Slashdot Mirror


Is Ubuntu Getting Slower?

An anonymous reader writes "Phoronix has a new article where they provide Ubuntu 7.04, 7.10, 8.04, and 8.10 benchmarks and had ran many tests. In that article, when using an Intel notebook they witness major slowdowns in different areas and ask the question, Is Ubuntu getting slower? From the article: 'A number of significant kernel changes had went on between these Ubuntu Linux releases including the Completely Fair Scheduler, the SLUB allocator, tickless kernel support, etc. We had also repeated many of these tests to confirm we were not experiencing a performance fluke or other issue (even though the Phoronix Test Suite carries out each test in a completely automated and repeatable fashion) but nothing had changed. Ubuntu 7.04 was certainly the Feisty Fawn for performance, but based upon these results perhaps it would be better to call Ubuntu 7.10 the Gooey Gibbon, 8.04 the Hungover Heron, and 8.10 the Idling Ibex.'"

544 comments

  1. Had went on? by rrohbeck · · Score: 0, Troll

    Why should I read this FA if the author apparently didn't finish high school?

    1. Re:Had went on? by bsDaemon · · Score: 3, Funny

      He would have been able to finish, had Ubuntu not been so slow that he was never able to finish his papers and turn them in on time.

    2. Re:Had went on? by just_another_sean · · Score: 5, Insightful

      Why should I read this FA if the author apparently didn't finish high school?

      The anonymous submitter and CmdrTaco's grammar skills have little to do with performance in Ubuntu. RTFA; it makes a good point and I for one hope that this observation is accepted by the Ubuntu developer's and something is done about it.

      --
      Creationist Textbook Stickers Declared Unconstitutional by CowboyNeal
    3. Re:Had went on? by caluml · · Score: 4, Insightful

      Why should I read this FA if the author apparently didn't finish high school?

      Because intelligence and wisdom have nothing to do with "finishing high school"? I've got nothing past GCSEs. Luckily for me, employers in the UK see past that.

    4. Re:Had went on? by Anonymous Coward · · Score: 2, Insightful

      Reasons include:
      1. Determining if the error is in the article or in the summary.
      2. Determining if the article is riddled with errors or if the summary highlights an uncommon occurrence.
      3. Determining if the article's substance is good despite presentation and communication problems.
      4. Not having a knee-jerk impulse to ignorantly flame strangers on the internet.

      The article could be written by a dyslexic, Sumatran orangutan, yet full of useful data. You'll never know.

    5. Re:Had went on? by Anonymous Coward · · Score: 0

      His name is Michael Larabel. He's the one who wrote the article. Yes, the grammar error is in the article. And what do you expect the Ubuntu developers to do? Rewrite the kernel? They only put the distro together. Maybe they should try GNU Hurd.

    6. Re:Had went on? by jabithew · · Score: 1

      Luckily for me, employers in the UK see past that.

      They used to. I'm not so sure any more...

      --
      All intents and purposes. Not intensive purposes.
    7. Re:Had went on? by rrohbeck · · Score: 4, Insightful

      I just read it and found it pretty devoid of information. It is one of those mindless performance reviews in as many pages as possible.

      Where are any measurements that look at where performance was lost? Running just the distros does nothing to isolate what got slower. Trying different kernels and different X servers would at least show an attempt at understanding what's going on. Why didn't they compare at least one Ubuntu version with a similar Fedora version, let alone Kubuntu or xubuntu?

      As I expected: If a site employs people who can't write and has no editorial control that would weed out a glaring error like this you can't expect anything but quick and dirty superficial work. If an error like this that just jumps at you isn't caught, how many subtle errors in the data should I expect?

    8. Re:Had went on? by tbannist · · Score: 1

      The lack of proper spelling and grammar detracts from the credibility of the article. I mean, if the author doesn't care enough to fix obvious grammar errors, how can he be trusted to have testing methodology that doesn't contain gross errors?

      It looks like there was no attempt to do some obvious longitudinal correlation. Once you've determined that Ubuntu is running slower in later releases, the next step is to check equivalent releases of Centos (or Fedora) and SUSE and see if it's an issue which appears in other major Linux distributions.

      --
      Fanatically anti-fanatical
    9. Re:Had went on? by Aphoxema · · Score: 1

      Did you think not he might just have a disability of mentality that keeps him from saying properly what he means to say, insensitive clod you?

      --
      "Most people, I think, don't even know what a rootkit is, so why should they care about it?"
    10. Re:Had went on? by Aphoxema · · Score: 1

      I dropped out in 10th grade for an entire realm of reasons, many of which were completely outside of my influence. I also skipped a lot of school my whole life and avoided authority as much as I could.

      I didn't do this because school was too hard or I didn't like the people, it just wasn't challenging, stimulating.

      Instead I spent my time playing with my 386, reading encyclopedias and a wealth of information on old hardware and stuff my father had, and speeding through the textbooks I had borrowed from school just to get all the work done and piss off my teachers.

      Why shouldn't you RTFA if the author didn't finish school? Imagine all the apparent 'morons' that built the internet and our complicated entanglement of computing businesses who never finished high school or college.

      --
      "Most people, I think, don't even know what a rootkit is, so why should they care about it?"
    11. Re:Had went on? by Daniel+Phillips · · Score: 1

      I just read it and found it pretty devoid of information. It is one of those mindless performance reviews in as many pages as possible.

      That is nonsense, this is Phoronix, a consistently reliable source.

      --
      Have you got your LWN subscription yet?
    12. Re:Had went on? by Bert64 · · Score: 1

      If the new kernel is considerably slower than the previous one, then they can continue shipping the old one until the new version catches up...
      Same with gcc - they can compile using the older faster version until the current version is capable of keeping up.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    13. Re:Had went on? by Daniel+Phillips · · Score: 1

      rying different kernels and different X servers would at least show an attempt at understanding what's going on.

      And if you had bothered to read the article, you would have noticed they carefully specified the kernel used in each version, the X server and the compiler.

      --
      Have you got your LWN subscription yet?
    14. Re:Had went on? by Bert64 · · Score: 1

      It's down to the individual company, the people doing the hiring and the job in question...

      Some places are snobs, and demand people who have a list of educational certifications... Others will have someone who understands the job doing the interview and use his knowledge of the field to gauge who is best suited to the job, while some will simply go based on experience.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    15. Re:Had went on? by tbannist · · Score: 1

      For me, impossible it is, to tell you, whether thinking of his, or speaking of his, broken is.

      --
      Fanatically anti-fanatical
    16. Re:Had went on? by Rennt · · Score: 1

      They carefully REPORTED the kernel, X server and compiler version, but they didn't specify anything. They just used the stock version that came on each release CD. They didn't even apply updates from offical repositories as far as I could see.

      The only thing SPECIFIED was the ATI drivers, but even then it was only because it was the easiest solution

      It would have been interesting to at least see comparisons done where they were all running the same kernel, if not X and GCC version

    17. Re:Had went on? by Aphoxema · · Score: 1

      Wait, what?

      --
      "Most people, I think, don't even know what a rootkit is, so why should they care about it?"
    18. Re:Had went on? by Anonymous Coward · · Score: 0

      I for one hope that this observation is accepted by the Ubuntu developer's ...

      Come on, don't leave us hanging. The developer's brain? The developer's secret club? The developer's mom? What is it, man?

    19. Re:Had went on? by Anonymous Coward · · Score: 0

      Well, how about that... a Brit just defended somebody who mangled the Queen's English. Now I be done seen 'bout everything.

    20. Re:Had went on? by LingNoi · · Score: 1

      It's not snobbish to expect someone to have GCSEs especially when some people I know couldn't even be bothered to turn up to the exam. Why the hell would you want to hire someone like that?

      If they couldn't be bothered to show up for the exam, or if they did then failed, they're either too lazy or too dumb to do any IT job.

    21. Re:Had went on? by Anonymous Coward · · Score: 0

      Because intelligence and wisdom have nothing to do with "finishing high school"?

      Intelligence and wisdom mean nothing without credibility.

    22. Re:Had went on? by mollymoo · · Score: 1

      Where are any measurements that look at where performance was lost? Running just the distros does nothing to isolate what got slower. Trying different kernels and different X servers would at least show an attempt at understanding what's going on. Why didn't they compare at least one Ubuntu version with a similar Fedora version, let alone Kubuntu or xubuntu?

      The subject of TFA is "Ubuntu 7.04 to 8.10 Benchmarks: Is Ubuntu Getting Slower?" and it answers that question. If the title was "Is Ubuntu Slower Than Other Distributions?" or "Why is Ubuntu Getting Slower?" you might have a point, but those were not the subject of the article. The article does exactly what it says on the tin, if you were expecting something else that's your problem. If you want to see the results of some further tests why don't you go and do them?

      --
      Chernobyl 'not a wildlife haven' - BBC News
    23. Re:Had went on? by Elektroschock · · Score: 1

      Install LXDE and it gets fast.

    24. Re:Had went on? by rrohbeck · · Score: 1

      That should have taken one page with one table that would take 30 seconds to digest.
      If they make me wade through 10 pages filled with data and bad grammar I expect some meat.

    25. Re:Had went on? by Anonymous Coward · · Score: 0

      Oi! Stop messing about on slashdot, you haven't got my fries yet.

    26. Re:Had went on? by Anonymous Coward · · Score: 0

      Just about anyone can mop floors, scrub toilets and empty garbages. There's nothing special about you, Calum.

    27. Re:Had went on? by Bert64 · · Score: 1

      Some places expect more than GCSE's, ie they expect you to have a degree or they won't even look at you...
      Many people are very smart, but either couldn't afford university, or went straight into work anyway, i was thinking elitist in this regard rather than plain GCSEs...

      That said, some people don't even have GCSE's for many reasons other than being too stupid to pass or too lazy to turn up, there are perfectly intelligent people who didn't pay any attention in classes due to peer pressure, there are those who missed the exams due to illness or similar, or who got far lower results than they should have due to other circumstances outside of their control, external pressures or major negative events during the exams and the last 2 years of school can affect peoples concentration, moving schools during the gcse years (last 2 years) can also have a major impact if the school teaches the curriculum in a different order, so you end up repeating some parts and missing some entirely.

      There are many reasons why someone would have little or no formal grades, and a lot of them are out of the control of the individual concerned.

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    28. Re:Had went on? by LingNoi · · Score: 1

      Many people are very smart, but either couldn't afford university

      Nonsense you can get a grant to go to university for free, you can also get a student loan from the government for up to £4000 a year for living expenses. The can afford excuse? There is none, however I do agree that not everyone needs a degree, however that's just a side effect of everyone having one, why take on the dross of society when another section actually put effort into their education.

      there are perfectly intelligent people who didn't pay any attention in classes due to peer pressure, there are those who missed the exams due to illness or similar

      Then you go to college and re-take them. Either you make the effort or you don't get a job because employers see your shitty effort.

      That's how the world works, no one owes you a god damn thing, you have to show you're willing to make an effort, not bothering to show up, not bothering to re-take shows only one thing. That's not being a snob, that's being smart and not being taken for a ride.

    29. Re:Had went on? by Anonymous Coward · · Score: 0

      And yet, based on your posts, you're very stupid.

      Kind of moots your whole point.

    30. Re:Had went on? by ShieldW0lf · · Score: 1

      The lack of proper spelling and grammar detracts from the credibility of the article. I mean, if the author doesn't care enough to fix obvious grammar errors, how can he be trusted to have testing methodology that doesn't contain gross errors?

      Clearly, those who are passionate about writing and grammar are the best people to go to for advice about the intricate nature of computer hardware. If you're not passionate about writing and grammar, clearly, you're an idiot whose ramblings don't deserve critical attention, but should simply be dismissed as the ramblings of an illiterate.

      I went to see a doctor the other day, and his penmanship was absolutely atrocious. Clearly, he doesn't pay attention to details, so I went to see someone else. I'm so glad I did... that person gave me a copy of their book, which was very well written. They advised me to start taking Echinacea, and if their medical advice is as good as their writing, I'm quite confident that my cancer will be gone soon.

      --
      -1 Uncomfortable Truth
    31. Re:Had went on? by Loibisch · · Score: 1

      Damn, a funny response and no mod points...haven't laughed as much in a looong while. ...reliable... *snickers*

  2. Re:Performance isn't its raison detre by Anonymous Coward · · Score: 5, Interesting

    Let's see how that statement works in this situation:

    It shouldn't be getting slower, but then again, performance isn't the reason Vista exists.

    If you really want performance, run FreeDOS. Otherwise, shut up and get used to progress.

  3. Ubuntu isn't getting slower, no. by Anonymous Coward · · Score: 0, Insightful

    Your laptop IS, on the other hand, getting OLDER, and while your hardware might not be changing the requirements put upon hardware are.

    Complete bullshit article, doesn't offer any useful information beyond a completely obvious conclusion -- the more features that are added to a given piece of software, the higher the demands on your PC. The only reason they've turned this into an "Ubuntu is getting slower" argument is precisely so that they can start debates like these and drag more people onto their ad laden site.

    Not a lot different from the tactics that Slashdot has been using for years, really.

    1. Re:Ubuntu isn't getting slower, no. by Anonymous Coward · · Score: 0, Insightful

      RTFA jackass. The laptop they used had a fucking core duo. It was a Lenovo T60.

    2. Re:Ubuntu isn't getting slower, no. by Timothy+Brownawell · · Score: 5, Insightful

      Complete bullshit article, doesn't offer any useful information beyond a completely obvious conclusion -- the more features that are added to a given piece of software, the higher the demands on your PC.

      I would think that those increased demands should be mostly in the form of slightly (a few MB) higher memory requirements to store the extra code for those features. Adding new functionality should not impact existing functionality. Haven't you heard of the zero-cost principle (idea from C++ and apparently Perl, "you don't pay for (as in take a performance hit from) what you don't use")?

    3. Re:Ubuntu isn't getting slower, no. by Anonymous Coward · · Score: 1

      Funny, replace "Ubuntu" with "Windows" in your statement, and it would closely resemble a crowd that has been chased out of the house.

    4. Re:Ubuntu isn't getting slower, no. by Sockatume · · Score: 4, Interesting

      It's not quite that simple. Performance in Java and media encoding was almost halved in the two newest versus the older versions of the OS. It's hard to imagine why that would be the case unless "more features" in Heron and Ibex are using up half the CPU time (and based on the other benchmarks, they ain't). I'd suspect test methodology rather than some oddity of OS performance but it's still something that needs to be addressed.

      --
      No kidding!!! What do you say at this point?
    5. Re:Ubuntu isn't getting slower, no. by DNeoMatrix · · Score: 3, Insightful

      This is the EXACT same reason people are having problems with vista. New OS = New requirements sometimes. If you criticize one for it you must criticize the other.

    6. Re:Ubuntu isn't getting slower, no. by gnud · · Score: 1

      A few MB extra code will cause cache misses, and thus slowdowns. That doesnt mean that there isn't some configuration problem with Ubuntu software, though.

    7. Re:Ubuntu isn't getting slower, no. by thePowerOfGrayskull · · Score: 1

      Core Duo... you mean the first generation that was succeeded by the Core 2 Duo in 2006? As I recall, they were steadily losing marketshare while the Core Duo was the best they had to offer.

    8. Re:Ubuntu isn't getting slower, no. by Clarious · · Score: 1

      Why modding parent down? He is right, they tested Ubuntu on one laptop.

    9. Re:Ubuntu isn't getting slower, no. by siride · · Score: 2, Informative

      Cache misses if the code is used. If it's not, it incurs no penalty except for the piddly bit of space it uses on the harddrive.

    10. Re:Ubuntu isn't getting slower, no. by KillerBob · · Score: 1

      On the other hand, that doesn't explain why different distros can be significantly faster on the same benchmarks. Admittedly, I'm biased when I say this, but I'm a long-time Slackware user, and more recently switched to Zenwalk for the package management system, but both of those distros completely blow Ubuntu out of the water in terms of performance. Switching from Win to Ubuntu is a big difference in speed of everything. Switching from Ubuntu to Zenwalk is that same difference all over again. Even when you're comparing apples to apples and using XUbuntu v. Zenwalk, the latter is noticeably less sluggish, and I'm running on a relatively modern system: Dell Insprion 1520 notebook... 1.66GHz Core2 Duo, 2GB of RAM, 7200rpm 120GB SATA hard drive, Intel 8945J wireless LAN, Intel sound, and a 256MB GeForce 7600M GT video card. Theoretically, it should be fast enough that Ubuntu would be zippy, no?

      Modern capabilities and demands on a system does *not* mean that it has to be bloaty or sluggish.

      --
      If you believe everything you read, you'd better not read. - Japanese proverb
    11. Re:Ubuntu isn't getting slower, no. by AmaDaden · · Score: 1

      I think the slowness is from the software changes Ubuntu has undergone. As I remember they changed the default audio driver recently, Java started using Iced Tea for 8.10, and they were using an ATI card with the new open source driver in 8.10. That's only what I can think of off the top of my head. From what I've seen Ubuntu has no problem switching to younger software if they feel it can improve the user experience and is the direction that the linux world is going. This is likely to cause some performance hits.

