking for all possible OS services to be up in a second... But I'm sure this could be improved greatly. It's all out there in the open. People want this.

--
FairSoftware.net -- work where geeks are their own boss

Re:Your Goal: One Second or Less

[beelsebob]

He already did that â" that's why OS X 10.3 took about 40 seconds to boot, and 10.4/10.5 take about 4 seconds. They fixed it by replacing initd with launchd, which schedules dependant tasks, and makes optimal use of resources.

Re:Your Goal: One Second or Less

[wizardforce]

here:

http://www.linuxdevices.com/news/NS5429881813.html
http://lwn.net/Articles/299483/

Re:Your Goal: One Second or Less

[rgbe]

I agree totally. 21.4 seconds is incredibly slow, and that's only to get to the login screen... which is typically only half way. I know that it is possible to boot Linux in 5 seconds for some special cases. However, the boot time should be even quicker.

Re:Your Goal: One Second or Less

[CannonballHead]

I think there's more to it than that, though, too. For example, you'd have to completely bypass all checking, device discovery, etc., on boot (it takes time to discover drives, PCI/PCI-E/ISA ;) /USB device. Yeah, you could just have that set up in BIOS or something and just use that configuration, but that could be a pain, too.

Now, if we're talking about post-POST boot-up, I think something could be done there. Even if it was having the option of, oh, 8GB of onboard memory dedicated to having a fast-boot operating system.

As far as the extremely fast-boot idea goes, though, isn't that sorta what Good OS's partnership of Cloud and GIGABYTE is supposed to be? The GIGABYTE Touch Netbook M912 to be precise. Link here. It was mentioned on slashdot a while ago as well.

Re:Your Goal: One Second or Less

[kcbanner]

Yea, I mean I run archlinux and I'm up and logged in, about 10-15 seconds...Dell laptop so nothing special

Re:Your Goal: One Second or Less

[gmuslera]

After reaching the goal of not having to reboot it at all for years, reaching the one of booting it in N seconds dont look so special.

To be fair, power outages happens, not always suspend in notebooks is the best choice, and kernel updates happens too, And sometimes you want to have downtime as minimal as possible.

Re:Your Goal: One Second or Less

[Anonymous Coward]

http://lwn.net/Articles/299483/

Intel engineers have significantly reduced the boot time. However, there's a lot of hacks that need to be done to do so. Hopefully, a lot of those changes will make it in. At 10 seconds, you've got a PC that boots really fast. At 5, you've got a PC you're more likely to shutdown than hibernate or sleep (Linux has session restore which, for the most part, counteracts some of the advantages of hibernating of shutting down).

Also, 1 second boot is probably not possible simply because it'll probably take that long just to read all the initialization files. Maybe with SSD it might become a possibility, but even still your system becomes a lot harder to maintain.

For instance, you have to have all driver support compiled in ahead of time, meaning either you have a very custom kernel or you have a kernel that is way bigger than you need.

Re:Your Goal: One Second or Less

[Anonymous Coward]

Maybe the workstations and servers do, but my 10.5 powerbook sure doesn't. Even when you get to the gui, it still takes another 10+ seconds before anything actually works.

Re:Your Goal: One Second or Less

[Cyberllama]

So in order to be a "visionary", I merely have to decide what consumers might want (not that hard being one yourself), and then ask people smarter than yourself to make it happen with no actual technical insight on how to make it happen yourself?

Re:Your Goal: One Second or Less

[vux984]

A 3 GHz dual-core processor can process 6 billion instructions in that first second. I know the disk is a problem. I'm not asking for all possible OS services to be up in a second... But I'm sure this could be improved greatly. It's all out there in the open. People want this.

Hard to say if there's really a point to booting up before the services are running.

What good is the PC being 'at the desktop' if the search service still hasn't started, the network still hasn't obtained an ip-address, half my tray icons aren't up? and the hard drive is still madly churning to get everything else running, so anything I try and launch is just going to be thrown into the queue and it probably will depend on something that hasn't started up yet anyway.

Seriously, how much stuff could you really -defer- to after seeing the desktop and have a useful system?

Remember the average hard drive moves under 50MB/s. Even a fairly modest Ubuntu desktop requires several times that much RAM. If the hard drive started loading data at maximum speed you've got maybe 50MB you can load in that time, and probably far less in actual practice. That means your kernel, drivers, HAL, desktop environment, localization, firewall, network, background, theme, etc has to ALL fit in under 50MB. And you'd need some sort of impossible situation where the cpu could run all the initialization code for all that in parallel, without waiting... nevermind that it almost has to be initialized in sequence due to the layer dependancies.

If you want instant on PCs, the only real solution is to never turn them off, waking from suspend to RAM is about as good as its going to get for the forseeable future.

Re:Your Goal: One Second or Less

[log1385]

Here's basically what happens when most computers boot up: the BIOS is loaded first, and makes sure that all of the important hardware is working. Then the BIOS looks for a bootable device (usually your hard drive). After that, the instructions of in the boot sector of your hard drive (or other bootable device) takes over and loads the OS.

That said, I think it's reasonable to ask for a 1 second boot time. Here's hoping.

Re:Your Goal: One Second or Less

[Kent Recal]

10.4/10.5 take about 4 seconds

On what hardware?
I have a brand new MacBook sitting next to me that takes pretty exactly 30 seconds from the push of the button to a usable desktop. Is it broken?

Re:Your Goal: One Second or Less

[Arterion]

I know the disk is a problem.

Absolutely. And if you want to have everything cached to RAM, it's significantly faster. In fact, someone already thought of that and engineered the ACPI S3 power state.

Re:Your Goal: One Second or Less

[neuromanc3r]

and 10.4/10.5 take about 4 seconds.

I'm sorry, but that is bullshit. I'm writing this from a MacBook running 10.5, which does not boot in anywhere near that time.

Re:Your Goal: One Second or Less

[diegocgteleline.es]

That's why suspend-to-mem exists.

Re:Your Goal: One Second or Less

[edbob]

You are absolutely correct. My first computer had a processor that ran just under 1 MHz and it booted up in maybe 2 or 3 seconds. If it didn't, I just turned it off and back on and it was up and ready. With today's dual- and quad-core processors running 3000 times faster, I think that the computer should be booted up and ready to go before I lift my finger from the power button!

Re:Your Goal: One Second or Less

[Tubal-Cain]

If it boots in less that 1/3 to 1/6 as much time as ext3... Surely there will be an improvement in overall performance?

Re:Your Goal: One Second or Less

[legirons]

He already did that â" that's why OS X 10.3 took about 40 seconds to boot, and 10.4/10.5 take about 4 seconds.

Nice try. After 4 seconds, a Mac with OS X 10.4 won't even have turned on its video output. After 2 minutes you might see the boot screen, but good luck getting any programs to run within the first 3-4 minutes.

It's simply not comparable to something like the asus eee, and the difference is noticable.

(being able to resume from sleep however, makes OS X 10.4 a world of difference better than Windows)

Re:Your Goal: One Second or Less

[Surt]

The hard drive is really the limiting factor. There's just no getting around pulling say a GB worth of data off of it (to cover all your device drivers and core operating system, etc). That alone will cost you 10 seconds, which is roughly the time that it takes to resume from hibernate. Now, you could have a staged boot, where you bring up services as needed, which might give the illusion of a faster boot, while actually making you wait the same amount of time to ultimately do anything.

Re:Your Goal: One Second or Less

[InlawBiker]

My old Atari could boot instantly! We could have the same now, except that nobody wants their OS on a chip.

Or do they? My guess is that nobody has put much effort into it. Maybe it's not considered a main selling point.

My machines actually boot pretty quickly. It's the waiting-for-desktop that's slow. Especially on the work laptop where the vendor has preloaded their useless helper apps.

Re:Your Goal: One Second or Less

[ruiner13]

This is one of my pet peeves: why can't computers boot in a second or less?

Why do computers need to reboot at all?

Re:Your Goal: One Second or Less

[RedHelix]

Asus has a few motherboards out there that offer an instant-on OS that'll take you to a web browser, skype client and a few other goodies. You aren't up in a *second* but it's pretty nice.

Re:Your Goal: One Second or Less

[MMC Monster]

I'll have to check how long my aluminum macbook takes to boot. Certainly more than 4 seconds. Probably less than 4 minutes, I think.

That being said, sleep & restore work flawlessly. So good that the grandparent poster probably got confused and thought it was doing a full boot in 4 seconds (which sounds about correct for restore from deep sleep).

Re:Your Goal: One Second or Less

[LWATCDR]

Part of the problem is that we now expect the computer to detect new hardware. If you plug in a new Video card or monitor you expect the computer to find the hardware and start using it.

Re:Your Goal: One Second or Less

[Tubal-Cain]

the network still hasn't obtained an ip-address

Well, with NetworkManager on a wifi connection, you don't obtain an IP address until after login anyways.

Re:Your Goal: One Second or Less

[Thinboy00]

That means your kernel, drivers, HAL, desktop environment, localization, firewall, network, background, theme, etc has to ALL fit in under 50MB. [snip]

Why not use DSL? (Damn Small Linux, not a second phone line)

Re:Your Goal: One Second or Less

[beelsebob]

Yes, but then, as we're comparing penis sizes, lets do it fairly. TFA explicitly states that they time from after the boot loader is finished, to when the login window appears. Boot your mac, and time between the grey apple with a spinner appearing (the grey apple is displayed while the boot loader does its thing), until the login window appears.

Re:Your Goal: One Second or Less

[Thinboy00]

Software bloats faster than hardware accelerates. It's a sad fact of life.

Re:Your Goal: One Second or Less

[MadnessASAP]

I have Compaq servers here and I'm lucky if the hard drives have finished spiing up after 5 minutes, much less thought about loadign the OS.

Re:Your Goal: One Second or Less

[Hmmm2000]

(being able to resume from sleep however, makes OS X 10.4 a world of difference better than Windows)

You can resume from sleep/hibernation in windows as well, that is as close to an instant boot as you will get with modern operating systems.

Re:Your Goal: One Second or Less

[Gat0r30y]

It takes a good while just to spin the HDD up (~ 8 seconds if you have a big 4 platter desktop drive). If we move to a hybrid solution (an HDD with a couple Gigs of Flash on it), we could really improve boot times.

Re:Your Goal: One Second or Less

[Kent Recal]

Sorry to break it for you but boot time is measured from the push of the power button to a usable desktop.
You may enjoy your 26 seconds of pretending that "this is not really happening" - most other people don't.

Re:Your Goal: One Second or Less

[auLucifer]

I find it depends on what it's booting from. I use a mac air now and use to use a pro. I found that if I just close the lid so it goes into a shallow sleep it is up and running in a couple of seconds. However if it's from power down and needs to boot fully it can vary from under a minute to a few minutes. From a deep sleep is faster then powered off and it gets the environment exactly how I left it. I almost never see the boot screen these days which I think is how it should be as I do much the same work every day.

Re:Your Goal: One Second or Less

[crazybilly]

This disk may be the problem, but there's lots of concurrency problems between the CPU and disk, though, too. Install bootchart and look at the comparison of disk usage to CPU sometime--there's a lot gaps where both under 50%.

Re:Your Goal: One Second or Less

[lindi]

My desktop uses nfs as its root filesystem so it is easy to measure how much data it will need to read on boot by measuring network traffic. A complete reboot with "shutdown -r now" generated only 44 megabytes of traffic (including both read/written data and ethernet overhead) so there is clearly no need to read a GB. The system runs debian gnu/linux 3.0 with linux 2.6.18-4-486.

Re:Your Goal: One Second or Less

[Chabo]

The boss of Volkswagen did this after they bought Bugatti. He said "let's build a car that produces 1000bhp and goes 400kph". Then it took years for the engineers to figure out how such a thing might be possible. In the end, they did it, and it's probably the greatest car ever made.

[/clarkson]

Re:Your Goal: One Second or Less

[theheadlessrabbit]

they already have one, its called DOS, it has extremely low system requirements, it boots in seconds, even on hardware that is nearly 20 years old!

Re:Your Goal: One Second or Less

[ReverendLoki]

So should we start accounting for the average time it takes the average user to type in the average password into boot up time benchmarks? Will we see manufacturers artificially slant times by doping testers with stimulants and instigating a strict maximum password length of 4 characters?

Re:Your Goal: One Second or Less

[Thaelon]

I'm certain they could if it were really made a priority.

You'd have to give up detecting devices and assume they were there unless put into a special "detect devices" boot mode (such as holding down a key).

And you'd have to have all the data the OS needs to start up sitting somewhere faster than a HD, or sitting in a very very organized way on a really fast HD.

And you'd probably have to completely replace the normal BIOS stage.

A good place to look would be digital cameras. Booting quickly has almost always been a very high priority for them, and these days they're more sophisticated than old computers. Mine does it approximately 1 second.

But personally, I want a machine that's so stable I don't have to reboot often enough to care how long it takes to boot.

Re:Your Goal: One Second or Less

[Just Some Guy]

So in order to be a "visionary", I merely have to decide what consumers might want (not that hard being one yourself), and then ask people smarter than yourself to make it happen with no actual technical insight on how to make it happen yourself?

That's correct. The fact is that certain things aren't actually impossible, just pretty difficult. No one has solved them because there's been no driving force to do so. There are certain people clever enough to ask if things are the way they are because they have to be or just because that's the way people have been doing them. I'm no big Apple fan, but I'd definitely put Jobs in that category.

Seriously, can you think of a compelling physical reason why a PC can't boot in a small number of seconds? Beyond certain fundamental limitations, such as a rotating-media drive having to spin up, I can't think of any.

Re:Your Goal: One Second or Less

[RCL]

Just put everything in ROM and don't care about the updates...

Re:Your Goal: One Second or Less

[pipatron]

What good is the PC being 'at the desktop' if the search service still hasn't started, the network still hasn't obtained an ip-address, half my tray icons aren't up? and the hard drive is still madly churning to get everything else running

I never use any "search service" since I know what a computer is so I know where "internet" will download the files. I don't need an IP address until I need to go online, and my thinkpad does not have a harddrive that churns.

Seriously, how much stuff could you really -defer- to after seeing the desktop and have a useful system?

I like to get a terminal and a text editor up, for example. Maybe I'd like to play a game, start listening to some music or watch a movie. All of them things you don't need to initialize.

Even a fairly modest Ubuntu desktop requires several times that much RAM. If the hard drive started loading data at maximum speed you've got maybe 50MB you can load in that time, and probably far less in actual practice. That means your kernel, drivers, HAL, desktop environment, localization, firewall, network, background, theme, etc has to ALL fit in under 50MB.

A huge percentage of that data is not loaded from disk, but rather either calculated data (tables that are initiated, allocated buffers, structures for GTK/QT windows etc). Another big amount of this may be actual data, which is not read but mmapped, and yes, the kernel, drivers, localization, firewall, background image, will all fit under 50MB.

And you'd need some sort of impossible situation where the cpu could run all the initialization code for all that in parallel, without waiting.

This magic is called DMA

Re:Your Goal: One Second or Less

[rossz]

In the business world that pretty much sums it up. You don't need to know how to do something. However, despite what you say, figuring out what consumers really want and are willing to pay for is damn hard. Companies spend billions trying to answer this question. Most of the results are complete failures. A few ideas make a few people very rich.

Geeks can be absolutely brilliant in their field. Given the right direction they can come up with the next big thing. However, most geeks spend their time on little pet projects that will never make a dime. The sad part is when the business man comes up with an idea and the geek implements it, the businessman usually doesn't give the geek enough credit, aka $$$.

The most rare of exceptions is when the geek comes up with an idea that becomes the next big thing.

Re:Your Goal: One Second or Less

[Dr. Spork]

You may enjoy your 26 seconds of pretending that "this is not really happening" - most other people don't.

Why yes, I do enjoy that very much. I call it "sex".

Re:Your Goal: One Second or Less

[Cyberax]

It's possible. And it has already been done with Linux: http://lwn.net/Articles/299483/

Now it'll just require some time to trickle down to production-quality distros.

Re:Your Goal: One Second or Less

[Trillan]

My 13.3" MacBook (the latest model) is 5 seconds from power key for the screen to light up, 13 seconds from power key to logo (which is when I can select the OS by holding down the option key), and 44 seconds from power key to show the login screen.

Measuring the way the article measures (grub->login), this looks like about 30 seconds.

Re:Your Goal: One Second or Less

[yttrstein]

BeOS 5.0 booted in 9 seconds on my Pentium III 500 with 128 megs of RAM.

Re:Your Goal: One Second or Less

[WeblionX]

Yes, but I must say I prefer a GUI of some sort.

Re:Your Goal: One Second or Less

[EvolutionsPeak]

Why does it take so long to discover the hardware? Is there some fundamental reason? Is it support for legacy hardware?

Re:Your Goal: One Second or Less

[xenocide2]

Actually, you can set wifi as a systemwide config as of Intrepid.

Re:Your Goal: One Second or Less

[Kalriath]

The BIOS. The BIOS is pretty much the sole reason PCs take so long to boot. For example, at home I have a Core 2 Quad Q8200. When I push power, I get the XFX logo up while the POST runs. This POST takes approximately 20 seconds alone to run because of the inherent slowness of actions like writing ones and zeroes to every byte of RAM and then reading them back to test whether any memory is faulty, or initialising the Video BIOS, so on. Power on to OS loaded (even if it's still spinning up services) is impossible in 1 second, because it takes about that time for the CPU itself to start!

Re:Your Goal: One Second or Less

[geekmux]

This is one of my pet peeves: why can't computers boot in a second or less?

Cripes, I'm all for innovation, but damn, if you're literally counting the half-seconds sucked from your obviously insanely demanding lifestyle waiting for your current OS to boot up, then what the hell are you doing reading Slashdot? ;-)

Hell, while we're on the topic of the damn-near unobtainable, I'd simply settle for true open-document standards, and a pop-up free Internet. Give me that, and I'll go get another cup of coffee while I wait for my OS to boot.

Re:Your Goal: One Second or Less

[Kalriath]

I like to get a terminal and a text editor up, for example. Maybe I'd like to play a game, start listening to some music or watch a movie. All of them things you don't need to initialize.

Good luck playing a game, watching movies, or listening to music when your sound subsystem hasn't started yet.

(WTF Firefox? Since when has "movies" not been a word? And Firefox?)

Re:Your Goal: One Second or Less

[xenocide2]

TFA explicitly states that they time from after the boot loader is finished, to when the login window appears.

Not quite. It's when the login window sleeps. Pretty close. Some people are arguing that this is too narrow sighted, and that we should wait for the gnome login process to sleep before punching the stopclock.

Re:Your Goal: One Second or Less

[Bryan Ischo]

Why does it take so long to discover those drives and other devices? Why does a CD-ROM drive take hundreds of milliseconds to be recognized during a POST? These things should happen basically instantly at modern hardware speeds, and yet they don't.

It reminds me of NFS timeouts. Years ago when I worked in an environment where everyone NFS mounted a shared filesystem, there would occasionally be outages on the server or in the network. My local system would lock up and hang for MINUTES while it timed out on requests to the NFS server. I could never understand why the thing didn't just time out in seconds rather than minutes. Even at that time, we were running 10 MBit or maybe 100 MBit network connections; if the remote system is going to respond, it's going to happen at MOST after a few second delay. Waiting for minutes just seems dumb.

The same sort of thing happens alot with web browsers too that wait far too long for servers to time out. If the server doesn't respond in 10 seconds, it's not going to respond. Ever. There's no reason to wait 30 seconds or longer to timeout an HTTP connection ...

Re:Your Goal: One Second or Less

[rnentjes]

Off course the geek doesn't get any money, any random geek would have come up with a working implementation when asked the right questions. The money is in the questions here, not the answers.

Re:Your Goal: One Second or Less

[pz]

But why does it take more than a few milliseconds to discover a device? I've never understood this. We have had CPUs that have had sub-microsecond execution cycles for DECADES now, and yet the timeouts for communicating with devices are still measured in seconds. Why?

Device discovery should take no more than a few milliseconds for an entire machine, with the possible exception of disk drives which presumably need to spin up and verify correct operating speed to report back on a self-check.

Re:Your Goal: One Second or Less

[Malevolyn]

Oh yeah? Well I optimized my sex down to 23 seconds. Top THAT!

Re:Your Goal: One Second or Less

[Cyberllama]

People have been asking this for decades though, the reason no one's done it isn't because no one's though of it -- just technical limitations.

It's one thing to motivate a team of people -- to challenge them -- to do something no ones done before by pushing current technology to the limit, but make no mistake the limitation is in the technology not in people's lack of vision.

If you could come up with a type of memory as fast as ram that maintained state with no power to it, then overnight we'd have instantly booting computers. Until then, we're stuck with the limitation of "how fast can your hard drive spin and how much data can it load?"

A new, faster File System is a great way to speed things up, but it's still working within the same limitations.

Re:Your Goal: One Second or Less

[Sponge Bath]

The PCI spec also has required delays from power good to reset negated and then another required delay from reset negated to first configuration access. The second delay alone is about 1 sec (2^25 clock cycles).

Re:Your Goal: One Second or Less

[soleblaze]

There's been work on getting an eeepc 901 to boot in 5 seconds to usable minimal XFCE. (Though I guess usable depends on your needs.. I find a minimal xfce rather usable). Right now My aspire one boots from off to usable desktop in ~17 seconds.. Unfortunately about 8 of those seconds is bios. Takes 8 seconds to get to a grub menu.. Kinda sucks considering my eee 701 takes about 1.

Re:Your Goal: One Second or Less

[bucky0]

You'd have to have some sort of auto-login setup, but it'd disingenuous to call your PC booted when it's just sitting at the login screen. On my ubuntu box I'd estimate a good 50% of my boot time is after the login screen before I'm able to do what I wanted to do.

Re:Your Goal: One Second or Less

[Al Dimond]

And that's the problem with just taking the vision of one guy. My 9-year old Ford is better than the Bugatti in most ways that matter to most people.

But let's put the fact that this "visionary" had a pretty damn useless vision aside. Bugatti may have made a really fast car, but that's hardly visionary. They might have developed some great technology with industry-wide implications to hit those marks (or maybe they didn't, I don't know), but that doesn't mean there's anything noteworthy about the goal of, "Go faster -- how much faster? A little faster than the last guy!"

Re:Your Goal: One Second or Less

[hailukah]

All this high technology, what ever happened to doing things by hand?

Re:Your Goal: One Second or Less

[bucky0]

legacy hardware. There's tons of timeouts.

Re:Your Goal: One Second or Less

[Just Some Guy]

The BIOS. The BIOS is pretty much the sole reason PCs take so long to boot.

Regarding Coreboot (was: LinuxBIOS):

The Linux BIOS replaces the normal BIOS found on PCs and other machines. The BIOS boot and setup is eliminated and replaced by a very simple initialization phase, followed by a gunzip of a Linux kernel. The Linux kernel is then started and from there on the boot proceeds as normal. Amongst many other things, it provides a very fast boot time: 3 seconds from power-on to Linux console

It doesn't have to be slow.

Re:Your Goal: One Second or Less

[whoever57]

It reminds me of NFS timeouts. Years ago when I worked in an environment where everyone NFS mounted a shared filesystem, there would occasionally be outages on the server or in the network. My local system would lock up and hang for MINUTES while it timed out on requests to the NFS server. I could never understand why the thing didn't just time out in seconds rather than minutes

Let me suggest that you go and look up the options for NFS-mounted filesystems. You should also read up why it is normal to have the nfs-mounted systems NEVER time out, and just let the client sit there waiting for the server to come back on line.

Re:Your Goal: One Second or Less

[enoz]

Because Installation of Automatic Updates has completed. Computer will automatically reboot in 5 minutes...

Re:Your Goal: One Second or Less

[harlequinade]

Why can't you put the OS on a flash drive and boot a PC from *that*? They're too unstable for use as an HD at present, but for storing and loading the OS [read only] they'd be ideal. You'll probably never achieve boot in under a second, but I'd bet you could easily be up and running in well under 10.

Re:Your Goal: One Second or Less

[abigor]

Er...bringing your laptop back from sleep/hibernate is totally unrelated to the boot process. Just so you know.

Re:Your Goal: One Second or Less

[nine-times]

People want this.

Do they really? I'd like to see some polls to figure out who wants this, how much they want it, and why. Most people I know don't reboot their computers very often. A 30 second delay every few weeks doesn't seem to annoy people that much.

Re:Your Goal: One Second or Less

[collinstocks]

I set up an auto-login for my Ubuntu laptop, and then have the session-manager lock the screen immediately after logging on (before the panel or nautilus have even loaded, so while the desktop is still unusable). This way, after pressing the power button, I don't have to interact with my computer at all until immediately before I want to use it (i.e. to type my password in order to unlock the screen).

Unfortunately, just putting `gnome-screensaver-command -l` into the session manager won't work because it doesn't seem to load immediately. Instead, I made it run a script that executes that command in between calls to `sleep 1` six or eight times. It works for me.

Re:Your Goal: One Second or Less

[samkass]

So in order to be a "visionary", I merely have to decide what consumers might want (not that hard being one yourself), and then ask people smarter than yourself to make it happen with no actual technical insight on how to make it happen yourself?

No, the "visionary" really just needs the first part. Of course, there are a lot of poor, unemployed visionaries. The trick to being an effective visionary is getting people to pay attention, communicating your ideas effectively, assembling the teams, putting together an infrastructure that lets them succeed, making sure the proper smart people all talk to each other appropriately, get the thing out, market it, and arrange for all necessary support. Then do it all again and again.

Re:Your Goal: One Second or Less

[Gerzel]

Uhm certainly not. You also have to factor out user login time because some of us actually use basic security features. By that measure many computers at the university computer lab may regularly take hours to boot.

Also if you count that why not count the time it takes to start up a webbrowser, file manager, and office suit along with whatever graphics intensive game you might play...oh and for the game better add in the time it takes to actually get that first frag of the day after all you're not REALLY playing until something dies.

Also "usable desktop" depends on the user. A good shell is usable for plenty of people out there while for others nothing but the sleekest shiniest eyecandilicious gui will suffice.

Re:Your Goal: One Second or Less

[roc97007]

When the time it takes to type in the login and password is a significant portion of the total bootup time, we will finally have made some progress.

Personally, I can't imagine any reasonable person would think that "boot up time" is something less than the time it takes from dead stop to when I can start work. Else, we start to play games like "wait to start all these system processes until after he's logged in" and my recent personal favorite, "let's go through the entire boot up process and build a suspend image while he's waiting for his laptop to shut down after the meeting".