      Personally I think this article is making a mountain out of a mole hill. If you look at the numbers in it you'll see that Ubuntu has NOT been consistently getting slower. Every few distros something will cause a massive speed drop and then it gets better and levels off. For me I can honestly say every new version of Ubuntu feels faster. Losing speed is a bad thing but I don't really think Ubuntu has shown it.

    12. Re:Ubuntu isn't getting slower, no. by westyvw · · Score: 1

      Maybe to get more hits, but one of my biggest complaints with Ubuntu next to how fragile it is, has been how slow it is. Every time I try it I am disappointed and its getting worse. Debian has been very good to me, along with KDE.

    13. Re:Ubuntu isn't getting slower, no. by cawpin · · Score: 1

      Agreed. While I haven't installed Ubuntu on a wide variety of machines, I have installed the last three versions on a PIII 500MHz/512MB machine that was my server. The last release, 8.04, actually gave me a noticeable INCREASE in speed over 7.10. It was a simple distro upgrade so all the same functionality was there, it was just a bit quicker. Actually, I take that back. I have installed it on a good variety of machines. My mom's Dell (Celeron 2.0GHz/256MB), my new server (Dell dual core Pentium/1GB), mother in law's Dell (Core 2 Duo/1GB), my Macbook Pro (2.4GHz C2D/4 GB). The last release, 8.04, has run quicker on all of them.

    14. Re:Ubuntu isn't getting slower, no. by Anonymous Coward · · Score: 0

      I think you confuse the status quo (due to sloppy programming, bloating, disinterest in optimization) with how things can and should be. Only new features that for some reason have to continuously drain system resources and run in parallel to your usual activities will put higher demands on your PC. In general, there is no logical or engineering reason why a new feature should have higher demands than any older one.

    15. Re:Ubuntu isn't getting slower, no. by gnud · · Score: 1

      If routines grow with conditionals for extra functionality, cache misses will occur.

    16. Re:Ubuntu isn't getting slower, no. by Computershack · · Score: 1

      Install Ubuntu 8.10. Use it for a while. Then install OpenSuSE 11 or Mandriva. Then come back and say that Ubuntu doesn't crawl like a slug through molasses.

      --
      I only please one person per day. Today is not your day. Tomorrow isn't looking good either. - Scott Adams
    17. Re:Ubuntu isn't getting slower, no. by Anonymous Coward · · Score: 0

      Yes, and you know why that is? Because compiz has been introduced, and as most people running Ubuntu with a decent machine have also been enabling compiz for a while, a lot of the CPU is tied up by desktop flash rather than actual work.

      Guess what happens when you turn it off?

    18. Re:Ubuntu isn't getting slower, no. by naasking · · Score: 1

      Cache misses if the code is used. If it's not, it incurs no penalty except for the piddly bit of space it uses on the harddrive.

      Of course it can cause cache misses! Introducing new code reorders existing code which can easily cause cache misses in the instruction stream, not to mention alignment issues and cache misses on the data due to new fields required for the new features.

    19. Re:Ubuntu isn't getting slower, no. by Sockatume · · Score: 1

      Actually if you bother to read the article, the benchmark was on a fresh install of Ubuntu with only the relevant graphics card driver and the benchmarking app installed. And if compiz was the reason, then the performance drop would not be so sharply limited to java and media encoding.

      --
      No kidding!!! What do you say at this point?
    20. Re:Ubuntu isn't getting slower, no. by Anonymous Coward · · Score: 0

      Haven't you heard of the zero-cost principle (idea from C++ and apparently Perl, "you don't pay for (as in take a performance hit from) what you don't use")?

      The hilarious part about that is how C++ doesn't make you pay for the feature... if you use C++ exactly in such a way as not to inadvertently trigger the new feature. It's a real minefield of implicit behaviors; the way modern languages like Java force you to spell everything out is actually a good thing.

    21. Re:Ubuntu isn't getting slower, no. by jlarocco · · Score: 1

      Your laptop IS, on the other hand, getting OLDER, and while your hardware might not be changing the requirements put upon hardware are.

      The point of this benchmark is to test the SAME requirements on different software setups. The whole point is that the requirements are not changing, but the time taken is increasing.

      There's no reason the same task should ever get slower on the same hardware. It just doesn't make sense. If the kernel and compiler optimizations are getting better/faster, then they should be better/faster on older hardware as well as newer hardware.

      New tasks that require more memory or faster processors may not perform well on old hardware, but existing tasks should stay the same or get faster. If that's not the case, then there's a problem somewhere.

    22. Re:Ubuntu isn't getting slower, no. by pseudonomous · · Score: 1

      Certainly true, of course, in this case, for most benchmarks the slowdown is happening gradually, but happening at 6 month intervals instead of one massive performance drop about every four years (MS release cycle). You know the "if you stick a frog in boiling water it jumps out / if you slowly bring the water to a boil w/ the frog in it, the frog dies" effect? I think this is the proper Vista to Intrepid Ibix comparison.

    23. Re:Ubuntu isn't getting slower, no. by Risen888 · · Score: 1

      Not exactly. This may be a case of "Does more stuff=Uses more resources" while Vista is blatantly a case of "Does less stuff=Uses more resources."

      --
      Hey, I finally got my first freak! Took you long enough!
  4. Re:Performance isn't its raison detre by Anonymous Coward · · Score: 0

    Ohhh! BURN!

  5. What hardware? by el_chupanegre · · Score: 5, Insightful

    Were they testing each distribution on exactly the same hardware?

    If so, that sounds completely fair to me that it would be slower. Go and (try to) install Vista on a machine that originally came with XP (pre-SP1) and see how much slower it is. Is that a fair test either? I think not.

    As software gets more useful (and Ubuntu has, Vista not so much) it gets bigger and thus gets slower on the same hardware. Hardware advances at the same time though, so in real terms they keep about equal. When you test new software on old hardware of course it's going to be slower though.

    1. Re:What hardware? by Spazztastic · · Score: 1, Interesting

      Mod parent up. This article is flamebait.

      For once I agree with a Gentoo/Slackware zealot who posted "This is what happens when you put out pre-compiled kernels, it makes it too easy for stupid people." And this is how questions like this arise. If you want performance, compile your own kernel with only your optimizations, then come back to us.

      --
      Posts not to be taken literally. Almost everything is sarcasm.
    2. Re:What hardware? by Anonymous Coward · · Score: 0

      WRONG!

      Just because you add features it should NOT have to run slower.

      People who think this are either not programmers, or real lazy programmers.

    3. Re:What hardware? by lolocaust · · Score: 4, Insightful

      Read the article, please. The exact same releases of software (such as the LAME encoder) shouldn't have a 2-3x decrease in performance.

      --
      Why does my post history abruptly stop? I want to laugh at the stupid things I posted as a kid.
    4. Re:What hardware? by Sockatume · · Score: 5, Funny

      Indeed, it is common knowledge amoungst Real Programmers that you can run an arbitrarily large number of instructions per clock, allowing you to introduce entirely new functionality with zero performance hit. You just need to write everything in asssembler and have the right enchanted oils to annoint the heat sinks. By such a scheme CowboyNeal famously calculated the highest possible prime and now lives forever in a magic castle full of unicorns which shit rainbows.

      (Hey, he used an absolute, I'm entitled to extrapolate it to its logical implications.)

      --
      No kidding!!! What do you say at this point?
    5. Re:What hardware? by nfarrell · · Score: 2

      Completely true. For example, the Ibex has much better support for various hardware, and comes with the software and drivers to suit. By default, beryl etc. is enabled for many graphics cards. Gnome's network manager is there too, to support the GSM connections, etc. etc.

      Apart from the fact the LTS releases mean you get security updates for years for certain older versions, there are a host of flavours explicitly aimed at low-end hardware, such as xubuntu.

    6. Re:What hardware? by Anonymous Coward · · Score: 0

      I suppose the guys at phoronix would have made sure they did not have unwanted services (at least those that show up on 'top') running during benchmarks.

      This means that the active task would consume 95-99% cpu. So bloatware in new OSes have a max 5% impact. But still, according to the benchmarks, _compilation_ and MP3 encoding tasks were taking more that 20% longer.

      Maybe I have to test this myself, but I hope the tests were incorrect. If it is indeed true, what is more worrying is that Phoronix noticed this before the Ubuntu team/community. Community includes myself.

    7. Re:What hardware? by gbjbaanb · · Score: 5, Informative

      When you test new software on old hardware of course it's going to be slower though.

      That's hardly a given, lots of software gets better as it ages - new features are added, but also performance tweaks get added.

      The problem is that software should be getting quicker on the same hardware, the alternative is bloaty apps that no-one wants to use. See Vista for the ultimate conclusion to that. You don;t want Ubuntu to end up the same, so its good that someone is pointing out performance issues. Hopefully the next release will have a few of these issues looked at and improved.

    8. Re:What hardware? by blind+biker · · Score: 4, Insightful

      I heard this argument a wee too often. Maybe software should be more useful while at the same time NOT getting slower? Maybe that would be a good thing, as it would then run well on netbooks as well, what do you think?

      --
      "The agriculture ministry is not in charge of Gundam" - Japanese ministry official.
    9. Re:What hardware? by RAMMS+EIN · · Score: 5, Insightful

      You make it sound like it is inevitable and acceptable that newer software is slower than older software. I disagree. For one thing, one way to improve software is to make it faster. This is actually done sometimes. Secondly, even if you add features to software (which is another way to improve software), that doesn't have to make the software slower. In some cases, this may be inevitable, but in many cases it is not.

      I personally see computers, and software, as tools for making life more efficient. When software becomes slower, efficiency is actually lost. When this isn't offset by providing me with a more efficient work flow, I lose efficiency. Since efficiency is the main reason I use computers in the first place, this is a big deal.

      --
      Please correct me if I got my facts wrong.
    10. Re:What hardware? by Anonymous Coward · · Score: 0

      Who the fuck is talking about Vista? Are you so insecure and pathetic that you had to throw in a quick Windows bash to make yourself feel better about your operating system as it slows down? Get the fucking dick out of your ass and get a life.

    11. Re:What hardware? by Anonymous Coward · · Score: 2, Interesting

      CPU intensive tasks like LAME should be affected that much by the OS. It's all CPU driven. So it's obvious there are architectural differences that actually hinder regular application performance. I don't consider that an improvement in any way.

    12. Re:What hardware? by e2d2 · · Score: 2, Insightful

      That would be true if software was given 100% CPU devotion. But software doesn't operate in a bubble like that and hence other needs are given CPU time, in turn slowing things like the LAME encoder down.

      It's something worth noting though, it's a real performance hit and perhaps something can be done about it in future releases.

    13. Re:What hardware? by Drakonik · · Score: 1

      But here you have the problem of the conversation of mass and energy. If you add functionality, the energy put into performing this new function must come from somewhere. Either you must subtract a small portion of energy from other functions or add a bit of load to the CPU. Either way, you can't create a new feature that magically consumes no resources.

    14. Re:What hardware? by Ma8thew · · Score: 1

      When I installed Leopard on my Mac, if performence didn't improve, it certainly stayed the same.

    15. Re:What hardware? by andy19 · · Score: 1

      Go and (try to) install Vista on a machine that originally came with XP (pre-SP1) and see how much slower it is. Is that a fair test either? I think not.

      Go and (try to) install Vista on a machine that originally came with Vista. Fair test? I think not.

    16. Re:What hardware? by dintech · · Score: 2, Insightful

      I hope there's a middle-ground somewhere in your new world order for users who want stable performance from release to release without having to compile a kernel.

    17. Re:What hardware? by siride · · Score: 1

      Those code paths don't have to be executed. So yes, they take up RAM (or maybe not if they never have to get paged in), but they will not slow other things down by simply existing.

    18. Re:What hardware? by e2d2 · · Score: 1

      But if it's improved wouldn't it be _new_ software?

    19. Re:What hardware? by lowlymarine · · Score: 1

      I was wondering what was going on with 7.04 getting vastly different results from the later versions, and I think you've hit the nail on the head: Compiz starts being enabled by default as of 7.10, and somehow I doubt the writers of this totally legitimate *snicker* article bothered to disable it before benchmarking.

    20. Re:What hardware? by Drakonik · · Score: 1

      I know that much. But the problem is, eventually, it will have to run. The processing power has to come from somewhere. You can optimize as much as you like, but there's a finite amount of clock time.

    21. Re:What hardware? by not+already+in+use · · Score: 2, Insightful

      Well, so much for the "it runs great on old hardware" argument. A million lonely blogger's "Top 10 reasons to switch to Ubuntu" just became top 9 lists.

      --
      Similes are like metaphors
    22. Re:What hardware? by siride · · Score: 1

      They used plain out of the box installs for each test. Ubuntu hasn't seen a significant increase in the number of services over the past few releases (if it has indeed had any increases at all), so that shouldn't be a factor.

    23. Re:What hardware? by siride · · Score: 1

      No it DOESN'T have to run if that feature is never used. I don't know why this is so hard to understand.

    24. Re:What hardware? by Anonymous Coward · · Score: 0

      In principal, you're making a sound arguments, except in practice it's plain silly.

      XP was released almost 7 years ago.
      Hardy was released, what, all of 7 months ago?

      I'd expect to take a huge performance hit from running Vista on a 7 year old PC.

      But to take a huge performance hit on hardware that was up to spec 6-7 months ago? Something is very wrong here.

      It's pretty comical when people shit on Vista on the basis that pushing hardware requirements (and thereby pushing upgrades) every, what, 6-7 years, but then spit out arguments like this. It's wrong when MS does it, though it's fine and dandy and completely reasonable to upgrade your hardware every 6-7 months to keep up with Ubuntu?

      It's all for the good of freedom and open source, right?

    25. Re:What hardware? by Anonymous Coward · · Score: 0

      The question is, what processes are running in the later versions that are responsible for the extra cycles -- and, do I need / want what the extra parts are doing? And, do I have a choice to turn them off?

    26. Re:What hardware? by Anonymous Coward · · Score: 0

      I heard this argument a wee too often. Maybe software should be more useful while at the same time NOT getting slower? Maybe that would be a good thing, as it would then run well on netbooks as well, what do you think?

      Problem is, you're completely ruling out the implementation of a lot of useful features. Last version of our OS only ran one program at a time. Multiple programs at once use more resources, which slow down the machine, so why should we implement that? We need to run as fast as possible!

      Computing power is still increasing at a pretty handy rate. If you want speed on the same old junker you've been using since 1995 you're going to need to focus on using software that values speed over functionality. The newest flashy Ubuntu is not how you do that. Meanwhile, the rest of us can upgrade our hardware and use the fancy RAM/CPU/disk-guzzling feature-filled (or at the very least, shiny) stuff.

    27. Re:What hardware? by Anonymous Coward · · Score: 0

      It's interesting to see that every article which doesn't hype Linux gets down modded, even if the article has some serious points - which might help to improve Linux/Ubuntu in the long run.

      Denying is the default solution for the loosers.

    28. Re:What hardware? by David+Gerard · · Score: 2, Informative

      OS X regularly gets noticeably faster each release on the same hardware.

      Mind you, they did start from the horribly unoptimised dog called 10.0.

      --
      http://rocknerd.co.uk
    29. Re:What hardware? by KillerBob · · Score: 1

      It's called Zenwalk

      --
      If you believe everything you read, you'd better not read. - Japanese proverb
    30. Re:What hardware? by westyvw · · Score: 1

      Then why is it I have the same functionality, although through a more powerful desktop environment KDE, and use basically the same packages, except its Debian, and my favorite distro installed, loads, and runs faster?

    31. Re:What hardware? by Spazztastic · · Score: 1

      It's called Zenwalk

      I'll be trying this at home today :)

      --
      Posts not to be taken literally. Almost everything is sarcasm.
    32. Re:What hardware? by Anonymous Coward · · Score: 0

      [Advertisement on Camera I saw at the store the other day]
      NEW AND IMPROVED!
      NO BATTERIES REQUIRED!
      (Manaual Advance Film Cam)
      Funny that.

    33. Re:What hardware? by Hoi+Polloi · · Score: 2, Insightful

      That assumption, that software gets slower as newer versions come out, is the "Windows Effect". People have grown up with Windows bloatware and assume that behavior is the norm. They are more concerned with adding new bells and whistles and not revisting existing code. When I work on software releases one of the main things we do is not only add new functionality but improve the performance of existing code, especially by taking advantage of new hardware/db tech features.

      --
      It is by the juice of the coffee bean that thoughts acquire speed, the teeth acquire stains. The stains become a warning
    34. Re:What hardware? by nschubach · · Score: 1

      Considering pretty much everything goes through a compiler that converts it all to common instructions, can software ever be "new"?

      --
      Every time I start to have faith in humanity, I ruin it by driving to work between 7 and 8 am.
    35. Re:What hardware? by dogmatixpsych · · Score: 1

      Exactly. That is what usually happens when a new version of OS X is released. I've read many reports from many people that their machines seem to run faster when a new version of OS X is installed. Whether those improvements are real or just perceived, I don't know, but that's pretty impressive for Apple to be able to add features without significantly slowing (and many times speeding up) the computer.

    36. Re:What hardware? by Anonymous Coward · · Score: 0

      but then again encoding an mp3 in lame taking TWICE the time (60s to 120s!) shows there's something COMPLETELY wrong going on.