Mind you, our Mac G4 is about 1/3 the clock speed of my Winders box, and the Mac is always up and usable first in a horse race from cold start. It's not a hardware issue, it's a software issue. It can't be solved by changing the definitions of "boot" and "up".

Re:Your Goal: One Second or Less

[jamesswift]

Almost, you merely have to be right too.

Re:Your Goal: One Second or Less

[ryanov]

And it even occasionally works!

Re:Your Goal: One Second or Less

[LordSnooty]

This is one of my pet peeves: why can't computers boot in a second or less?

It's not just the computers on our desk, it applies to any device with integrated circuitry these days. It struck me when I bought a DAB radio - for every gain like channel bandwidth there's a loss like it taking five seconds to boot up. Not much, but compared to the traditional wireless, which was.. instant! It's poor. You start to wonder how much time you're wasting as you wait for tech to wake up. Of course, it would seem unavoidable, as the machine must test all of its components before loading the signal processing software. But like artefacts on digital video, delayed startup seems to be just one of those things we have to put up with. The period post-valves but pre-heavy reliance on ICs will be seen as a tech golden age in the future!

Re:Your Goal: One Second or Less

[Fred_A]

Imagine a visionary like Steve Jobs

Aaaaaah !
My eyes !

The googles ! They do nothing !

Re:Your Goal: One Second or Less

[icebraining]

Yeah, at least XP rarely works, but Linux, when patched with TuxOnIce never failed to me! Of course, you have to be able to compile your own kernel, so it's not for the Me-Loves-Apple-Hates-Cli people.

Re:Your Goal: One Second or Less

[fast turtle]

Sounds about right for my Vista64 Business system to wake from S5 STR mode. Course this is on an Intel Board with a C2D and 4gb of memory.

Re:Your Goal: One Second or Less

[anexkahn]

Some people say it's not real server unless it takes at least 5 minutes to boot :)

Re:Your Goal: One Second or Less

[dns_server]

It Is a selling point, that's why i bought an Asus motherboard with splashtop. Splashtop is a pre bios linux boot loader that allows me to have firefox in 5 seconds after turning on the computer.

Pheonix has announced a bios level hiperviser that will allow you to boot both linux and windows at the same time, you will be able to boot linux and launch firefox while you wait for windows to load and switch between the two with a keyboard shortcut.

There is also the coreboot which is a replacememnt of a bios with the linux kernel that you may be able to replace your bios with.

Re:Your Goal: One Second or Less

[quanticle]

Yeah, you could just have that set up in BIOS or something and just use that configuration, but that could be a pain, too.

It might be a pain for desktops, given that they're supposed to somewhat customizable, but, I don't see why having a fixed configuration burned into the BIOS would be inappropriate for laptops. After all, its not like users are going to go in and change the motherboard or graphics chipsets on a laptop.

Yeah, you'd have to allow for external add-on hardware like USB sticks, etc., but, even today, the OS takes care of that, rather than the BIOS.

Re:Your Goal: One Second or Less

[quanticle]

They might have developed some great technology with industry-wide implications to hit those marks (or maybe they didn't, I don't know), but that doesn't mean there's anything noteworthy about the goal of, "Go faster -- how much faster? A little faster than the last guy!"

The Bugatti Veyron is a little more than a little faster than the next fastest car. Let's put it this way: the Veyron can hit 200, before the next fastest car (the McLaren F1) can hit 150.

Re:Your Goal: One Second or Less

[Kent Recal]

Use your common sense.
"Boot time" obviously refers to the time that the user is waiting for the machine and not the other way round.
And "usable desktop" is obviously the point in time when the user can begin launching his applications without significant slowdowns from boot-tasks still grinding in the background.

This is obviously always an apples-to-oranges comparison but with just a tiny bit of common sense it can still be more meaningful than "OSX boots in 4 seconds".

Re:Your Goal: One Second or Less

[vux984]

I never use any "search service" since I know what a computer is so I know where "internet" will download the files.

On OSX I use spotlight to locate things.
On Vista I use the the new start menu search for almost everything.
etc

Not because I don't know where stuff is, but for anything not pinned to the start menu or taskbar or dock or the top menu in ubuntu... its FASTER than browsing hierarchical menu or filesystems.

I don't need an IP address until I need to go online,

The first thing I usually do after logging in is launch either my browser or my email.

and my thinkpad does not have a harddrive that churns.

Whether you can hear it or not is irrellevant to the conversation. Its still going to pushing data as fast as it can.

A huge percentage of that data is not loaded from disk, but rather either calculated data (tables that are initiated, allocated buffers, structures for GTK/QT windows etc). Another big amount of this may be actual data, which is not read but mmapped, and yes, the kernel, drivers, localization, firewall, background image, will all fit under 50MB.

I agree a big part of the ram usage of an operating system is mmapped and in memory allocated tables and buffers etc, however I think you are mistaken that the sum of all the binaries and configuration files that are completely loaded is less than 50 MB in a modern system startup sequences that hasn't been seriously seriously gutted.

This magic is called DMA

Care to explain how DMA is supposed to allow the PC to initialize the kernel, wifi, sound, network, printer monitor/spooler, firewall, all simultaneously, apparently while letting you run and use other programs that somehow manage to avoid depending on the above list?

Re:Your Goal: One Second or Less

[anexkahn]

Have you heard of the distrubtion called Damn Small Linux? Abbreviated DSL:
http://en.wikipedia.org/wiki/Damn_Small_Linux
The entire distribution is 50 MB and it includes network, Gui, etc... Based on your numbers above we should be able to load this entire distro into Memory within a second or two, maybe another 2-3 additional seconds if you want to add a 3d desktop like Compiz. Other services could potentially be loaded in the background after the login screen and/or desktop are available.
I see little reason why an OS like Ubuntu can't reduce boot times down to the sub 10 second range with a little work. It's all about scheduling.
I do think that 1-2 second boot time is a bit unrealistic at this time without a faster hard drive, but I would not say impossible.

when the computer comes back from hibernate it can be faster, or slower, depending on the situation. This is because it does not have to do all the initializations during startup. If the OS has to do a bunch of initializations every time it start up, why cant it just do a memory dump after those initializations, then only load the ones that change every time the computer starts?

If the user makes a change that would cause a change to one of these cached values, the computer could modify these start up files at that time the result would be a hybrid between suspend and shutdown where your computer does not have to do a complete memory dump when you shut down, only a partial one, and boot times would be significantly reduced.

The same goes for applications running on the machine, no reason you couldn't accelerate the startup of applications the same way...but that would have to be on a per application basis.

Re:Your Goal: One Second or Less

[drijen]

What a load of crap. I have no idea why boot speed is suddenly all over the media like it's some big thing..

My old desktop (Athlon XP era) does POST -> KDE, including me typing in my login and password, in under 6 seconds. None of this hibernation junk or w/e doohickey crap you youngins are coming up with.

Re:Your Goal: One Second or Less

[5865]

"Premature optimization is the root of all evil" - Donald Knuth

Re:Your Goal: One Second or Less

[anexkahn]

The OS could do the following:
1. Find all files required for startup, and put them in the same region on the disk.
2. Put these files in order of when they are needed for the startup process
3. At startup, the OS loads all files marked as required for startup into memory, in order.

The hard disk is now able to load everything into memory in one big motion, and at it 's maximum speed without having to do any seeking. The computer could run a scheduled task to optimize startup files...this process would be part of that.

After following the process above, all the files would be in memory, which should be a lot faster than the hard drive, and hopefully reduce bootup times dramatically. As files are no longer needed they could be flushed from memory.

This process would require a lot of memory, but that shouldn't be a problem, especially if you are running 64 bit. I just picked up 8 GB of memory at fry's for $70 2 weeks ago.

You could also have an algorithm which calculates the values you specify ahead of time, and caches them to eliminate some of these concurrency problems. When you tell the computer to shut down the computer could verify all these cached values, and recalculate/recache if needed.

Re:Your Goal: One Second or Less

[SanityInAnarchy]

On mine, it may take 50% for everything to start, but easily within five seconds, I can already be launching programs. I wonder if we should be counting the time it takes to load all needed programs?

Re:Your Goal: One Second or Less

[Chabo]

Right. At least one car (the SSC Ultimate Aero) has beaten the Bugatti's speed record for a production car, but the Bugatti is simply an engineering marvel. Most "really fast" cars are designed to hit their speed limit a few times, and F1 cars are designed to do a couple races, but the Veyron is designed to last 20 or 30 years of road driving.

The Top Gear presenters kept comparing it to Concorde. That's how big of a leap forward it was.

Re:Your Goal: One Second or Less

[trolltalk.com]

... but well before then, you can log into a terminal on one of the tty consoles. THAT is what you should count. X, etc., are NOT "part of the boot process".

Re:Your Goal: One Second or Less

[BikeHelmet]

Your dream is only doable with highly optimized BIOS code. I'd look to LinuxBIOS/Coreboot/OpenBIOS for the answers. You might also investigate what makes VIA boards and many laptops POST way quicker.

My "enthusiast" board posts in 22 seconds, and boots the OS in 35; that's another 13s. My VIA slow-as-shit board boots to the desktop in 14-15 seconds. I think POSTing only takes about 4-5. It's already past detecting drives(and letting you enter the BIOS) by the time the monitor flicks on.

One interesting thing that slows down booting is Sound Drivers. Realtek sound drivers, in particular. I noticed a 7 second drop in boot times by disabling sound in the device manager, on boards with Realtek HD integrated audio. With the VIA board, there was no drop at all from disabling sound.

I have heard that some operating systems can remember device configurations, to speed up booting. By default this is disabled in most linux distros, and enabled in windows, which is why swapping motherboards is a PITA for WinXP and down, and usually requires a repair install unless you prepare for it. ;) It's also why an unclogged WinXP is one of the fastest booting OS's out there.

Re:Your Goal: One Second or Less

[SanityInAnarchy]

How about this: Have a dedicated device, of a reasonable size. Boot off of that, always, and don't bother with anything else. Once that's booted, it can start handling the rest -- in parallel, if it's smart. Make it a removable standard, like SD, in case there are problems.

Modern OSes, once they have the correct drivers, tend to ignore the BIOS anyway.

My kernel is 2.3 megs, uncompressed. My initramfs is 9.1 megs, compressed. Give me, oh, 100 megs of flash to boot from, and guarantee nothing more than that device is available when it's booted from.

But I do believe post-POST is where the problem is. After all, I've seen computers whiz through POST in maybe two seconds, and then take several minutes to get to a usable desktop. Hibernate is cool, but it's still not instant-on.

Re:Your Goal: One Second or Less

[andrikos]

"Premature ejaculation is the root of all evil" - Viagra

Re:Your Goal: One Second or Less

[SanityInAnarchy]

Well, let's see:

One problem is POST, because so many things have to be checked for, including legacy hardware. Apple doesn't have to care -- they supply the hardware, they can change how it works. They could even use the OS X kernel as the firmware, or have the firmware know nothing other than Firewire target mode and the internal hard disk.

Once you've got that kernel, the rest is partly about the memory, yes -- but with solid state, maybe it won't be as fast as RAM, but it will be fast enough that you could have a usable system immediately.

So, it's really more about software design. It's about, give me a login screen RIGHT NOW, and bring everything not related to that login screen up in the background -- or better, as-needed. My current Ubuntu does that in the wrong order -- there's tons of shit, like CUPS, which is started before the display manager. I rarely ever print -- CUPS should be started on demand, not on boot.

After logging in, it's the same story -- giving you a panel and a launcher should take priority. Instead, every OS I've used either throws up a splash screen or spends several seconds loading random crap before I get those things, and then maybe 30 seconds afterward loading everything else.

My current OS could take me to a usable desktop in 20, maybe 25 seconds. It gets better as the hardware gets faster. But the stuff packed into my initramfs says that the system is plenty usable for certain things long before I get that login prompt. The fact that once all of that is ready, Konqueror takes less than a second, and Firefox takes less than five, suggests that at least I could have a browser while I wait for everything else.

Re:Your Goal: One Second or Less

[Fri13]

That should be possible when the new RAM comes to market. Dont now remember what it was called but it stores the information without needing a power. So you dont actually need to sleep/hibernate anymore, you just turn computer ON and OFF from switch and it just works.

All the data stored to SSD and OS + system services + desktop + applications what were running are on RAM.

Re:Your Goal: One Second or Less

[vux984]

First, the main thrust of my post was really to comment that getting to the desktop BEFORE everything else is running is a victory simply not worth fighting for.

I.e.deferring things to startup AFTER you arrive at the desktop to give you the appearance of a faster boot time is pointless if you need those things to actually use it... or even if the fact that those things are still starting up is pegging your cpu/hard drive making it essentially unsusable even if you aren't loading something dependant on the items still loading.

That said...

The entire distribution is 50 MB and it includes network, Gui, etc... Based on your numbers above we should be able to load this entire distro into Memory within a second or two, maybe another 2-3 additional seconds if you want to add a 3d desktop like Compiz.

I think potentially, yes, this is theoretically possible. There is a laptop out there, for example, with an instant on Linux distro flashed into the BIOS, that you can use to quickly browse the web etc, without having to boot up the OS off the hard drive.

http://www.itnews.com.au/News/77281,asus-laptops-to-offer-instanton-linux.aspx

So this absolutely -can- exist. I'm not sure just how instant, instant-on is here, but it sounds like its in the 3-5 seconds range.

Other services could potentially be loaded in the background after the login screen and/or desktop are available.

I think this is a bad idea. See above, for why.

I see little reason why an OS like Ubuntu can't reduce boot times down to the sub 10 second range with a little work. It's all about scheduling.

Sure, I agree 10 seconds is quite concievable conceivable.

However much beyond that and I think coping with querying the hardware itself will take longer than that. Just querying all the buses etc to make sure nothing has changed will probably take a few seconds.

If the OS has to do a bunch of initializations every time it start up, why cant it just do a memory dump after those initializations, then only load the ones that change every time the computer starts?

Why bother reinventing the wheel? We ALREADY have "suspend to RAM" and "suspend to Disk" and that is basically already what it does. Trouble is, the device drivers have to support it for it to work properly. And it turns out that, for suspend to disk at least, that reading in the big ballooned out memory image to and from disk is usually SLOWER than just booting clean because of all the extra data involved.

And on top of that you STILL have to wait for a pile of device initialization because simply loading in your network/video/audio/etc driver to a particular ram image state doesn't do a thing towards actually putting the network/video/audio/etc device into a suitable state.

(This is in fact precisely why you need a dedicated protocol to communicate you are going into and out of suspend and device driver support for it.)

Re:Your Goal: One Second or Less

[linhares]

meh, I type my 10+ character password (containing lowercase, uppercase, and numbers) so often that it takes less than 2 seconds.

IDIOT! Now everyone knows your passwd is Ubuntu904

Re:Your Goal: One Second or Less

[Fri13]

There is other one too, it is modern operating system and it is Unix-like. It's name is Linux.
Whole OS in 10 million lines of code. Fast boottimes and exelent hardware support and lots of available software for it, most of are even from GNU projekt so you can get easily a GNU/Linux development platform for old machines too and make applications without worring to buy new one...

Re:Your Goal: One Second or Less

[HermMunster]

Though booting in 21 seconds is nice, they need to shut down in 1/10 that time. Right now it seems to take as long to shut down as it takes to start up. Close my files, prep the file system, and shut down in 2 seconds or less. That's it. It isn't enough to boot in 21 seconds as most leave their computers on for days at a time. What's best is to shut down faster. I don't need every daemon shut down in reverse order. Just close my files and boom, end it.

Re:Your Goal: One Second or Less

[HeronBlademaster]

The question is, aside from hard-drive-read-time-related concerns, why not? Given that modern processors can do so much in one second, why can't we write our code such that we have a usable GUI after that one second? Maybe we'd have to cache some of the files in a flash buffer somewhere so we can ignore the hard drive for a bit... not really a problem, in the grand scheme of things.

Even resuming from Hibernate is faster than a clean boot, on many (all?) laptops. This makes zero sense to me, since hibernate reads your entire RAM from disk, whereas a clean boot does not.

Re:Your Goal: One Second or Less

[HeronBlademaster]

I rarely have problems with Suspend/Resume or Hibernate in XP... come to think of it, I can't remember ever having a suspend or hibernate problem that wasn't my fault (e.g. I let it suspend so long the battery died).

Linux, on the other hand, is so finnicky when it comes to suspend and hibernate that I just gave up entirely and disabled both features.

Re:Your Goal: One Second or Less

[linhares]

You are absolutely correct. My first computer had a processor that ran just under 1 MHz and it booted up in maybe 2 or 3 seconds.

After that you typed win and waited for 8 minutes.

Re:Your Goal: One Second or Less

[Eil]

You know how everyone wanted a Linux-based operating system that "just worked" on a wide variety of hardware with drivers for everything? And didn't throw a shit-fit if you moved the hard disk to a completely different machine and tried to boot it up?

That's why Linux takes so long to boot these days. You can have very good hardware compatibility or you can have very good boot speed. You can't have both. (Well, until someone invents persistent RAM.)

Why does it take so long to discover those drives and other devices? Why does a CD-ROM drive take hundreds of milliseconds to be recognized during a POST? These things should happen basically instantly at modern hardware speeds, and yet they don't.

The CD-ROM does respond to the BIOS very quickly. What takes forever is the BIOS checking each controller, chain, and bus location for a device. Waiting for those probes to time out is what takes so long. This isn't just the BIOS either, it's the Linux kernel too and any OS that might want to speak to whatever hardware might happen to be there.

. Even at that time, we were running 10 MBit or maybe 100 MBit network connections; if the remote system is going to respond, it's going to happen at MOST after a few second delay. Waiting for minutes just seems dumb.

Seems dumb to you, the user. Didn't seem dumb to the programmers who wrote NFS and whatever application you were using. Why? NFS is 1) a block device, and 2) largely a hack. The way UNIX was designed, block devices just don't disappear from the system. Just like wheels (ideally) don't go flying off your car while you're driving down the road. But when NFS, a block device can suddenly go unavailable and as far as the OS is concerned, that's just really really bad for all sorts of reasons. The programmers figured that in order to make the system as robust as possible, they'd extend the timeout as long as tolerable to reduce the chances of data loss and corruption as much as possible. It's conceivable that a large number of problems could be resolved in a matter minutes (say, somebody tripped over the power cord for the network switch), thus preventing the loss of what could be very valuable data.

The same sort of thing happens alot with web browsers too that wait far too long for servers to time out. If the server doesn't respond in 10 seconds, it's not going to respond. Ever. There's no reason to wait 30 seconds or longer to timeout an HTTP connection ...

You click a mouse button. This initiates a request which, after all of the appropriate nameservers have been consulted, hops from your machine over dozens of routers, switches, and cables owned by different countries and corporations. It travels thousands of miles away to some place you can't even pronounce. Once there, the server recognises the request and acts on it, sending you back a mix of static content, images, and database content several orders of magnitude greater in size than your original request. The content then travels back to you another few thousand miles, perhaps via a different path until it eventually reaches your machine where it is processed and displayed in a mostly-legible fashion. And you have the gall to complain that sometimes it takes longer than 10 seconds for all of this to happen?

Good. Fucking. Grief.

I'm continually amazed that it works at all and I'm a sysadmin at a web hosting company. Almost every day I run across a site I want to visit that takes longer than 10 seconds to respond in full. There are lots of very good reasons that a website might take between 10-30 seconds to load in your browser. The authors of the HTTP protocol, web server software, and web browsers having a personal grudge against you sure isn't one of them.

Re:Your Goal: One Second or Less

[HeronBlademaster]

So... turn off the extensive memory test. You don't need it on every boot. My laptop spends maybe five seconds at most in POST, unless I changed a BIOS setting. Many desktops spend much less than that in POST.

You have to admit, it's your own fault you're waiting 20 seconds every POST.

You're right about one thing though - it does take a while (relatively speaking) for the voltages inside the CPU to stabilize. It's less than a second, though, afaik. Anyone know for sure?

Re:Your Goal: One Second or Less

[mhall119]

Coreboot got an X desktop loaded in just a few seconds by having it load from the BIOS firmware.

But if you want a modern desktop, you'll either need more firmware space of a faster hard drive.

Re:Your Goal: One Second or Less

[ryanov]

Kubuntu 8.10 is the first one that has worked well enough for me to rely on suspend. I can qualify that a bit by saying that the others worked out of the box too, but subsequent hacks to get parts of my machine to work undid that apparently.

Re:Your Goal: One Second or Less

[Just Some Guy]

legacy hardware. There's tons of timeouts.

So why wait for them sequentially? Even better, why not supply optional kernels tuned to modern hardware? I don't own a system with an ISA bus (even a faked-up one on the southbridge or similar), so let me skip probing for one.

Yeah, yeah, compile my own and all. But surely big distros like Ubuntu could make a legacy-free kernel available that skips ISA, serial, parallel, etc.

Re:Your Goal: One Second or Less

[ryanov]

I don't think I've ever owned a laptop where suspend in Windows was reliable. Hibernate worked pretty regularly though.

Re:Your Goal: One Second or Less

[Sebilrazen]

meh, I type my 10+ character password (containing lowercase, uppercase, and numbers) so often that it takes less than 2 seconds.

IDIOT! Now everyone knows your passwd is Ubuntu904

I know your post was supposed to be funny, but while you passed the uppercase, lowercase and number criteria, you failed to pass the 10+ characters criteria mentioned.

Re:Your Goal: One Second or Less

[DaleCooper82]

Well, actually, in Vista and KDE world I would count that until it is usable after logon.

I mean I took Vista for a test drive (that lasted 10 months) last year and the boot time (stage 1 - to login screen) was +- comparable to Kubuntu however after logon it kicked off some bloody files indexing and that rendered the PC quite unusable. I assumed it would do this while screensaver was running but no, it pretty much had impact on usability during the first 40minutes or so - until I turned that completely off.

Similarly, now that I returned to Kubuntu, after logon I get system services starting and then apps/applets like KGPG, Kopette etc. These of course depend on what was running before shutdown.

After all this I can only start using the PC "seriously" for my work.

I should also say I am not big fan of "minimize the boot time", I just simply do not see the point. The OS needs to boot various services and stuff and I prefer it does that upfront and then goes out of my way. Also, maybe it is because currently it all takes less than a minute on my Kubuntu and... so what?

Re:Your Goal: One Second or Less

[HeronBlademaster]

A gigabyte of data read on boot? I don't know where you're getting your data, but I have a really hard time believing that.

Windows may load every device driver known to man before bothering to decide what your computer actually needs - that is, in fact, what Windows Setup does. But that's ridiculous - there's no reason it should need to load all of them.

Linux doesn't do this (at least, my laptop doesn't, because it doesn't have any extraneous device drivers). My kernel is 2.9MB (compressed, I suppose) and includes every device driver for my system except the nvidia video driver.

So, a gigabyte? That's ridiculous.

Re:Your Goal: One Second or Less

[HeronBlademaster]

Who modded you flamebait? I think you have a valid question. Device drivers can be unloaded and reloaded on the fly; I've even heard about running a secondary kernel that will take over in the event that the currently running kernel dies or shuts down - why can't the secondary kernel be the updated one, and have control transferred to it without a reboot?

Why can't Windows update itself without rebooting? Because rewriting it such that reboots are unnecessary costs more than they're willing to pay - even if they're supposedly starting from scratch.

Re:Your Goal: One Second or Less

[HeronBlademaster]

But there's no reason the computer shouldn't function before the "real" drivers are installed - we have standards for a reason. Ever notice that your video card works before you install drivers for it? Windows is using a standard API to access the card.

Re:Your Goal: One Second or Less

[linhares]

you don't see the space there?

Re:Your Goal: One Second or Less

[HeronBlademaster]

People who boot their laptops two or three times a day want it. (Suspend doesn't work in Linux on my laptop right now, and my battery is almost kaput anyway. Shutdown then boot later is my only option.)

Re:Your Goal: One Second or Less

[HeronBlademaster]

Along the same veins, I find it unbearable when TVs take two or three seconds just to change the channel. How can I channel surf if I have to wait so long just to see what's on the next channel?

Re:Your Goal: One Second or Less

[ConceptJunkie]

Well, clearly there is no "vision" at places like Microsoft where the OS takes longer to boot that it did 10 years ago and the same app takes as long to start up as it did 20 years ago.

Frankly, if you told me 20 years ago that computers would in the late 2000s take 1-2 minutes to boot (for the short time I used Vista on a very capable machine it took 7 minutes to boot after a single OS update), and that Word takes between 30-60 seconds just to load or that Outlook freezes for 15-30 seconds every time I unminimize it or click on a folder, I would have told you you were crazy.

(Of course, I also would have expected to see fully-blown natural language processing, voice recognition and Duke Nukem Forever...)

Despite all the advances in speed and capability, our computers today are doing many of the same tasks no faster, and even more slowly than they did when they ran at 100th the speed with 1000th the memory and 10000th as much disk space.

Re:Your Goal: One Second or Less

[BigGerman]

Yes. Pretty much that what it takes.

Re:Your Goal: One Second or Less

[SteelX]

From the original poster:

Imagine a visionary like Steve Jobs (by the way, enjoy your leave of absence and please come back). He goes to his team and says "I don't care what it takes, build me a computer which boots in one second".

From http://www.linuxdevices.com/news/NS5429881813.html:

Yet, most BIOSes available for x86 chipsets were built for the desktop market, and thus have not been optimized in this area, according to Steve Jones, General Software CTO.

When Steve Jobs is not around, leave it to Steve Jones!

Re:Your Goal: One Second or Less

[jonasj]

You're reading your parent post wrong. It never mentioned anything about sending the whole response back in 10 seconds; only for the server to accept the connection in that timeframe.

Re:Your Goal: One Second or Less

[icydog]

2^25 is around 33 million. Surely 33M cycles isn't a second these days?

Re:Your Goal: One Second or Less

[BetterSense]

Which is extremely annoying and stupid. I was shocked when I discovered that my (Ubuntu) computer fails to connect to the internet if I don't start X. Most GUI widgets tend to be front-ends for some 'real' process and I don't understand why the Ubuntu network managers aren't just ifconfig frontents.

Re:Your Goal: One Second or Less

[russ1337]

So in order to be a "visionary", I merely have to decide what consumers might want (not that hard being one yourself), and then ask people smarter than yourself to make it happen with no actual technical insight on how to make it happen yourself?

Did the king that asked for pyramids know how to build them?

(I guess in that case you'd need to define the Consumer)

Re:Your Goal: One Second or Less

[thtrgremlin]

Is that the proper nerd way to benchmark nerd time? Hit the power button and if you and your computer finish at the same time then you know things are going well.

Re:Your Goal: One Second or Less

[StreetStealth]

For some, doing the task manually is even faster!