    37. Re:What hardware? by ari_j · · Score: 1

      I thought xubuntu was aimed at people who like XFCE. Oops.

    38. Re:What hardware? by badpazzword · · Score: 1

      Why bother adding features that nobody ever uses?

      And yes, adding lines of code that are never used does take some time and resources. Just to mention one, the OS still needs to read the extra lines of code from disk (which is pretty slow in comparison to execution) and to store them in memory.

      --
      When ideas fail, words become very handy.
    39. Re:What hardware? by Gulthek · · Score: 1

      That doesn't have to be.

      On my 800Mhz G3 iBook, OS X got faster and more responsive from Jaguar to Panther to Tiger.

      Sadly, Leopard doesn't support my hardware.

    40. Re:What hardware? by Chaos+Incarnate · · Score: 1

      If you're trying to benchmark the typical end-user experience, then it's completely fair—arguably expected—that they don't tweak the OS at all and run it as it comes out-of-the-box.

      --
      Benford's Corollary to Clarke's Law: "Any technology distinguishable from magic is insufficiently advanced."
    41. Re:What hardware? by msuarezalvarez · · Score: 2, Funny

      So adding features does not have performace implications provided those new features are never used. Great.

      Would you make us a list of software in the development of which you've been involved, so that we can avoid it?

    42. Re:What hardware? by pato101 · · Score: 1

      I'm guessing if they have left beagle/tracker on when doing the encoding. That might be a possible cause, if tracker/beagle is trying to index the resulting file which is large and constantly changing.

    43. Re:What hardware? by Sockatume · · Score: 1

      Another absolute! Obviously it's not completely wrong, for it is actually encoding the files as MP3s and not, say, summoning dark creatures with non-Euclidean geometries. If winged rats from hyperbolic space started appearing then yes, I would concede that something COMPLETELY wrong was going on.

      --
      No kidding!!! What do you say at this point?
    44. Re:What hardware? by morgan_greywolf · · Score: 1

      Furthermore, if you don't want the new functionality, don't use it. You can customize Ubuntu and simply not install the features you aren't using.

      Duh. Good luck getting this work with Vista. ;)

    45. Re:What hardware? by jefu · · Score: 1

      Even for experienced linux users (I've been using linux since the early 90s), compiling a kernel is not always fun. More than once I've had odd problems either with the configuration (sooooo many options, and deciding which ones are good can be more than a bit confusing) or with the install process. Sure, I then get a snappier kernel without all the cruft, but it is not a process I'd suggest for anyone not interested in such things.

      It might be nice for distros to provide an option to download and install kernels with good default options for specific classes of machines. If there were also a small application that could probe the hardware and use some expert-system type rules to specify the best one of those packages to install, the whole thing could then be automated so even newbies would be able to tune their kernel for better performance.

    46. Re:What hardware? by siride · · Score: 1

      The real cost of new features is when they are actually used. Which they may be, may not be. They may not require any extra overhead above what they already cost as is. The cache miss stuff is trivial on a modern desktop: not so much because computers are fast, but because programmers really don't do much to mitigate cache issues. But let's go more to the point. Let's say I have a command line program that does some sort of text manipulation, like sed. And let's say I add a new command line flag that makes sed do a completely different operation on its arguments. If that flag isn't used, the new code isn't used (except for the command-line option parser) and it's entirely possible that it may never even be paged in in the first place. Even if it is, it'll never make it to the cache. Yeah, it might change the cache usage of other parts a little, but that will be trivial compared to the cost of running sed on input using the old code. Let's say I have an option that modifies the behavior of the existing code. Okay, that'll make some changes. There will probably be more if-statements. This will cause some branch prediction failures and some cache misses. Certainly, no doubt. But again, compared to the cost of running the regular code on the input, it's going to be insignificant. I'm sure you can imagine some edge cases where you get pathological behavior. I highly doubt all Ubuntu software is running up against a conspiracy of edge cases all across the desktop stack. The point is, you can add features in a variety of ways (far more than I've mentioned here) without decreasing performance in any meaningful way. In fact, it's possible to add features while increasing performance. If the new features change the structure of the program in such a way that it's easier and more efficient to do the old stuff, then you will get a speed up. Refactoring can be a good thing.

    47. Re:What hardware? by vagabond_gr · · Score: 1

      The legend says that his program could reach 16.5 ilps (infinite loops per second).

    48. Re:What hardware? by Markspark · · Score: 1

      seriously, not all features are for everyone. I might use a feature which you don't. and i sure as h*ll don't use compiz.

      --
      i find your lack of faith in science disturbing!
    49. Re:What hardware? by Anonymous Coward · · Score: 0

      Faggot.

    50. Re:What hardware? by Anonymous Coward · · Score: 0

      Maybe software should be more useful while at the same time NOT getting slower? Maybe that would be a good thing, as it would then run well on netbooks as well, what do you think?

      Completely. Also, it should have an incredibly small footprint. And fit on a 1.2 floppy. I should also be able to access it from anywhere and have it completely secured. And I should be able to use it in space. And stop a meteor.

    51. Re:What hardware? by default+luser · · Score: 1

      CPU intensive tasks like LAME should be affected that much by the OS. It's all CPU driven. So it's obvious there are architectural differences that actually hinder regular application performance. I don't consider that an improvement in any way.

      Absolutely agreed. LAME is actually pretty memory-dependent, because your typical WAV file is several dozen MB in size. The shocking drop in memory bandwidth benchmarks mirrors the huge drop in LAME performance.

      What the hell did they change that hit the usable memory bandwidth so hard? Is this just a motherboard driver issue, or is this deeper? Unfortunately, since they did not try another motherboard (preferably non-Intel), we cannot begin to fathom what piece of the code is responsible.

      --

      Man is the animal that laughs.
      And occasionally whores for Karma.

    52. Re:What hardware? by mollymoo · · Score: 1

      There were big strides in OS X performance from 10 to 10.3, but that's more about the earliest versions being crappy than about the later versions being great. 10.3 to 10.4 was a bit better, but I didn't notice any difference when going from 10.4 to 10.5. Now it seems to be a case of more features without a slowdown, which is good, but not as good as more features and more speed. Hey! That cake I ate yesterday has gone!

      --
      Chernobyl 'not a wildlife haven' - BBC News
    53. Re:What hardware? by azgard · · Score: 1

      I believe it is not inevitable, but it's generally true that newer software is slower and no one I think knows what to do about it. The problem is that the modern pieces of software are large systems, and we don't know who to predict complexity of and optimize such large systems.

      I think most of the bloat is accumulation of very small problems, for example in libraries. These problems may seem fine, but can be amplified in another layer of indirection. One would need to optimize through all the layers, and it would make impossible to write a complex program (or at least to write it quickly).

      Maybe in the future, there will be compilers capable of that. But I don't see them now.

    54. Re:What hardware? by jonaskoelker · · Score: 1

      By such a scheme CowboyNeal famously calculated the highest possible prime

      I've got some special pot that can make them even higher...

    55. Re:What hardware? by setagllib · · Score: 1

      LAME also has significant IO speed and memory speed components, which seem to be slowed down in general in the test environment. Besides that, the new scheduler gives generally lower priority to batch tasks. And GCC might be worse. Compiz is running. So there are a few slowdowns that really stack up. I'm not sure if they account for the whole 50%, however.

      --
      Sam ty sig.
    56. Re:What hardware? by mgblst · · Score: 1

      It is about time. If they have to make it perfectly, then they probably could do this, depending on the functionality. I mean, if you are adding a file indexer, this is always going to slow down the computer. But for other stuff, there isn't always time to create the perfect software. Thing have to be released.

    57. Re:What hardware? by Draek · · Score: 1

      There's a matching set of quotes floating around the 'net that are apropos:

      There is no complexity problem in programming that cannot be solved by adding a layer of abstraction.

      There is no performance problem in programming that cannot be solved by removing a layer of abstraction.

      Now, it's easy to see why newer software is usually slower. First you start with a simple converter, say, from txt to rtf. Then you want to accept HTML files as input. Abstraction. Then output to PDF. Abstraction. Then separate the CLI from a library, so you can write a GUI in the future. Abstraction. Then a spell checker, just because everything has one now. Abstraction. And now, to do the one thing you originally could in a single step, your program has to go through a bazillion layers of abstraction, with the obvious performance hit.

      But on the other hand, your little program can do a lot more than it originally could, and if someone wanted to extend it to output to, say, PNG images, it'd be *much* easier for him now than if he were to start with the original codebase.

      Me, I'll take the performance hit. Computers are getting faster, and my list of "must-have" features is only getting bigger, but YMMV, depending on both the type of software and hardware in question.

      --
      No problem is insoluble in all conceivable circumstances.
    58. Re:What hardware? by Cato · · Score: 1

      It does run on low-end software with the right Ubuntu flavour - Ubuntu for up to 200 MB RAM, or Ubuntulite / Fluxbuntu for up to 100 MB RAM.

    59. Re:What hardware? by not+already+in+use · · Score: 1

      I can say without any hesitation that Ubuntu runs far slower than XP on a computer with 256mb ram. The linux community are a loud bunch of folks, and based on their ramblings I assumed that Ubuntu would run great on a comp with 256 meg ram, so I converted my brother over to Ubuntu. It was painfully and ridiculously slow. I would venture to argue that it was even slower than XP with a bunch of crapware installed. It is beyond me as to why. I thought at first video drivers, but either way in Linux or Windows, the video memory is shared. He stuck with it for about a month, but in the end asked me to put Windows back on it.

      --
      Similes are like metaphors
    60. Re:What hardware? by Cato · · Score: 1

      I meant to say Xubuntu for up to 200 MB RAM, but you are right - it does run quite slowly on such systems particularly for large apps like Thunderbird and Firefox, but once launched they run OK.

  6. Re:Performance isn't its raison detre by Linegod · · Score: 5, Insightful

    It's nice to see that the Ubuntu fanboys have moved so quickly to 'shut up and like it'.

    It took Windows fanboys a decade to get there...

    --
    -- I care not for your foolish signatures.
  7. Re:Performance isn't its raison detre by TheSpoom · · Score: 5, Funny

    See, there's this thing called an analogy. It's kinda like a car...

    --
    It's better to vote for what you want and not get it than to vote for what you don't want and get it.
    - E. Debs
  8. Maybe by nawcom · · Score: 1, Insightful

    Because Ubuntu uses generic kernel builds and starts up unneeded shit at boot time. You also have frontend apps for a lot of apps that don't really need it - that can explain the reason the memory is being eaten up. Suggestions? Learn how to compile a kernel, or use a distribution that doesn't have a list of memory eating apps specific to itself, like Slackware for example. I've never had issues with it, and I've gotten the kernel to finish booting in 6-7 seconds with only the device support and services i only need.

    Yeah I know - all these new Linux users don't like Slackware. It's so.. Linux like, and not Windows like. Perhaps Ubuntu can work on optimization and take care of the problem.

    1. Re:Maybe by nawcom · · Score: 1

      fyi to compare specs i reach the kdm login screen with wireless ip address found in 6-7 seconds with a dell latitude d420 (1.2ghz core 1 duo, 1 gig of memory).

    2. Re:Maybe by Anonymous Coward · · Score: 1, Interesting

      Bullshit. It might resume from standby that fast, but I have a D420 here and it takes that long just to post up and hit the bootloader.

    3. Re:Maybe by vigour · · Score: 2, Insightful

      Because Ubuntu uses generic kernel builds and starts up unneeded shit at boot time. You also have frontend apps for a lot of apps that don't really need it - that can explain the reason the memory is being eaten up. Suggestions? Learn how to compile a kernel, or use a distribution that doesn't have a list of memory eating apps specific to itself, like Slackware for example. I've never had issues with it, and I've gotten the kernel to finish booting in 6-7 seconds with only the device support and services i only need. Yeah I know - all these new Linux users don't like Slackware. It's so.. Linux like, and not Windows like. Perhaps Ubuntu can work on optimization and take care of the problem.

      Because Ubuntu uses generic kernel builds and starts up unneeded shit at boot time. You also have frontend apps for a lot of apps that don't really need it - that can explain the reason the memory is being eaten up. Suggestions? Learn how to compile a kernel, or use a distribution that doesn't have a list of memory eating apps specific to itself, like Slackware for example. I've never had issues with it, and I've gotten the kernel to finish booting in 6-7 seconds with only the device support and services i only need. Yeah I know - all these new Linux users don't like Slackware. It's so.. Linux like, and not Windows like. Perhaps Ubuntu can work on optimization and take care of the problem.

      what a way to waste my mod points but anyway here goes...

      There's no need to be so smug and condescending. If you're a supporter of F/OSS, then you should be happy people are trying out some distro. Of course you'll never get the same performance out of the one-size-fits-all approach of Ubuntu, but that's not the point of the more newbie friendly distros. The whole point of them (i'd include openSUSE, Mandriva and Scientific Linux) is to 'just work', or with as little mucking about in the terminal as possible. It takes time to learn the intricacies of a different OS, especially if you've never used any *NIX OS. Personally I started on a Sinclair Spectrum :)

      If you have the time, desire, or need to have a highly optimised kernel, the you have the choice to use something like Gentoo or Slackware. For the rest of us, precompiled binaries work fine, thank you (with exceptions of course), and that kind of negative attitude puts people off trying out, and learning about GNU/Linux.

    4. Re:Maybe by ruckc · · Score: 1

      I believe the majority of slow downs are the addition of new "fancy" services such as Tracker. When my new ubuntu installation has more processes running than a fresh windows xp laptop from dell something is wrong.

      Speaking of generic kernels... anyone had any luck building your own kernels in ubuntu. I usually do in gentoo but i am not sure on how ubuntu will handle it.

    5. Re:Maybe by nawcom · · Score: 1

      Yeah I know I was being a little bitch :) I think there has to be another technique of dealing with precompiled binaries in an organized fashion. I actually am happy that Linux is reaching a point where it "just works" - it just frustrates me at times when "it doesn't work" and I have to fix the issue. The last thing I can think of is I had to help someone out with getting their broadcom ucode installed. the binary package from the repository wasnt working, so i just gave him a shell script that took care of everything and did the dirty work. I had to upload a binary of b43-cutter though for him to grab via curl.

      But yes, i do admit being harsh and unfair in this post. perhaps i made this because my day is starting shitty, who knows :-P someone mod this person up.

    6. Re:Maybe by westyvw · · Score: 1

      Except that other distros arent as noticeably slow, even compiled for one size fits all. And when people try the new kid (Ubuntu) and it feels slow (Gnome? ok not that flame war), they might not come back.

    7. Re:Maybe by vigour · · Score: 1

      I feel your pain! and there are valid arguments about how bloated Ubuntu is getting. I stick to vanilla Debian because of many old dependencies of some computational physics packages I use.

    8. Re:Maybe by vigour · · Score: 1

      fair point :P

    9. Re:Maybe by bioglaze · · Score: 1

      I'm using Kubuntu and I always compile a kernel from kernel.org. I don't even make a .deb, just install it manually.

      --
      Who is John Galt?
    10. Re:Maybe by setagllib · · Score: 1

      He might mean from bootloader to KDM login, which is still absolutely absurd because just loading that much executable data off the D420's disk takes more than 10 seconds. I run a tight Debian Lenny on my D420 and it takes longer than that just for *console* login.

      --
      Sam ty sig.
  9. "That's quick" by G3ckoG33k · · Score: 1, Informative

    "That's quick" was the phrase my girlfriend after an update of Debian Sid to include KDE 4.1 and OpenOffice.org 3.0 from Experimental. "Wish my slow machine at work was this quick".

    You don't have to guess what OS she is using there...

    Anyhow, once you replace 3.5.x with KDE 4.1 you will notice a difference. At least I did. (No, I didn't read the article first... Bad boy.)

    1. Re:"That's quick" by iworm · · Score: 4, Funny

      How many geeks have heard such phrases: " "That's quick" was the phrase my girlfriend after..."

      Alas.

    2. Re:"That's quick" by apodyopsis · · Score: 1

      Agreed, from my gut level appraisal I thought KDE4.1 was a lot faster - but I returned to 3.5 because it seemed more polished, completed and well rounded. Little things like how items were implemented, widgets all seemed to lack some finesse.

      Having said that in another few months I will definitely make the switch - because I really liked what I saw and the direction it was heading in.

      You expect more functionality to run slightly slower - that is if you assume the original implementation was part way optimal, but I guess it depends on what kind of functionality has been added.

    3. Re:"That's quick" by doti · · Score: 1

      You're doing it wrong.

      You should just say "That's what SHE said."

      --
      factor 966971: 966971
    4. Re:"That's quick" by MightyYar · · Score: 1

      I usually hear, "What was that?"

      --
      W..w..W - Willy Waterloo washes Warren Wiggins who is washing Waldo Woo.
    5. Re:"That's quick" by Anonymous Coward · · Score: 0

      If you consider jacking it to be practice, he practices daily. Meanwhile, I'm tapping his girlfriend's ass. All that stuff she won't do with him -- bareback, anal, rimming, fisting, blumpies, etc.

    6. Re:"That's quick" by Anonymous Coward · · Score: 1, Funny

      You're doing it wrong.

      That's what she said.

    7. Re:"That's quick" by Anonymous Coward · · Score: 0

      ...geeks...my girlfriend..."

      hypothetically speaking, of course.