Re:Your Goal: One Second or Less

[thtrgremlin]

The odd thing reading through this stuff is that I know the majority of my boot up time is just getting to the point where the grub screen is shown. after that it is 10 seconds for everything to be loaded. My bios and grub together take the longest, so thir really makes me curious what is going to be faster.

Re:Your Goal: One Second or Less

[EotB]

33 million cycles of the bus clock, so for 33MHz PCI bus 33M cycles = 1s.

Re:Your Goal: One Second or Less

[rgravina]

Bullshit. An idea is useless until it is implemented in the same was that any other potential is useless until it is realised. Once implemented, its success is a combination of how good the idea was and how well it was executed. Giving either party too much credit for the outcome means not giving the other enough respect.

I'm a big believer in paying for both talent and effort. If your idea took you 10 minutes to come up with, but took a programmer six months fulltime to flesh out and implement, you don't deserve 90% of the credit no matter how good your idea is. But you don't deserve 10% either, if your idea was informed by years of experience *and* you were an active participant in the development process.

Finding business people who understand this, as a programmer myself, is harder than it should be.

Re:Your Goal: One Second or Less

[thtrgremlin]

Yeah, I don't even count the time after the boot loader it is so fast. The bios and grub are what take by far most of the cold start to desktop time. Measuring time from after boot loader to GDM is like measuring your penis from tip to asshole.

Re:Your Goal: One Second or Less

[WhatDoIKnow]

If the server doesn't respond in 10 seconds, it's not going to respond

Unless maybe you are connected to the internet over cell phone based broadband .

Re:Your Goal: One Second or Less

[fractoid]

The sad part is when the business man comes up with an idea and the geek implements it, the businessman usually doesn't give the geek enough credit, aka $$$.

That's because the business man provides the money and takes the risk. The geek's happy because he's working on something interesting, and being given enough bananas that he's not hungry and can afford his home internet bill. It's not difficult to come up with a business plan and to be the business man, but it's boring compared to fiddling with computers for a living, even though in the end it pays a LOT more. The sad thing is that us geeks very seldom *bother* to do the leg work required to turn an idea into a saleable product.

Re:Your Goal: One Second or Less

[camperdave]

All I see is **********. Maybe someone else could post their password so we can check it.

Re:Your Goal: One Second or Less

[fractoid]

And sadly, if a 1000-to-1 bet has 1000 punters, chances are that one of them will win it. And will be called a visionary, and people will crap on about how he was amazingly insightful, and a genius, and all that, when in actual fact he just happened to be the one chump who got lucky.

Re:Your Goal: One Second or Less

[lysergic.acid]

what the hell kind of "basic" security measures are you using that takes hours for the system to become ready? even if you used two-factor strong authentication it shouldn't take more than a minute to log in. heck, you could provide voice + fingerprint + retinal identification, use multiple synchronous dynamic password tokens, implement the two-man rule, and still complete authentication in under 10 minutes.

what, is your computer lab in some sort of subterranean bunker locked behind a 55-ton door that can only be unlocked by providing a small blood, urine, and sperm sample, and reciting the first 8000 digits of Pi or something?

boot-up time is how long it takes for a computer to boot into the OS and become ready for use. it's not how long it takes for the video card drivers to load or how long it takes to render a partial GUI. and it's also not how long it takes for you to start up your browser or video game. that would be the startup time for that application--which is likewise how long it takes from when the user first initiates the program to when it is first ready to be used. what's so difficult to understand about this?

i mean, do you consider a program fully loaded the second it displays a splash screen?

Re:Your Goal: One Second or Less

[4D6963]

I totally agree. Being just a geek with good ideas and the skills to implement them, I thought to myself, why not just use them to make a commercial product that I would sell. The whole "making it" part was easy, that was my thing, algorithms and all that. Time consuming, but easy. Then came the "selling it" part.

All I can say is this, you have no idea how hard it is to market and sell something until you try it yourself. That's the hardest part, for someone like me, hands down. You need to know your market, how to market to them, know how to do, all of that stuff...

This is the commercial project I'm talking about. People were excited about the original idea, people are excited about the implementation, but the people who've even heard of it are scarce. Right now I'm just like an hypothetical Steve Wozniak without a Steve Jobs to whom people would ask "hey nice, this Apple I thing, but what am I gonna do with it?". The Apple I was the next big thing, but if it wasn't for Steve Jobs to explain to the world why it was the next big thing, there wouldn't have been an Apple II.

Re:Your Goal: One Second or Less

[linhares]

All I see is **********. Maybe someone else could post their password so we can check it.

You have a good idea there. Please, moderators, post paypal.com passwords so that we can check them for your full safety from internet pirates.

Re:Your Goal: One Second or Less

[debatem1]

For cold boot to be as fast as hibernate, it would have to be able to run through the bootloader, do hardware detection, and generate a system image as fast as your system could copy one into ram. It's like trying to write a book versus trying to read one. Well parallelized init processes, cool hardware tricks, and bootloader shenanigans can get you pretty close to it, but its still not exactly a fair fight, and as another poster pointed out, the two have very little in common from a technical perspective.

Re:Your Goal: One Second or Less

[jwhitener]

""The CD-ROM does respond to the BIOS very quickly. What takes forever is the BIOS checking each controller, chain, and bus location for a device. Waiting for those probes to time out is what takes so long. This isn't just the BIOS either, it's the Linux kernel too and any OS that might want to speak to whatever hardware might happen to be there.""

Why can't an operating system default to "use the last settings for the controller, chain, bus location, etc..." and not have to check everything every time?

If you add new hardware, you should be able to boot up in a 'scan mode' or something. Otherwise, it never checks on your hardware...

Re:Your Goal: One Second or Less

[HeronBlademaster]

What if the OS just stored a "clean" version of a hibernate file, so that a cold boot is just a resumption of a "empty" hibernation state?

Re:Your Goal: One Second or Less

[Bryan Ischo]

Don't get your panties in such a bundle. I said that if an HTTP request takes longer than 10 seconds to service, it's very, very, VERY likely never going to be serviced. This is based on years and years of using a web browser every single day.

Maybe 10 seconds is too short a time, I'm not really trying to quibble about the specific timeout. I'm just saying that in general, I find that programmers are too conservative when they choose timeouts. I'd rather prematurely terminate 1% of my requests that take longer than 10 seconds but would have been serviced if I'd just waited long enough, than wait 30 seconds for the other 99% of the requests that really were never going to complete.

In my NFS case, we *never* had a failure that resolved itself in a matter of minutes. If things went down, there was no reason to sit and wait indefinitely for them to come back up. Which is my point - on a fast network with a dedicated server, if a request takes longer than 10 seconds to service, you really should immediately timeout, because in such a situation, you can be 99% sure that it's never going to be serviced. If it was going to be serviced, it would have been within 10 seconds. A similar thing happens occasionally when I configure a new Linux system and mess something up and the network doesn't work, and some service on reboot wants to do a DNS lookup or something and it totally hangs my entire boot for MINUTES while it times out. In this day and age, minutes is too long to wait to timeout.

I wonder if the same concept applies to these devices that have to be queried. What hardware cannot respond nearly instantaneously to a discovery query? Why do bus controllers wait multiple seconds for responses? They should time out after like 20 milliseconds or some really low amount. Is there any device that doesn't respond nearly instantaneously? If so, WHY?

Re:Your Goal: One Second or Less

[debatem1]

Well, that's pretty much what actually happens with the initial ramdisk. The hard thing is that it isn't (right now) possible to make those golden ramdisks very powerful AND very portable. A team made a Eee boot in something on the order of five seconds not too long ago because they expanded the role of the initrd at the expense of hardware detection (I'm simplifying), but if you try to copy one of your system's hibernate images onto a virtual hard disk and run your virtualization of choice on it, it'll crash horribly. The solution so far is to simply make the ramdisk's scope small, so that the OS still does all that time-consuming stuff afterwards. Until we get a good solution for all of that, or move it down into hardware, we're pretty much going to be stuck parallelizing the init process.

Re:Your Goal: One Second or Less

[wmac]

Microsoft and some others already did it using Hibernate. Have you tried that?

Re:Your Goal: One Second or Less

[perbu]

To be a 100% sure you see everything on an old SCSI bus you reset the bus and wait 15 seconds to let the devices settle before scanning the bus. Most devices settle within a couple of seconds but some odd tape drives (or robots, I can't remember) took ages to settle after a reset.

I guess modern hardware is not that different although the 15 seconds timeout are gone.

Re:Your Goal: One Second or Less

[camperdave]

And "usable desktop" is obviously the point in time when the user can begin launching his applications without significant slowdowns from boot-tasks still grinding in the background.

If there are boot tasks still grinding away in the background, then the machine has not finished booting, regardless of how usable the desktop is.

Re:Your Goal: One Second or Less

[Bert64]

Some laptops, like the eee 901, has a feature in the bios to cache the results of these probes... Makes sense in a laptop where the hardware is unlikely to change much, perhaps a usb device or two...

Re:Your Goal: One Second or Less

[Bert64]

NFS by default doesn't time out, it just waits indefinitely for the server to come back... I believe this is designed so that you don't lose data.
There is a mount option called intr (interrupt) which will let you interrupt a process that's locked due to a dead NFS server.

Re:Your Goal: One Second or Less

[ILuvRamen]

It's a macbook, it comes broken. One mouse button is missing! Hehehe. Anyway, my Pentium 3 laptop running windows ME boots in 13 seconds. It shuts down in 4.5 seconds. And everyone knows that's an unstable, service heavy, pile of crap. How is Linux not beating that when it's ME on a freaking pentium 3?! It's pathetic!

Re:Your Goal: One Second or Less

[ToasterMonkey]

I think it's fair to consider time to resume from hibernation when talking about boot times. On some systems, hibernation (or "Safe Sleep") is automatic, anytime the system sleeps. Because of that, portable devices can have very high hibernation resume to cold boot ratios. Honestly, I can't think of a good reason there shouldn't be many more sleep/hibernation resumes than cold boots for any PC, unless the OS really sucks at doing it, or the user is a masochist. - NOT reasons to discount it for systems that do it well.

Anyway, especially for portables, hibernation resume time at least should be highly relevant. Cold booting a PC should be up there with jumpstarting a car nowadays. As in, as long as it's a somewhat reasonable time, who cares. If you have to do it often, you have greater problems.

Re:Your Goal: One Second or Less

[Elladan]

Boot time for an OS is measured from boot loader to usable system, not from power button. This is a different measurement from boot time for a whole computer system.

Why the distinction?

Well, for example, Linux can't do anything if your BIOS sucks. It could boot in half a second, but it'll still take 90 seconds to get to the boot loader. What this means is that you use the OS time when comparing operating systems.

When comparing a Mac to an Dell computer, you compare the OS and the hardware boot time together, because Apple and Dell can actually fix that BIOS stuff if you complain enough.

So, the poster above was right to compare OSX to Ubuntu using the time from boot loader.

Re:Your Goal: One Second or Less

[xtracto]

That's why Linux takes so long to boot these days. You can have very good hardware compatibility or you can have very good boot speed. You can't have both. (Well, until someone invents persistent RAM.)

I agree that detecting and configuring all the devices might take time. What I do not understand is why should it be doing at booting time and, why should the OS wait until the devices are detected.

IMHO, only the keyboard, mouse, screen , hard disk and memory are absolutely needed for booting. All the other things can be loaded afterwards or even in parallel (of course, giving the proces a higher than 0 nice value).

Re:Your Goal: One Second or Less

[RMH101]

My password is 10 asterisks, you insensitive clod!
Apologies to Dilbert.

Re:Your Goal: One Second or Less

[xtracto]

This [photosounder.com] is the commercial project I'm talking about. People were excited about the original idea,

Despite your shameless ad plug, let me give you a tip. You may have better luck by implementing your idea as a plug-in for one of the many sound editors available like SoundForge, Pro-Tools, GarageBand and whatnot.

Re:Your Goal: One Second or Less

[Corporate Troll]

Very very different market. Have you seen the *inside* of a Veyron? It's not some go-cart that can go 400kmh. You should be able to use the Veyron as a daily car. I don't see you doing that with an Ariel Atom.

No doubt that the Atom is extremely fun to drive, but I'd see it as a competitor of Lotus. (Didn't check the specs, but Lotus also goes the "ultra-light + powerful engine = insane acceleration" way.)

Re:Your Goal: One Second or Less

[4D6963]

Thanks for the advice, I get it a lot, but how would you go about integrating it in a way that would make it more useful than as a standalone program?

Re:Your Goal: One Second or Less

[frenchbedroom]

Including foreplay and shower ?

Re:Your Goal: One Second or Less

[xorsyst]

Last I looked you can't hibernate safely if you are using on-the-fly whole disk encryption. Which you should be.

Re:Your Goal: One Second or Less

[TarpaKungs]

NFS is 1) a block device

No, it's a Network File System (hint, the initials). NFS supports filesystem primitive operations - see: http://tools.ietf.org/html/rfc1094 Networked block devices are possible, but NFS isn't one of them. However, some of the points are still valid - unix doesn't particularly like it when mounted filesystems go away unexpectedly.

Re:Your Goal: One Second or Less

[RiotingPacifist]

To save power and because some hardware sucks (e.g my hdd is screwed and can take my system down)

Re:Your Goal: One Second or Less

[petermgreen]

but with GUI unlikely.
Anyone remember the archimedies machines with riscos 2.x. The screen flashed up

risc os 2048K
acorn ADFS

and then almost immediately the desktop popped up.

Of course having the OS in rom, the configuration in some kind of solid state storage (not sure if it was EEProm or battery backed ram) and a relatively known hardware setup helped a lot.

Re:Your Goal: One Second or Less

[bepe86]

The PCI bus is 33MHz. 33 Million cycles equals about 1 second.

Re:Your Goal: One Second or Less

[petermgreen]

Afaict every PC has either an ISA bus or something that looks like one to software (e.g. the intel LPC bus).

Re:Your Goal: One Second or Less

[petermgreen]

I presume he means PCI bus clock cycles not CPU clock cycles.

Normal PCI slots are and always have been 32 bit 33mhz, there are faster variants of PCI but they have remained niche interfaces (mostly used in servers).

Re:Your Goal: One Second or Less

[petermgreen]

Though IIRC when they finally got stig to drive it round thier own test track and put it on the leaderboard they were surprised how poorly it performed.

Re:Your Goal: One Second or Less

[j_sp_r]

Go to BIOS -> Fast boot -> on (or whatever it is called)

It skips the RAM check.

Also disable any secondary disk controllers that you don't use (and boot from LAN)

Re:Your Goal: One Second or Less

[Laurence0]

Best idea, I think, would be for grub to give two options - fast boot and scan boot. Have a timeout of, say, 2 seconds then it'll fast boot by default. When it fast boots, it assumes all the hardware is the same as last time (most things which change frequently will be USB and therefore hotplug anyway, so this shouldn't matter) and doesn't bother to scan everything.

If it hits any errors during the fast boot, or if anything doesn't appear to be right then it can do a full scanning boot.

Ideally, there'd be a flag you can set before you restart or shut down as well, so if you know you're about to add a hard drive, you click "shut down and scan on next boot" and you don't have to worry about catching grub.

Re:Your Goal: One Second or Less

[petermgreen]

Ever notice that your video card works before you install drivers for it?
For sufficiantly small values of works.

Re:Your Goal: One Second or Less

[TheRaven64]

Nice try. After 4 seconds, a Mac with OS X 10.4 won't even have turned on its video output. After 2 minutes you might see the boot screen, but good luck getting any programs to run within the first 3-4 minutes.

Not sure what hardware you're testing this on. I only usually reboot my Mac after a security update, but last time I did without one it loaded the window server in under 4 seconds and completed the progress bar and displayed the log in window in about 10. Launching all of the applications I have set to start when I log in took a little over a minute, and was by far the biggest chunk of the boot time. Getting the applications back to the state they were in when I shut down took even longer, which is why I usually just use sleep mode.

Re:Your Goal: One Second or Less

[BlackPignouf]

but the Veyron is designed to last 20 or 30 years of road driving.

Too bad that oil reserves are not designed to sustain 20 or 30 years of Veyron's road driving! :D

At full throttle, it consumes around 2mpg or 115l/km, and its fuel tank empties in a whopping 12mins : assuming it could really drive at top speed during 20 years, it would consume ~0.5 million barrels.

Re:Your Goal: One Second or Less

[Rich0]

The Top Gear presenters kept comparing it to Concorde. That's how big of a leap forward it was.

And about how practical it was as well... The Concorde may be a very appropriate analogy.

Yes, it works. Yes, some could afford it. Yes, it is actually somewhat practical provided you're willing to pay 100X the cost of other options for getting from point A to point B.

Yes, it will also sell about as many units as the concorde did. Yes, it will have about as much impact on driving as the Concorde had on aviation. No, chances are you'll never ride in anything like it.

Who knows, maybe somebody will come up with a more practical alternative - both for the plane and this car.

Re:Your Goal: One Second or Less

[TheRaven64]

BeOS booted to a usable GUI in under 5 seconds on my 200MHz Pentium with 32MB of RAM. This was commodity x86 hardware with no special BeOS support. Going through the BIOS took longer than going from the bootloader to the GUI. I've not looked at Haiku, but I'd imagine it's similar. Thinking about it, Haiku would probably make a good OS for a netbook, considering how responsive it is on very low-end hardware.

Re:Your Goal: One Second or Less

[TheRaven64]

Pretty much. The (old?) CEO of Palm got a block of balsa wood cut that fitted in his shirt pocket and gave it to his engineers as their guide, telling them to fit as much as they could into the system but keep it within that size. The Palm sold well because executives could easily fit it in their pocket. Only a tiny fraction of the things that are technically possible are ever built. Working out which ones are worth building is difficult.

Re:Your Goal: One Second or Less

[Rich0]

Frankly, what linux really needs is a working network filesystem.

NFS is just a real mess. It has all kinds of security issues, and has no concept of users beyond the local machine. If /etc/passwd isn't synced across the network all kinds of stuff goes wrong. It also has a lot of limits around permissions/etc.

Probably the next closest usable network filesystem for linux is samba - which really isn't ideal (for one it is almost entirely reverse-engineered and depends on a spec that isn't open). That filesystem does handle many of these issues, but not in a linux-friendly way.

I just can't believe that nobody considers it important to have file sharing over a network that "just works". Windows has had fairly simple to set up file serving from the desktop for at least a decade. Sure, like all things windows they had a number of security bugs, but that has settled down a bit.

OpenAFS seems to have some potential, but nobody uses it, and it is a bit of a pain to set up since it has so many layers to it. Is anybody working on a simple way to have users on a network be able to mount their home directories remotely with working permissions and non-synced /etc/passwd?

Re:Your Goal: One Second or Less

[TheRaven64]

Some free advice for you:

You can't sell that as a product. It's a cool demo, but it leaves you thinking 'now what?' It's a fun toy, but does it have any real use? As a stand-alone product, it just isn't compelling. As a plug-in for another system, it might. When you've done the OS X port, try using the code as import and export plugins for things like iTunes and iPhoto, or even for Photoshop or some high-end audio editing apps. Once you've got that working, take it to the makers of the original app, and ask if they want to buy it. Then all you have to do is convince them that it's better to hire you than to get their own people to reimplement it. Maybe iLife '10 will have sound-as-image editing across the board.

The other option is to open source it under a permissive license, make it a library, and get it added to other open source projects, like the GIMP, and use it as marketing for yourself. This won't make money directly, but being able to point to something like that that everyone is using and writing about will make it much easier for you to demand a high salary from the company you choose to work for.

Re:Your Goal: One Second or Less

[RichardJenkins]

I'd really like my computer to automatically log me in if it detects a bluetooth device with a particular signature (ie my phone) close to the machine.

Re:Your Goal: One Second or Less

[fprintf]

I don't know the order of operations for all the items loading on system boot, but I can say the goal would be to have a usable system as quickly as possible. One thing I wouldn't want, which exists today, is a GUI that looks ready to use but has an hourglass or otherwise makes me wait until it is done. On my current Ubuntu installation once I actually get through loading all the stuff, and type in my login, my desktop comes up pretty quickly, but then it takes a few more seconds until I can do anything. Those few seconds are spent waiting in anticipation for the ability to click or type on something.

If I wasn't unsure of the power usage of my laptop during sleep I'd probably never shutdown. Resume from suspend is really super fast.

Re:Your Goal: One Second or Less

[Shamenaught]

It's talking about the bus's clock speed, not the computer's. PCI operates at 33 MHz. Nowadays there's also PCI 2.2 (66 MHz) and PCI-X (133 MHz), for which the timeout would be lower, but I'm unsure if that means they can eliminate the legacy delay without removing legacy compatibility.

Re:Your Goal: One Second or Less

[borizz]

Mine does it perfectly, every time. Everything comes back properly. It's an MSI 271 on XP Pro.

Re:Your Goal: One Second or Less

[MarkKB]

All I see is **********. Maybe someone else could post their password so we can check it.

You can go hunter2 my hunter2-ing hunter2!

Haha, does that look funny to you?

Re:Your Goal: One Second or Less

[Ginger Unicorn]

im not posting this to be smug, but i run firefox with adblock plus and noscript installed and i've barely seen a single ad for 3 years, and absolutely no pop ups. Noscript also blocks flash and java apps from loading till you click on them which saves time and bandwidth. If you're sick of ads on the internet then I highly recommend it - when i use IE on someone else's PC the difference in clutter and obnoxious flashing junk in web pages makes me wonder how i ever put up with it.

Re:Your Goal: One Second or Less

[eggstasy]

I believe you are looking for the standby button.
Why do you even turn it off? Ecological reasons? Noise?
Even Windows can be up forever these days... and with the dawn of unlimited broadband and large *cough* downloads... most people just leave their computer on all the time.

Re:Your Goal: One Second or Less

[hairyfeet]

Why not a ROM? That is how the machines were so fast in the old days. Why not have say, a Linux OS and all the basics written onto a ROM and have it load from there, and then have the users apps on a SSD? It seems like for something like a Netbook(where the hardware isn't going to change) that having the OS as a ROM image would not only speed up booting but also make it pretty hack proof, as the OS is in a read only state. And since most folks use the Netbooks as a "browser in a box" anyway by having the OS locked into ROM they can have their apps while having the OS idiot proofed against them screwing something up.

It just seems to me that a ROM connected by something like Hyper transport to the CPU and RAM would make even the slower Netbook chips feel very fast and responsive while at the same time reducing the risk that the user might screw something up. Or is there something I'm missing?

Re:Your Goal: One Second or Less

[DrgnDancer]

The first guy was wrong (I WISH my MacBook booted in 4 seconds), but you're being equally ridiculous in the other direction. I installed the Windows 7 Beta on my boot camp partition last night to check it out, and I feel pretty confident saying it took nothing like 3-4 minutes for systems to become usable in either Windows or OS X in any of the several boots I made in a short time. It took about 10 seconds to get to the Boot Camp OS choosers (Which is about twice as long as usual because I had a bootable DVD in the Drive. Without the DVD it's more like 5 seconds). Once the OS is chosen, I'd say it was may 20-30 seconds for a usable OS X desktop, and a tiny bit more for the Windows Desktop. It's no more than a 30 second, and maybe as little as 25 second process if I don't try to use the Boot Camp loader (For whatever reason it takes a bot longer to use Boot Camp).

Re:Your Goal: One Second or Less

[Sj0]

Why the hell should I be using on-the-fly whole disk encryption?

My machine's physical security is high, so it's unlikely I'll have it stolen. In the unlikely event that it is stolen, I've got bigger problems.

So it seems to me that whole-disk encryption is a waste of processor power, killing partition access CPU utilization for a security element that will go entirely unused and will ensure if my machine has a hardware failure I won't be able to access my own files from another PC.

Re:Your Goal: One Second or Less

[Sj0]

You're comparing apples and oranges, that's why.

MS-Dos beats all these fancy pants operating systems. On modern hardware, it will boot so fast you won't have a chance to hit the stopwatch.

If you wanted, you could put a copy of MS-DOS 7 and arachne on a (PATA) SSD, load it up on the fastest single core processor available with cache maxed out, and your boot time would be the bios start time, nothing else. You'd have some basic computer functions, and great boot time, but good luck getting anything done.

My GPS runs Windows CE 5, and it boots in the time it takes the start-up sound and animation to play. Guess what it does? Exactly ONE thing.

Re:Your Goal: One Second or Less

[xorsyst]

I've had a computer stolen. There's nothing quite like that feeling of some random criminal being able to look at all your files to spur you to action. In my case my laptop was the only thing stolen and I had backups of important data, so it _was_ the biggest of my problems. With modern CPUs you almost can't notice the CPU overhead and you certainly don't see a speed difference in day-to-day use. Hell, I've got a pentium 3 800Mhz PC fully encrypted and you don't notice it. All the encumberance you have is the requirement to type a password on bootup. As for the hardware failure, that's fluff. Stick the hard drive in another PC and install the same software and you'll be able to mount the drive and access the files. There are negatives to whole disk encryption but they are small compared to the peace of mind.

Re:Your Goal: One Second or Less

[hesaigo999ca]

back in the day...red hat7.0, mandrake7.0, would all boot within about 5 to 10 seconds...
maybe its all the add ins these days to stay competitive to M$???

Re:Your Goal: One Second or Less

[Sj0]

Maybe I'm naive, but why is everyone so intent on the boot time of their PC?

Personally, I only have to boot up my computer every few weeks. In the meantime, either it runs, or I'll send it to suspend. Booting isn't something I spend a lot of time doing.

Re:Your Goal: One Second or Less

[LWATCDR]

Yea as a simple VESA framebuffer!
I hate it the fake fast boot. Vista kinda does that. We had a bunch of customers that would try to run our software and get a strange error and then it worked fine. The reason? A device driver that the program needed wasn't loaded yet and it gave an error.

Re:Your Goal: One Second or Less

[Doug Jensen]

Why the hell should I be using on-the-fly whole disk encryption?

My machine's physical security is high, so it's unlikely I'll have it stolen. In the unlikely event that it is stolen, I've got bigger problems.

So it seems to me that whole-disk encryption is a waste of processor power, killing partition access CPU utilization for a security element that will go entirely unused and will ensure if my machine has a hardware failure I won't be able to access my own files from another PC.

Uh, because your employer requires on-the-fly whole disk encryption? Mine has a security guard at all doors who checks every outgoing laptop to be sure no one takes one out without encryption.

Re:Your Goal: One Second or Less

[ckaminski]

Using IBM SGE on a thinkpad, yes, I did notice the difference. I consume every available drop of CPU power in my machines, and the 5-10% +/- I lost when I had to convert the disk with SGE, it hurt. Then again, I'm not your typical road warrior...

Re:Your Goal: One Second or Less

[Doug Jensen]

Mostly because people are carrying around laptops.

Re:Your Goal: One Second or Less

[Medievalist]

back in the day...red hat7.0, mandrake7.0, would all boot within about 5 to 10 seconds...
maybe its all the add ins these days to stay competitive to M$???

Look up "Wirth's Law".

Re:Your Goal: One Second or Less

[cb88]

ArchLinux does indeed rock! I can boot a 100% uptodate archlinux even on a 300mhz PC with only 64mb ram ... where as Ubuntu probably wouldn't even boot in that little ram

Re:Your Goal: One Second or Less

[darkonc]

If you care about the results of the IO to the server that's timing out, then aborting is just going to cause you problems. Although rebooting your local system and THEN waiting for your server to come back makes it seem like you're doing something, it's actually going to take a few seconds longer for you to get your wanted results than if you just switched to another window and did something else while waiting for your server to respond.

Personally, I think that the Windows default of timing out and trashing whatever transaction you wanted to do is bad netiquette. Even if it takes hours for the server to come back, it's usually better if you can safely complete your pending operation when it does and continue on.
If the server's never coming back, then somebody is gonna have to tell you and you can then take appropriate action. .

Re:Your Goal: One Second or Less

[CarpetShark]

Correction: fully usable and fully secured desktop. All of that after-login antivirus loading on windows needs to be measured too, if you're judging fairly, since a) other benchmarks (performance and security etc.) of the desktops are done when services/daemons aren't actively loading. b) The desktop isn't truly usable until it's necessary services are loaded.

Re:Your Goal: One Second or Less

[Sj0]

I can see the problem. I'm the sort of person who would happily buy a piece of innovative software over the internet, but I have no reason to want your software.

"You can draw sound!", you say. So?

You mention it's unexpectedly useful on the website, but you say anything to make me want to even try the demo. It looks like something abstract and strange, that I won't be able to wrap my head around.

Your problem isn't marketing as such, it's a communications problem. You say you're excited about the product, and you say others are excited about the product, but you don't bother telling anyone why. This is a brand new paradigm, you're going to have to tell prospective customers why they should be remotely interested.

Re:Your Goal: One Second or Less

[Sgs-Cruz]

So design the system so it can only handle new hardware. Just clearly post on the website for One-Second Linux that it won't handle your ISA scanner controller card from 1992. Tonnes of people would still be interested.

Re:Your Goal: One Second or Less

[Sj0]

flash drives are painfully slow to run from. Ask an EEE PC or Acer Aspire One user after they convert from the super optimized version of their OS to something generic like Windows XP.

Seek times become negligible, but throughput is substantially less than a good HDD.

Re:Your Goal: One Second or Less

[Sj0]

That actually gives me an idea. Have a login screen that shows up the moment the framebuffer device is available. Allow the user to enter his or her username and password, then begin pulling information off the network (after the network comes up), while the rest of the OS loads. Considering the pathetic rate at which most people type, I think it'd be an excellent illusion that could work pretty well to make people think the OS is booting faster than it really is, while removing the user bottleneck from the boot process later down the line.

Re:Your Goal: One Second or Less

[Sgs-Cruz]

Humanity put several people on the moon. It's bullshit to think that a one-second boot is simply out of the question. Maybe with the current PC architecture with all the legacy support that entails.

Start with a clean slate. PCI express only (no slow PCI bus, no quasi-ISA bus), a mandatory flash drive on the motherboard which stores the 500 MB of "boot-up" data that you need to get to a desktop. So maybe it takes another 2 seconds to get an IP address; that's okay because it takes the user that long to double-click the Firefox icon anyway. But you should be able to click the Firefox icon and start the program loading in one second.

And I don't know shit about computer architecture. Whoever it was up above that said this needs a Steve Jobs to inspire the geeks to just figure out how to do it was right on.

Re:Your Goal: One Second or Less

[thtrgremlin]

I think it has something to do wit the motherboard I got. It has a LOT of stuff on it. I hit the power button, and the fans come on quietly. after a couple seconds, the computer beeps, and the fans get really loud for maybe 3 seconds before they slow down, and the screen lights up, and the animation for the bios appears. 3-4 seconds later, the screen goes black, some text appears and disappears in less then a second, and a cursor sits there for 7-8 seconds before the grub menu appears to do its 2 second count down. I can hit enter right away, or even just while the cursor is blinking, and when the menu appears, it goes to the Ubuntu loading screen within a second, and 12-15 seconds later, the GDM appears. After name and password thing, the desktop is ready to use in about 6 seconds, but a bit sluggish for about 20-30 seconds depending on how many things I try to open up right away.

I wish I could get LinuxBIOS cause I know that would fix some of it, but not exactly being legacy hardware, it isn't supported. I have tried grub2, and that takes MUCH longer. So my entire boot up time, if I got it right, is about 38 seconds. Like I said, I don't get why grub is so slow, but I figure all in all, life is too short to spend it being impatient.

Re:Your Goal: One Second or Less

[MCZapf]

NFSv4 was supposed to take the best of NFS and AFS and combine them, but development of it never got far, apparently.

Re:Your Goal: One Second or Less

[4D6963]

Well, yeah, that's one big problem, I think part of the problem is that it doesn't really do one thing but makes lots of things possible. It's actually not that abstract, you may even consider that it's the opposite of abstract, as it dabbles directly with what constitutes "sound features", that's more like a low level but very potent approach.

I think low level is the right word, the only thing lower level than this is directly setting your sample values by hand (but it would be impossible to accomplish something precise this way). Actually, that's it, that's the explanation of what it does, it may be confusing, but that's the most "low level" approach to sound edition or creation, the assembly language of sound if you will.

I'll admit, it's a hard concept to explain, let alone sell, but that's it, and that's why it's so potent. I could mention what I know can be done with it but the thing is, I only know a few things, when other people experiment with it I learn of new possibilities, I don't actually know the full extent of what can be done with it, because the answer to that is, pretty much any possible sound.

But okay, that doesn't tell you why you should buy it. So here's one : you can try new effects with it, for example, by turning a few knobs or pushing a few buttons you can obtain effects that weren't possible before. You can export the image to work with it in Photoshop/GIMP and experiment with other effects and transformations, for example you could try taking speech, exporting the image, moving the image down a bit and blurring it and mixing it with the original sound to give it a sort of bass resonance. You could try warping and moving stuff around to change how something progresses or make it sound weird, like possibly changing the tone/mood of a voice or, I don't know, I just thought these things up.

You can as well spend your time breaking samples down into separate "layers" of instruments, or even breaking instruments themselves into layers, alterate them separately and arrange them into something different, or go through your collection of cloud pictures to look for the perfect eagle-ish cry, it's hard to tell you what you'll want to do with it really.

I think it's one of these things that you need without knowing you need, to use the Apple I example again, back then people might have thought "What would I need a personal computer for?" and now everybody knows they need that. I think my program answers to a lot of problems, one of them being that lots of people who make music or sound effects don't know where to turn anymore to find new and novel sources of sound. Photosounder is an answer to that problem in that every kind of sound can be created from scratch with it. I'll concede that it won't be the easiest thing for someone to create just what they want by drawing it from scratch, that's not even how it goes (my approach is look at the sounds that sound close to what you want, try to reproduce them, then try to do it again differently in a way that matches what you want), but on the other hand that's the one thing you can do anything with. Not everything there is hard either, if you look at the video that deals with drawing a beat, then it seems relatively simple to create a satisfactory drum beat. Surely you might prefer using your TR-808 to make a drum beat, but at this point it's like all the possible sounds it could make are on wax/CD now, so here's your opportunity to try something new, and it's not even too hard, you only have to draw a curve for a bass drum, and a large blob cut in half for a snare drum, it's easy to experiment from that point.

How do I get all these points across, I don't know, I don't even know if I got them across to you, I'm notorious for failing at explaining my novel ideas.

Re:Your Goal: One Second or Less

[cHiphead]

You must have really strong arms.

Re:Your Goal: One Second or Less

[Shokac]

You forgot other characters... Password: Ubuntu904! There, I fixed it for you.

As usual, I forgot to login before writing post.

Re:Your Goal: One Second or Less

[skeeto]

On my ubuntu box I'd estimate a good 50% of my boot time is after the login screen before I'm able to do what I wanted to do.

A couple of years ago in college I used the computer labs for printing, and the nearest one had some brand-new macs. I would prepare a PDF on my own machine, then e-mail it to myself where I could access it from the lab. The frustrating part was that logging in took over a minute. The computer was already booted but I still had to waste a minute before I could do anything with it.

Re:Your Goal: One Second or Less

[SBrach]

Hibernate, suspend, and other power saving features is really working well on my Dell inspiron 6400 with Win 7 Beta Build 7000. It worked good in Vista also but I did have a few problems with what it would do if I closed the lid with it plugged in and then unplugged it. Win7 also adds a "Dim the Display" interval.

Re:Your Goal: One Second or Less

[DrgnDancer]

To be fair, while Microsoft's file sharing system doesn't have all of the weaknesses of NFS, it does have a fair few. Unless you are using AD you have to sync users in MS land as much as in NFS (and if you are using AD, then to be fair you should consider the *nix systems to be using LDAP or NIS to share user directory information, which alleviates the same problem as AD does). Lost Windows shares, while they don't hang the file system handle forever like hard NFS mounts, do take several minutes to time out and often take out ALL file system access while they wait. A broken NFS handle will take out the application trying to reach the remote file system, but a new shell or another program can get to the rest of the file system. Broken Windows pipes kill all instances of Explorer, My Computer, and any open file browsing windows for other apps, and hang any new attempts to open something else until they finally time out. There can be little doubt that Windows SMB sharing is more secure than NFS if properly configured though, which is a bit disheartening.

Re:Your Goal: One Second or Less

[HeronBlademaster]

I didn't say you'd be able to play Fallout 3, I just said it works. You can certainly browse the web well enough to go find the proper driver ;)

Re:Your Goal: One Second or Less

[Chabo]

You assume that you'd drive it at full throttle all the time. As the Top Gear presenters also showed, if you drive a Toyota Prius at full throttle, you get about 17 miles to the (imperial) gallon.

I mean the Veyron's not exactly a gas-sipper; if Wikipedia is to be believed, then it gets 10 miles per (US) gallon with "combined" driving. But don't assume that it'll be using all 1000 horsepower all the time.

Re:Your Goal: One Second or Less

[swillden]

2^25 is around 33 million. Surely 33M cycles isn't a second these days?

It is on a 33Mhz PCI bus. It's half a second on 66Mhz PCI.

PCI-X 1.0 can run at 133Mhz, so a quarter second, assuming they didn't revise the wait time to 133M cycles, or something.

PCI-X 2.0 can run at 266Mhz and 533Mh, so 1/8th second or 1/16th second, again assuming they didn't revise the wait time cycle count upwards.

Re:Your Goal: One Second or Less

[HeronBlademaster]

I'm trying to get used to Linux, since the job I start at in May will require me to use Linux for all of my development. It's not a big enough deal to me to get suspend working in Linux, because come May this laptop is going to be my wife's and I'll probably wipe Linux since she won't use it.

Furthermore, suspend wouldn't do me a whole lot of good, since I can't be sure that when I open the machine again I won't want to be in Windows (I like Left 4 Dead...), and rebooting after waiting for the machine to resume from suspension takes longer than a cold boot.

In other words, it's not worth my time.

Re:Your Goal: One Second or Less

[swillden]

If it boots in less that 1/3 to 1/6 as much time as ext3... Surely there will be an improvement in overall performance?

I seriously doubt the major factor in boot time improvement is the file system. They're also continuing to work on Upstart, their replacement for the SysV init daemon, and one of Upstart's primary goals is to increase parallelism in the boot process. The traditional boot process is quite linear and as a result spends a lot of time waiting around.

Re:Your Goal: One Second or Less

[swillden]

Unfortunately, just putting `gnome-screensaver-command -l` into the session manager won't work because it doesn't seem to load immediately. Instead, I made it run a script that executes that command in between calls to `sleep 1` six or eight times. It works for me.

Another option is to use KDE, which has automatic login to a locked screen as a built-in feature. Go to the login tab of the control panel and turn it on.

Well, in KDE 3.5, anyway. Dunno if that feature is in 4.x yet.

Re:Your Goal: One Second or Less

[Chabo]

It wasn't that it was slow, it's just that it didn't reach the top. James May guessed it was too heavy to go around the corners quickly.

Ironically, if Bugatti had let them run the Veyron around the track when it was first launched in 2005, it would've been at the top of the Top Gear Power Board for almost a year, when it would've been beaten by the Koenigsegg CCX2, with the wing added. Then the Ascari A10 a year and a half after that, then the Gumpert, Zonda F Convertible, and the Caterham R500 in 2008.

Re:Your Goal: One Second or Less

[Vexorian]

Slightly OT: In windows, even after logging in, there's just a huge spam of seconds (about 30) in which it is starting yet more programs, and you really can't expect firefox or IE or your favorite game to even start until this period goes off. Really annoying, and thus I consider this time as well when measuring windows' boot time.

Re:Your Goal: One Second or Less

[edbob]

Actually, if you entered "win" you would get an error. This was a TRS-80 Color Computer. The OS was made by Microsoft, though.

Re:Your Goal: One Second or Less

[Chabo]

I think both of these accomplishments are very important. They may not be practical, but they push the boundaries, and they help progress our society. I certainly don't expect our society to progress after seeing the newest version of the Hyundai Accent, or even the Mustang, do you?

Forty years ago, supercars like the Ford GT40 and Ferrari Daytona had around 350 horsepower. Now you can buy that much horsepower for under $30,000 new. Maybe forty years from now we'll be able to get 1000 horsepower in normal road cars!

Re:Your Goal: One Second or Less

[Skjellifetti]

That's why Linux takes so long to boot these days. You can have very good hardware compatibility or you can have very good boot speed. You can't have both. (Well, until someone invents persistent RAM.)

Or you could just write the list of modules required by a machine to a file during the initial install and look it up during later boots. I installed Ubuntu about 3 months ago on my home pc (older VP6 MB, 2x PIII processors and some ide disks). Ubuntu takes 20 minutes to boot. Further, since the cups system locks up the machine every time someone tries to print, the machine needs frequent reboots. No, I've never had these problems with any previous versions of linux on this box. Ubuntu is going to be replaced as soon as I have a spare weekend and will never again be installed on any machine I control.

Re:Your Goal: One Second or Less

[Sj0]

I'm sorry, you didn't answer my question.

"Why the hell should I be using on-the-fly whole disk encryption?"

Your phobias and neuroses aren't my problem.

Re:Your Goal: One Second or Less

[Weasel Boy]

I have owned quite a few computers that could boot in about a second. One of them is even a PC-grade machine with a GUI and a hard drive. What they all have in common is the OS burned into fast ROM.

Re:Your Goal: One Second or Less

[Jamie Lokier]

I've often had useful web pages take 40 seconds or more to come up. It's annoying, it's relatively rare, but it would be very annoying if they didn't come up at all.

I've also had ping times in excess of 60 seconds. Ho hum, it's useful to know the machine is still there at least :-)

Re:Your Goal: One Second or Less

[Kalriath]

So Coreboot doesn't bother doing any POST at all? Screw that. I'd rather be assured that no memory is faulty because my BIOS has tested it, and I'd rather have a test establish whether my CPU can handle a given clock speed/voltage setting and undo any changes.

No, Coreboot is just a bad idea. Not doing a Power On Self Test (and sanity-checking configuration) is completely idiotic.

Re:Your Goal: One Second or Less

[emlyncorrin]

OK, mine is hunter2

Re:Your Goal: One Second or Less

[Jamie Lokier]

The point isn't PCI bus cycles, though.

It takes time, physically, for devices to get from power and reset good to charging up all the little bits internally before expecting them to behave sensibly. It also provides time for devices to boot themselves logically - many of them have their own software to boot, and/or FPGAs to load up. PCI bus speed is irrelevant to that.

If PCI devices could just block a configuration request until their internal boot was ready, that would be better. But some PCI devices don't have any logic at all until they're booted - those with programmable logic interfaces direct to the bus. The boot delay makes it possible to build those without an extra chip in between.

Re:Your Goal: One Second or Less

[Jamie Lokier]

How do you do that?

Re:Your Goal: One Second or Less

[Kalriath]

Actually, I've got Quick Boot enabled, so it only does a single pass of memory. To be perfectly honest, I don't understand why it takes so long for the rest of it. In fact quite a bit of that time isn't spent in POST, but in the BIOS nonetheless (updating the configuration tables perhaps? PnP initialisation?)

And if it helps any, no I didn't enable PXE.

Re:Your Goal: One Second or Less

[Kalriath]

It doesn't actually skip the RAM check, just reduces it to a single pass using a much faster test method. I do have it enabled, and PXE disabled, yet it still takes a while. To be honest, I think it's likely got more to do with PnP initialisation and updating configuration tables.

Re:Your Goal: One Second or Less

[Jamie Lokier]

Suspend does work on my laptop.

But on the other hand, there is no Ubuntu kernel which simultaneously works with Bluetooth 3g broadband and can write to a CD on my laptop.

So I reboot mine just to change between working kernel feature sets quite often.

I also run out of battery quite often, when working in a cafe. Then I switch to a second battery, but that requires a reboot too...

Re:Your Goal: One Second or Less

[Jamie Lokier]

Hibernate also compresses your RAM, and it's usually quite compressible.

The major delay is probably rotational disk seek times. After the login screen on Ubuntu Intrepid, I still have to wait ~10-15 seconds before I can launch anything from the desktop, while it's loading a few things onto the task bar, starting NetworkManager (the applet), loading menus and icons, etc. The disk is seeking heavily in that time, and I strongly suspect that's what makes those things slow.

Try booting from an SSD (flash replacement for hard disk at a decent speed, not the cheap USB ones). I don't have one, alas. I would be surprised if it doesn't boot much faster.

Re:Your Goal: One Second or Less

[HeronBlademaster]

I've never seen a Windows hibernate file smaller than my physical RAM size. It might not use all the space, but it certainly reserves the full size.

Re:Your Goal: One Second or Less

[Jamie Lokier]

I don't know about Windows (don't use it, except in virtual machines :-)

But Linux hibernate is similar, it requires as much space as RAM to be reserved (in swap, doesn't need a separate file), but it won't use all the space if it can compress RAM contents when storing it.

Re:Your Goal: One Second or Less

[nxtw]

my EFI system (Intel Mac) takes just as long to start loading as any other recent PC I've used, regardless of the OS being loaded.

Re:Your Goal: One Second or Less

[Just Some Guy]

I'd rather be assured that no memory is faulty because my BIOS has tested it,

So let the booting kernel test it. It can't take more than a second or two to read/write a few gigs of data.

and I'd rather have a test establish whether my CPU can handle a given clock speed/voltage setting and undo any changes.

Ah, I see the disconnect. I don't think the coreboot people have any interesting in pushing their hardware out of spec just to see what happens.

No, Coreboot is just a bad idea. Not doing a Power On Self Test (and sanity-checking configuration) is completely idiotic.

Good point. Those knuckleheads at Google, AMD, and VIA should bow to your superior wisdom.

Re:Your Goal: One Second or Less

[isorox]

My laptop runs fluxbox, which is pretty quick post-login (I think, I don't login much). It's a pain starting up though, 85 seconds according to boot chart. Most of that's disk activity at 30MBytes/second, it kicks in at T+12.5 (just after "resume" finishes), and stays at 100% until T+75.

Looing at the boot-time, it appears that "flumotion" is responsible for blocking the process for about 20 seconds in iowait. apt-cache is another bottleneck

Re:Your Goal: One Second or Less

[nxtw]

Digital cameras and many other portable devices don't have to deal with annoying things like:

• detecting RAM & CPU
• detecting system fans (on systems with speed-controlled fans)
• wait for VGA BIOS initilization (remember video cards that displayed a BIOS screen?)
• enumerate any busses that may be used for booting the system (PCI, USB, etc)
• run any option ROMs (network and storage cards)
• wait for hard drive spinup
• choose a boot device

Re:Your Goal: One Second or Less

[Sj0]

Well, let's start at the beginning.

What do you mean by "seeing sound"? Is it like a fourier transform on one axis and time on the other axis? If I paint a long horizontal line, what will I hear; a tone? How can I figure out where each frequency is? Are there tools to make that easy?

Now let's continue, but keeping it simple. I'm a DIY artist. I'm doing this for fun. I don't know what a double feedback bypass demilog filter is, and I don't care.

If I wanted to make a guitar-like sound, would it look like a fuzzy line that fades out? If I imagine what a guitar-like sound "looks" like, that's what I think of. If it shows what I'm "thinking" of, then it's very intuitive, unlike waves, or the strange words people use to describe filters and wave patterns.

Finally, a glimpse of something more advanced. If I wanted to make an echo, would I just have to copy/paste a transparent block of something? If I wanted to make something sound fuzzy, could I use a blur? If I wanted to sharpen the sound of something, could I erase some blur?

Putting it like that, I could see myself trying it out, and that's the first step to buying it.

Re:Your Goal: One Second or Less

[spitzak]

He meant that the machine may sit there for hours before somebody wants to use it and decides to login.

Re:Your Goal: One Second or Less

[xenocide2]

Right click on the nm-applet, pick "edit connections". Select your connection of interest, click edit. Check the "system setting" box and then "Ok."

Repeat for every connection you'd like to consider at startup.

Re:Your Goal: One Second or Less

[bucky0]

you have problems with the in-kernel filesystem structures getting out of sync with what is on the disk in that case.

Re:Your Goal: One Second or Less

[SanityInAnarchy]

One thing I wouldn't want, which exists today, is a GUI that looks ready to use but has an hourglass or otherwise makes me wait until it is done.

I wouldn't mind that, for two reasons:

First, at least it lets me input further actions. So, while I wait, I can click Firefox, Kopete, a terminal, a text editor, etc. Chances are, by the time I'm done doing that, at least one of them will be open and ready to use.

Second, it removes the useless animation/sound/dance/whatever that happens with a splash screen, which, with a sufficiently fast disk/filesystem, is actually going to be slowing things down, because maybe it would actually be usable in a few seconds. The worst are the ones that automatically stay on screen for a few seconds, even if the program in question is already usable.

If I wasn't unsure of the power usage of my laptop during sleep I'd probably never shutdown. Resume from suspend is really super fast.

Try it. I suppose it depends how well your system is configured, and what kind of hardware you've got, but a Mac in sleep will last days.

I don't sleep so much because I never know if I'm going to be booting it with or without a second monitor, and my own resume from sleep (or hibernation) is actually pretty annoyingly slow.

Re:Your Goal: One Second or Less

[Rich0]

True, I've seen some of these issues on Windows as well.

However, it doesn't require AD to address the security issues on Windows. An NT4 domain also works fine (I'm running one over Samba for my windows machines at home).

Oh, and it would be nice if MS had implemented roaming profiles in a more sane way. Actually, half the problem is stuff like Firefox that dumps 100MB of cache in the wrong profile folder so that it roams.

If I could get openafs working better I'd seriously consider using it across my network. If I could get my roaming profiles to work over it that would be even better - I wouldn't mind having 100MB of cache in my roaming profile if it replicated across all the machines on the network more efficiently. However, getting both linux and windows to play nice with krb5 sounds like a real pain.

Re:Your Goal: One Second or Less

[Heather D]

There's a difference with web browsers. People with dial up or satellite connections would have trouble with that. Back when I was still on dial up I noticed that it took much longer for some servers to respond than would be easily explained by the slower speeds involved. Sometimes adjusting the timeout settings would allow me to reach a server that would otherwise return a timeout error.

Re:Your Goal: One Second or Less

[toddestan]

I would just skip the ram test entirely, as those cheesy single pass tests don't verify anything, including whether you have the DIMM seated completely in the slot (I've seen it pass where you could look at the DIMM and see that half the connector was clearly not touching the pins). If you suspect you have bad ram, you're going to have to boot into something like memtest86 to let it churn so just disable the BIOS test save yourself the seconds.

Otherwise, it could be a crappy bios, or a motherboard with a ton of bells and whistles that need to be initialized. I have a Dell where the POST takes ~1 second, so by the time the monitor warms up you're looking at the Windows XP loading screen. And it's nothing special, some old P3-based Celeron system.

Re:Your Goal: One Second or Less

[bucky0]

It'd be great, but the problem is that a lot of the timeouts are necessarily serial processes because of the way the devices used to be written. (There's a lot of devices that depend on certain interrupts to be caught for their probing), even if you wanted to, you couldn't.

I agree there should be some sort of caching though to skip probing after the first boot. Maybe some people smarter than me are working on it?

Re:Your Goal: One Second or Less

[4D6963]

"Seeing sound" refers to the sound's spectrogram. It's frequency on the vertical axis and time on the horizontal axis (see the axis on the screenshot). Yes, an horizontal line will sound like a tone, a vertical line like a click, a diagonal like a chirp. You can see which frequency and pitch a feature matches to by referring to these axis.

If you want to make a guitar sound, it'll look like thin horizontal parallel straight lines that fade out, yes. The height of the lowest line defines which note is played, the distance between each of the parallel line and each of their respective intensity defines how the instrument sounds, defines how warm, metallic, harmonic/anharmonic something sounds.

It's more intuitive than some other approaches, but it's a bit more technical than that. You can't just obtain the sound you want by thinking about it hard and letting your hand do the rest ;-) you have to know how to do what you want. As I said, it's pretty low level, but if you want to create a guitar sound, you'll have to look at one first (that is open one with Photosounder and look at the image), then you can "let your hand do the job" by reproducing what you saw/what you remember seeing, and from that point on you can try tweaking things, doing things a bit differently, making some lines or parts brighter, spacing lines differently, adding a sort of glow to make it more noisy, and so on.

And yeah it's more intuitive than looking at waves, looking at waves informs you about little more than the sound's envelope, which is its intensity regardless of frequency. A spectrogram (what I refer to as a sound's image) is more informative in that it shows everything you can hear in a very descriptive way. Although once again it's very low level, it takes a bit of deciphering, depending on what you're looking at. It's very much like a musical score, actually if you loaded a musical score it would sound quite like the tune it's supposed to represent, if you omit the bars. To use a programming analogy, a piano score is like code in a high level language. The compiler (the interpret, or the synthesiser) turns it into sound. Photosounder disassembles this sound into an image, which is a low level representation of the tune, but if you compare it to the original "high level" score, you'll find out that basically the "compiler" turned the score into music by replacing graphically each dot (representing a note) into a set of parallel horizontal lines which are more or less long horizontally, depending on the length of the note interpreted by it. The interest of disassembling is that you can modify anything you want, i.e. you can shift a note up by moving it up, you can shorten one, or you can change the distance in parallel lines to make it sound like a more or less different instrument, and then reassemble the modified result into a sound, by pressing the Play button. Of course it might be more practical to just change the original score and reinterpret it, but maybe you're an electronic musician who wants to modify a bit a specific piano sample.