    8. Re:"That's quick" by Anonymous Coward · · Score: 0

      How many geeks have heard such phrases: " "That's quick" was the phrase my girlfriend after..."

      Alas.

      phrase my girlfriend after..."

      my girlfriend

      wut

  10. Re:Performance isn't its raison detre by Anonymous Coward · · Score: 1, Interesting

    Complete and utter crap.

    Most likely the performance decrease is to do with some unoptimized kernel feature. That kernel feature should be identified and optimized. Linux actually has a nice history of maintaining status quo or getting faster between releases. Atleast when you track it over say 10 releases.

    I don't understand this, quite frankly, Windows user mentality of just accepting the state of things.

    I would love to see Phoronix do a retest with some of the major patchsets removed and see if they can find the one or ones that cause performance decreases.

  11. Re:Performance isn't its raison detre by kv9 · · Score: 2, Funny

    that's OK, I'm sure the next version, Moronic Monicker, is going to be a LOT faster.

  12. How significant? by Max+Romantschuk · · Score: 1

    As we add complexity and layers of abstraction things tend to slow down in general. If hardware keeps up, and actual human productivity increases, do we have an issue?

    I'm all for lean and mean, but it's quite possible to optimize a distro for speed as well. Ubuntu getting slower is not a good thing, but slower is better than harder to use. Netbook distros can be optimized for the hardware in question, after all...

    It would be interesting to see how these tests perform across distros, or with a kernel optimized for certain tasks. (Ubuntu Studio for example has a RT-optimized kernel to keep audio from skipping. In theory at least...)

    --
    .: Max Romantschuk :: http://max.romantschuk.fi/
    1. Re:How significant? by siride · · Score: 1

      They haven't added any layers of abstraction over the past few releases. That isn't the problem anyways. I run Gentoo with the same bits of software, but it's not slow at all. And I'm not just talking about boot time. I mean, the UI is actually snappy and programs start up quickly. I don't know what Ubuntu does to make it so shitty, but they have done a really good job of it. Fedora on our workstations at work also seems to have suffered some severe performance regressions over the past few releases. Fedora 7 was pretty fast when we put it on. Now, at Fedora 9, the UI is painfully laggy and program startup is slow.

    2. Re:How significant? by samkass · · Score: 1

      I'm glad someone made the point. It really is irrelevant how fast the OS performs microbenchmarks. What matters is how fast the user gets things done. If you spend all day encoding MP3s so be it. But for a lot of people, a kernel that's half as fast but makes some complex things simple is the way to go.

      Anyway, that's Apple's philosophy, and why you see Apple not caring so much about kernel benchmarks. That being said, every version of MacOS X has been faster than the last one the same hardware despite adding new functionality. One can argue that that just shows how Godawful slow 10.0 was, but all the releases have been very usable and made the user very productive.

      --
      E pluribus unum
    3. Re:How significant? by RAMMS+EIN · · Score: 2, Insightful

      ``As we add complexity and layers of abstraction things tend to slow down in general. If hardware keeps up, and actual human productivity increases, do we have an issue?''

      You have that exactly right. Software getting slower is bad, but it's ok if it is offset by other changes, such as faster hardware or new, more efficient ways to perform tasks. In the end, it's our productivity that counts. Now the real question is, how do we measure that, how has it developed over time, and what changes have had the greatest impact (both positive and negative) on it?

      --
      Please correct me if I got my facts wrong.
    4. Re:How significant? by PeterBrett · · Score: 1

      Fedora 7 was pretty fast when we put it on. Now, at Fedora 9, the UI is painfully laggy and program startup is slow.

      Well, let me add another datum: I have a Fedora 7 system and a Fedora 9 system sitting in front of me. Fedora 9 is definitely faster to boot and the UI feels generally snappier. *shrug*

    5. Re:How significant? by siride · · Score: 1

      We have sub $500 Dell workstations with integrated Intel graphics. That may be the main problem.

  13. xubuntu by kisak · · Score: 3, Insightful

    Would have been interesting to have the same benchmarks for Xubuntu, since that is the distribution that is targeted for computers where performens increase/decrease is very noticable.

    --

    --- guns don't kill people, people with guns kill people ---

    1. Re:xubuntu by Mr.Ned · · Score: 5, Informative

      Xubuntu's performance targeting appears limited to choice of desktop environment, which was a small component of what these benchmarks tested. The big performance increases the article talks about were in databases, compilers, encryption, memory access, and audio/video encoding/decoding, none of which really have much to do with the desktop environment.

    2. Re:xubuntu by t482 · · Score: 1

      I just upgraded an Thinkpad x24 yesterday. My experience is that Xubuntu saves about 130 Mb in RAM usage. The laptop has 640 Mbs RAM and under vanila Gnome in was 90% used on boot. Under Xubuntu it was about 68% used. SWAP was 1.3 GB on each. And neither were swapping.

      Seems heavy to me - but still usable. Maybe I need to look at using a lighter version of the windowing manager or migrating back to 6.10.

    3. Re:xubuntu by jedidiah · · Score: 1

      Every little bit of crap that's running at the same time as
      your compile or your transcode is going to impact your
      "benchmark". This is especially true if the "extra crap"
      manages to hammer a common bottleneck.

      That said... the speed of LAME encodes ultimately doesn't
      mean a whole lot. The parts that the end user directly
      interact with are what matter. The desktop itself shouldn't
      be seen to slow down, especially when that slow ass LAME
      encode (and 3 of it's friends) are running.

      --
      A Pirate and a Puritan look the same on a balance sheet.
    4. Re:xubuntu by BlackCreek · · Score: 1

      Thw slow down since Feisty I am aware of has nothing to do with desktop environment: https://bugs.launchpad.net/bugs/131094

    5. Re:xubuntu by msuarezalvarez · · Score: 1

      Where did you get the 90% figure? top? ps? free?

    6. Re:xubuntu by Anonymous Coward · · Score: 0

      There was also specific GTK tests, however, they were tested on binary-blob ATI drivers, which could be a source of problems.

    7. Re:xubuntu by pizzap · · Score: 1
      Slowdowns most likely have their cause in:
      • kernel
      • gcc
      • trackerd
      • pulseaudio

      A clean X session might give more accurate results.
      Sure, trackerd, deskbar-applet and pulseaudio are memory-/cpu intensive, but they do provide a richer desktop experience and can be turned off easily. Kernel and gcc changes on the other hand have nothing to do with the desktop environment and are more difficult to change.

    8. Re:xubuntu by setagllib · · Score: 1

      Try Debian Lenny with xfce. Mine boots to under 30 MB RAM for console login, 160 MB RAM for GUI, which is pretty heavy but almost all of that is just the X server which is getting extremely large. Console login is still rathger small and that's including a couple of services in the background, so it's not like the kernel and shell are bloated.

      --
      Sam ty sig.
    9. Re:xubuntu by setagllib · · Score: 1

      That bug reports seems to only apply to upgraded installs, not fresh installs. Perhaps it doesn't address the issues in this benchmark.

      --
      Sam ty sig.
    10. Re:xubuntu by BlackCreek · · Score: 1

      That bug report also applies to fresh installs. If you read the whole thing (which is fairly long, I know); you will see that the problem also hits new installs and different file systems.

    11. Re:xubuntu by setagllib · · Score: 1

      Thanks, and my apologies for tl;dr behavior.

      --
      Sam ty sig.
  14. Real men by Anonymous Coward · · Score: 3, Funny

    Real men don't upgrade their OS for exactly this reason. In fact, we don't even use OSes. To get maximum performance we write all operations directly to RAM in machine code, while the machine is running, using a needle and a car battery.

    1. Re:Real men by Tridus · · Score: 4, Funny
      --
      -- "So they told me that using the download page to download something was not something they anticipated." - Bill Gates
    2. Re:Real men by Anonymous Coward · · Score: 1, Funny

      You had a battery? Spoiled brat.

    3. Re:Real men by ari_j · · Score: 3, Funny

      A car battery? Real men wear wool socks and use static electricity.

    4. Re:Real men by Anonymous Coward · · Score: 0

      Back then I wore an onion on my belt, which was the style at the time...

    5. Re:Real men by Anonymous Coward · · Score: 0

      I didn't know that real men could post as Anonymous Coward.

    6. Re:Real men by ipb · · Score: 1

      Ha, Real men stand outside in a thunderstorm waiting for the lightning strike.

    7. Re:Real men by Anonymous Coward · · Score: 0

      seriously real men can change the course of the time space continuum, by scratching their heads.

  15. software versions? by LingNoi · · Score: 1

    Some of theses tests such as the SQLite test I am wondering if they used SQLite within ubuntu or they build and run it on the system they were testing.

    This matters because Ubuntu comes with different versions of SQLite.

    However if there is a problem then I hope they report it on launchpad. I have noticed any slowness myself.

    1. Re:software versions? by Anonymous Coward · · Score: 1, Informative

      They were using the Phoronix Test Suite. By saying they ran 1.4.0 Beta 1 it can be found all of the versions they were using. The Phoronix Test Suite syncs to specific versions and builds them from source no matter the OS.

  16. Security Patching? by Ironsides · · Score: 4, Informative

    Ok, the article completely ignores this (as do most of the above posters it appears). Each version of Ubuntu tested seemed to have different kernel builds. How much of the slowdown is due to the kernel being patched for security and bugs and how much is due to the software that has been added?

    --
    Fly me to the moon Let me sing among those stars Let me see what spring is like On jupiter and mars
    1. Re:Security Patching? by Atzanteol · · Score: 1

      Why would they bring it up? They did a test that shows newer Ubuntu releases are slower than old ones. Does the reason matter?

      --
      "Ignorance more frequently begets confidence than does knowledge"

      - Charles Darwin
    2. Re:Security Patching? by Anonymous Coward · · Score: 0

      Yes, the reason does matter. They are pitching the statistics to imply that the newer releases are slower due to bloat in the distribution or something similar. However, if it is due to security patches in the kernel and elsewhere, it is something that is beyond the control of the Ubuntu packagers. Bounds checking, to prevent buffer overflow, is an example of a patch that will slow down the kernel.

    3. Re:Security Patching? by thePowerOfGrayskull · · Score: 4, Interesting
      I think that's part of the point, yes? The concern is that for whatever reason - whether it be the new scheduler system in the kernel or the bloatware that Ubuntu includes, performance is degrading.

      Good follow up would be to figure out specifically where - is it due to kernel changes? Then the problem may not be Ubuntu...

    4. Re:Security Patching? by LWATCDR · · Score: 5, Interesting

      I would like to see if this is an Ubuntu issue or Linux in general.
      What about Fedora, OpenSuse, and Debian? How do they compare to Ubuntu?

      --
      See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
    5. Re:Security Patching? by recoiledsnake · · Score: 1

      Slowdowns due to security didn't prevent everyone here from berating Vista for being slow.

      --
      This space for rent.
    6. Re:Security Patching? by Atzanteol · · Score: 0, Flamebait
      No, it does not matter. The point of their test was not to determine cause. There could be thousands of various causes.

      What they showed is that it is getting slower. Period. Their research appears to be correct and accurate. If it's purely caused by bounds checking in the kernel then so be it. But that's for somebody else to show. But until it is proven this is only a hopeful assumption by people who can't stand hearing bad things about their pet operating system.

      --
      "Ignorance more frequently begets confidence than does knowledge"

      - Charles Darwin
    7. Re:Security Patching? by Aphoxema · · Score: 1

      Most of the slowdown was UAC asking users for the n-thingmillionth time what the Hell the user wants to do, "(A)llow virus to destroy computer / (D)on't run this program that you really want to use"

      --
      "Most people, I think, don't even know what a rootkit is, so why should they care about it?"
    8. Re:Security Patching? by Yahma · · Score: 1

      The article does not completely ignore the issue of kernel versions. In fact, they noted the different kernel versions in use between the various versions of Ubuntu they tested.

      It is likely, that regressions in performance were introduced in new Linux kernels.

      Zbr noted huge regressions in tbench benchmark resulting in a loss of 100MB/sec performance between Kernel 2.6.23 and the latest 2.6.27 on the same exact hardware!

      It was reported recently that tbench has a long history of regressions, started at least from 2.6.23 kernel. I verified, that in my test environment tbench 'lost' more than 100 MB/s from 470 down to 355 for 8 threads between at least 2.6.24 and 2.6.27. 2.6.26-2.6.27 performance regression in my machines rougly corresponds to 375 down to 355 MB/s.

    9. Re:Security Patching? by funfail · · Score: 1

      Then why is the title "Is Ubuntu Getting Slower"? It should be "Is Linux Getting Slower" or "Is Ubuntu's Slowness Comes from Linux Kernel".

    10. Re:Security Patching? by jc42 · · Score: 1

      They did a test that shows newer Ubuntu releases are slower than old ones. Does the reason matter?

      The reasons matter if you view the slowdown as a problem that you want to fix.

      About 10 years ago, a bunch of guys in the local linux/unix users group noticed that the latest linux was noticably slow on several popular motherboards that used a certain chipset. One guy managed to extract the boot ROM and they analyzed it. They discovered a little loop that went through the kernel looking for a specific short sequence of instructions, and if found, it turned off the CPU's cache.

      They found this sequence in the code generated by gcc in a specific linux kernel routine. They edited it, making a few inconsequential changes (moving around commands whose order didn't matter), recompiled - and that kernel ran a lot faster.

      They spread the word that this boot ROM was sabotaging the latest linux kernels, that routine was modified in the archived, and machines sped up when upgraded.

      It's useful to be on the lookout for things like this. All sorts of things can cause slowdowns. One common reason is security enhancements, of course. Added features often add code to toplevels, slowing everything down. And we also have to be on the lookout for sabotage by manufacturers and OEMs who are trying to bias the results of things like benchmarks, to make one vendor's software look better than others'.

      --
      Those who do study history are doomed to stand helplessly by while everyone else repeats it.
    11. Re:Security Patching? by Cygnus78 · · Score: 1

      It did not _completely_ ignore it, it mentions on the last page that many changes have been made to the kernel.

    12. Re:Security Patching? by knails · · Score: 1

      Yes, the reason does matter. On the surface, the tests were done to test the speeds of Ubuntu releases. But once you find out that newer ones are slower, then the reason is important. If something isn't working right which causes the slow down, then the community will want to fix it.

      --
      "I disapprove of what you say, but I'll defend to the death your right to say it" -Voltaire
    13. Re:Security Patching? by mollymoo · · Score: 1

      They are pitching the statistics to imply that the newer releases are slower due to bloat in the distribution or something similar.

      They're not implying, you're inferring.

      However, if it is due to security patches in the kernel and elsewhere, it is something that is beyond the control of the Ubuntu packagers.

      Which version of an upstream package Ubuntu decides to include is fully in their control. Previous versions of software often get the same security patches as the latest version and Debian have demonstrated that it's possible to maintain older versions and backport security fixes.

      --
      Chernobyl 'not a wildlife haven' - BBC News
    14. Re:Security Patching? by Anonymous Coward · · Score: 0

      Or the title should be "Benchmarks indicate that newer versions of Ubuntu Linux are performing slower than older versions".

    15. Re:Security Patching? by recoiledsnake · · Score: 1
      That really shows how much you know about the architectural changes from XP to Vista. For one, from your GP post:

      Bounds checking, to prevent buffer overflow, is an example of a patch that will slow down the kernel.

      A similar thing was made compulsory in Vista and wasn't turned on much in XP.

      --
      This space for rent.
    16. Re:Security Patching? by thePowerOfGrayskull · · Score: 1

      I disagree; the testing was done at the system level - so the title of "Ubuntu getting slower" was accurate from the perspective of the tests performed.

    17. Re:Security Patching? by Anonymous Coward · · Score: 0

      Interesting you should pose that question.
      I've noticed slowness in disk I/O and glxgears performance from 2.6.22.9 and 2.6.26.3

      2.6.26.3 is slower than the older kernel. I was going to git-bisect the kernel and see at what point the issue appears, but not had time yet.

      Everything else has remained the same (besides some minor kernel options that are non-existant in newer kernels etc).

    18. Re:Security Patching? by Atzanteol · · Score: 1

      The reasons matter if you view the slowdown as a problem that you want to fix.

      Which they don't. That wasn't the point of their test (and my point). It's up to others to use this information. The person I was responding to was making excuses for the slow down ("Oh, it's probably because it's more secure!").

      --
      "Ignorance more frequently begets confidence than does knowledge"

      - Charles Darwin
    19. Re:Security Patching? by Anonymous Coward · · Score: 0

      i dont even think its the kernel i think it has something to do with gnome. Kubuntu has run about the same speed through all of the versions i have tried (6.04 - present). The gnome versions on the other hand have been getting slower and slower with each new release. I had decided to ditch kde all together and do all of my development using gnome ( for gtk programming) but ubuntu 8.04 was so slow it was almost unusable and thats even after i turned off all of the desktop effects. I installed kubuntu 8.04 and it runs just like feisty did. And before you even ask these are clean installs on the same laptop.

    20. Re:Security Patching? by Anonymous Coward · · Score: 0

      That really shows how much you know about the architectural changes from XP to Vista.

      I'm sure it does, but don't be disappointed, there's just some things that are really worth not knowing, like what urine tastes like, what it feels like to get mauled by a bear, or exactly what your parents do when they lock the bedroom door.