There are many kinds of echoes. The mountain kind (the one that repeats what you say) can be achieved by just duplicating many times the sound's image, shifting the copies to the right, toning them down a bit (making it darker) and blending the whole. You can get creative and achieve something new by also for example shifting the copies downwards a bit to get something progressively diminishing, or blur them progressively, or rotating them a bit, or whatever you can think of really. A more room-like echo tends to look like random "scanlines" overlaid on the original image.

Yes you can make a sound fuzzy by blurring it, although if you want to make it just more fuzzy in time (i.e. make it sound kind of slower while still playing at the same rate) you could just use an horizontal motion blur. Yes you can sharpen the sound you can erase some blur, that's actually what I do to fix the blur on some

Re:Your Goal: One Second or Less

[Just Some Guy]

It'd be great, but the problem is that a lot of the timeouts are necessarily serial processes because of the way the devices used to be written.

I'm not that much of a hardware guy (which should be obvious to anyone reading my questions), but couldn't the kernel do something like fire off several probe requests and then sit back and wait for them at the same time?

Re:Your Goal: One Second or Less

[xtracto]

Thanks for the advice, I get it a lot, but how would you go about integrating it in a way that would make it more useful than as a standalone program?

I do not know a lot about sound processing but I used to play with Sonic Foundry SoundForge (before it was bought by Sony) and my brother is a music producer (record, master, etc).

From what I have seen, the people that use those things professionaly base their work on a specific tool (protools, for example) and then use the available plugins to modify a specific section of their projects.

Moreover, when you offer it as a plugin for one of these more "professional" (read expensive) software, you are aiming at a different userbase (market).

Finally, although the idea of your program seems interesting (I have to admit I only had a quick look at your website), you should think about offering something /useful/. Right now, the product seems like a gimmick. Do not take it the wrong way, it seems like an interesting idea and the software seems fine, but if you really want to profit, it should have a purpose (maybe mixing it with other instruments [a channel in Protools or other prof. software]).

It is in some way similar to Phtoshop. You can find have several free programs that do image processing, but graphic designers happily buy Photoshop plugins that achieve the same results, just because such plugins follow their workflow.

Re:Your Goal: One Second or Less

[4D6963]

Yeah, I know what you mean, that's why I spend much of my time working on examples to try and demonstrate what it can actually be used for, and that it's not just a gimmicky outlandish idea. It's not easy, but on the other hand, at the rate of one sale for every 300 demo downloads, maybe I'm not doing that bad, I guess I need more volume (i.e. more people to visit) but I'm not quite sure how to achieve that either..

Re:Your Goal: One Second or Less

[Kalriath]

Uh, you can't disable the single pass memory test, only the three pass test which takes a couple of minutes on its own. Although you may be onto something with bells and whistles - it seems that the cost of a motherboard is directly proportional to the size of its feature set, and mine wasn't cheap. XFX (which is rebranded MSI) if it helps any.

Re:Your Goal: One Second or Less

[bucky0]

It'd be nice, but a lot of legacy hardware's probing goes something like:

1) Enter protected mode on the CPU
2) Put some data at a specific memory location
3) Set up an interrupt handler on a specific interrupt
4) Throw an interrupt (ping)
5) Wait for the hardware to return with another interrupt handler (pong)

Unfortunately, since it's an asynchronous system, you have to have a timeout to wait for the hardware to do it's thing. Double-unfortunately, a lot of hardware uses the same memory locations/interrupts, so you can't paralellize the process.

Re:Your Goal: One Second or Less

[Just Some Guy]

Double-unfortunately, a lot of hardware uses the same memory locations/interrupts, so you can't paralellize the process.

Ah, I see how that'd make a difference. Thanks!

Re:Your Goal: One Second or Less

[Peepsalot]

The PCI bus is based on a 33MHz clock

Re:Your Goal: One Second or Less

[Peepsalot]

If you RTFA, you can see they show the difference in boot time between the exact same build of 9.04, with only the filesystem differing.

Re:Your Goal: One Second or Less

[Peepsalot]

You know how everyone wanted a Linux-based operating system that "just worked" on a wide variety of hardware with drivers for everything? And didn't throw a shit-fit if you moved the hard disk to a completely different machine and tried to boot it up? That's why Linux takes so long to boot these days. You can have very good hardware compatibility or you can have very good boot speed. You can't have both. (Well, until someone invents persistent RAM.)

BIOS configuration is peristent...

What if a BIOS existed that could learn your hardware configuration, and optimize itself for that configuration. The first time you boot, it would try to detect every possible piece of hardware under the sun to find out what you have. Once it figured that out, it could possibly write to itself to remember those settings. Every subsequent boot would skip all the nonsense and only initialize the hardware you need. If you later need to change your hardware configuration, the BIOS could have an option to run super detect mode again by pressing a key during boot.

I'm not a BIOS programmer, but it at least seems to me that something like this could be technically feasible with today's technology.

Re:Your Goal: One Second or Less

[Jamie Lokier]

That is _such_ a poor name for the option.

I often wondered what it did; thanks for clarifying.

(I won't be using the option, as I regularly have to kill and restart NetworkManager (the daemon) anyway, since it doesn't manage bluetooth 3g internet, which I use half the time. But it's nice to know the option exists.)

Re:Your Goal: One Second or Less

[swillden]

If you RTFA

What's that?

Re:Your Goal: One Second or Less

[quickOnTheUptake]

I don't think the original poster was talking about fully loading. He was talking about replying. And your description of the process of loading a page is awfully inadequate in key ways, especially for a sysadmin at a web hosting company: The server doesn't send all the content after the first request. It sends a page which then causes the client to send a series of new requests (for images, stylesheets, linked-in JS etc). The original post was saying that if the first page isn't returned in 10 sec. it isn't coming.

Oh YEAH?

[Xtense]

Well my TABLE LAMP boots in 50ms! Beat THAT!

(And to all you electrotech-people, yes, i live in Europe, 50Hz here. You may laugh now.)

Re:Oh YEAH?

[Ungrounded Lightning]

Well my TABLE LAMP boots in 50ms! Beat THAT!

Not using those new-fangled compact fluorescents, are you sonny?

Re:Oh YEAH?

[Xtense]

I didn't mean the "really rich and posh part of Europe" though... :)

Re:Oh YEAH?

[melikamp]

Does it run Linux?

Re:Oh YEAH?

[vamidus]

Your table lamp is kinda slow. But does it run Linux?

Re:Oh YEAH?

[natebarney]

Maybe my math is wrong, but wouldn't that be 1 / (50 Hz) = 20ms?

Re:Oh YEAH?

[cwesley]

1S / 50 = 20mS

Re:Oh YEAH?

[Xtense]

Well there you go, ruin all my fun with your silly math thingamabob. Thanks a lot!

Oh, and to whoever modded OP as insightful:
God damn it man, you need to leave your house once in a while.

Re:Oh YEAH?

[CannonballHead]

In a completely off-topic post...

Staring The United States as Weimar Germany.

Why is the US staring at Weimar?

Re:Oh YEAH?

[Cthefuture]

I have to admit, posting from your table lamp is quite impressive. What browser does it run?

Re:Oh YEAH?

[Xtense]

OmniWeb, obviously!

Welp, that about makes it for today's bad joke limit. I retreat to my underground sanctuary...

Re:Oh YEAH?

[Anonymous Coward]

Lighthouse 1.8

I'd like to see Navigator not crash on the rocks without Lighthouse. The Safari would never happen. The Explorer would stay at home.

Re:Oh YEAH?

[Tubal-Cain]

Does Weimar have oil?

Re:Oh YEAH?

[CannonballHead]

Maybe Weimar has lots of schnitzel?

Re:Oh YEAH?

[Idiomatick]

Hes referring to the weimar republic. You know when germany went through hyperinflation and had to use wheelbarrows to carry around money, i'm sure you've seen pictures.

Re:Oh YEAH?

[evanbd]

It does not reach full brightness in one cycle. 50 ms will get it nearly there, perhaps 100ms or a little more to get all the way. I'm afraid I'm away from my scope right now...

Re:Oh YEAH?

[Linker3000]

Wow, imagine a chandelier of those!

Re:Oh YEAH?

[mr good]

he was using the metric system

Re:Oh YEAH?

[Rick Bentley]

50Hz, even if you had a wait a full cycle, is 20ms NOT 50ms. Assuming the filament is considered lit when the voltage first peaks then you only have to wait a maximum of a half of a cycle.

So really your lamp is even faster than you said, it has a FSL (Front Side Lamp) Speed of 100Hz and takes only 10ms to light!

Re:Oh YEAH?

[DMUTPeregrine]

Yes, but the filament has to heat up. At least if it's an incandescent. Thus, it probably takes even longer than 50ms.

Re:Oh YEAH?

[zippthorne]

If it only takes 1 cycle to reach full output, It's going to have some awful headache inducing flicker.

Re:Oh YEAH?

[mrslacker]

About 300ms - at least according to Mythbusters. Fluorescent can take as long as 30 seconds.

Re:Oh YEAH?

[pjt33]

I've read that it's the only time in human history that grandmothers were legal tender.

Re:Oh YEAH?

[MichaelSmith]

It does not reach full brightness in one cycle. 50 ms will get it nearly there, perhaps 100ms or a little more to get all the way. I'm afraid I'm away from my scope right now...

Thats why LEDs running on DC are so much better.

Re:Oh YEAH?

[ChunderDownunder]

But aren't most dwellings running on AC? For a 'table lamp', isn't there an overhead associated with AC/DC conversion?

Re:Oh YEAH?

[Jugalator]

Right, but it ain't over 'til the fat lady sings the Opera.

Re:Oh YEAH?

[MarkKB]

A Beowulf chandelier?

Re:Oh YEAH?

[drachenstern]

God damn it man, you need to leave your house once in a while.

You saw the South Carolina article too did you?

Re:Oh YEAH?

[Jamie Lokier]

Over here, fast booting light bulbs are now only available on the black market!

I booted up ubuntu just to post

[colin_n]

Booted into Ubuntu 9.04 just to say "first post". Let's just say the ubuntu folks still have some work to do.

Backwards Compatible?

[Thelasko]

Is EXT4 backwards compatible with EXT2 and EXT3? (3 is backwards compatible with 2) I'm asking because there are only Windows drivers for EXT2, and this could cause problems for those that dual boot.

Re:Backwards Compatible?

[linuxkrn]

For most users, no it will not work. One of the major features of ext4 is extents, which basically reserves space for a file to continue writing at a later date. This will decrease file fragmentation and improve performance.

If however, you disable extents, then yes you can mount it as ext3. And as you know, ext3 can be mounted as ext2 without the journaling.

I agree that the win32 ext2 drivers need updating. I would hate to lose access to ext partitions for dual boot systems.

Re:Backwards Compatible?

[compro01]

I think ext4 will mount with ext2fsd, if and only if you don't use extents. With extents on, there is no backwards compatibility to ext2 or 3.

Re:Backwards Compatible?

[hunteke]

Is EXT4 backwards compatible with EXT2 and EXT3?

It's partially backwards compatible. You can mount it as an ext3 filesystem up to the point you use extents.

Ext4 Features: Compatibility

Re:Backwards Compatible?

[timeOday]

"One of the major features of ext4 is extents, which basically reserves space for a file to continue writing at a later date. This will decrease file fragmentation and improve performance"... for the next couple years until SSD takes over, that is. I recently put the Intel SSD in my latop, particularly to run VMs from (since they add another layer of fragmentation), and let me tell you, I have seen the future. Finally I can run my filesystem at 99% full (which I often end up doing anyways) without a huge performance hit. At this point, tweaking filesystems to accommodate not-really-random-access media seems like backwards thinking.

Re:Backwards Compatible?

[Spikeles]

I'm asking because there are only Windows drivers for EXT2

If you feel like doing a little manual work, you could use coLinux to load those obscure file systems like Ext3, ReiserFS and XFS.

Re:Backwards Compatible?

[DragonWriter]

At this point, tweaking filesystems to accommodate not-really-random-access media seems like backwards thinking.

Over the next couple of years, SSDs performance benefits : price premium ratio may increase to the point where they are usually the primary and often the only drive on new desktops and laptop systems, but Linux is more than an operating system for the the newest desktop/laptop hardware. Its also for servers, and older hardware, and...

And, of course, ext4 is hardly the only supported filesystem.

Re:Backwards Compatible?

[Jerry Smith]

Is EXT4 backwards compatible with EXT2 and EXT3? (3 is backwards compatible with 2) I'm asking because there are only Windows drivers for EXT2, and this could cause problems for those that dual boot.

Well, it'll probably be be limited to the ext3-options (like ext3 compared to ext2), but it'll work. But I'd wait for a couple of months before implementing it, you never know.

Re:Backwards Compatible?

[jaavaaguru]

And, of course, ext4 is hardly the only supported filesystem.

Yeah, we have some killer options out there

Re:Backwards Compatible?

[calc]

How fast will that 1% of flash wear out with you leaving the rest occupied with data, so no wear leveling can take place... Flash is only reliable up to a few hundred thousand write cycles. So you can keep your SSD's to yourself.

Re:Backwards Compatible?

[mR.bRiGhTsId3]

-1 Beating a Dead Horse

Re:Backwards Compatible?

[RiotingPacifist]

im sticking with the killerfs unless ext4 can fsck in a reasonable amount of time, ext3 is ridiculous.

Re:Backwards Compatible?

[swilver]

It has always been claimed Ext was so good at keeping files defragmented that there was no need for a defragmentation tool. I call bullshit on that. Ext fragments just like any other filesystem, and it can get very bad. The worst cases are log files and slowly progressing downloads. I usually keep a separate partition just for those purposes since IT WILL fragment.

This bullshit with "reserving" space for files so they can be extended later can only go so far. For files that are just copied (and thus will never be extended), reserving this extra space is unnecessary and will end up fragmenting free space instead (which arguable is very important to keep unfragmented as well). For files that do keep growing slowly, how much do you reserve? 10k? 100k? 1MB? So I download 2 DVD's at the same time, and I end up with 2 files that consist of how many fragments?

At some point, stuff gets fragmented, it's completely unavoidable once the disk is nearing full capacity. Every little bit of space the Filesystem skipped before for "extending files later" will get filled by that last huge file you copied to the drive, splitting it up into thousands of fragments. Even if you clean up the drive to make some more space (as it was full), it is unlikely you'll remove the last few files you copied to it (which are fragmented all over the place). This in turn will make the fragmentation worse and worse.

This happens, even on ext, and any other filesystem that does not actively defragment itself.

more appropriate title would be

[ani23]

boots 3.1 seconds faster with ext4 over ext3

Re:more appropriate title would be

[dirtyhippie]

And the question that I want to know the answer to is *why* would linux boot faster with ext4? Is the filesystem layout really that much more optimized for sequential reads? Or is this just a fluke? In absence of any purported explanation, I have to say I'm guessing fluke, much as I'd like to believe ext4 can make that much of a difference.

Re:more appropriate title would be

[RiotingPacifist]

whats it like compaired to XFS or reiser though?

reformat?

[doti]

Converting an ext2 file system to ext3 takes a simple command, that runs instantly. It basically just add a flag that enable journaling.

Will ext4 be so different that it will not be possible to convert without reformatting?
That's would be a pain for the half-terabytes partitions we have today.

Re:reformat?

[Shikaku]

http://en.wikipedia.org/wiki/Ext4#Forward_compatibility

Mount ext3 as ext4.

Re:reformat?

[wild_berry]

I'll add another URL: http://kernelnewbies.org/Ext4, which summarises the online, live conversion tools available to you today. But, seeing as you're backing up your data anyway, for safety and especially with a new filesystem, you could just rsync it off to the backup disk and then rsync it back onto a reformatted ext4 partition.

Re:reformat?

[doti]

I see.. but to get all the advantages you need a costly rewrite of all files:

"All your existing files will continue using the old indirect mapping to map all the blocks of data. The online defrag tool will be able to migrate each one of those files to a extent format (using a ioctl that tells the filesystem to rewrite the file with the extent format; you can use it safely while you're using the filesystem normally)."

source: http://kernelnewbies.org/Ext4

Why reformat?

[McNihil]

if by reformat you mean fdisk followed by an explicit mkfs.ext4 after....

You should be doing the following:
http://kernelnewbies.org/Ext4#head-3891522e0601162aab24c73c1f148a1e28c6a9d4

I hate to say it but....

[CdBee]

.. that is more than twice as long as Windows 98SE took to boot on my Athlon 1ghz in 2001. 8 years development and we're still ass-whipped by 90s technology. Way to go....

Re:I hate to say it but....

[larry bagina]

ehh, BeOS on my Pentium II 233 was faster.

Re:I hate to say it but....

[UnknowingFool]

You kids and your fancy under a minute boot times. Back in my day, it took all day to boot up. And that's after we had to wait a few hundred thousand years for the sand to become stone so that we could create our stone punchcards.

Re:I hate to say it but....

[makapuf]

try FreeDOS + GEM on your Athlon and discuver the real power of 80's tech power.

Re:I hate to say it but....

[RulerOf]

8 years development and we're still ass-whipped by 90s technology. Way to go....

I know you jest, but this almost amazes me too. On the converse side of it, ever try booting Win98, 95, or 3.1 on modern hardware?

Even as a person who shuts his computer down when he leaves it for more than a couple hours, while I would welcome faster booting machines, I'd prefer faster logons. Bootup in less than 30 seconds, which I've even got with Vista, is fine with me.

Logon, on the other hand, is finished whenever your machine decides to do all those things you told it to do that got queued up while you were waiting for it to become responsive, and is much, much more annoying.

Re:I hate to say it but....

[plisskin]

I had a Tandy 1000 HX with DOS 2.11 in ROM which booted in about 3-4 seconds. And that was in 1988.

Re:I hate to say it but....

[Idiomatick]

I can boot to dos in like 1/100th of a second from windows in my 3Ghz machine......

Re:I hate to say it but....

[El_Oscuro]

My Commodore 64 can boot almost instantly on a 25 year old 1Mhz machine...

Re:I hate to say it but....

[NatasRevol]

Kinda like the Prius & mpg. Still gets whipped by a lot of cars from 1982. Just now getting back to the 50 mpg barrier. 25+ years later.

http://www.mpgomatic.com/2007/10/09/1982-a-banner-year-for-high-mpg-cars/

Re:I hate to say it but....

[tlawson]

My Commodore 64 can boot almost instantly on a 25 year old 1Mhz machine...

PRESS PLAY ON TAPE

Re:I hate to say it but....

[rbanffy]

And that is more than 10 times longer than my 1MHz Apple II did in the early 80s. And it booted from a floppy disk that's orders of magnitude slower than even the hard-disks we had then.

Re:I hate to say it but....

[DrgnDancer]

Yeah, but "Pool of Radiance" takes 5 minutes to load from floppy, and then you have change disks every 15 steps or so.

Comparison times from article

[mgkimsal2]

* Ubuntu 8.10 with EXT3 filesystem boots in 31.8 seconds (on the AMD Sempron system);
* Ubuntu 9.04 Alpha (Build 20090112.1) with EXT3 filesystem boots in 28.3 seconds (on the AMD Sempron system);
* Ubuntu 9.04 Alpha (Build 20090112.1) with EXT4 filesystem boots in 23.1 seconds (on the AMD Sempron system).

* Ubuntu 8.10 with EXT3 filesystem boots in 26.8 seconds (on the Intel Core 2 Duo system);
* Ubuntu 9.04 Alpha (Build 20090112.1) with EXT3 filesystem boots in 24.5 seconds (on the Intel Core 2 Duo system);
* Ubuntu 9.04 Alpha (Build 20090112.1) with EXT4 filesystem boots in 21.4 seconds (on the Intel Core 2 Duo system)!

Re:Comparison times from article

[ruiner13]

So roughly a 20-25% improvement, when you update to Ubuntu 9 and EXT4. Why couldn't the summary say that, instead of giving "21.4 seconds" with no mention of the time being compared to or any mention of hardware. They could have just swapped a 5400RPM drive to a 10K. I know the editors can't be expected to read the articles, but can they at least make the posts somewhat meaningful?

Re:Comparison times from article

[Just Some Guy]

* Ubuntu 9.04 Alpha (Build 20090112.1) with EXT3 filesystem boots in 24.5 seconds (on the Intel Core 2 Duo system);
* Ubuntu 9.04 Alpha (Build 20090112.1) with EXT4 filesystem boots in 21.4 seconds (on the Intel Core 2 Duo system)!

So what about reading EXT3 is so inherently slow that it adds that much overhead? Surely that whole 21.4 seconds wasn't spent waiting on the drive, so let's subtract 10 seconds from each number. That's 15s on EXT3 and 11s on EXT4. Does EXT3 really have a 30% read overhead? I call shenanigans.

Re:Comparison times from article

[NullProg]

* Ubuntu 8.10 with EXT3 filesystem boots in 31.8 seconds (on the AMD Sempron system);
* Ubuntu 9.04 Alpha (Build 20090112.1) with EXT3 filesystem boots in 28.3 seconds (on the AMD Sempron system);
* Ubuntu 9.04 Alpha (Build 20090112.1) with EXT4 filesystem boots in 23.1 seconds (on the AMD Sempron system).

* Ubuntu 8.10 with EXT3 filesystem boots in 26.8 seconds (on the Intel Core 2 Duo system);
* Ubuntu 9.04 Alpha (Build 20090112.1) with EXT3 filesystem boots in 24.5 seconds (on the Intel Core 2 Duo system);
* Ubuntu 9.04 Alpha (Build 20090112.1) with EXT4 filesystem boots in 21.4 seconds (on the Intel Core 2 Duo system)!

I would say this is a testament on how bloated Gnome is getting.

XUbuntu 8.10 on my work Athlon 2400 boots to desktop 20 seconds. On my home Pentium IV system it boots in about 15 (Both ext3).

I bet if the Gnome team were to rethink the design of some of those services they could shave off 10 seconds easily.

My opinion,
Enjoy,

Re:Comparison times from article

[Kanasta]

That's all very fine, but how long does it take from the login screen to a usable PC?
for me in 8.04 it takes about 1.5min

Re:Comparison times from article

[jaavaaguru]

Really? On my Dell Mini 9 netbook, it takes 41 seconds from poweron to login prompt, and 23 seconds from hitting enter there to having a usable desktop (Ubuntu 8.10 netbook remix). Coming out of standby just takes a second or two.

Re:Comparison times from article

[buchanmilne]

Does EXT3 really have a 30% read overhead?

The btrfs wiki has a link to some benchmarks which included ext3 and ext4 (and btrfs, but it's got a way to go still ...), which show that ext4 is substantially faster than ext3 in a number of the workloads.

For this use case, this is probably about the most representative benchmark (but also one of those in which ext4 really shines), in which ext4 is about 75% faster than ext3.

Re:Comparison times from article

[Kanasta]

Coming out of standby is pretty fast, but hibernation is really bad.
I see my apps, but when you click on them, it loads for 5-10s. Then the next app, etc

I thought maybe ti's not ready yet. So next time I had it come out of hibernation 30min b4 I used it. Same thing, the apps are still not awake!

Also: does "shred" work with it?

[Ungrounded Lightning]

Is EXT4 backwards compatible with EXT2 and EXT3?

Also: Does "shred" work with it? (It works with EXT3 but only if journaling is disabled - which seems to defeat the purpose of the filesystem...)

Alternatively: Is there a replacement or upgrade for "shred" that would work with EXT4 and/or other journaling filesystems, or a "shred the free blocks" IOCTL?

Re:This booting thing is overrated.

[snl2587]

You mean the suspend for Linux that Actually Worked that I just woke my computer out of ten minutes ago?

And that's with Ubuntu 8.04.

A huge boost to my workplace productivity

[trouser]

They've shaved 10 seconds off the boot time? In a typical working week that buys me 50 seconds more work time. I'll be so much more productive.

Re:A huge boost to my workplace productivity

[ruiner13]

Yeah, but you have to think long term. After 10 years, you'll save 433 minutes! That's almost a full day's work! If you make 100K/yr, that's almost $400 of salary time wasted! The HORROR!

Re:A huge boost to my workplace productivity

[Idiomatick]

Multiply that by the millions of computer users.......

Re:A huge boost to my workplace productivity

[trouser]

Yeah, I considered that. But my morning ritual is turn on the kettle, turn on the computer, make the coffee, login. Since the kettle already takes longer to boil than the computer takes to boot there is no real benefit to reducing the boot time.

Also, I then waste at least 30 minutes reading Slashdot and drinking the coffee before I do any work. 30 minutes per day, 5 days a week, every working week for the last I don't know how many years. Hmmmmm......

Re:A huge boost to my workplace productivity

[Random BedHead Ed]

They've shaved 10 seconds off the boot time? In a typical working week that buys me 50 seconds more work time. I'll be so much more productive.

Every night I go to bed wishing I had just ten more seconds to make the day productive. Now I can get that extra time, and all I have to do is convert to a new filesystem. (And here I thought I'd have to stop posting on Slashdot or something.)

Re:A huge boost to my workplace productivity

[glitch23]

I'm productive at work by not ever logging out or shutting down the machine unless absolutely necessary. I never shutdown my machine at home either unless absolutely necessary. At home I usually have to about once a month because PowerDVD stops playing DVDs and a reboot is the only way I've found to fix the problem. At work I currently have an uptime of about 60 days. Both work and home machines run WinXP. Why are you shutting your machine down every day? To reopen 5-10 tabs in FF, 2 IE (v6) windows, a couple My Computer windows, Notepad for notes, a couple Word docs, maybe a couple Excel files, Outlook (TBird at home) and currently a PowerPoint presentation, I'd lose more than just 10 seconds a day.

Re:A huge boost to my workplace productivity

[trouser]

The PC uses power, even when its asleep. I pay the bills and I have some concern for the environmental impact of my behaviour. Sure I leave the machine on when I stop for lunch, but I turn it off when I finish work.

Actually, here's a thought....we need to add the 10 seconds back to the boot time and use those 30-40 seconds of boot time everyday riding exercise bikes attached to generators connected to the grid. If enough geeks are willing to pedal 40 seconds a day we could reduce our dependence on fossil fuels without having to buy into any of this 'clean coal' nonsense.