      I somehow equate Vista to all of these.

    21. Re:Security Patching? by WhiteHorse-The+Origi · · Score: 1

      Every year these articles come out and every year people find "issues" with the tests and every year it's discovered the testers work for Microsoft. Get over it. Ubuntu has stellar performance on all my systems and with every release I add more programs and features. I have not noticed any performance issues whatsoever, only performance gains. Well, except for that pesky Intel graphics card...

  17. Slower? by Anonymous Coward · · Score: 0

    I'm running the 8.10RC on my laptop and i feel the new kernel is much more responsive than previous versions. Gnome is becoming more robust, but thats the natural order of things.

  18. Look carefully at the power management by Peter+Desnoyers · · Score: 5, Interesting

    If you look closely you'll notice that (a) the benchmarks were run on a Thinkpad T60 laptop, and (b) there were significant differences on some benchmarks like RAM bandwidth that should have little or no OS components.

    This sounds to me like the power management was dialing down the CPU on the later releases...

    1. Re:Look carefully at the power management by trumplestone · · Score: 5, Interesting

      Mod parent up.

      Many of the benchmarks (such as the lame, Java and RAM bandwidth benchmarks) are CPU-bound, and will run the majority of time in userspace. As the kernel should only be invoked for timer ticks, interrupts, TLB misses, etc (which would probably account for less than 1% of the CPU time), and change to the kernel should have minimal impact on the benchmarks.

      The parent's comment that power settings have been misconfigured sounds spot-on.

    2. Re:Look carefully at the power management by jabithew · · Score: 1

      Interesting thought. Mod parent up. Did they check the power settings?

      --
      All intents and purposes. Not intensive purposes.
    3. Re:Look carefully at the power management by Cthefuture · · Score: 2, Informative

      Yes! In particular check the "ondemand" CPU scaler. That thing just doesn't work very well. It takes too long to trigger the higher clock speed and if you have multiple CPU's and/or are running lots of quick processes then the clock will constantly be shifting between speeds. This totally kills the performance.

      I turned off the CPU scaling on my Ubuntu workstation and I disable it on my laptop when I need maximum performance.

      This can be fixed with two changes to the ondemand profile. First it should bump all CPU cores to the higher speed no matter how many processes/threads are requiring performance. This is necessary because the kernel shifts threads between cores and you don't want to keep switching speeds on the various cores as it does this. Second is to add a delay before dropping the speed of a CPU. This allows time so that new threads/processes have full speed immediately.

      --
      The ratio of people to cake is too big
    4. Re:Look carefully at the power management by Anonymous Coward · · Score: 0

      Why is this modded interesting?
      Did everyone not read the first page of the article where they said that "Enhanced Intel SpeedStep Technology (EIST) were disabled during testing"?

    5. Re:Look carefully at the power management by Bert64 · · Score: 2, Informative

      Most CPUs cant allow cores to run at independent speeds...
      On the other hand, AMD quad cores do, and i'm glad to have one core running full speed processing a single threaded program, and the other 3 cores as slow as possible to handle the background OS tasks..

      --
      http://spamdecoy.net - free throwaway anonymous email - avoid spam!
    6. Re:Look carefully at the power management by jabithew · · Score: 1

      They may have done it wrong? It doesn't matter that the article says they disabled it, the info still doesn't add up.

      --
      All intents and purposes. Not intensive purposes.
    7. Re:Look carefully at the power management by Anonymous Coward · · Score: 0

      In that case it sounds like a bug, as it shouldn't be dialing down on AC power when something CPU intensive is running.

      Looks more like a background process that is causing a cache trashing that considerably slows down the benchmark.

      For graphic benchmarks, yep, probably a binary radeon driver.

    8. Re:Look carefully at the power management by wirelessbuzzers · · Score: 1

      They said they thought it was a compiler bug. But I think they're wrong. It would be a pretty severe compiler bug to give ~50% performance drop in both Java- and C-based benchmarks. But you'd expect almost exactly 50% performance drop in I/O-bound benchmarks if the CPU and FSB were throttled to half speed...

      --
      I hereby place the above post in the public domain.
    9. Re:Look carefully at the power management by CAIMLAS · · Score: 1

      Yep. Unless the processor is being utilized at or near 100% on 8.04, the P3M 1.2GHz in my Thinkpad X30 does not leave the 'powersave' mode - which downcycles to ~800Mhz. This happens even when the power adapter is plugged in.

      As far as actual, realized performance, the biggest thing it did was make the system slightly more latent than it should have been when doing something while something with a high CPU load ran. In this case, I noticed it while running gxemul (which uses a fair amount of CPU) and web browsing. I should note that I ran the laptop for a good 3+ months with Ubuntu 8.04 (fresh install, previously running debian) without noticing this.

      On a system which has more than one CPU scaling frequency alternative, I can see how this would (directly) downwardly impact performance of applications. I'm not sure what the newer CPUs have for available power states, but I suspect it correlates closely to 1/4, 1/3, and 1/2 maximum designed clock.

      For what it's worth, I 'fixed' this glitch on my system by adjusting the over-aggressive processor down-throttling. All I did was:

      echo 1 > titan:/sys/devices/system/cpu/cpu0/cpufreq/ondemand/ignore_nice_load

      Now the system will cycle up under a heavy load (say, over 20% - I've not actually tested it) CPU load, such as under common non-youtube browsing activities, programming, word processing, and the like.

      --
      ~/ssh slashdot.org ssh: connect to host slashdot.org port 22: too many beers
    10. Re:Look carefully at the power management by Guspaz · · Score: 1

      Furthermore, many of the benchmarks show things getting faster in newer releases, not slower. In many benchmarks, 7.04 is a distant last place. So it's not so cut and dry. It seems that if the benchmarks were done correctly (as in, figuring out why memory bandwidth was affected), the opposite of the claimed findings might be found to be true.

    11. Re:Look carefully at the power management by Anonymous Coward · · Score: 0

      I have experienced this problem also with ubuntu 8.04 + lenovo laptop, I had to manually disable the frequency scaling to achieve adequate computing performance. Using on-demand frequency scaling option didn't help.

      The trick was to execute "sudo chmod +s /usr/bin/cpufreq-selector" and after that it was possible to manually select the frequency from frequency scaling applet.

    12. Re:Look carefully at the power management by lorenzino · · Score: 0

      Or .. wait.
      Maybe they just configured the power settings right.
      Just think about it. It's a laptop, and on highly stress full thing they usually heat up a bit.

      A good interface with the hardware, acpi and whatnot would sensor this and slow down the cpu.

      Was the power cable connected ?
      Because in that case you can start talking about battery savings ..which are supposed to get better and better. :)

    13. Re:Look carefully at the power management by Max+Littlemore · · Score: 1

      But you'd expect almost exactly 50% performance drop in I/O-bound benchmarks if the CPU and FSB were throttled to half speed...

      That's exactly what I get on my lappy.

      An easy thing to do to test this kind of issue is install Ubuntu-Tweak which gives you, amongst other things, an easy way to adjust power management on the fly. Set the power slider down and the policy to power save and encode something. Then set the slider up and the policy to "power" and encode the same thing, observing the cpu temp suddenly sky rocket.

      I tend to set the policy to on demand - giving me good performance when I need it but also maximising power savings. Sometimes when I need to do something intensive, I turn things up. Gotta love that tool.

      --
      I don't therefore I'm not.
    14. Re:Look carefully at the power management by Anonymous Coward · · Score: 0

      The article notes that SpeedStep is disabled so the processor likely isn't running at a lower speed at any point during the testing.

    15. Re:Look carefully at the power management by Magada · · Score: 1

      I'm curious as to what you think this accomplishes, beside putting undue thermal stress on your processor.

      --
      Something bad is coming when people are suddenly anxious to tell the truth.
    16. Re:Look carefully at the power management by ChuckDriver · · Score: 1

      This sounds to me like the power management was dialing down the CPU on the later releases...

      "Compiz and Enhanced Intel SpeedStep Technology (EIST) were disabled during testing."

      This sounds to me like they controlled for power management when testing.

  19. Re:Performance isn't its raison detre by pipatron · · Score: 2, Interesting

    It's more likely the Ubuntu/GNOME moving all apps to run in mono. I doubt the kernel have anything to do with this.

    --
    c++; /* this makes c bigger but returns the old value */
  20. Ubuntu? No way. by Scholasticus · · Score: 3, Funny

    "Ubuntu" -- an African word, meaning "Slackware is too hard for me".

  21. Comment removed by account_deleted · · Score: 5, Insightful

    Comment removed based on user account deletion

  22. I'm not convinced they know what they're doing by mlwmohawk · · Score: 4, Interesting

    Some of the benchmarks were hardware testing, and those showed variation. They should not, unless the compiler changed the algorithms used to compile the code between distros.

    Benchmarking a multi-tasking system like Linux is a tough thing to quantify. The Linux kernel recently had a big scheduler change, this alone could account for shifting benhmark numbers. It may not actually "slowing down," but running multiple programs more evenly. The effective work is the same or better, which would mean "faster," but an almost useless benchmark would look slower.

    1. Re:I'm not convinced they know what they're doing by Anonymous Coward · · Score: 0

      Indeed, but supposedly the rest of the system was mostly idle, it shouldn't have slowed down a CPU-bound calculation by 50%, otherwise it's a scheduler regression.

      Something else is happenning, but they didn't even try to deduce what was the problem.

      I would: try different kernels, try in text mode without all those pesky gnome processes or binary ati drivers. I'd use a same binary and supporting libraries on all distros ( to check for gcc variations), run without acpi, etc.
      Finally, they should test on desktop machine, since I don't even believe that ACPI, on Linux, will reliably turn off all powersaving modes (so that may vary between kernel versions).

      But Windows mentality seems to be copied all over those benchmarks (meaning, they don't mess with the system).

    2. Re:I'm not convinced they know what they're doing by mlwmohawk · · Score: 2, Interesting

      Indeed, but supposedly the rest of the system was mostly idle, it shouldn't have slowed down a CPU-bound calculation by 50%, otherwise it's a scheduler regression.

      The supposition is not part of the facts. My point was there are some benchmark number that should have remained constant but showed variability.

      The benchmark is utterly useless until they can explain the variability in tests that should be constant.

  23. Considering I got Hardy running with 256mb RAM by AlbinoClock · · Score: 1

    I'd say it's doing ok in terms of performance. So maybe it won't run on a toaster anymore. Oh no.

  24. Re:Performance isn't its raison detre by Anonymous Coward · · Score: 0

    No, he's talking about linux. Linux Is Not UNix, remember? Stop confusing the two.

  25. Snap Son! by Anonymous Coward · · Score: 0

    Canonical just got told!

  26. Re:Cars! by TaoPhoenix · · Score: 5, Funny

    Analogies are like matchbox cars full of chocolates... you never know how much spillover chinese paint you're going to get.

    --
    My first Journal Entry ever, in 8 years! http://slashdot.org/journal/365947/aphelion-scifi-fantasy-horror-poetry-webzine
  27. There's no spoon. by draxredd · · Score: 1

    The problem is not speed or benchmark scores.
    The problem is usefulness and usability.

    we tend to focus on the former because it is "objective" while the later is highly subjective.

    --
    --- Back to the trees, back to the trees !
  28. I see what you did there.... by paultag · · Score: 2, Interesting

    I may be a bit biased here. Full disclosure: I am an Ubuntu Member, User, Abuser. I think that Ubuntu is one hell of a Distro, and GNU/Linux is one hell of an OS. Ubuntu, however is not geared for the market where we squeeze every CPU cycle we can. For that you have to do a _LOT_ of cleaning, replace your kernel to something a bit more fit for a server environment. Ubuntu is, and will always be a distro that is Easy to use first, even if that comes at the expense of some kruft. Each distro is becoming more bloated, but one great feature in Ibex ( 8.10 ) is the "System Cleaner" ( for all you GNOME users, Applications --> System Tools --> System Cleaner ) that checks for unused packages. This may not be a whole lot, but before bashing speed, or claiming its fat, take a hard look at the distro really.

    --
    This is not a viral sig. Copy it at your peril.
    1. Re:I see what you did there.... by recoiledsnake · · Score: 1

      I think that Ubuntu is one hell of a Distro, and GNU/Linux is one hell of an OS. Ubuntu, however is not geared for the market where we squeeze every CPU cycle we can. For that you have to do a _LOT_ of cleaning, replace your kernel to something a bit more fit for a server environment. Ubuntu is, and will always be a distro that is Easy to use first, even if that comes at the expense of some kruft. Each distro is becoming more bloated...

      So... You mean to say Ubuntu is becoming like Windows then?

      --
      This space for rent.
    2. Re:I see what you did there.... by Anonymous Coward · · Score: 0

      Yes, insofar as it is intuitive, Just Works when you plug your brand new camera in, has easy-to-install software, etc.

    3. Re:I see what you did there.... by AvitarX · · Score: 1

      Mod parent up.

      I rant the tool, it identified lack of relatime being set, and things went WAY faster while disk activity happening.

      --
      Wow, sent an e-mail as suggested when clicking on "use classic" banner, and got a fast response that addressed my msg
  29. wut about 6.06 by Anonymous Coward · · Score: 0

    is 6 any good anymore? or should I just toss it.

  30. Yes, absolutely! by Phreakiture · · Score: 1, Informative

    Yes, Ubuntu is getting slower, absolutely, without question on my part.

    My single biggest complaint against 8.04 was that it could not get out of its own way to play an MP3 on my somewhat modest hardware (Via MII-12000). It runs fine on my wife's machine, however (AMD Sempron on Via MoBo).

    Now, it is possible that the slowdown is only with 32-bit versions. My wife's machine is running the 64-bit version, and seems to run pretty well. In the mean time, I have reverted to Slackware, which has always been my refuge.

    --
    www.wavefront-av.com
    1. Re:Yes, absolutely! by Aladrin · · Score: 1

      My biggest complaint is disk access. While something is using the HD hard, other things will freeze up. Try unrar-ing a 5gb file and using firefox. I even had Pidgin lock up while unrar-ing the other day.

      It's honestly getting to where I'm back doing most audio/video stuff on the Windows machine because I know it won't freeze in the middle of it and annoy the crap out of me.

      As for flash... Well, you don't even need to be doing anything else to make it freeze repeatedly on even the simplest video clip.

      --
      "If you make people think they're thinking, they'll love you; But if you really make them think, they'll hate you." - DM
    2. Re:Yes, absolutely! by tayhimself · · Score: 1

      They need to have a tuned for decompressing DVDs and large torrents "Pirates Edition".

    3. Re:Yes, absolutely! by crimsun · · Score: 1

      By default, 8.04's PulseAudio configuration uses a "nicer" but more resource-intensive resampler. Many people have simply turned off/deinstalled PulseAudio without really understanding the culprit.

      By default, 8.10's PulseAudio configuration uses a less resource-intensive resampler (at the cost of some "quality"), so you shouldn't see the regression that you did from 7.10 -> 8.04.

    4. Re:Yes, absolutely! by Phreakiture · · Score: 1

      Well, I am planning to try it when it drops officially. I haven't given up hope yet.

      --
      www.wavefront-av.com
  31. You're vastly overestimating the situation by CaptainPatent · · Score: 1

    Hell, how many geeks can say such phrases: "...was the phrase my girlfriend..."

    baby steps man... baby steps.

    --
    Well, back to rejecting software patent applications.
  32. Re:Ubuntu? No way. by Anonymous Coward · · Score: 2, Funny

    "Slackware" -- an African word, meaning "Gentoo is too hard for me".

  33. OpenGL / Gaming by martin_henry · · Score: 1

    What is important to note is that the gaming performance hadn't dropped with the newer releases.

    Mystery solved. My games will run just about the same.

    --
    www.purevolume.com/martyd
  34. Re:Performance isn't its raison detre by speroni · · Score: 4, Funny

    An analogy is a simile, metaphorically speaking.

    --
    Eschew Obfuscation
  35. Re:Ubuntu? No way. by Anonymous Coward · · Score: 0

    ROFLMFAOOMGBBQ

  36. Slackware? No easy way. by Anonymous Coward · · Score: 1, Insightful

    Doesn't "Slackware" mean "I am too lazy to make it easy to install." ?

  37. Re:Performance isn't its raison detre by doti · · Score: 1

    I don't understand this, quite frankly, Windows user mentality of just accepting the state of things.

    That's the effect of having a Linux distro to be used by Windows-minded users.

    --
    factor 966971: 966971
  38. ha! by thatskinnyguy · · Score: 1

    Don't tell me Ubuntu is getting slower. On my eee PC with a stock hardware configuration, I go from a cold boot to the desktop in around 40 seconds whereas with XP it was taking about 3 minutes on average. Granted the sub-distro (if you would call it that) that I am using is optimized for my setup.

    --
    The game.
    1. Re:ha! by destroyer661 · · Score: 0

      That's insane, my EEE 701(very first generation) boots in ~15-20 seconds to XP, granted it's NLited. I still don't see how your XP would take 3 minutes. I used to lurk around the eeeuser.com forums a lot and haven't ever heard of someone having that long of a boot time. Throw up a video of it taking 3 whole minutes and I'll believe you.