I believe!!!

I'm waiting for the 10.10 version

[Anonymous Coward]

Sure to be named Masturbating Monkey

Aww....

[Tubal-Cain]

Not enough time to backup my /home and install before class starts...

What does ext4 have to do with it?

[Lord Byron II]

My understanding is that ext4 provides some very nice features, but faster data access isn't necessarily one of them. I'd imagine that an ext2 fs, which doesn't have journaling to slow it down, should be even faster.

Re:What does ext4 have to do with it?

[Anonymous Coward]

You understand journaling only applies to writes? Sure, there's some log file writes on startup, but not much. Oh, and ext4 should be better about keeping things defragmented, which could explain the difference.

disappointing...

[sofar]

This is a truly disappointing news item. Instead of setting the bar higher and truly trying to reduce boot time, they have not done much more than shave seconds off the existing boot time.

For a generic desktop distro, 20+ seconds is still terribly long. 10 seconds should realistically be easy to achieve, especially as it took Arjan and me only a few months to get to 5 seconds on a netbook. We sure cut some corners, but we did not even use ext4 on those netbooks, and we still had buggy X starting times of 1.5 seconds, something which we can probably do in 0.5 seconds with kernel modesetting.

I hate to see everyone settle down with "20 seconds" being "the next 5 second boot". This is really not progress at all, but rather, complacency.

Re:disappointing...

[steveha]

This is a truly disappointing news item. Instead of setting the bar higher and truly trying to reduce boot time, they have not done much more than shave seconds off the existing boot time.

I just checked, and it does seem that a fast boot time was one of the goals that Mark Shuttleworth set for Jaunty.

There are some specific goals that we need to meet in Jaunty. One of them is boot time. We want Ubuntu to boot as fast as possible - both in
the standard case, and especially when it is being tailored to a specific device. The Jackalope is known for being so fast that it's
extremely hard to catch, and breeds only when lightning flashes. Let's see if we can make booting or resuming Ubuntu blindingly quick.

https://lists.ubuntu.com/archives/ubuntu-devel-announce/2008-September/000481.html

Given that, I must confess that I'm also a bit disappointed that the boot time isn't closer to five seconds.

I love your work with the 5 second boot, and I look forward to that technology being implemented widely. On a modern super fast CPU with a solid-state hard drive, I should hope that a desktop computer could boot as fast as a netbook. (And I'd be willing to install Coreboot to get that speed.)

steveha

Re:disappointing...

[PitaBred]

How specially tuned to the specific hardware was your netbook boot, though? Would it work on every x86 machine with a BIOS? Or were your cut corners including compiling in all the modules to the kernel that your machine would need?

Don't get me wrong, I admire the 5 second boot time. I'd love to have it. But it needs to be a general-purpose speedup for it to work with a major distro.

Re:disappointing...

[sofar]

In the 5 second demo we had a kernel that supported pretty much all intel chipset based systems. It would be trivial (for a distribution) to make the kernel optimized in exactly the same way for a different chipset and install a specific kernel for each system. There would only be 3 or 4 kernels to distribute to cover most of the generic systems out there.

Re:This booting thing is overrated.

[Walpurgiss]

suspend works for me just fine. FC9 on an acer laptop. Suspends correctly via lid, menu selection, and sleep key. Wakes correctly from any keypress. Hibernate also works fine.

One thing though, the first time I tried it after clean install from disc, coming back from resume the terminal screens were mangled and could only see a corner of them, the rest being off screen. At some point in the last 4-5 months that stopped happening for me.

Great

[kanweg]

So people will work less time with Linux.

Bert

Who cares?

[bigredradio]

I don't mean to troll, but I could care less about boot up times. What I care about is uptime!

With Windows, you are always having to reboot the system due to everything from software installs to changing a network connection.

On Linux, I never have to reboot. Basically my desktop stays on unless I am taking a long weekend. I understand that efficiency is good, however, a fast boot-up does not seem like news to me.

Re:Who cares?

[bigstrat2003]

With Windows, you are always having to reboot the system due to everything from software installs to changing a network connection.

No, you aren't. This hasn't been true since Windows XP, at least. I can get uptimes of months at a time on my Windows box, the only time it comes down is for hardware changes or OS updates.

Re:Who cares?

[Jeremy Visser]

Not everyone gets their electricity for free, or lives in a country where the cost is negligible. Argentina, for instance.

Re:Who cares?

[Quarters]

You haven't had to restart Windows due to a networking configuration change in almost 7.5 years. You haven't had to restart Windows due to a driver change for almost 3.5 years now. Please get your facts correct.

Re:Who cares?

[Aladrin]

"On Linux, I never have to reboot."

Only if you never install updates. I just decided to try Gnome for a while and it's had an icon in the tool tray since the day after I installed it telling me to reboot.

Not that it's really any different than KDE, mind you. It's just funny that I -just- updated everything and installed Gnome and the next day I have to install more updates and reboot again.

Re:Who cares?

[sofar]

server maintainers care, because people pay them a ton of money to get a guaranteed 99.999% (extreme case, like NY stock exchanges etc.) or more uptime. That's only 5(!) minutes of downtime a year, and if you can boot in 5 seconds (and lets say shutdown in 5 as well), you can reboot 30 times a year for security updates. If you reboot in 30+30 seconds, that's only 5 reboots.

imagine having a scsi raid array which takes 1 minute to initialize. a 20+20 boot+shutdown time would give you barely 3 boots per year. A 5+5 boot+shutdown almost gives you 5 reboots in the same time.

you care for netbooks. The batteries are small, if you waste one minute at boot, and a minute at shutdown, at which the cpu and ssd (or worse: hard disk) are working hard, you lose two minutes of battery time, which translates into 5+ minutes idle or browsing the internet time. Reboot your netbook to quickly send a blog update from the airport a few times, and you've lost half an hour or effective work time.

bottom line: shorter boot (and shutdown) means more _net_ work time available, for both a/c connected and mobile devices.

Re:Who cares?

[crazybilly]

You don't have a laptop, huh? Even w/ suspend-to-RAM, you still run the risk of data loss/corruption if you leave your computer on for a long time w/o plugging in, a risk I'm not willing to take while traveling.

Give me fast boot times--I don't care a lick about uptime.

Re:Who cares?

[Archangel Michael]

"I can get uptimes of months" is diametrically opposed to "OS updates"

Patch Tuesday and the various out of band patch releases, predicates that at best you can get is about 1 month between updates, IF you apply patches when they are released.

Now if you don't apply patches when they are released, but wait "months" before apply them, then all bets are off, because at that point reboot cycles becomes arbitrary.

Many hardware installs on windows require "reboot" as well.

Re:Who cares?

[kisielk]

Many software updates still require a reboot on both XP and Vista. While I can usually go a week or two without rebooting, eventually either a Microsoft patch or patch for some other software I have installed forces me to reboot. The worst thing about the Microsoft ones is that they will reboot your computer even if you say no. There's been numerous times where I've woken up in the morning only to see the message "Your system was recently restarted by Windows Update" on my desktop when I log in.

Re:Who cares?

[armanox]

actually, with kexec, you don't even have to reboot to change the running kernel.

Re:Who cares?

[mugurel]

Exactly! Real men (m/f) don't boot! ... But admit, we all have our weak moments.

Re:Who cares?

[Random BedHead Ed]

With Windows, you are always having to reboot the system due to everything from software installs to changing a network connection.

No, you aren't. This hasn't been true since Windows XP, at least. I can get uptimes of months at a time on my Windows box, the only time it comes down is for hardware changes or OS updates.

Windows XP? That sounds like a description of Windows 95/98/ME/NT to me. Decent uptime was possible on XP and 2000, IMHO. (That is, if you didn't reboot to load security patches.)

Re:Who cares?

[Anonymous Coward]

I call that bullshit! I didn't have to reinstall Windows after I installed OpenVPN for example, which creates a "fake" network connection. The thing about the restarts is mostly the fault of the applications, because all software vendors write apps with one thing in mind: "Windows = desktop OS," which means the user can reboot as often as they want. Then we go around bitching that Windows sucks because you have to reboot. Damn it man, that thing about the restarts ended almost a decade ago, when we got Win 2k.
It's ignorants like you that give the OS a bad name, it's rarely the actual software that's built into it. There are server editions of Windows, you know... There's Windows Server 2003 which kicks Linux's ass big time, but nooooooohohohooooo, you're not thinking about using it, because Win 95 sucked, so why bother and try something new when you can just whine about the poor overall quality of MS products by giving examples from what a single product you've tried a long time ago?
You're nothing but a spoiled brat! [yeah, ad hominem, but that doesn't mean I'm not right]

Re:Who cares?

[drsmithy]

On Linux, I never have to reboot. Basically my desktop stays on unless I am taking a long weekend. I understand that efficiency is good, however, a fast boot-up does not seem like news to me.

Note that for most people, "having to reboot" is irrelevant, since they turn their computers off every night anyway.

Further note that for most of those who are left, the practical difference between "reboot" and "restart all your applications and services", is zero.

Re:Who cares?

[Risen888]

Your OS has to reboot to run updates! LULZ!

Re:Who cares?

[Anonymous Coward]

How did this get modded insightful?

a) If you garantee five nines than there is no way around redundancy, so you can just reboot a few machines at a time and let the load balancer take care of the rest

b) the problem isn't so much boot time as it is the software. your server may boot in a few secconds or in a few minutes, getting e.g. a warm cache will probably take hours anyway

Re:Who cares?

[Yfrwlf]

I'd like to see a kernel that could do things like coping with new drivers without a reboot, and even find a way to "move things over" to a new kernel before shutting down the old one instead of having to reboot for that. I'm sure it's possible somehow, to be able to "hot swap" everything running on top of the old kernel to the new one, somehow. The expectations for creativity need to be raised higher. ^^

Re:Who cares?

[PitaBred]

If you're guaranteeing 5 nines, you'd be stupid to be using a single machine. Update a test environment, verify it works, then take down your cluster a machine at a time updating each one. No downtime if you do it right, that way you can "bank" your downtime to deal with network outages and such that are outside your control.

Re:Who cares?

[bigstrat2003]

What's funny is that you consider uptimes of months to be significant. I know Linux geeks that will laugh at that. Apparently they don't consider anything measured in units smaller than years to be significant.

It's not funny at all. For a home machine, any uptime longer than a couple weeks is dick waving. For servers, it's a different story, but I'm not running a server. My home machine comes down sometimes once a month just to swap IDE drives, and we want to talk about uptimes of years? Not important.

Re:Who cares?

[Sinbios]

That only happens when you choose to install updates automatically. I'm not sure where you said no in this process. Silence is consent! :P

Re:Who cares?

[kisielk]

No, I have automatic update install disabled. It seems there are still some really high priority upgrades that get forced through regardless.

Re:Who cares?

[ion.simon.c]

I'd like to see a kernel that could do things like coping with new drivers without a reboot,

Would you clarify this statement? 'Cause, ATM I think that rmmod OLD_DRIVER && modprobe NEW_DRIVER generally inserts a new driver into the kernel w/out a reboot.

Re:Who cares?

[KasperMeerts]

I'd like to see a kernel that could do things like coping with new drivers without a reboot, and even find a way to "move things over" to a new kernel before shutting down the old one instead of having to reboot for that. I'm sure it's possible somehow, to be able to "hot swap" everything running on top of the old kernel to the new one, somehow. The expectations for creativity need to be raised higher. ^^

Ever heard of kexec?

Re:Who cares?

[sproot]

Places achieving 4 or 5 nines are using clusters or at least redundant servers.
They don't care about shaving 30secs of the boot time.

Having said that, I loved the work on the netbook. Would be great to see it in the mainstream distros :)

Re:Who cares?

[Hucko]

And BSD measures in decades...

Re:Who cares?

[Hucko]

Then why the hell did my wife's constantly updated computer request a reboot for a driver update just recently? (it may not have been accepted, but it did request it.)

Re:Who cares?

[toddestan]

I've never seen this myself. What will happen is that if you install updates, and Windows asks you to reboot, and you push 'no', Windows will eventually force the issue. But otherwise, I set my computers to download but let me choose when to install them, and so long as I ignore that little yellow shield I can go weeks/months without rebooting.

ON WHAT HARDWARE?

[amn108]

ON WHAT HARDWARE? That number is meaningless unless we know the specifications.

Re:ON WHAT HARDWARE?

[amn108]

Nevermind, the article explains everything in detail. I was trigger happy there for a moment.

Re:ON WHAT HARDWARE?

[Idiomatick]

You could RTFA.... or comments further up the page....

Re:This booting thing is overrated.

[bigredradio]

You sir get a star. http://haacked.com/images/haacked_com/WindowsLiveWriter/IConfigMapPathIsInaccessibleDueToItsProt_1446B/works-on-my-machine-starburst.png

What exactly is the definition of boot?

[VinylRecords]

What exactly is the definition of boot?

When I start up my IBM ThinkPad (1.5ghz single processor, 512RAM, garbage video card) running Windows XP, it takes roughly 10-15 seconds to get to the user log-in interface from the moment the power button is pressed.

But, once you log in, you are talking two to three minutes where background applications and processes are opening, explorer is loading, and applications that launch at start are loading.

After you log in does that time count as boot time? Considering it takes me 10-15 seconds to get to the sign in screen, not that much time, but after logging in it takes well over two minutes for me to be able to actually run anything at normal capacity.

Re:What exactly is the definition of boot?

[bored]

Thats about the same amount of time it takes my machine, but I can type my password in ~1 second and a second later be at the desktop. Ie generally will start in another second or two. This is because I have disabled most of the crap that puts itself into the system tray. Everything continues to work, except when I want to change my desktop resolution I actually have to right click on the desktop instead of the system tray... anyway most of that crap that rattles your disk after login is probably useless and you won't notice if you disable it. Bottom line, my XP machine goes from power on to useful work in ~15 seconds, but it generally is at S3. From S3 to useful work is really more like ~2 seconds while the monitors come back on and the HD spins back up. The machine consumes ~6W in S3, but all the USB, networking, monitors etc running in standby bump it to ~13W. With everything running but idle its ~210W (monitors included).

ext4

[digitalhermit]

However, the good news is that the EXT4 filesystem was implemented in the upcoming Ubuntu 9.04 Alpha 3 a couple of days ago and it will be available in the Ubuntu Installer, if you choose manual partitioning.' I guess it's finally time to reformat my /home partition..."

From what I understand, there's no need to reformat. Similar to how EXT3 was layered on top of EXT2, EXT4 should just be another mount option as long as the kernel supports it.

I have a couple EXT4 partitions I'm testing... It's been rock-solid so far...

Re:ext4

[phantomlord]

I migrated some of my non-critical partitions over to EXT4 and hit a race condition that corrupted my filesystem and resulted in data loss (the bug has since been fixed). I'm waiting a little longer before converting my important partitions over.

Switching from EXT3 to EXT4 is as simple as a flag change and a remount. HOWEVER, your existing data will still be laid out without extents and thus you'll miss you on a lot of the improvements in EXT4. Eventually, an online defragmenter will be written to defrag your drive while mounted and convert the old data to use extents, but Ted T'so says there are problems with the existing implementations and a working one isn't on the near horizon. Your best bet would be to move the data to another partition and back to convert it to using extents in the meantime.

Also note that if you're using extents, you can no longer mount the partition as EXT3.

Re:ext4

[VValdo]

Don't forget that ext4 wants your inode size to be 256 not 128, so you may want to increase the size. This can be done with the tune2fs -I command, but I hear it takes like 20 hours and I don't know how reliable it is (anyone?)

W

Re:ext4

[phantomlord]

I converted a ~40GB partition and it took a while (I didn't time it... between 1 and 2 hours).

Per Ted Ts'o The fix is checked into e2fsprogs 1.41.4 but release is waiting for a couple other high priority bugfixes to be released. If you want it now, you can pull it from his git repo

Re:ext4

[VValdo]

Hey thanks!

W

Ask the right question . . .

[bogidu]

and you may get a relevant answer.

The question is what do you want the computer (or appliance) to be able to do after that four second boot? All the supporting software to do all the whizbang stuff (games, graphics, etc) takes time to load.

I decided awhile back that I simply wanted a device that I could power up quickly, surf the web, check email, oh, and had to be portable. Just last month I got such a device. It powers on instantly, finds and connects to wi-fi rapidly, the browser opens quickly, etc. It does exactly what I needed it to do. It's an IPOD Touch. First Apple product I've ever owned, so I'm not exactly an Apple advocate, but it does the bloody job.

So, back to the OP, I'll ask the same question that I ask every person who comes up and asks me what kind of computer they should get. . . . . what do you want it to do?

Re:Ask the right question . . .

[Kalriath]

A minute is QUICKLY? The iPod Touch may well be the slowest booting device I've ever seen! I end up staring at that Apple logo for longer than my phone makes me stare at the Windows Mobile logo!

Let me get this of my chest...

[jernejk]

I've been using Ubuntu for I year. I was quite happy with the 8.4, but unfortunately I've switched to 8.10 64bit (to support 4GB RAM). You know what? I couldn't care less about how fast it boots. I do, however, care about these things:
- switching from dual display to presentation (clone) and back totally messes up x config, I have to uninstall and reinstall nvidea drivers
- in dual screen mode, nautilus opens on the first display. I have to open terminal and run nautilus& to lunch it on the second display
- in dual screen mode, keyboard keeps focus in the previous screen. I have to minimize/maximize a windows on the "new" screen to move keyboard focus
- RDP client crashes X windows in some cases (it does not close the drop down list of used servers... and bang)
- oh and NO it's not AN ERROR if I close the RDP window. If I want to reconnect, I will, don't hide under my active windows and bring RDP windows back in 30 seconds. That's just plain stupid.
- java and window decorations don't play well together (popups without buttons etc.)
- How about opening a connection to a new server in a new tab, not in a new nautilus window?
- Flash stops working. I just see a gray square where flash is supposed to be.
- Firefox is not very stable.
- Windows would become gray and unresponsive when there's a lot of disk activity.
- I've seen ubuntu crash on my much more times than I've seen BSOD on the same HW.
- If i lock my computer, I want it to be locked. I don't want it to be locked for a minute or so and than display what was last on my desktop. Sure, you'd have to log in to get access, but there could be things for my eyes only on that screen. So don't you ever roll eyes on windows security, ok? You've got your own issues.

I could probably think of more but this is just a list of things I remember from the top of my head. Sure, you'll be downmoding me and say I'm trolling. Maybe I am. But my point is: there are MUCH more IMPORTANT things to fix than the FUCKING BOOT TIME. Who the fuck even cares about boot time?? Can't you just grab a coffee while it boots? What kind of idiotic metric is this?

I guess SW development is hard and complex. And we've reached a point where maintaining these beasts is hard, for either open source or commercial products.

Re:Let me get this of my chest...

[Shikaku]

Almost all of those issues are from third party software.

Nvidia is notorious for awful drivers, especially for dual display. The screensaver issue is also probably from bad Nvidia drivers.

Adobe Flash is unstable in Firefox, especially on 64 bit systems. Open Source alternatives are also very outdated and slow.

Third party plugins like Java and Flash can make Firefox have to wait. The Mozilla team needs to design much less dependency on plugins and more of a sandbox model, so that a crash or hang of a plugin will not freeze all of Firefox.

Programs become unresponsive due to a lot of disk activity for reasons of speed; DiskIO has more priority. This is a GUI design problem and it should be decidable/configurable easily on or after install.

Re:Let me get this of my chest...

[bloodninja]

Fire up the system without the proprietary Nvidia blob and file bugs! Ubuntu, like all FOSS software, needs _you_ to file the bugs so that the kinks can be worked out. Do not assume that what you see is what the devs see.

You like Ubuntu and FOSS? Great! Help make it better.

Re:Let me get this of my chest...

[Anonymous Coward]

Who are you defending? Blaming other people/projects doesn't make linux any better.

Re:Let me get this of my chest...

[pavon]

Almost all of those issues are from third party software.

And it is Canonical's job to test that software and choose which version they are going to ship with. The last release of Ubuntu, all sorts of software broke on my computer that used to work before. This is their fault for choosing to package bad software.

Also, for what it's worth, I've been having the same problem that he is having with Flash when using Gnash and swfdec as well. It seems like ndiswrapper has some issues in the latest Ubuntu that were not a problem in previous releases, beyond the fact that the flash plugin sucks.

Re:Let me get this of my chest...

[freddy_dreddy]

This is exactly why I left the Linux scene and consider it a failure.
Filing bug reports isn't answered with a solution or bug fix, but with one of these:
- bug report already exists
- You're doing something wrong, it's not Ubuntu/Linux
- It's your hardware, not Ubuntu/Linux
- It's because of these evil hardware companies, not Ubuntu/Linux
- You have the source code, fix it yourself

Re:Let me get this of my chest...

[theCoder]

Your comment about a locked screen showing the contents of the desktop is probably related to a screensaver taking a screen capture and using it in the screen saver. This can be fixed by running xscreensaver-demo (I think) program to get the configuration for the screen savers. You may have to install the xscreensaver package to get this program. When it comes up, it will warn you that xscreensaver isn't running -- that's OK, just ignore it. You can still set the options for the various screen savers.

This is necessary because about a year ago (or so) Gnome decided to replace the standard xscreensaver with their own screen saver system. In good Gnome fashion, there are almost no actual configuration options. However, it still actually uses the xscreensaver programs, which have a lot of configuration options, including where to get image sources (that's on one of the tabs). Tell it not to use a screen capture. Personally, I have it use images from my camera directory, so I'll see pictures I've taken in the various screen savers.

Considering that xscreensaver has had this options dialog for a long time, I don't know why Gnome couldn't either use it, or make one like it. But I guess that would expose useful options to users.

You're right that Ubuntu has been fairly bug filled. My personal annoyance with Ubuntu 8.10 comes with Gnome 2.24: they released a version of Gnome with a half re-implemented session management. They ripped out the mostly working implementation and replaced it with a version that cannot save or restore session state! Now, I'm all for replacing poor implementations, which as I've read the previous session management code was. However, it's really bad to take out mostly working code and replace it with a "** (gnome-session-properties:4618): DEBUG: Session saving is not implemented yet!" printout.

Re:Let me get this of my chest...

[Wrath0fb0b]

Nvidia is notorious for awful drivers, especially for dual display. The screensaver issue is also probably from bad Nvidia drivers.

I don't know what you are talking about, but Nvidia makes the only usable Linux drivers on the planet. Go to any linux computer that's not using the Nvidia drivers and run glxinfo (http://www.xfree86.org/4.4.0/glxinfo.1.html) and tell me if it supports framebuffer, pbuffers, GLSL or redirected direct rendering. In fact, if you press the MESA people hard enough, they will disclose that their full openGL compatibility comes with a software render engine.

At the bottom of all this, is the fact that the nvidia driver has a real unified memory manager -- something that I've tried to get with MESA on an Intel/ATI card at great length (I needed fast openGL for some projects). You can do it, but you have to get crazy experimental drivers and patch your kernel in a few dozen places. Oh, and ACPI doesn't work.

TL;DR version: nvidia is "notorious" for having the only working linux drivers in production status, nothing more.

Re:Let me get this of my chest...

[jaavaaguru]

You get the same responses filing bug reports for Windows :-)

Re:Let me get this of my chest...

[freddy_dreddy]

The problems I encounter in Windows are a few orders of magnitude smaller than those in Linux. I think it's been since the early 90's that I had problems with sound, video or internet on MS products.
The reason is fairly simple: no matter how complex these peripherals have become, they are ubiquitous: we're 2009 ... an OS failing to provide basic multimedia ? Big Failure.

I see the humour in your reply, but to be honest: Ubuntu makes me feel ashamed of being in the S/W business.

Oh, one more thing I forgot to mention/troll about: MS doesn't kill my hard drives as Ubuntu did. About a year ago there was this small peculiar bug which made HDs sleep and wake up constantly. It's not something you notice until you read through the bug reports, but it killed one of my drives and I had to throw away the other one. The store paid it back, after I changed my story from "Linux" to "MS", btw.

Re:Let me get this of my chest...

[sciurus0]

How many of these problem are solved by going to System->Preferences->Appearance->Visual Effects and choosing None?

Re:Let me get this of my chest...

[sciurus0]

Sure, there haven't been any video problem on Windows recently.

Re:Let me get this of my chest...

[freddy_dreddy]

Read the article before posting. You're really scraping the bottom of the barrel on this one.

In Windows, there are cases of things going wrong.
In Linux, there are cases of things going right.

Re:Let me get this of my chest...

[tqft]

"You like Ubuntu and FOSS? Great! Help make it better."

I do - but it doesn't help when stuff regresses - Ubuntu broke - and hasn't yet fixed - some uvcvideo/webcam and bluetooth in the change from 8.04 to 8.10. Ubuntu changed the software (kernel patches, uvcvideo stuff and bits of bluez) for some reason and some stuff has broken.

Bugs are filed, devs questions answered, fix yet to appear.

I am yet to diagnose the intermittment problem with nuvexport, but I think it might be related to HD recordings and the version of ffmpeg vs nuvexport.

Re:Let me get this of my chest...

[Kjella]

Bitching to people about things they can't fix never helped. You just didn't get the support line filter you usually get calling a help desk.

Noone but nVidia can fix their blob.
Noone but Adobe can fix flash.
Unsupported hardware means just that - will NOT work with Linux.
Fixing crashes in reverse engineered drivers is searching in the dark.

I'm sorry your experience hasn't been better, but I've found people to be helpful about things they can fix:
- Supported open source drivers
- Desktop environments
- Open source applications

Most of the complaints I hear are about things that "should not" work. Except someone has made a halfassed attempt and it somehow, somewhat runs unstable on that combination anyway. There should be a big honking "This is for Linux hackers, not your typical desktop user" every time you did that and there's not. Too bad, because Linux is quite pleasant when you use hardware and software that works well with Linux. The problem is mostly "newbies" with configurations even I wouldn't touch that get dragged into compiling dark magic rather than pick up some easy, stable hardware.

Re:Let me get this of my chest...

[Vexorian]

And it is Canonical's job to test that software and choose which version they are going to ship with

I doubt ubuntu ever shipped with flash or nvidia's graphics stuff.

Re:Let me get this of my chest...

[Vexorian]

I've switched to 8.10 64bit (to support 4GB RAM).

Never, never , never do 64 bits Linux, at least not in the next few 10 years... Some tips:

• Remove 2GB, you didn't need those 4 GB to run Linux, ya really.
• Install windows or vista or whatever that actually runs fine in 64 bits.

I mean really, I am the Linux zealot but I don't think any human being should ever dare to use 64 bits on Linux.

Re:Let me get this of my chest...

[freddy_dreddy]

Hey,

Currently running Vista to work on. Next to that I still use BeOS for multimedia, but that's mainly because it runs a homemade mediaplayer. I also have a laptop with Minix due to my 16-something years of experience in assembly.
The other box with RedHat and Fedora was thrown from the 2nd floor straight in a container.

You didn't miss a thing. Most of us are tech fiddlers, or have grown to become one. We have fun with this stuff. But Ubuntu is aimed at the mainstream and is not what it's portrayed to be.

Get Intel hardware.

[Jeremy Visser]

Not intending to troll, but if suspend doesn't work for you, it's your fault for buying hardware that doesn't support ACPI in a standard enough fashion to make it work.

Go Intel. If you get an Intel-based chipset with Intel graphics, suspend works beautifully. NVIDIA nForce chipsets also work, but are slightly less reliable.

Try and not blame Linux for suspend not working. Suspend already works. If it doesn't work, your motherboard manufacturer most likely hasn't bothered testing with Linux, and finding and squashing ACPI bugs.

Re:why the obession with boot time?

[Shados]

Especially if you use Sleep, or to a lesser extent, Hibernate... a pseudo-boot time of 2.5 seconds is much nicer than 21 :)

Re:This booting thing is overrated.

[hattig]

I got it to work in Ubuntu 8.10 on a HP 2133 netbook which has a crappy VIA chipset!

Sometimes Networking comes back up as disabled, and I need to re-enable it, and wait for it to connect to the wireless again.

Re:Also: does "shred" work with it?

[Jeffrey Baker]

All you need to do is write gibberish to a file until the device is full then unlink the file, and do that repeatedly until your paranoia is satisfied.

Comment removed

[account_deleted]

Comment removed based on user account deletion

Re:Also: does "shred" work with it?

[elgaard]

No, but at least the people make "wipe" are paranoid too:

From the wipe man page
==
NOTE ABOUT JOURNALING FILESYSTEMS AND SOME RECOMMENDATIONS (JUNE 2004)
              Journaling filesystems (such as Ext3 or ReiserFS) are now being used by default by most Linux distributions. No secure deletion program that
              does filesystem-level calls can sanitize files on such filesystems, because sensitive data and metadata can be written to the journal, which can-
              not be readily accessed. Per-file secure deletion is better implemented in the operating system.

              Encrypting a whole partition with cryptoloop, for example, does not help very much either, since there is a single key for all the partition.

              Therefore wipe is best used to sanitize a harddisk before giving it to untrusted parties (i.e. sending your laptop for repair, or selling your
              disk). Wiping size issues have been hopefully fixed (I apologize for the long delay).

              Be aware that harddisks are quite intelligent beasts those days. They transparently remap defective blocks. This means that the disk can keep
              an albeit corrupted (maybe slightly) but inaccessible and unerasable copy of some of your data. Modern disks are said to have about 100% trans-
              parent remapping capacity. You can have a look at recent discussions on Slashdot.

              I hereby speculate that harddisks can use the spare remapping area to secretly make copies of your data. Rising totalitarianism makes this
              almost a certitude. It is quite straightforward to implement some simple filtering schemes that would copy potentially interesting data. Bet-
              ter, a harddisk can probably detect that a given file is being wiped, and silently make a copy of it, while wiping the original as instructed.

              Recovering such data is probably easily done with secret IDE/SCSI commands. My guess is that there are agreements between harddisk manufacturers
              and government agencies. Well-funded mafia hackers should then be able to find those secret commands too.

              Don't trust your harddisk. Encrypt all your data.

              Of course this shifts the trust to the computing system, the CPU, and so on. I guess there are also "traps" in the CPU and, in fact, in every
              sufficiently advanced mass-marketed chip. Wealthy nations can find those. Therefore these are mainly used for criminal investigation and "con-
              trol of public dissent".

              People should better think of their computing devices as facilities lended by the DHS.
==

DS Linux

[goombah99]

On my 300 Mhz 192mb Pentium 2 fuji lifebook Damn small linux boots on 25 seconds. It hardly makes a diffenence if I use the CD or install it on the HD.

Interestingly, on a modern computer it takes almost the same amount of time.

So I'm guessing most of that time has to do with a combination of device initializations and timeouts, and disk I/O

ubuntu takes about 15 minutes to boot off CD on the computer.

So one reason you don't have a 1 second booting computer is that if you ever did get the boot under 25 seconds all that is going to happen is that since that is plenty fast, people will lard the thing back up with new features to bring it back to 25 seconds.

One thing I always enjoy about boot ups is that it sort of reminds me of the way a developing fetus goes through all these steps that resemble the vestigial characteristics of more primitive animals. On my mac you see the hardware checks then the boot loader then the services come on, and the screen change to one color of blue, then a slightly different hue, then it gets graphics. etc.... It's like recapitulating it's development in a way that is alive at every point along the way.

If you want something faster I think you need to discard the layered boot. Have a direct boot to the final state instead of bootstrapping one layer from the next.

How does an iphone pull it off? Afterall it is running OSX and on a slow PCU to boot (pun).

Re:DS Linux

[geekboy642]

The iPhone emphatically does NOT pull off a quick startup. I've just timed mine, it took 43 seconds to go from pushing the power button to having the springboard appear. It does, however, sleep and wake as quickly as any other mac computer.

Could care less?

[Finallyjoined!!!]

Idiot.

The expression is "I Couldn't care less" i. bloody e. You are expressing utter contempt, there is nothing below whatever you are professing not to care about.

"I could care less": Yes this is almost, maybe 30%, possibly 73.2%, up my list of hates, apart from purple feathers though, but then I'm a bit airy-fairy anyway.

PS

With Windows, you are always having to reboot the system due to everything from software installs to changing a network connection.

Only if you don't know what you are doing, I'm actually posting from a winXP machine, which has been up since April 2004, because updates are turned off, all paths to Microsoft are blocked by my firewall, I don't install stuff I don't need & use Opera & Eudora.

Re:Could care less?

[Maestro485]

How can you call someone an idiot while simultaneously admitting to not updating your machine for 5 years?

Re:Could care less?

[Risen888]

I'm actually posting from a winXP machine, which has been up since April 2004

Screenshot or it didn't happen.

Seriously. If that's true, I'm hella-impressed.

Re:Could care less?

[Hucko]

He would show the screenshot, but he has to install the software and that will require a reboot...

preallocation on ext3

[heroine]

You can preallocate on ext3 by doing

dd if=/dev/zero of=$OUT_PATH bs=1024 seek=$SIZE count=1

No funky system call required.

Re:preallocation on ext3

[XanC]

...But that requires you to wait for all those zeroes to be written. EXT4 can preallocate without actually doing the write.

Re:preallocation on ext3

[Craig Davison]

That doesn't allocate any space at all. It just creates a sparse file.

Besides, even if you did write all the bytes to disk, it still wouldn't be the same as an extent, because extents are contiguous (not fragmented).

Re:preallocation on ext3

[Craig Davison]

Oops, I mean it only allocates/take up enough disk space to hold the single byte you've written.

Re:why the obession with boot time?

[heroine]

Probably because sleep modes don't work on almost all laptops.

Re:This booting thing is overrated.

[IMightB]

Actually the full hibernate/suspend/resume cycle just started working correctly for me for the first time ever under linux.

This is on a Dell Inspiron 9400 with F10.

Errm OK well it started working after a few weeks of having F10 installed and I did a yum update.

Booting is so last century, it's all about suspend

[atmurray]

In daily use I use mac, linux and windows. 99% of the time these are in some form of suspend mode. Booting your computer is virtually a thing of the past for me, I couldn't care less if it took 60 or 6 seconds to boot really. Things like hybrid or safe sleep remove the risk of data loss due to power failure too. The most important thing to me is that sleep is instant and reliable and it has taken OSs a long time to get there. Mac OS I find to be the most reliable and quickest (but given the significant advantage of limited hardware set and the fact that apple write all the kernel drivers), Windows is fairly good with most hardware and Vista is almost as quick as Mac OS. Linux is catching up fast, I have an EEE PC running Ubuntu 8.10 and it is *almost* instantaneous (still get a blinking cursor before the X server comes back to life) but I've found reliability to still be a bit of an issue.

Easy!

[Cyberax]

http://www.reuters.com/article/pressRelease/idUS144846+15-Apr-2008+PRN20080415

"QNX Achieves Boot Times in Milliseconds on the Intel(R) Atom(TM) Processor Z500 Series"

Re:why the obession with boot time?

[timmarhy]

it probably doesn't work on linux...

Re:This booting thing is overrated.

[MobyDisk]

I've never seen suspend work on anything. Even my MacBook Pro can't do it without USB devices losing connectivity after it wakes up. Blame the hardware manufacturers.

*cries*

[GFree678]

FFS people, just hibernate or if you really need minimal boot time, put it on standby.

Don't give me that rubbish about how these power saving modes don't work reliably. They work fine in Vista (for me anyway, heck they work fine for me in Ubuntu), and resuming from hibernation is much faster than Ubuntu's startup time. If they don't work on your Linux rig, well sucks to be you, but that doesn't mean your hardware isn't capable of it, just your distro for whatever reason.

Flamebait post, perhaps, but I wish people would appreciate the fact that fast boots are completely feasible these days, just not necessarily from a cold boot.

Re:*cries*

[FrankSchwab]

Okay, you're right; resuming from power savings modes works perfectly in Vista.

Now, run a test for me. Attach a secondary monitor, and place it to the LEFT of your laptop. Configure everything to work well. Reboot, and notice everything is still good. Open a few applications, move them to the secondary monitor, then close them. Something mainstream, like Outlook, will do.

Now, suspend your laptop. Undock it, and walk to a conference room. Wake it up. Note that many applications now open on the (non-existent) second monitor. Including mainstream applications from major software companies, as an example Outlook.

Suspend it. Take it back in and dock it. Wake it. Notice that Vista now believes that your secondary monitor is on the RIGHT of your laptop.

Heaven help you if you connected your laptop to the conference room projector when you were there.

Yep, Vista works exceptionally well for all common usage scenarios with suspend/hibernate.

That's why I'm interested in boot times. /frank

Re:*cries*

[Erikderzweite]

This issue was very frustrating in XP at work (I don't use windows at home). Was hoping they'll fix this issue someday.

Still using 98?

[KalvinB]

Changing the network configuration hasn't required a reboot since Windows desktop OSes merged with NT. Windows 2000 didn't require a reboot just to change the network settings.

Most software also doesn't require a reboot either.

Even OS X requires a reboot when you install system updates.

Unless you can restart the OS without restarting the computer every OS is going to require occasional reboots when major changes are made to the OS.

I have Windows XP on a Dual Core Athlon 64bit processor and it boots in under a minute. Most of the boot time for a system is spent loading up all kinds of user software and drivers.

Re:Still using 98?

[Fulcrum of Evil]

Even OS X requires a reboot when you install system updates.

Maybe if you update the kernel; not using it much, I can only extrapolate from my linux experience, where updates almost never require a boot cycle.

Re:Still using 98?

[nxtw]

Maybe if you update the kernel; not using it much, I can only extrapolate from my linux experience, where updates almost never require a boot cycle.

For the past few weeks, I've been ignoring a "Digital Camera Raw Compatibility Update" that requires a reboot. Past QuickTime updates have also required system reboots.

(Why are you making comments about stuff you don't have experience with?)

Um... Why do we care?

[Basilius]

One of the great things about Linux is that you hardly ever have to boot at all... My home server has gone months at a time without any rebooting. Shaving 5 seconds off boot (as it seems would happen moving from 8.10 to 9.04 on a Core 2 Duo system) will save me around 5 seconds/month. That's one extra slashdot summary.

Now, an unstable O/S like Windows, then boot time matters. You might be saving a few minutes a day with those kinds of numbers. There's a serious productivity gain.

Re:Um... Why do we care?

[Zorque]

I'm gonna go ahead and guess that you haven't used Windows since 98, since I'm on Vista and my uptime is nearing 2 months now. The last time I restarted was simply to allow a small driver update to complete, so it wasn't really a necessary reboot either.

Re:Um... Why do we care?

[narcc]

Turn on automatic updating.

Re:Um... Why do we care?

[freddy_dreddy]

Same here, boot about every 3 weeks on Vista but that's because I'm out for a few days without the laptop. My last linux experience was Ubuntu 8.10 where I had to reboot pretty much every day. Reasons: not waking up from a screensaver (!), windows decorations messing up, losing wifi, losing sound, ...
Spare me the command-line fixes, I've given up on that just before getting RSI from that.

Re:Um... Why do we care?

[Zorque]

Yeah, exactly. And besides that, most of the updates rolled out are Defender updates, which don't require or even request a restart.

Re:Um... Why do we care?

[NotBorg]

I think the bigger news is EXT4, which affects more than just boot time.

Bingo!

[Ungrounded Lightning]

You know when germany went through hyperinflation and had to use wheelbarrows to carry around money, i'm sure you've seen pictures.

Precisely.

I particularly liked the one with the kids' snow fort built out of bales of money. Though the woman feeding her furnace with bundles of money because it produced more heat than the fuel it could buy was classic.

Re:Bingo!

[Idiomatick]

You should change it to Yugoslavia though. They had to make a 500billion dollar bill :p

Re:Also: does "shred" work with it?

[Ungrounded Lightning]

All you need to do is write gibberish to a file until the device is full then unlink the file, and do that repeatedly until your paranoia is satisfied.

Which risks hanging things every time the partition fills up.

XP

[caitsith01]

I have XP on my desktop, and I'd say I have a usable work environment in less than 20 seconds from the moment I hit the power button. That's on a Core2Duo with 2 gigs of RAM.

Re:XP

[Mista2]

On My Mac Mini 2Ghz I have the following:
OS X 10.5
Vista Home Premium

On my 1.8GHz HP TC4000 I have XP and Sse Linux 11.1

Over all OS X on the mini boots faster than any of the other systems I have, less than 20 seconds from Apple logo to login. Login however seems to take an age as it starts all my user processes, maps drives etc. Probably another 40 seconds until you can actally do anything. But from sleep - I tap a key, screen wakes up and I enter my password - viola!

Vista - 45 seconds from Apple to Login prompt when set as default boot option. Then 30 seconds to login and load desktop environment.
Similar times for my TC4400 with XP, but I use less on the desktop so this comes p pretty quick.
SuSE11.1 - takes about 40 seonds to get login prompt, but KDE 4.1 desktop loads in 15 seconds as most of the environment is preloaded prior to login, hence the longer boot time.
Sleep for XP does not work for me. On Vista, 10 - 15 seconds to wake up, and I still sometimes have to reconnect my network connection on wake up. ABout the same for waking SuSE 11.1 up too.

Over all the Mac Mini boots faster, but has long desktop load time, but sleep is the best. I almost never have to reboot unless I need to get into Vista on it, but thanks to Fusion, it can run a VM using the bootcamp partition, so I haven't booted Vista on it for a couple of months.
I'm not an OS X fanboi, I just use the OS that does the job the fastest, and with least fuss, the most.
Note: XP, and Ubuntu boot as VM Guests on the mini in under 30 seconds, so I'm pretty sure most of the delays in boot time is the hardware getting itself sorted out.

Re:This booting thing is overrated.

[porjo]

I've got it working *most* of the time under Ubuntu 8.10...does that count? It irks me how it will work 9 times out of 10 and then fail on the 10th for no apparent reason. Just work or don't work damn it!! Actually, come to think of it the suspend part always works - at least in so far as it always goes to a power-off state without errors. The resume is another thing though.

Not much of an ubuntu user are you?

[ebbomega]

I probably have to reboot every month or two. Or have you forgotten of the concept of a kernel update?

Re:Also: does "shred" work with it?

[_Hellfire_]

Recovering such data is probably easily done with secret IDE/SCSI commands. My guess is that there are agreements between harddisk manufacturers
                            and government agencies. Well-funded mafia hackers should then be able to find those secret commands too.

Seriously WTF?

Break out the tin-foil hats everyone. Those mafia hackers are on the loose!

People should better think of their computing devices as facilities
                lended by the DHS.

Ah huh. My computer is trying to send me to gitmo. Got it.

Worse is that this tripe is in the wipe manpage. rense.com-style conspiracy theories have no place in an OS trying to establish its market credibility.

Re:Also: does "shred" work with it?

[Jeffrey Baker]

Yeah well you can do it offline, or you can probably cross your fingers and hope to survive a transient ENOSPC condition. It happens to desktop users all the time, accidentally.

Don't see how it's worth the main page.

[Erikderzweite]

21,4 seconds is about the time I get on my notebook from grub to kdm login (kernel 2.6.28, openrc and ext4). I have Core2Duo 7250, 2Gb RAM and 5400rpm hard drive. The latter is a show-stopper here. I don't think I'd boot much slower with slower CPU or less RAM, but I'd get much better result with a fast SSD.

Re:Don't see how it's worth the main page.

[WhatDoIKnow]

I also don't see what the big deal is for a 20-some second boot to LOGIN MANAGER! I have an old 1.4Ghz Pentium M notebook with an equally old hard drive and and it boots ubuntu 8.04 to gdm login in about 25 seconds. (I must admit that this was with a little tweaking, in a fresh install as I recall it was in the low thirty second range)

NFS hard mounts

[Anonymous Coward]

My local system would lock up and hang for MINUTES while it timed out on requests to the NFS server. I could never understand why the thing didn't just time out in seconds rather than minutes.

This was the default setting: hard mounts. If the server went away the NFS client was told to hang so that any program trying to access the export would block to minimize the chance of data corruption. Once the NFS server came back, things unblocked.

You can of course configure NFS clients to use soft mounts, so that an error is returned to the process that was calling read() or write(2), and you would simply hope that the application code did error checking.

It was possible to mix hard and soft mounts on one clients. Soft mounts were often used on read-only file systems (/usr/local), since you didn't have to worry about corruption of write()s.

The time it took for an NFS client to stop retrying and throw an error was also configurable per site policy (as were/are most Unix system things).

Re:And what's wrong with that?

[Lord Bitman]

If a web browser/office suit/file manager takes four seconds to be usable, something is wrong with it.

If that is an open source project which is bundled with a "fast" operating system, that bundle should take the speed of its most frequently-accessed bundled apps into account when declaring itself speedy.

Who really cares though?

[macshome]

Honestly, who shuts a computer down anymore? Even if you do, is starting it back up in the morning such a huge drain?

I just let mine go to sleep...

Re:This booting thing is overrated.

[CyDharttha]

Thankfully it works for me on every machine I've touched in the past 3 years (many.. I've installed Linux on a few hundred machines in that time). I only restart my work laptop for kernel updates. It's a Dell Latitude D600, getting up there in years. Otherwise, I suspend on avg. 5 times a day. My desktop and TV PCs at home with recent nvidia chipsets suspend, hibernate and wake without issues as well.

Not truthworthy article

[Fri13]

So this is just question about the default configuration of software system when testing Ubuntu's next release.

The OS boot time on Ubuntu is under one second, no matter of what. But the whole system to boot to login screen, it depends from what system services you are running what INIT needs to start and how it process them.

Well, this test was not about "hey! Ubuntu boots under Xx.x second" but just that Ext4 speed is better than Ext3 and you can get much better tests for that from web. You do not need Ubuntu FOR THAT!

So the whole article is just Ubuntu marketing, not about Linux OS speed on Ubuntu system, not about Ext4 file system what hit stable on 2.6.28 version of Linux OS, but only about how great "improvement" is about on Ubuntu. You get much faster boot to login when you turn off few system services and especially the GDM/KDM and you take only the shell with the OS to start. It is question about how you configure your system on boot, not how you configure your Linux OS on that system.

The modern OS's boots damn fast, it gets started by bootloader, loaded as whole to RAM and then it starts first userspace application and after that, you just count seconds OR minutes of the system start time to login screen or desktop...
Everyone could make that easily worst boottime just by enabling few applications as system services and start before other important system services. Like place the Open Office to to start as service and you get 10 seconds added to boottime. Then place computer to calculate heavy mathematic calculations and you can even push the system boot time over 10 minutes if you want and all that time the OS was booted in one second.

Your eee might be using bootbooster

[Sits]

Bootbooster caches various bits of the BIOS probing to the disk and allows the Eee to finish the BIOS phase faster than normal (providing you are booting off the internal SSD). The link is slightly inaccurate in so much as it doesn't speed up grub but instead cuts the time it takes to get to grub.

Cheap hardware

[rsilvergun]

The more tolerance you build into the hardware the cheaper it is to mass produce (because you throw out less during QCing). So you run a loop to keep trying so people don't return something that sorta works.

Umm I think you might be a bit off base

[Sits]

You know how everyone wanted a Linux-based operating system that "just worked" on a wide variety of hardware with drivers for everything? And didn't throw a shit-fit if you moved the hard disk to a completely different machine and tried to boot it up?

Driver probing isn't really why it's slow. Even without Arjan van de Van's fastboot kernel work probing for hardware* and loading drivers is not usually not what takes the most time during boot.

Typically boot time is dominated by the time it takes to actually read data from the disk (so disk seek time is a factor - especially if you aren't using readahead). Secondary to that is the time to start services. Assuming you are booting something graphical desktop environments (GNOME/KDE) take a bit of time to start too followed by X itself (this might be fixed now if you have kernel mode setting).

The above is excluding the BIOS boot time by the way - those can vary dramatically too. I've seen some servers that take many times longer to boot from cold past the BIOS than a typical desktop system takes to finish booting from poweroff all the way to a started Firefox.

* So long as we aren't talking about servers which may have large numbers of spun down disks or a big SCSI bus that takes time to enumerate etc.

That's probably got to be uncompressed

[Sits]

I admit I haven't checked this recently but isn't DSL using squashfs which is a compressed filesystem? If so that 50Mbytes is going to have to be uncompressed and that's going to take up precious time in that one second boot...

Re:That's probably got to be uncompressed

[martin-boundary]

Decompressing files on the fly is often faster than reading them uncompressed. This is actually a common trick to speed up processing of large datasets: even if they're not compressed to begin with, you still compress them. The reason this works is because the cpu is stalling while the data comes in, so the decompression code is executed "for free", and overall you're loading less data, so it's faster.

You have a point - slowly improving

[Sits]

It's absolutely true that Linux has a terrible time suspending to RAM (or coming back from suspend) on certain hardware. It has DEFINITELY been improving over the past two years though (one of my systems was fixed around 9 months ago).

First up make sure you are using the latest kernel you can (new fixes for suspend issues seem to have gone into each of the past few kernel releases including the very latest one). If you have the time you might be able to use the OpenSUSE instructions on debugging suspend to RAM to isolate where the fault lies.

Assuming the problem is more than monitor being off (i.e. the system is completely hanging without any binary only drivers being loaded) if you know how to run the very latest kernels (a prerelease 2.6.29) could you file a bug report over on Kernel Bugzilla after you've checked out the Linux kernel suspend debugging howto?

(You were right about USB and pciehp)

[Sits]

A while back you mentioned about the USB probing could be made faster (and asked why I was using pciehp). Well you were right USB probing DID get faster (boot tracing SVG of EeePC 900) and I no longer need pciehp on my EeePC 900.

I bet that making the kernel more asynchronous than it already is (with the current async patches) won't save any more significant amounts of time on this particular setup though! : )

Re:(You were right about USB and pciehp)

[sofar]

With a kernel boot time of 1 to 2 seconds (still) on most systems, there is still more time to be won :)

It's still broken.

[Grendel Drago]

It sounds like the default configuration uses screenshots as input for the various screensavers. This is not a good default, if it's really set up like this. On the other hand, there's a reported bug where nVidia drivers cause the screensaver to be transparent--the desktop is visible behind it. So maybe it's that particular issue. (If it is, someone who's seeing it should confirm that it's still present.)

Hey, we try.

[Grendel Drago]

Nearly everyone else working on it is a volunteer doing it in their spare time. We're working on it, I assure you. If a bug report exists, that's important to know. If there's a workaround, it may still be that there's a usability issue and that's valid. If it's a problem with your hardware, what on earth do you expect them to do about it? If you can live without your shiny 3D eye-candy (or buying an Intel graphics card), you don't run into the evil-hardware-company issue.

And lastly, the quickest way to fix an issue is to provide a patch. That's not really fair, but, given that you're not paying anyone for the software, that's the way it is. (That doesn't mean that someone who tells you that the only reason you're not a happy user is that you haven't written enough patches isn't a tremendous jerk.) I've gone from filing bugs, to confirming and testing them, to writing my own patches and testcases. It's rewarding, in its own way, to make the system better, bit by bit.

Honestly, the situation on the Ubuntu tracker isn't that bad. Yes, there are still people who drop into ignored bug reports, ask "Is it still present?" and set the bug to expire if someone doesn't write back that, yes, the bug is still present in the current version, as (in plenty of cases) the owner could see if they just took five minutes to test it. Yes, there's no good way to escalate a bug or get it triaged with a quickness, even if it's something that's really damned important. Given how bad things are at the GNOME bugzilla (bugs wait forever there), I'm pleased in comparison.

Given all this, it's understandable that Linux isn't for everyone. Hell, look at the state of audio support. It's a damned tragedy. You have to really love it at this point. I'm motivated by the fact that it's worlds better than it was only a few years ago: suspend/resume actually works sometimes, a major vendor (Intel) actually maintains bleeding-edge open-source video drivers as part of the X.org distribution, and there's a lot more polish on things--all the little usability details that sound like nitpicking when you enumerate them, but add up to a good or bad user experience, in the final evaluation. You may have left Linux--and, really, if you're not willing to put up with quite a bit at this point, it's not for you--but it's not a failure.

Re:Hey, we try.

[freddy_dreddy]

Thanks for taking the time to reply.

I'm sorry to say that this is not a satisfactory answer. Look around the net - hey, just stay on this site and click on a random link - and you'll see that various Desktop versions of Linux are portrayed as nothing less than perfection. Me and the people around me who have tried the same software (I'm talking 13 years in my case) must be subjected to some strange phenomenon where the most simple tasks seem to require RSI-inducing terminal typing. Fix one thing, something else breaks. Update and things go bonkers. Disable update and ... it updates. Reboot to get sound. Recompile to get wifi. Repeat.

It's 2009, people. Multimedia is ubiquitous, no matter how complex these peripherals are, each and every common-sensed consumer expects them to work without a problem and does not wish to lose time on it. Truth is that the best Ubuntu experience is the one where you have a Windows box running a foot further on the desk, so you can scroll through an avalanche of third-rate forums and half-arsed wikis to find some obscure sequence of terminal commands. In most cases you're trying to find out how to run a windows look-a-like tool or package in your newly found time consumer. Consumers want this and that. Don't discuss with them whether or not they need it. I'm hinting at a very wide range of issues here, going from drawing squares in Gimp to having flash whilst browsing.