      --
      #define true false // Have fun debugging!
    2. Re:ha! by deadbeefcafe · · Score: 1

      Heh, I put Ubuntu on my eee 900 but found the ~40 sec boot time way too slow for me :) Firefox would keep freezing and greying out which made it feel very bloaty.

      So I put Arch on and it boots in around 20. Firefox problems almost gone, a lot better than before. eduke3d runs better too, but still keep running out of RAM after playing for a bit though. :( Probably should stick it to those SSD lifetime naysayers and have some swap...

    3. Re:ha! by thatskinnyguy · · Score: 1

      You could probably attach an external hard drive and get some ad-hoc swap space that way. Just an idea...

      --
      The game.
    4. Re:ha! by Hatta · · Score: 1

      What does boot time have to do with anything? My Apple II GS boots to ROM in a couple seconds.

      --
      Give me Classic Slashdot or give me death!
    5. Re:ha! by lamapper · · Score: 1

      ...On my eee PC with a stock hardware configuration, I go from a cold boot to the desktop in around 40 seconds whereas with XP it was taking about 3 minutes on average...

      On my Asus EeePC (the Surf, a 901 w/out the webcam...should have got the webcam, running debian Linux 2.6.21.4-eeepc i686) I boot up between 10 - 23 secs. I will double check that and try to remember to come back and edit it this post.

      On my Windows 2000 PC I counted to 110 seconds (1001, 1002, 1003 ... 1109, 1110) before I had the desktop up. I am getting up a couple of Linux servers, perhaps after I get them up I will reinstall my Windows 2000 from scratch to speed it back up. But I doubt that it will boot up faster then 50 - 60 seconds no matter what I do to it.

      My windows 2000 box is only an IBM 6565 w/ 667 Pentium III processor, I got three of them for about $100 per quite a few years ago. With cheap, cheap memory they make a very workable Linux servers for under $250.

      Heck i just got 1 GB or RAM (2 X 512 MB) on sale for $39.00 and a 500 GB drive for $109, looking forward to setting up my own distributed server machine and see how many different distros I can run on it at the same time. I have been pleasantly pleased what I can do with only 128 MB of RAM, bump that up to either 256MB or only 512MB of RAM and you have a very workable home server.

      Thanks to Linux, I can turn one of them into a very fast custom DVD player, (maybe even put in my own internal fiber), currently running a gigabit network which of course is faster then my ISP provided router and connection anyway.

      Not a big fan of WINE and would like to run a virtualized Windows 98, 2000 or XP in its own slice one day. Put it in its own sandbox just for the heck of it. (Either that or just stop buying Microsoft games all together. That is all I ever use Microsoft at home for anymore anyway is Microsoft specific games; fortunately I am not a big gamer anyway).

      Hoping that my next job will have me in either a Linux only or at least a dual Linux / Microsoft environment. Best of all would be a Linux / Mac / Microsoft environment, but that is probably too much to hope for!

      --
      Is your Internet Throttled? Install DD-Wrt, OpenWRT or Tomato to learn the truth! Google: 1Gbps/1Gbps: 5 Communities
  39. Power management by Anonymous Coward · · Score: 0

    One thing I have noticed for a long time now is that the power management CPU scaling in Linux really sucks when you have more than one CPU/core. The problem seems to be caused by the kernel moving processes around to different cores. What this does is cause the various core to constantly speed and slow down, since the trigger for increasing speed is not instantaneous then you end up with crappy performance.

    Note that this is when there is just one process sucking up a lot of CPU. If you have multiple processes using a lot of CPU then everything is OK because all the cores get bumped up in speed and stay there.

  40. Stats? by Sockatume · · Score: 1

    A few minor quibbles: there are no error bars (at just 3-5 runs for "most" benchmarks and 1 for the remainder, I have to wonder how significant the results are), the beta benchmark app was used for some reason, and the benchmark application automatically downloads any needed dependencies for each test, so I have to wonder if the methodology became inconsistent somewhere. On that last point, the most striking and inexplicable performance hits were for Java and media encoding, so I'd be more satisfied with the results if they could reassure me that they'd used a comparable virtual machine and encoder on each OS.

    --
    No kidding!!! What do you say at this point?
  41. "fair"? by Paradigm_Complex · · Score: 1

    No, Ubuntu. Just because Microsoft's parents let them do something naughty it doesn't mean you can get away with it yourself, young man. We expect better of you.

    Neither Vista nor Ubuntu have improved sufficiently over their previous versions to justify the decreased performance. Ubuntu's focus is on user-friendliness, and it so it isn't exactly counting calories. Vista doesn't really have any excuses.

    Ubuntu was the distro that completely replaced Windows for me a bit over two years ago. Since then my desktop has just gotten faster and faster and left more and more RAM and disk space for the applications as I've found more bloated things I can take out of Ubuntu. It seems pretty evident to me that Ubuntu could keep its weight down if the developers put some focus there. Ubuntu's target audience is largely Windows users who are used to the bloat so Ubuntu can get away with it, but that doesn't make it right.

    --
    "A witty saying proves nothing." - Voltaire
  42. GCC is getting slower by hattig · · Score: 1

    From those benchmarks the one thing that stuck out was that GCC is getting slower. This could be down to the Ubuntu default compilation parameters changing to use more optimisation, for example, which takes longer.

    In addition the media portion of Ubuntu or Gnome has become incredibly slow compared to 7.04. Encoding is far far slower, and that's simply embarrassing.

    Next up for review: How fast is Kubuntu (Linux + KDE), how fast is FreeBSD (+Gnome and +KDE), and how does it compare to Linux + Gnome.

    1. Re:GCC is getting slower by jamei · · Score: 1

      From those benchmarks the one thing that stuck out was that GCC is getting slower. This could be down to the Ubuntu default compilation parameters changing to use more optimisation, for example, which takes longer.

      Surely if GCC is using more optimisation we should be seeing performance of the other tests increase.

    2. Re:GCC is getting slower by cpghost · · Score: 2, Interesting

      From those benchmarks the one thing that stuck out was that GCC is getting slower.

      That's a well-known fact for us source-based distributions/OS users. Compiling everything from source on Gentoo, or the BSDs took a severe performance hit since GCC got more and more slow (for no apparent reason), esp. the C++ backend... But what's slowing Ubuntu down is probably the quality of ASM code generated by GCC, as well as programs being writting more and more sloppily by developers with very fast machines.

      Maybe the source of the problem is actually GCC getting slower. This forces developers to use faster machines to shorten the compile runs; but those faster machines also hide the problem of software getting slower. Many devs simply don't care anymore for slower machines, because they simply don't see the problem on their own boxes. To them, the software is "fast enough."

      --
      cpghost at Cordula's Web.
  43. Re:Performance isn't its raison detre by Anonymous Coward · · Score: 0

    The funny thing is that the OP name is BadAnalogyGuy...

  44. Performance Problems AREN'T Where You Think... by Anonymous Coward · · Score: 5, Interesting

    ... they are.

    Seriously.

    I can see several problems with the testing methodology as is:

        * The test suite itself: The Phoronix test suite runs on PHP. That in itself is a problem-- the slowdowns measured could most likely be *because* of differences in the distributed PHP runtimes. You can't just say "hey, version Y of distro X is slower than version Z! LOLZ" because, WTF. You're pretty much also running different versions of the *test suite* itself (since you have to consider the runtime as part of the test suite). Unless you remove that dependency, then sorry, you can't measure things reliably. Which brings me to my second point...

        * What exactly are they testing? The whole distro? The compiler (since most of the whole of each distro version is compiled with different versions of GCC)? The kernel? If they're testing the released kernel, then they should run static binaries that *test* the above, comparing kernel differences. If they're testing the compiler, then they should build the *same* compiled code on each version and run said compiled code (which is pretty much what I gather they're doing). If they're testing the utilities and apps that came with the distro, then they should have shell scripts and other tools (which run on a single runtime, not depending on the runtime(s) that came with the distro version). Because if you don't, you have no fucking clue what you're testing.

    Honestly, I was unimpressed by the benchmarks. I happen to do performance benchmarking as part of my job, and I can tell you, you have to eliminate all the variables first -- isolate things to be able to say "X is slow". If you rely on a PHP runtime, use *exactly* the same PHP runtime for all your testing; otherwise, you'll get misleading results.

    1. Re:Performance Problems AREN'T Where You Think... by BlackCreek · · Score: 4, Informative

      Disk access has been slowing everything down.
      Variable elimination has been done, to varying extent, by multiple people here:
      https://bugs.launchpad.net/bugs/131094

    2. Re:Performance Problems AREN'T Where You Think... by Anonymous Coward · · Score: 1

      PHP isn't causing the problem. It's not like the tests are written in PHP or anything. PHP just "glues" it all together and then issues shell_exec cmds to run the tests. Your claims are ridiculous.

    3. Re:Performance Problems AREN'T Where You Think... by Anonymous Coward · · Score: 0

      Because if you don't, you have no fucking clue what you're testing.

      There is some value in *not* controlling for the execution environment; the benchmarks in the article may have been useless, but in some other cases it _does_ make sense: then you're testing how fast each out-of-the-box release is at doing whichever task you put it to. In other words, you're trying to answer the question "what's the real-world practical performance?" for each of the releases.

      I think sysadmins care very little about benchmarks saying libc-gutsy is 10% faster than libc-intrepid. They care about how fast the setup they have to support is going to be, so they can buy the correct amount of servers. Sure, they can wiggle the versions a little, but dependencies from applications might constrain you to only the versions that come out of the box.

      somewhat-half-a-counterpoint-somewhat-orthogonal'ly yours, --Jonas K

    4. Re:Performance Problems AREN'T Where You Think... by bjourne · · Score: 1

      ... they are. Seriously.

      Seriously. Yes they are. Any computer literate long time Ubuntu user that has been running Ubuntu since Feisty can testify that it has been getting slower. The performance problems are all over the distro.

      I can see several problems with the testing methodology as is: * The test suite itself: The Phoronix test suite runs on PHP. That in itself is a problem-- the slowdowns measured could most likely be *because* of differences in the distributed PHP runtimes. You can't just say "hey, version Y of distro X is slower than version Z! LOLZ" because, WTF. You're pretty much also running different versions of the *test suite* itself (since you have to consider the runtime as part of the test suite). Unless you remove that dependency, then sorry, you can't measure things reliably. Which brings me to my second point...

      Or you could just RTFA instead of speculating about possible test methodology failures. The suite is GPL:ed and anyone can reproduce their results if they have the inclination to do so.

      Honestly, I was unimpressed by the benchmarks. I happen to do performance benchmarking as part of my job, and I can tell you, you have to eliminate all the variables first -- isolate things to be able to say "X is slow". If you rely on a PHP runtime, use *exactly* the same PHP runtime for all your testing; otherwise, you'll get misleading results.

      And I'm honestly unimpressed by you, your fake appeal to authority and the rest of the commentators who apparently has no idea about what testing is. The PHP runtime is used for aggregating test results. If it takes 60 seconds to encode an mp3, then what version of the PHP runtime used won't matter.

      For example, one possible culprit is CFS which were added in, I think 8.04. A feature that on a micro scale might look very attractive (run 10 infinite while loops simultaneously without them starving each other), but which on a macro scale might have disastrous consequences (mouse pointer movement becomes sluggish when unpacking rar files).

  45. Re:Ubuntu? No way. by aproposofwhat · · Score: 2, Funny

    No - it's Yoruba for "Come back after the rainy season - it'll probably have compiled by then, but if not, there'll be plenty of dried wilderbeest to snack on."

    --
    One swallow does not a fellatrix make
  46. Re:Ubuntu? No way. by thatskinnyguy · · Score: 5, Insightful

    "Ubuntu" -- an African word, meaning "I'm sick of fucking with Linux in order to get it to do what I want but I really don't like the alternatives."

    Yeah, I rocked Gentoo for a couple of years. I just want something that is fast, easy to use and gives me as little of a headache as possible. Linux is Linux and most of the knowledge learned in one distro will carry-over to another.

    --
    The game.
  47. A Wizard Did It by picard2600 · · Score: 1

    Why should I read this FA if the author apparently didn't finish high school?

    Because intelligence and wisdom have nothing to do with "finishing high school"? I've got nothing past GCSEs. Luckily for me, employers in the UK see past that.

    Well, I've got nothing past OWL, but fortunately employers in the US see past that.

  48. Comment removed by account_deleted · · Score: 4, Insightful

    Comment removed based on user account deletion

  49. Ricers... by Anonymous Coward · · Score: 0

    Ubuntu needs ricing...They should be targeting performance instead!

  50. Re:Ubuntu? No way. by Anonymous Coward · · Score: 0

    rofl!!

  51. Re:Performance isn't its raison detre by dintech · · Score: 5, Interesting

    A project gets late one day at a time. There's probably a similar proverb for this too.

  52. Re:Performance isn't its raison detre by Anonymous Coward · · Score: 0

    "all apps" being F-Spot and Tomboy, neither of which are running by default?

  53. Is ubuntu getting slower? by Anonymous Coward · · Score: 0

    I dunno? I use Arch & Gentoo. LOL!

  54. Arch Ubuntu by Whiteox · · Score: 1

    That's the end point if you think about it.
    Linux is one of those OSs' that should be built around the hardware.
    Arch Linux forces you to do that and consequently is much faster than a vanilla install of Ubuntu.
    Not having to use resource hogs like KDE/Gnome etc helps as well.
    You don't need overbuilt gui to work a good desktop.
    But Ubuntu is for everyone and Arch linux is for real geeks.

    --
    Don't be apathetic. Procrastinate!
  55. Re:Performance isn't its raison detre by Anonymous Coward · · Score: 0, Funny

    As a Windows fanboy, .NET consultant I make more then 700 USD/day :)
    I drive a nice car, decline sex to desperate women who compliment me endlessly each weekend, while I'm up egostroking and renewing my knowledge and try to make more monnies instead.

    I used to be a Linux fanboy, I watched porn, had long hair and boy, I was hot on the internet. I would be up all night *trying* to get the damn thing running, without getting paid.

  56. Don't be that guy by moteyalpha · · Score: 1

    Hey let's run valgrind on it and if a random number thingy is slowing it down we can remove it. DOH! A desktop machine does not need to scream. If you want speed use another distro like Arch if you don't want CPU cycles wasted on pretty visual cues.

  57. Re:Performance isn't its raison detre by Whiteox · · Score: 4, Funny

    I think that's a rhetorical tautology.

    --
    Don't be apathetic. Procrastinate!
  58. Re:Performance isn't its raison detre by RemoteSojourner · · Score: 1

    Linux is not an acronym. I think you are talking about GNU is Not Unix.

  59. GTK performance stats? by siride · · Score: 4, Insightful

    The GTK+ statistics are mind-boggling slow. That's what I notice most when I use Ubuntu or Fedora. On my non-Ubuntu laptop, I get the following results for GTK performance:

    GtkDrawingArea - Pixbufs: 3.73s (on mine) vs 43-55s (Ubuntu)
    GtkRadioButton: 13s vs 29-60s

    I just think that's ridiculous. What did they do to GTK+ to make it so slow?

    1. Re:GTK performance stats? by mike_sucks · · Score: 1

      I remember seeing a thread on planet gnome that the majority of GTK perf problems were caused by (bad) themes.

      So, try changing themes and see what difference that makes. Or better yet, get out a profiler like sysprof and see precisely where the difference is.

      /Mike

      --
      -- "So, what's the deal with Auntie Gerschwitz et all?"
    2. Re:GTK performance stats? by siride · · Score: 1

      Yes, I'm aware of the theming issues. I use the same theme on Ubuntu and Gentoo (Clearlooks "Classic"). On Ubuntu, it definitely feels slower. I didn't run GtkPerf on my own Ubuntu to get a true comparison yet, but I'd be surprised if the results were close to what I get on Gentoo.

      Profiling would be worth it if I really cared about using Ubuntu. So far, for me at least, it offers almost no benefits, and a number of drawbacks, among them performance problems. I'll stick with Gentoo for the time being.

      And no, I'm not a ricer. I just happen to find that Gentoo fights against me less than some other distros and that's why I keep using it. My CFLAGS are quite sane.

  60. Re:Ubuntu? No way. by MightyYar · · Score: 1

    Some of us possess the ability but not the will.

    --
    W..w..W - Willy Waterloo washes Warren Wiggins who is washing Waldo Woo.
  61. Re:Ubuntu? No way. by Anonymous Coward · · Score: 0

    'Anonymous Coward' -- African words, meaning 'sometimes funny, sometimes not.'

    'Ucuntu' -- an African word, meaning 'Unix for c*nts.'

    'Ukuntu' -- an African word, meaning 'KUnix Kfor K*nts 4.1.'

    'Hurd' -- an African word, meaning 'Much promised, but little gained.'

    'Microsoft' -- an African word, meaning 'Help, I'm being anally violated by an angry, bald man with a chair fetish, who keeps shouting "Yeah! Give it up for me!" while bending me over.'

    'Slashdot' -- an African word, meaning 'When Digg sucks cock, turn here.'

  62. What should then we install on our laptops? by VincenzoRomano · · Score: 1

    Which Linux distro?
    Or even more radical, which OS?

    --
    Maybe Computers will never be as intelligent as Humans.
    For sure they won't ever become so stupid. [VR-1988]
  63. Is it just Ubuntu? by torry_loon · · Score: 1

    I would like to see the same comparison on other GNOME based distros before I make any judgement. Is Debian Lenny slower than Etch? Is Fedora 10 slower than Fedora 8? Is OpenSUSE.... oh, wait!

  64. Comment removed by account_deleted · · Score: 3, Interesting

    Comment removed based on user account deletion

  65. I just ran my own test.... by SIR_Taco · · Score: 0

    Turns out that DOS loaded and ran 100x faster on my machine than Windows 95, 98, 2000, XP, Vista AND all the distributions of Linux I have kicking around!

    c:\no\shit\

    Maybe we should all just switch back to DOS.

    --
    I say don't drink and drive, you might spill your drink. Before you get behind the wheel just stop and think.
    1. Re:I just ran my own test.... by badpazzword · · Score: 1

      But have you loaded it from its floppy disk?

      --
      When ideas fail, words become very handy.
  66. OH NOES!! by inaneframe · · Score: 1

    Well I guess if I ever plan on encoding an extensive number of movies and music, I'll just reinstall 7.04. Hmmm, this couldn't have anything to do with the many applications that do not yet take advantage of the latest kernel features could it? There have been major advancements in kernel development in the last 2 years, too many to list just go here and read a little: http://kerneltrap.org/

    --
    "Creationists make it sound as though a 'theory' is something you dreamt up after being drunk all night." -Asimov
    1. Re:OH NOES!! by Aphoxema · · Score: 1

      The damnedest thing is I just saw these figures after I started a conversion of ~11,000 MP3's into OGG's this morning. Does this mean that if I were using 7.10 instead of 8.04 it would take me 3 days instead of 6?

      --
      "Most people, I think, don't even know what a rootkit is, so why should they care about it?"
    2. Re:OH NOES!! by inaneframe · · Score: 1

      Yup, seems that way. Apparently, you're just wasting your time.

      --
      "Creationists make it sound as though a 'theory' is something you dreamt up after being drunk all night." -Asimov
    3. Re:OH NOES!! by Aphoxema · · Score: 1

      Oh, God, what am I to do with myself?! My computer will be doing the same thing for twice as long even though I've already intended for it to do it that long and I can still do a million things with it even while the conversion software is eating up my processor without a bit of lag thanks to the CFS!

      --
      "Most people, I think, don't even know what a rootkit is, so why should they care about it?"
    4. Re:OH NOES!! by inaneframe · · Score: 2, Informative

      Sounds like it has something to do with to whether an application takes advantage of multi-threaded enhancements or how it does.

      --
      "Creationists make it sound as though a 'theory' is something you dreamt up after being drunk all night." -Asimov
  67. Re:Ubuntu? No way. by Kjella · · Score: 1

    No - it's Yoruba for "Come back after the rainy season - it'll probably have compiled by then, but if not, there'll be plenty of dried wilderbeest to snack on."

    I could swear that's spelled "Gentoo"...

    --
    Live today, because you never know what tomorrow brings
  68. Re:Ubuntu? No way. by Eli+Gottlieb · · Score: 1

    "Ubuntu" -- an African word meaning "I only just found out that Gentoo has moved KDE 4.1 into the stable Portage tree, and will upgrade soon".

  69. Flexibility and freedom are its raison d'Ãtre by Dystopian+Rebel · · Score: 4, Insightful

    I currently have three Ubuntu-based systems. I have customized the GNU/Linux distribution for each system. Anyone can do it.

    One of the computers is an old (1998) Thinkpad notebook that doesn't have the video capability to run the full Ubuntu-Gnome GUI well. I do a minimal installation (the minimal CD is an official distribution of Ubuntu), then install about 25 select packages using a script that I call "Thinbuntu". This gives me a very functional GNU/Linux desktop for the old Thinkpad with the Long Term Support of the Ubuntu package system, including updates. It has all the features I need.

    Microsoft simply can't compete with this.

    --
    Rich And Stupid is not so bad as Working For Rich And Stupid.
  70. Re:Ubuntu? No way. by elrous0 · · Score: 2, Funny

    "Linux"--the Finnish word for "Many confusing forks"

    --
    SJW: Someone who has run out of real oppression, and has to fake it.
  71. Re:Performance isn't its raison detre by Anonymous Coward · · Score: 5, Funny

    So the soul-removal procedure went well, I see.

  72. Re:Performance isn't its raison detre by LWATCDR · · Score: 5, Insightful

    Why do you think this? I mean do you have a good reason or is it just because you don't like mono?
    Take a look at some of the tests.
    "Computational: The Dhrystone 2 performance within the BYTE Unix Benchmark was also the fastest on Ubuntu 7.04. There was approximately a 20% drop in performance between 7.04 and 7.10 that remained consistent even in the 8.04 and 8.10 releases. "
    This is NOT in mono.
    "Database: In our SQLite test of measuring the time to perform 2,500 SQL inserts, the performance hadn't dropped off after Ubuntu 7.04 but instead after 8.04 LTS. In this performance drop it was over 2.5x slower. "
    SQLlite isn't written in mono.
    "but in our compilation benchmarks we spotted major performance losses following the Feisty Fawn release. It was noticeably slower to compile Apache, PHP, and ImageMagick in the 7.10, 8.04, and 8.10 releases."
    GCC isn't written in mono.
    I could go on and on but many of the benchmarks have nothing to do with mono at all.
    Heck I am not a big fan of mono but your statment is baseless.

    --
    See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
  73. Progress vs Motion by vtcodger · · Score: 1

    ***Otherwise, shut up and get used to progress.***

    Progress is motion toward a goal. Remove the goal and what is left is just motion. A lot of people confuse motion with progress. What goal do you perceive Ubuntu to be moving toward? Why isn't faster operation/lower overhead part of the goal?

    --
    You can't see ANYTHING from a car, You've got to get out of the goddamned contraption and walk...Edward Abbey
    1. Re:Progress vs Motion by tomatensaft · · Score: 1

      Maybe because feature-rich and functional means a lot of overhead and thus slower operation? Not always, of course, but mostly.

    2. Re:Progress vs Motion by CheShACat · · Score: 3, Funny

      See Ubuntu bug #1 for your answer.

    3. Re:Progress vs Motion by mweather · · Score: 2, Insightful

      What goal do you perceive Ubuntu to be moving toward

      Fixing bug #1, of course.

  74. these results are bizarre by buddyglass · · Score: 1

    I can think of only two explanations, assuming their CPU-related benchmarks aren't memory-constrained. If they compiled the benchmark suite on each system, then perhaps there's some major regression in the newer versions of gcc that's negatively affecting the test results on all newer versions of ubuntu. This seems pretty unlikely.

    The other is that it's a scheduling issue. Why would a benchmark that *only* exercises the CPU run slower on one version of Ubuntu compared to the other? Probably because it's not getting all of the CPU. Or being interrupted continually, etc. They should run the tests w/ "nice" turned all the way up, to suggest to the OS that they get *all* the CPU they want. I suspect if they did, that all the CPU-bound tests, at least, would have near-identical results. I/O is a different matter entirely.

  75. You're completely wrong by ciroknight · · Score: 4, Insightful

    In this case, anyways. The benchmarks had everything to do with X, GCC and Kernel performance numbers, which all slid over the intervals measured.

    In other words, Ubuntus' not getting slower. The software that Ubuntu bundles is getting slower.

    It's likely due to GCC's epic failure to better optimize code as 4.x progresses, but I'm not putting my money on anything until they've tested at least one other distro which builds with a different GCC version.

    --
    "Victory means exit strategy, and it's important for the President to explain to us what the exit strategy is." G.W.Bush
    1. Re:You're completely wrong by mollymoo · · Score: 1

      In other words, Ubuntus' not getting slower. The software that Ubuntu bundles is getting slower.

      Ubuntu (Canonical) pick what goes into Ubuntu. Older versions of everything Ubuntu bundles are still available, there is nothing forcing them to use a new version if it sucks donkey balls.

      --
      Chernobyl 'not a wildlife haven' - BBC News
    2. Re:You're completely wrong by Khashishi · · Score: 4, Informative

      In other words, Ubuntus' not getting slower. The software that Ubuntu bundles is getting slower.

      Ubuntu is the software that it bundles.

    3. Re:You're completely wrong by Anonymous Coward · · Score: 0

      Ubuntu is the software that it BUNdles, Thank U.

  76. Re:Performance isn't its raison detre by recoiledsnake · · Score: 1

    Wow, I was wondering how quickly Microsoft would be somehow blamed for this. Didn't take long at all!

    --
    This space for rent.
  77. Well, hurra for choice. by Blice · · Score: 2, Interesting

    Ubuntu isn't really suited for low end machines anymore, IMO. Has to be at least 2ghz, with at least 512mb RAM.. Anything lower and it's going to be pretty slow.

    I'm sure some people here are going to be comparing Ubuntu to Vista in regards of getting slower with release, but because it's Linux we have a few more choices.

    They make distros that are meant to be lightweight- Anyone with a machine that's a little old, I urge you to try one of them. You'll be pleasantly surprised and maybe find a new favorite window manager/desktop environment in the process. Before you start talking about how joe sixpack doesn't want to try another distro or learn anything about Linux- I'm not talking to Joe sixpack here, I'm talking to you, a slashdot lurker.

    Try one of these distros and be amazed at how fast everything is:

    Crunchbang Linux

    KMandla's GTK 1.5 Remix

    Or, if you want to be more adventurous, get Arch Linux and grab a window manager like Openbox or PekWM. If you go that route, take a look at this Openbox guide that'll show you a nice panel to use, file navigator, and generally hold your hand through the process, here. But if you want your hand held even more, someone packaged a panel and file navigator and theme chooser and stuff like that together with Openbox already- Called LXDE. You can just grab that too, should be in any repository.

    I do think it's unfortunate for joe sixpack that it's getting a little slower- But for them it's still faster than Vista and XP, right?

    You know what they should make? They compile pretty much everything in the kernel as a module, and then they probe hardware and load the right modules each time you boot... It would be cool to be able to do a "Speed up my computer" boot where it loads the modules, and then compiles a kernel with the modules for the hardware it finds compiled into it. Disable things that it hasn't seen their computer use, etc., and then just still probe the hardware to fall back on another kernel if things have changed.

    OR, how about loading modules when you actually need them..? And this goes for daemons, too. When you go to listen to something, and it returns that there's no module loaded for sound, how about loading the module then, and then starting the alsa daemon. Have you ever looked at the daemon list for Ubuntu? It's huge. I know I don't need all of those- I know because on the distro I'm on now I only run 3 daemons on boot, and everything works fine.

    I don't know. Maybe that's not the solution. But those guys are clever, I'm sure they can come up with something to get rid of the extra daemons and modules running without sacrificing usability. Anyone here have any good ideas..?

    1. Re:Well, hurra for choice. by fprintf · · Score: 1

      How "old" are we talking about for the recommendations for moving to a more lightweight Linux distribution? Do people use these machines for actual CPU intensive work?

      I have a Thinkpad x31 at home running Ubuntu Hardy Heron. I have managed to bloat it up quite a bit adding in all kinds of services, including a recent misguided foray into Ubuntu Studio. It will not run Compiz due to the lack of a graphics card. Regardless, it still runs like a champ but I only really use it as a web browser, video player and the occasional Pyton script checking.

      --
      This post brought to you by your friendly neighborhood MBA.
    2. Re:Well, hurra for choice. by SanityInAnarchy · · Score: 1

      It would be cool to be able to do a "Speed up my computer" boot where it loads the modules, and then compiles a kernel with the modules for the hardware it finds compiled into it.

      This would be both dishonest (it's not just "speeding it up", it's also binding that install to that particular hardware configuration) and pointless (loading modules is simply not a performance bottleneck).

      OR, how about loading modules when you actually need them..?

      Tricky to get right, and also not necessarily what you want. When your computer goes to suspend or hibernate, you pretty much want every driver loaded, so you can tell it to suspend itself.

      There's also the problem where not everything can be triggered that way. Even if the software allowed it (which it doesn't, always), there's the stuff you want to be triggered from outside that computer. Trivial example: Bluetooth. I use a Bluetooth mouse, and I want it to just work when I tell it to connect. I shouldn't have to go type, or worse, click something, to get the bluetooth stuff loaded.

      And this goes for daemons, too. When you go to listen to something, and it returns that there's no module loaded for sound, how about loading the module then, and then starting the alsa daemon.

      Same applies -- with Bluetooth, you need that daemon running.

      Some of this is done -- on KDE3, the arts daemon is only started when something needs it, and is stopped when it's idle for 30 seconds. But some of it is equally pointless.

      Now, sure, for lower end machines, I wouldn't use a full Ubuntu. I'd use something stripped down -- either xubuntu, or ubuntu-minimal/server plus fluxbox. But I think the focus has rightly been how Ubuntu works on newer hardware -- that is, after all, the future.

      --
      Don't thank God, thank a doctor!
    3. Re:Well, hurra for choice. by dotancohen · · Score: 1

      Have you ever looked at the daemon list for Ubuntu? It's huge.

      Enlighten us Ubuntubabies. How does one look at the daemon list? And where is a plain-English (or Hebrew, please) document that explains what they all do so that an id10+ like me can decide what to disable?

      --
      It is dangerous to be right when the government is wrong.
    4. Re:Well, hurra for choice. by Blice · · Score: 1

      Hi dotancohen, You should take a look at KMlanda's guide on speeding up Ubuntu- It's filled with little tweaks and hacks and you're going to end up learning a lot more about Ubuntu, and linux in general. You can find it here.

      I forget where the list is, but it'll be in the guide somewhere. Just google each daemon and get an understanding of what it does.

    5. Re:Well, hurra for choice. by dotancohen · · Score: 1

      Thanks, Blice. As the site is geared towards Ubuntu 7.10 I will ask on the Ubuntu mailing list about specific tweaks in 8.04. Also, I see that the info is available as a Zim notebook. That's great, I love Zim. Thanks!

      --
      It is dangerous to be right when the government is wrong.
    6. Re:Well, hurra for choice. by Blice · · Score: 1

      Ack... Didn't link you to the most recent one, sorry about that.

      howto-set-up-hardy-for-speed

      If you look at his "howto" section of his blog, there's some other things for Ubuntu too.

    7. Re:Well, hurra for choice. by dotancohen · · Score: 1

      Yes, I've been going through it. And what a noob- I posted asking about updating the document for Ubuntu 8.04 and two minutes later I found it!

      --
      It is dangerous to be right when the government is wrong.
  78. I blame all the python rubbish ... by Anonymous Coward · · Score: 0

    ... that they are shovelling into the system, the dodgy init replacement, the weird wlan drivers

  79. Re:Ubuntu? No way. by FrozenFOXX · · Score: 5, Insightful

    Mod the parent up, I'm in the exact same boat. Ran Gentoo for three years, loved it, but realized that I was spending a lot more time tinkering with the OS than actually USING the OS. That's perfectly fine for a server of some variety where really the tinkering IS the interaction, but for a desktop or something you're going to use more interactively on a daily basis it became too much of a pain.

    Ubuntu gives me some of the strengths I liked (such as a simple, straightforward package manager, wide amount of customization without too much screwing around) without too many of the weaknesses (compiling all software, praying emerging the world doesn't break my desktop, so on and so forth).

    It's not a bad distro at all and it's tiring to hear of people slamming it for not being Slackware or Gentoo. This may come as a revelation, but Linux is about choice.

    --
    "Just a fox, a whisper."
  80. Re:Performance isn't its raison detre by Anonymous Coward · · Score: 0

    So the soul-removal procedure went well, I see.

    It was well compensated with a bonus ;)

  81. Too much Python by Anonymous Coward · · Score: 1, Insightful

    Subject says all *AC hides under rock*

  82. OpenOffice.org and Microsoft Office: slower too? by ahziem · · Score: 2, Informative
    Here are the numbers

    It's called Wirth's Law

  83. Regression Speed Test by IYagami · · Score: 1

    Why there isn't a suite of speed tests that are automatically done by Canonical about the speed of several usual operations? (maybe there is this kind of test and Phoronix test suite is very different)

  84. Re:Arch Ubuntu by Blice · · Score: 1

    er... Sorry to burst your bubble here, but Arch is not built around the hardware for your computer. Arch comes with a binary blog for a kernel with everything compiled as a module just like Ubuntu does, the only different is daemons that run at boot, which has little to do with your hardware. I agree that Arch is faster, depending on what you install, but saying it fits your hardware and that it's for "True geeks" is sort of ridiculous.

    Perhaps you meant Gentoo?

  85. Xorg, mainly. by ciroknight · · Score: 3, Informative

    Namely, graphics hardware. ATi graphics hardware and the FGLRX driver. FGLRX is known to have crappy 2D performance relative to its very strong 3D performance and the 2d performance that the open source driver excels at.

    Meanwhile, 2D performance on Intel's hardware is smoking everyone else's pipe.

    --
    "Victory means exit strategy, and it's important for the President to explain to us what the exit strategy is." G.W.Bush
  86. Ubuntu = Bloatware by Anonymous Coward · · Score: 0

    I know the OSS freaks like to copy everything from Windows, but they've gone too far this time!