One of the things I find most troublesome, however, is that not Linux but Windows is the best platform for running FOSS. For some reason (shall we call it "API" ?) these programs run with a minimum of problems on my MS. And don't dare to tell me "We're working on it". I used to "work" on writing Linux software, you spend most of the time peeling back layer after layer of shitty abstraction to find the true reason: bug obfuscation. Ubuntu is probably the worst hack on a hack in existence.

Most of the energy isn't put into bug resolving, btw. The average Ubuntu dev works on polish and gloss, like this incredible progress in boot time. If there's one thing that's possible to learn in the history of operating systems: apps kill competition, not OSs. No applications ? No carrot! The recent Open Orifice debacle is a very nice illustration that when the going gets tough the hard-working volunteers have better things to do.

Excuses are the trademark of FOSS, not freedom, not liberty, not whatever third-world slogan they can come up with. Excuses are what you normally get for complaining about FOSS. They vary from "your fault", "worksforme", "their fault" to "give it time". Ubuntu is what you get for following this open-minded anti-corporate philosophy: a very shiny car where everything is held together with sticky tape. After a decade and a half I think it's time to reflect on a few facts:
- Things which work in Desktop linux a la Ubuntu work for one of two reasons: 1) the linux support is provided/pushed by the corporate world or 2) there is a very strong user base on the windows version.
- The Desktop Linux devs apparently lack know-how when it comes to hardware
- People will pay for something if it saves them time

Last August I simply formatted my Linux installation for the last time. I still have a partition with BeOS and one with Minix when I want to fiddle around. I went to the store and bought Vista. 4 hours later I had installed everything which was on my linux system and more ... for free. The amount of time I saved since then is beyond comprehension. I have time for three extra hobbies now. The standard rhetoric to this from ye olde linux base is "stop bitching". Sorry mate, wherever I roam on the net to read up on H/W or S/W specs there's some bleeding ass punk rambling on about Ubuntu as if it's God's gift to humanity, and I feel like I need to correct that. The Desktop Linux promotion chitchat is nothing more than a bu

May want to do more than just mount as ext4

[spaceturtle]

You may be able to mount ext3 as ext4, but you won't get the cool new features like extents. From Converting an ext3 filesystem to ext4, it is easy to enable these features without a reformat. You may want a plain ext3 /boot partition for compatibility with older versions of GRUB, but it looks like the GRUB for Jaunty will support ext4, so a separate /boot partition may not be necessary.

Let me post some comments in reply

[Sits]

This post is TOTALLY offtopic. Really you need to split these up and file them as bug reports over on launchpad. I'll post a couple of comments answers but I'm not going to follow up on any of this (even if you answer any questions I ask).

- switching from dual display to presentation (clone) and back totally messes up x config, I have to uninstall and reinstall nvidea drivers

Talk to NVIDIA (Linux web forum) about this. It's their code you're running and they are probably the only ones who are willing to fix it.

- in dual screen mode, nautilus opens on the first display. I have to open terminal and run nautilus& to lunch it on the second display

You can't drag it? I don't quite understand...

- in dual screen mode, keyboard keeps focus in the previous screen. I have to minimize/maximize a windows on the "new" screen to move keyboard focus

Are you using desktop effects? (Do windows fade and slide etc?) If so this sounds like a bug in compiz...

- RDP client crashes X windows in some cases (it does not close the drop down list of used servers... and bang)

Hmm. I'm really curious now as to whether you are using compiz. Regardless your best bet with this one would be to be to see if you can capture a backtrace of the crash with debug symbols and to file a bug report against the RDP client (I'm guessing you're using tsclient) in launchpad.

- oh and NO it's not AN ERROR if I close the RDP window. If I want to reconnect, I will, don't hide under my active windows and bring RDP windows back in 30 seconds. That's just plain stupid.

I guess file an enhancement request on tsclient in launchpad.

- java and window decorations don't play well together (popups without buttons etc.)

I really would like to know whether you are using compiz. If you are I have a feeling this was a known "bug" in the Java bug database for a long time but the fix is not yet in Ubuntu.

- How about opening a connection to a new server in a new tab, not in a new nautilus window?

Hmm probably best to file an enhancement request over on the GNOME bugzilla.

- Flash stops working. I just see a gray square where flash is supposed to be.

64 bit Firefox using 32 bit Flash via nspluginwrapper I'm guessing. There is a 64bit Linux Flash plugin that is in very early beta that MAY work better for you (I've heard mixed things mind). Also make sure you're using Flash 10 whatever route you are taking.

- Firefox is not very stable.

Might be because of extensions or plugins or you may have found a problem page or your memory might be faulty or Firefox might be buggy or... You are going to have to sit down and capture the issue in Firefox this then file a bug report in launchpad.

- Windows would become gray and unresponsive when there's a lot of disk activity.

You're using compiz aren't you? The greying is compiz telling you that the window HAS become unresponsive! As to why this is happening on I/O it probably varies from program to program. Too little information to many possibilities to say more.

- I've seen ubuntu crash on my much more times than I've seen BSOD on the same HW.

Quite possible. I've seen Linux stable on some computers and fla

You can't run what is compressed though

[Sits]

Your boot is going to have to be sequential until you have decompressed ALL the data though. At only 50Mbytes your CPU is probably going to be stalled longer doing the decompression than waiting for 100Mbytes off your storage medium in our mythical one second boot (unless single CPU speeds get incredibly fast). You also can't make use of execute in place.

Using compression in the fashion you are suggesting is only a win if you are avoiding the slowness of reading in the data (e.g. by avoiding disk seeks due to having less data). As it stands you are probably going to cycle the RAM caches with all that decompression too...

Re:You can't run what is compressed though

[martin-boundary]

You're right, but remember that the full 50MB is not strictly needed just for booting anyway (kernel, some static libs, xserver, all depends on what the definition of booting is). Someone with time on their hands could probably lay out an incremental decompression system that optimizes cache use in the first second. No way to know without real profiling data, tough.

Why the Debain Symbol?

[pseudonomous]

OK, it's not that important, but why the Debain symbol? This is a story about Ubuntu, yes it's a Debain derivative, but this article is not about some pre=release of Debian booting in x seconds it's about an Ubuntu alpha.

there's still to much legacy rubbish in the BIOS

[ross.w]

And when we get to Zippy Zebra sometime in 2017, the world will end and it will still take more than 20 seconds to boot.

What's the problem?

[Troberg]

Why this obsession with boot speed? I have a very, very simple solution: I never turn my machines off. Linux is stable enough to run 24/7, there is no reason to turn them off. Problem solved.

Re:What's the problem?

[HoaryCripple]

Ever hear of a laptop?

Re:What's the problem?

[Troberg]

Yep, and I have a solution for that as well. I pick the laptop out of the bag and place it on the table and turn it on. Then I get the other stuff out and start connecting power, network cable, mouse and get out a mouse mat. When I'm done with that, boot is already complete. No lost time, you just have to do things in the smart order.

Although, nowadays, when laptops are cheap, I've switched to simply having one in each place where I use a laptop (bedroom, TV room, kitchen, at work, server room (as a remote desktop client for managing the servers) and so on), and simply revert to the habit of not turning them off. I have one laptop when I travel, as I obviously can't have a laptop in each hotel room, but for that, I revert back to the order of startup outlined above.

It's all a matter of planning

LTS

[Britz]

You switched from LTS to normal. I also installed 8.10 for my dad, because I wanted OpenOffice 3. Now I have some strage problems as well. The Xserver resolution is all messed up at the login. I can't even read anything on the login screen. It was easy enough to change screen resolution on the desktop. But for some reason Ubuntu 8.10 does not apply that resolution to the login.

To make a long story short: I got OpenOffice 2.4 on Ubuntu 8.10 even though I only have 8.10 sources in my sources list. Crappy, isn't it?

Re:LTS

[Skuto]

8.10 does not include OpenOffice 3, you have to get it from external sources. And the latest versions is completely broken on KDE.

On 8.04 it's the same situation except it actually works :)

In other news...

[getuid()]

...a name change to "Insta-Buntu" is currently being evaluated for the upcoming release 9.04 :-)

Not impressed

[gzipped_tar]

My Arch Linux boots in 33 seconds (that's from GRUB beginning to boot the kernel, to the login prompt). And it's in a virtual machine.

Wake me up when you managed to boot in 3 seconds.

Linux On The Non Geek Desktop

[cyclomedia]

Do you also leave your shower running, and subsequently not need to care about the warm-up time?

Do you also leave your TV on, so you can see what's on the instant you walk in the room?

Did you throw out your electric kettle and replace it with a thermostatic boiler so that you don't have to wait a minute for a hot drink?

If we're ever to get our Year of Linux on the Desktop then the 99.9% of the population that aren't geeks, care about the environment and their electricity build will turn on their computer when they want to do something, and turn it off when they're finished.

Defragmentation included

[martijnd]

From the same wikipedia post:

Ext4 has an online defragmenter. Even with the various techniques used to avoid it, a long lived file system does tend to become fragmented over time. ext4 will have a tool which can defragment individual files or entire file systems.

Peter Norton is going to love this! Norton Defrag for Linux coming in 3,2,1...

What do you mean reformat?

[holizz]

Use tune2fs (assuming you're currently using ext2/3): http://ext4.wiki.kernel.org/index.php/Ext4_Howto#Converting_an_ext3_filesystem_to_ext4

Also, you can convert an ext3 filesystem in-place to Btrfs: http://btrfs.wiki.kernel.org/index.php/Conversion_from_Ext3

Too long time

[wye43]

My 3 years old Windows XP PC boots in 6 seconds (from power button to desktop usable and hard drive has stopped being busy!). No fancy hardware except maybe a dumb stripping raid0 from 2x250gb SATA2 drives and a customized/optimized OS configuration. This is a computer used for large range of activities: gaming, movies, music, internet, programming. On a modern/good/sshd PC and if I put my hands on it seriously I bet I can reduce it under 4 seconds.

I remember a guy who customized his Gentoo 6 years ago to boot under 10 seconds as well.

So whats with this hype for a lousy 21.4 on today's hardware?

Re:Too long time

[LodCrappo]

My 3 years old Windows XP PC boots in 6 seconds (from power button to desktop usable and hard drive has stopped being busy!).

bullshit. youtube, or it didn't happen.

My NEC MobilePro 900 WinCE boots in 1 second

[Doug Jensen]

1 second later I am typing in Word. That, and email, are pretty much all I use it for, but I use it for that all the time. I bet there are still some journalists out there using the Radio Shack 100 (or whatever it was called) with a 1 or 2 line display, booting in 1 second, running on 2 AA's forever. When I can get a netbook that boots in a second or 2, running linux I presume, than I'll buy one. Don't try to tell me about standby (uses battery power) or hibernate (still takes a long time to resume).

'I guess it's finally time to reformat my /home.."

[fudgefactor7]

This kind of always irked me. Why can't there be a way to convert a partition to Ext4 from whatever-it-might-be without having to reformat? For instance, FAT can be converted to NTFS (I know this example sucks, but give me a break, I'm reaching here), so why can't there be a convert for Ext2/3 to 4? Surely, this has to be something that is "doable."

even more appropriate title would be

[pablomme]

Ubuntu 9.04 boots 13% faster with ext4 than with ext3

or using the AMD Sempron data:

Ubuntu 9.04 boots 18% faster with ext4 than with ext3

If the boot time with ext3 was one hour, 3.1 seconds would be pretty unimpressive, and if ext3 took 3.2 seconds ext4 would be the bestest filesystem ever. Absolute differences are mostly meaningless.

Have a login window as a bootsplash.

[spaceturtle]

If we define boot time by when the login window sleeps, we could technically boot really quickly if the bootsplash was a login window... (although this isn't necessarily a bad idea, so you can login at your leisure without slowing the boot.)

Re:Have a login window as a bootsplash.

[xenocide2]

By login we mean GDM. Which depends on X. Which depends on having your kernel modules installed.

15 seconds here

[Xabraxas]

Grub to GDM for me is about 15 seconds. My desktop is pretty heavy to it takes another 15 seconds to completely load after I login.

Converting Ext2/3 to 4 is doable.

[spaceturtle]

Or at least that would explain why half a dozen previous posters scratched their head and pointed to http://ext4.wiki.kernel.org/index.php/Ext4_Howto#Converting_an_ext3_filesystem_to_ext4

One minor issue is that this will leave existing data in the previous format, so you might get better performance if you reformat, or at least copy your old data around.

Re:Also: does "shred" work with it?

[Al Dimond]

Clearly, then, the people that wrote the wipe manpage aren't working on an OS trying to establish its market credibility.

Free Unix (all the software: Linux, the BSDs, GNU, KDE, Perl, etc) isn't trying to do anything. It's just code. People that write code and contribute it have their own reasons and motivations. People working for Red Hat and Novell are trying to establish the market credibility of Free Unix (or at least their brands of it). But there's no reason that just because those companies use some geeks' code that those geeks share the marketing goals of the companies. If Red Hat wants their manpages to sound professional, they should review and rewrite as necessary. It's (usually) allowed by the license. Same thing for Ubuntu if they want user-friendly manpages. If either group can't get changes adopted upstream, they can either fork or maintain a patch set. Neither is as much work as writing the software from scratch, so they've still managed to benefit from the work of the paranoid geeks. Just as those same geeks likely benefit from their work.

It's improving. Really, it is.

[Grendel Drago]

I'm sorry to say that this is not a satisfactory answer. Look around the net - hey, just stay on this site and click on a random link - and you'll see that various Desktop versions of Linux are portrayed as nothing less than perfection. Me and the people around me who have tried the same software (I'm talking 13 years in my case) must be subjected to some strange phenomenon where the most simple tasks seem to require RSI-inducing terminal typing. Fix one thing, something else breaks. Update and things go bonkers. Disable update and ... it updates. Reboot to get sound. Recompile to get wifi. Repeat.

You may not believe this, but few things piss me off more than mindless fanboyism, and I say this as a Linux diehard. It leads to terrible disappointment for users who find out that, yes, it's still got a ton of problems.

It's 2009, people. Multimedia is ubiquitous, no matter how complex these peripherals are, each and every common-sensed consumer expects them to work without a problem and does not wish to lose time on it. Truth is that the best Ubuntu experience is the one where you have a Windows box running a foot further on the desk, so you can scroll through an avalanche of third-rate forums and half-arsed wikis to find some obscure sequence of terminal commands. In most cases you're trying to find out how to run a windows look-a-like tool or package in your newly found time consumer. Consumers want this and that. Don't discuss with them whether or not they need it. I'm hinting at a very wide range of issues here, going from drawing squares in Gimp to having flash whilst browsing.

I know I may seem like a broken record here, but the developers can't fix the problem if they're unaware of it. Developers get into habits of using a program a particular way, and without users providing feedback in the form of bug reports with explicit instructions on reproducing the bug, they won't become aware of it.

I know, I know, most people have better things to do with their time than file bugs, and this is why I don't recommend the OS to people who don't fiddle with computers as their primary occupation.

As for Flash... well, that's the problem when you try and mix proprietary and free software. X.org/PulseAudio/Firefox developers point the finger at Flash as the source of the suckiness, while Flash developers (presumably) point at X.org/PulseAudio/Firefox. The best free solution is reportedly swfdec, but it's not going to be as good as proprietary Flash, at least not for a while. (Reportedly, it plays YouTube, at least.)

Most of the energy isn't put into bug resolving, btw. The average Ubuntu dev works on polish and gloss, like this incredible progress in boot time.

I don't think that's quite fair--it's a false dichotomy. The major change in boot time comes from enabling the new ext4 filesystem; it's not plausible to assume that Ted Ts'o and everyone else who's been working on the kernel would be making your Flash work properly if they hadn't been doing that.

If there's one thing that's possible to learn in the history of operating systems: apps kill competition, not OSs. No applications ? No carrot!

We don't have a working browser in Ubuntu because everyone warped the system enough to get Internet Explorer working via IEs4Linux; we have a working browser because of Firefox. Similarly, the Flash problem won't be solved by any amount of fiddling on Adobe's part, or by any amount of work from the PulseAudio crew. It'll be solved when swfdec (or maybe Gnash) can finally step in as replacements.

Excuses are the trademark of FOSS, not freedom, not liberty, not whatever third-world slogan they can come up with. Excuses are what you normally get for complaining about FOSS. They vary from "your fault", "worksforme", "their fa

Re:It's improving. Really, it is.

[freddy_dreddy]

I hope you like to read ...

You may not believe this, but few things piss me off more than mindless fanboyism, and I say this as a Linux diehard. It leads to terrible disappointment for users who find out that, yes, it's still got a ton of problems.

Excellent. Unfortunately they have strength in numbers.

...but the developers can't fix the problem if they're unaware of it. Developers get into habits of using a program a particular way, and without users providing feedback in the form of bug reports with explicit instructions on reproducing the bug, they won't become aware of it.

That's not the root of the problem. There's a real problem in hardware support - computer hardware in Desktop Linux' qualification model is the Ultima Thulle for devs. Each dev has a (very) limited amount of platforms to test on. A product which relies on bug discovery after product release can not survive on this consumer base. The vast majority of computer users find it troublesome to program a granny-proof car stereo - we sell them (I'm in micro-electronics) a technology which can only be described in superlatives. All these people, from the first to the last, will cease and desist if they encounter a bug which is either repetitive or blocking. Yes, reporting has advanced - it has even become an advance in software management on its own - but all these people expect them to be resolved before release. It's what they're used to, it's the standard, it's what they expect, they know it exists under a different name.

Did you talk to someone who has a workflow on Vista recently? I haven't had a bug since I installed it - ever. I reboot about every 2 or 3 weeks, hibernate 2 or 3 times a day. I'm now watching Digital terrestrial TV via a TerraTec dongle which installed in 8 minutes (9 clicks) and all my software is freeware. Currently typing on Chrome (trust me: Mozilla has no reason to exist when these guys will release extensions). It's free and comes bundled with google.com (or the other way around). Desktop Linux isn't the platform that enables choice, Windows is: currently have 5 browsers (webdev for fun), tried 3 ftp clients, have 3 media players (kept VLC, killed iTunes & Winamp), a number of programming & debugging environments for embedded (IDA, MicroAsm and even my Borland debugger runs).

We don't have a working browser in Ubuntu because everyone warped the system enough to get Internet Explorer working via IEs4Linux; we have a working browser because of Firefox.

True. Can't help but make a correction: you have a working browser because it was pushed/provided by the corporate world (a point in my previous post). Firefox will not survive. The biggest & most popular website ever, which is the homepage of the best marketing/advertising company ever, has released a browser which is simply too sweet for words. They'll continue to fund FF, to no avail. It's not difficult to see Mozilla as one of Google's little outsourcing trials, btw. I would probably be pushing your buttons by calling Mozilla Foundation a sweatshop where the majority isn't paid. Open Office is also stalling, and both Goog and MS are in the opposition. MS is always portrayed (on the mainstream Linux front, not just fanboys) in the light of IE and their OS, accompanied by predictions of failure. Don't forget that they 'just decided' to jump in the gaming console market, and won. In response to Java they 'just decided' to deliver an alternative and java is going down, in favor of .NET, C# and JS (which is google's domain). MS also decided to jump on the embedded video wagon recently, thanks to the Olympics they were able to boost their shares. Forget ogg and the video tag. MS recently delivered a very peculiar progress on the static photography front (photosynth

That's a bug; please file it.

[Grendel Drago]

X is supposed to be bulletproof--it's supposed to at least get you to a working login screen no matter what. File a bug; the developers may not be aware of the situation. They'll likely ask for some more information and testing on your part, so please keep an eye on the report after you file it.

I know I'm asking you to do some work, but good bug reports are the primary means by which problems get fixed. The developers probably don't know that the system is broken on your hardware, because they don't have a copy of your hardware on which to test it. You're not just helping yourself out; you're helping everyone else who might run into the same problem in the future.

Re:Also: does "shred" work with it?

[skeeto]

Unless you have highly trusted human security guards posted around your computer 24-hours a day, you don't need to worry about secret IDE/SCSI commands planted by the government. If an organization as powerful as a government or large criminal organization (heh, same thing, right! :-P) wanted your data, they would have them. While you are away, they have physical access and could easily bug your computer, install a hardware keylogger, and so on, to get access. Or worse, they get out the rubber hoses.

Locks won't work either. You need real trusted, human eyes at all times. Locks are inconvenient enough to keep away a petty thief -- or make them gain access to your belongings by a cheaper method like smashing a window -- but are generally trivial to pick.

If your data are so important that it is under live guard, it's probably so important you would physically shred the hard drive anyway, secret IDE/SCSI be damned.

Re:Also: does "shred" work with it?

[swillden]

Besides the rest of the tinfoil-hattery at the end, this part:

Encrypting a whole partition with cryptoloop, for example, does not help very much either, since there is a single key for all the partition.

... doesn't make any sense. There's nothing wrong with having a single key as long as it's a good key and kept well-secured. If you do this, and also take care to make sure that your swap is encrypted, then you eliminate all of the issues with hard drive sector remapping, etc. You also eliminate any need for "wipe". Just destroy all copies of the key, and the data is gone.

Of course, there is no protection from the DHS-installed transmitter in your CPU.

Re:Also: does "shred" work with it?

[WillDraven]

Of course, there is no protection from the DHS-installed transmitter in your CPU.

Faraday cage? Only bringing data to and from the computer via sneakernet?

Of course I'm not that paranoid. I don't bother with encrypting my files or anything like that. The most sensitive info on my computers is some MAFIAA content I could get sued over. Anything I really don't want getting out (Ideas for grant proposals, passwords I can't be bothered to memorize, etc.) goes straight from my brain to a little book (that I handmade from a cigarette pack, reflective tape, and old computer paper) which I keep in my wallet that's never more then a couple yards away from me.

This especially works for me because I'm a bit OCD about keeping track of my stuff. I'd still be on a 12 year old wallet if I hadn't decided to finally replace it a few weeks ago (and I still kept the old one in my arts and crafts drawer to be reincarnated as part of something else).

Why ignore the past?

[nurb432]

it all depends on how you define boot, and how many layers of a 'modern' GUI, and services, you want on top that OS.

I don't think you are ever going to get sub 1 second boots from *power down*, but if you were around in the old days, booting an Atari ST to full GEM desktop didn't take but a few seconds from ROM. Not much longer from disk.

Re:Also: does "shred" work with it?

[_Hellfire_]

Clearly, then, the people that wrote the wipe manpage aren't working on an OS trying to establish its market credibility.

And that's their prerogative. I couldn't care less if they wrote a manpage that included ASCII-art pr0n.

People working for Red Hat and Novell are trying to establish the market credibility of Free Unix (or at least their brands of it). If Red Hat wants their manpages to sound professional, they should review and rewrite as necessary. It's (usually) allowed by the license. Same thing for Ubuntu if they want user-friendly manpages

Which is precisely my point. Ubuntu *is* trying to establish Linux's market credibility, and this paranoid delusion is included in the Ubuntu manpage for wipe.

No Competition

[SoopahCell]

In my opinion it's just lack of competition. The Windows standard for running apps needs to become a fully documented public, free work so that competing OS manufacturers can finally start showing up. Then you'll get great boot times.

Vista is a crime but in a competitive market it would have been a blip everyone ignored for other, better competitors. Ubuntu with Wine is almost there but still too complex for your average user with a Windows software install CD and without a clue.

Re:Hibernate/suspend is not for everyone

[Sj0]

Ah, the environerd. You know, a nerd is the perfect person for that lifestyle.

Only a nerd would take pride in a life of complete social isolation, where the only connection with the outside world is a glowing terminal screen.

Of course, you're not a very good nerd, because you fail at science. A laptop in sleep mode and a laptop turned off completely will burn negligible amounts of energy. If one watt is going to cause you to lose sleep, your power system is badly designed.

This isn't surprising, since you also appear to fail at statistics. lithium batteries DO NOT typically burst into flame. A very small number of them fail due to defects in manufacturing. This is not typical.

Thankfully, I actually just buy renewable power from the hydro company. This means I don't need to stay awake at night worrying about whether I've left enough power in my energy budget to heat my home overnight.

Re:This booting thing is overrated.

[Hucko]

Doesn't work on mine :(

21? pfah.

[AdamWill]

My new desktop boots current Mandriva Cooker x86-64, from lilo to gdm, in 16 seconds. And a full working GNOME desktop with Firefox running takes about 25.

Of course, it's a heavily overclocked quad-core system with a pretty fast disk drive and 4GB of RAM.

Point being, random raw numbers (21.4 seconds!!!) don't mean an awful lot. Hardware is rather important. The original story's a good one - it explains the hardware used to test, and carefully compares ext3 and ext4, which is what they were actually trying to do. But Slashdot's abstraction of it is crap.

MacOS does not have GDM.

[spaceturtle]

But MacOS X doesn't have GDM (or ironically, X) so this thread still hasn't defined a "fair" comparison between Ubuntu and MacOS X boot times.

Samba/CIFS currently "easiest"

[Sits]

Experience says that Samba usually is the "easiest" way (especially if some smidgen of security beyond IP address is needed). Don't forget if Samba notices another Samba system at the other end it enables some extensions. The alternative is going down the NFS v4 massive configuration route.

If you don't care about speed and are in Linux (or OSX... kinda) only environment and you mean ad-hoc (rather than log in) shares, sshfs is pretty easy to get going with recent distros. However, the moment you also need to share to Windows it's better to do the Samba dance.

POST is bad for spotting memory errors

[Sits]

The POST a typical non-server grade system does some very weak memory tests. In all the cases I've seen of bad RAM in the past few years the POST never said anything. Skipping the memory check would be no big loss.

If you suspect your memory really is bad, run a proper memory checker. They catch far more memory faults than the POST does,

Re:MacOS does not have GDM.

[xenocide2]

OSX does have a login window. I don't know what it's called, but either way, I'm looking at it right now.

Require Linux use GDM but allow MacOS any login?

[spaceturtle]

This tread is going round in circles. It isn't fair to require that Linux load GDM (rather than just display a login window onto the frame buffer with out loading X, kind of like a bootsplash), while allowing MacOS X to display anything that look like a login window. Now maybe Mac OS X is at an equivalent stage in the boot process when it reaches the login window as Linux is when it reaches GDM, but it is very hard to define an equivalent stage of the boot process between two entirely different OS's. FYI, I suggested replacing the bootsplash with a tiny program like a login screen: http://brainstorm.ubuntu.com/idea/10735/