  87. It partly due to the inclusion of alpha software by FreeGamer · · Score: 1

    I think it was 7.10 that first included Tracker turned on by default. At this point Tracker was still alpha software, many users reported it taking up all available resources, and complaints were simply brushed aside. After suffering from my Ubuntu installation constantly freezing due to trackerd, and being unable to disable it without removing it (I disabled it in 3 separate places but it still persisted on restart) I installed Fedora and never looked back. It may not be quite so "polished" but it hasn't sprung any nasty surprises on me.

  88. After RTFA... by Clarious · · Score: 1

    ... I see that tests that requite high computational power like BYTE Unix or Compilation is getting slower with each version, while disk benchmarks show that newer version is better. I am not an IT student so I can't really say anything about it, but maybe there are changes in kernel/Ubuntu that make disk operations faster while adding new works for the CPU to do.

    1. Re:After RTFA... by Anonymous Coward · · Score: 0

      Well it showed that SQLite is now a lot slower, that's a disk benchmark, so what you said doesn't make sense.

  89. I blame scrollkeeper... by BobMcD · · Score: 1

    In my completely unscientific tests, I've found that stripping off scrollkeeper in favor of rarian makes some of my older systems A LOT happier. Scrollkeeper was actually pegging and overheating a Celeron PC that used to run Feisty without complaint...

  90. Re:Flexibility and freedom are its raison d'Ã by dave420 · · Score: 4, Interesting

    Apart from XP Embedded, where you can choose which parts of the OS are installed. It's not designed for home-use, but I've used it to make a 140MB version of XP that is blazingly fast. It can play media, has internet access, games, everything I need.

  91. Re:Flexibility and freedom are its raison d'Ã by kiwimate · · Score: 1

    Microsoft simply can't compete with this.

    Really? It's not open-source, sure, but from your post you're not rewriting or recompiling. (Nor are most of the people commenting on this article.) You're selecting which components you want to install.

    I can disable services I don't want. I can uninstall (or simply not install at the outset) components I don't need. I have the control panel to customize, e.g., the video depth and turn off video-hungry components.

    In all seriousness: at this level (not the recompile level), what's the difference?

  92. Re:Ubuntu? No way. by thatskinnyguy · · Score: 1

    Amen to all of the above. I grew disillusioned with the fact that Open Office along with all the JRE crap took about 5 hours to compile (conservative estimate) on my laptop with limited specs. For a set-and-forget setup (server) it is a great OS. But for regular everyday use, it's just plain tedious.

    --
    The game.
  93. Re:Performance isn't its raison detre by Anonymous Coward · · Score: 0, Troll

    I drive a nice car, decline sex to desperate women who compliment me endlessly each weekend, while I'm up egostroking and renewing my knowledge and try to make more monnies instead.

    You know, you could have just written "I'm gay" instead.

  94. Better & Faster Damnit! by Anonymous Coward · · Score: 0

    The blatant acceptance of new functionality always meaning increased resource consumption is crap.

    Other's mention below that as software matures it should get faster and I whole heartedly agree with this view point.

    In my few years as a Mac enthusiast (pre-OS X) I always remember looking forward to the latest OS release because it almost always added functionality AND ran more efficiently on the same HW.

    I wish the rest of the SW industry would follow suit. I sympathize, I know it's hard work - but I don't think it's an unreasonable expectation.

  95. Article is NOT flamebait by BlackCreek · · Score: 1
    Many here without any specific knowledge of problem are giving it 10 seconds thought, waving hands and saying that the article is a flamebait.

    Being a Ubuntu user since Feisty, I believe that the article is spot on. The reason I say this is because there is a huge problem with desktop responsiveness, and it started after Feisty: https://bugs.launchpad.net/bugs/131094

    This is NOT the plausible small hit due to increase in hardware support. The problem reported by so many on that bug report above is about having a modern system hanging for 1 or 2 seconds (or more), for operations that were "instantaneous" on Feisty.

    FWIW, in the LWN comments about 8.10 there is yet another person mentioning the same issue: problems with kernels after Feisty http://lwn.net/Articles/304710/

  96. And here's the problem with Slashdot by RulerOf · · Score: 1

    If the majority of Slashdot users can read what you wrote (which I agree with, I may add) and consider you insightful, or even worse, accept this as inevitable and necessary, then you have just solidified a very large number of people as hypocrites.

    It could only be more sweetly ironic if the next version of Ubuntu shows to be slower than the current one, and then a .1 release fixes all of the problems... Which the MS Fanbois will continue to cite (and make fun of) for years after they're fixed... which will lead to negative public opinion about a perfectly fine operating system.

    --
    Boot Windows, Linux, and ESX over the network for free.
  97. Re:Performance isn't its raison detre by MikeDirnt69 · · Score: 1

    Perfect

    --
    Am I eval()? - http://www.monst3r.com.br
  98. Re:Performance isn't its raison detre by Bert64 · · Score: 1

    Ubuntu being demonstrably slower than older versions of itself is not exactly a selling point, especially because linux is often considered for lower powered machines...
    But having Ubuntu running typical tasks slower than windows is just inexcusable, i saw some benchmarks where open source tools such as ffmpeg were slower on ubuntu than xp... There really is no excuse for that.

    --
    http://spamdecoy.net - free throwaway anonymous email - avoid spam!
  99. Am I the only one... by Jazz-Masta · · Score: 1

    That doesn't see the huge concern over "progress" resulting in slightly slower software on the same hardware?

    It's been happening with every new version of almost ANY software. The only time speed increases with a subsequent version is when the original program was either hacked together or not properly optimized for specific hardware.

    No one opens up Photoshop and says "Oh no, CS3 is slower than Photoshop 4...WTF? How is this possible?"

    And thanks to Vista (tm), memory is now DIRT CHEAP. I can stuff 8GB into a machine for less than $150!

    If you really want to complain about violating some law of nature in regards to software engineering - complain to Adobe about ACROBAT!

  100. Ad Hominem by Anonymous Coward · · Score: 0

    Because you are falling prey to a logical fallacy?

  101. Is PIE the primary cause? by calc · · Score: 4, Interesting

    I believe that PIE (position independent executable) along with some other security enhancements were turned on in Ubuntu around the time the slowndowns showed up. This would definitely cause at least some slowdown on the 32bit version since there aren't enough registers to begin with. I'm not sure if it causes any noticeable slowdown on the 64bit version, since the amd64 architecture has a lot more available registers, which would correlate with the person mentioning earlier that the 64bit version seemed fast.

  102. Re:Performance isn't its raison detre by Bert64 · · Score: 1

    And as a Linux/Unix consultant, seeing as how there's less of us and we generally work more efficiently than our windows counterparts, you could probably be making a lot more than $700/day.

    At least all the positions i've seen, unix consultants command more money, have less interview competition, and still provide better value for money to the companies who hire them, whereas windows consultants are ten a penny and most are very poor at their job, having very little real experience being fresh off an mcse course or similar.

    --
    http://spamdecoy.net - free throwaway anonymous email - avoid spam!
  103. Insightful my ass! by Phreakiture · · Score: 1

    Performance may not be the raison d'etre, but performance must still be adequate. Linux in general has, as one of its "selling points", the fact that it runs well on older hardware. With Ubuntu 8.04, that point went to hell.

    --
    www.wavefront-av.com
    1. Re:Insightful my ass! by WormholeFiend · · Score: 1

      Performance may not be the raison d'etre, but performance must still be adequate. Linux in general has, as one of its "selling points", the fact that it runs well on older hardware. With Ubuntu 8.04, that point went to hell

      You get what you pay for!

      Seriously though, as a relatively new linux user, if Ubuntu ever gets so slow that I get frustrated, I'll switch to another distro.

      For me, the ability to switch distros is a linux selling point. Perhaps the best one.

      If it wasn't for Knoppix and a corrupted HDD boot sector, I'd probably still be using some form of windows.

    2. Re:Insightful my ass! by Computershack · · Score: 1

      Seriously though, as a relatively new linux user, if Ubuntu ever gets so slow that I get frustrated, I'll switch to another distro.

      I did. Went from Ubuntu 8.04 to OpenSUSE 11. It was quite an eye opener. Felt unbelievably quick compared to Ubuntu. I had thought that Ubuntu felt slow because it was on a 2yr old lappy but SuSE showed that wasn't the case. Tried 8.10 but it's actually worse. I put that down to being in BETA but after the comments here, I'm not sure that's the case.

      --
      I only please one person per day. Today is not your day. Tomorrow isn't looking good either. - Scott Adams
    3. Re:Insightful my ass! by Phreakiture · · Score: 1

      As I posted further down in this topic, I did. Over the decade or so that I have been a Linux-only user, Slackware has always been the one I go back to when I get sick of $other_distro. As such, I currently run Slackware 12.1.

      That said, I have the following points to make:

      • The question was: is Ubuntu getting slower. The answer is yes. "Switch to FreeDOS" is a bullshit non-answer. I see no reason why ease of use and acceptable performance cannot coexist as they did in earlier Ubuntus.
      • I have, for my own systems, taken an action that was appropriate to me and moved back to Slackware. Already done. Don't need help. Thanks anyway.
      • In my side-line as a computer consultant, I have converted a number of customers from Windows to Linux. My pitch has been based on a triangular approach consisting of security, ease-of-use and better performance than Windows. Ubuntu has been key to the first part of that pitch. For Ubuntu's performance to go to hell is a big concern for me.

      Now, in fairness, I have found that the 64-bit version of Ubuntu does not seem to have gotten trashed, and still performs well. It is just the 32-bit version that sucks.

      --
      www.wavefront-av.com
    4. Re:Insightful my ass! by davolfman · · Score: 1

      Switching to another distro is likely to be a mixed bag. If the problem is in the underlying components then everybody using them in their distro should be slow.

    5. Re:Insightful my ass! by Anonymous Coward · · Score: 0

      Awesome next time there's an article on slashdot about SUSE i'm going to fill it with Ubuntu comments.

      Why don't you just stop trolling and fuck off Computershack? I'm sick of your trolls..

      WHY HAVEN'T THEY FIXED WINDOWS NETWORK BROWSING?? (Score:1, Flamebait)

      Windows Network browsing is STILL BROKEN in Nautilus. They fucked it up in 8.04, as it was working in 7.10, and they've still not fixed it.
      Also, the hard drive load cycle bug has only been half fixed. Why? They know how to fix it - people have submitted full fixes in bugtraq.

      Fuck Ubuntu. Getting a bit pissed off with the shitty quality control they now have due to being hellbent on releasing every 6 months on the dot despite all the show stopper bugs.

      Re:The Chicken and the Egg (Score:1)
      *yawn*

      BULLSHIT

      Re:Stop paying MS for bad software... (Score:1) ...right up until you put it alongside a Windows PC running the same hardware and wonder where half the graphics effects have gone and why you're only getting half the FPS with the Linux box. HL2 with a DX9 card under Linux is like running it in DX7 in Windows.

    6. Re:Insightful my ass! by danieltdp · · Score: 1

      I could swear your post was about the insighfullness of your ass. My mistake! Sorry

      --
      -- dnl
  104. Let's see now. When was Mono introduced? by Anonymous Coward · · Score: 0

    Does the rest of the *buntus (sans U) show a similar decline in performance?

  105. Re:Flexibility and freedom are its raison d'Ã by Dystopian+Rebel · · Score: 1

    everything I need

    I have no experience with it but it sounds cool.

    I read that the next version of XP Embedded will have Genuine Advantage Embedded.

    --
    Rich And Stupid is not so bad as Working For Rich And Stupid.
  106. Uh... by Larryish · · Score: 1

    Yes.

  107. Re:Ubuntu? No way. by Anonymous Coward · · Score: 0

    You sure that's not gentoo you're thinking of? :D

  108. Re:Ubuntu? No way. by SanityInAnarchy · · Score: 4, Insightful

    I was pretty much exactly the same.

    Turning point for me was realizing that I was compiling more and more in, just in case I needed it, because rebuilding world just to enable that new USE flag was getting kind of old.

    In other words, I was using it like Ubuntu. The only advantage I had was I would compile for -march=i686, and other optimizations which produce binaries which only work on recent CPUs (the '686' class) -- whereas Ubuntu was -mtune=i686, if I remember, so it was possible to run on a 486, but would run best on a 686.

    And, hey, there were other things I would turn on that were Athlon XP specific, and so on... then I realized that, on amd64, the optimizations were basically exactly the same -- merely compiling for x86_64 gave me all the benefits anyway. At which point, what the hell -- Ubuntu would necessarily be at least as optimized as my Gentoo.

    And, more recently, I've realized that since switching to Ubuntu, I spend much more time actually using the OS, rather than tweaking it. Despite having it already much more customized than any version of Windows ever was, I still don't spend as much time tweaking it as it takes to maintain Windows, let alone Gentoo.

    --
    Don't thank God, thank a doctor!
  109. and? by Salem+Willow · · Score: 1

    ok so Ubuntu is getting slower..by how much? and most importantly: do we really care? it's still a damn sight faster imho than windows or ( the blasphemy of IT: ) mac.

    --
    this is a virtual insanity that always seems to be governed by our love for this useless twisting of our new technology.
    1. Re:and? by phrostie · · Score: 1

      i just switched from a straight debian install to Ubuntu.

      it's faster and more responsive by far.
      my only gripe is the whole F'd up superuser/root thing.

    2. Re:and? by Anonymous Coward · · Score: 0

      sudo passwd root ...and you're back in business, are you not?

    3. Re:and? by EvilIdler · · Score: 1

      It's not effed up at all. It's actually quite smart.
      As the other reply says, you can easily set a password..but it's just as easy to just do sudo -s.

    4. Re:and? by setagllib · · Score: 1

      The biggest user-visible difference between Debian and Ubuntu kernels is that Debian kernels are always "-server" style kernels, that is, no preemption and low clock rate. This lowers responsiveness but can vastly improve throughput for server tasks. So yes, Debian is still a server distribution, although on modern hardware the difference is more subtle.

      --
      Sam ty sig.
  110. Re:Ubuntu? No way. by Fallingcow · · Score: 1

    Another ex-Gentooer-turned-Ubuntuer here. I had no idea we were so common.

  111. Re:Flexibility and freedom are its raison d'Ã by AlterRNow · · Score: 5, Funny

    In which orifice?

    --
    The disappearing pencil trick. Let me show you it.
  112. Re:Performance isn't its raison detre by RyuuzakiTetsuya · · Score: 1

    But not like a truck?

    --
    Non impediti ratione cogitationus.
  113. Mac by Anonymous Coward · · Score: 0

    Time to migrate to NetBSD and get a Mac now that Linux is becoming a Vista-wannabe resource hog.

  114. Modular Design by Anonymous Coward · · Score: 0

    One of the best parts ive always enjoyed with linux in any form is the modular design, that is, that any feature isn't exactly hard designed into the OS, you can pick and choose features and disable the ones you don't need. New software usually increases these options but that doesn't mean you need to use said new features. As for increased performance demand in updated software, if it is a concern i believe the option is always there to not update. Besides, read about the update to find out why it consumes more and what was fixed and you might find yourself no longer regretting that increased performance demand. The same goes for kernel differences, if there has been a change there is likely a good reason for it, do some research and find out for yourself.

    Garrett

  115. Re:Flexibility and freedom are its raison d'Ã by _Sprocket_ · · Score: 2, Insightful

    I can disable services I don't want. I can uninstall (or simply not install at the outset) components I don't need. I have the control panel to customize, e.g., the video depth and turn off video-hungry components.

    In all seriousness: at this level (not the recompile level), what's the difference?

    Linux (and Unix for the most part) tends to be a lot more modular than Windows. Windows does provide options. But not to the same degree. If you want to really dig in to a Windows system, it takes a lot more shennanigans than it does with a Linux distro (and then you're at the risk of losing all your changes at the next service pack).

    Note that this isn't an Open Source thing. Proprietary Unix environments tend to work much in the same way.

    One final point - Linux is no silver bullet. You still have to make trade-offs. There are still dependencies involved and removing something might mean removing a desired application. However, I've rarely run in to a situation where that decision is all that difficult or unexpected.

  116. Re:Performance isn't its raison detre by Fast+Thick+Pants · · Score: 5, Funny

    FYI, actually having sex is usually more fun than declining it.

  117. Re:Flexibility and freedom are its raison d'Ã by mollymoo · · Score: 1

    You might want to check out nLite, which makes it easy to customise XP installations. You can remove the bloat, slipstream drivers and patches, run fully unattended installs etc. AFAIK under the hood it uses the tools Microsoft provide for corporate sysadmins and the like, but with a more user-friendly interface and useful defaults. Very popular with the Windows-loving section of the Netbook community and great for VM installs.

    --
    Chernobyl 'not a wildlife haven' - BBC News
  118. Re:Flexibility and freedom are its raison d'Ã by _Sprocket_ · · Score: 2, Insightful

    Apart from XP Embedded, where you can choose which parts of the OS are installed.

    I've heard of such a beast but never have seen it. I would hazard to guess that XP Embedded isn't accessible to most folks; it's not an option with their standard XP install. In contract, anyone who has access to Ubuntu could go this "Thinbuntu" route.

  119. Re:Flexibility and freedom are its raison d'Ã by Computershack · · Score: 2, Insightful

    Whereas with Linux, I have to keep 180MB of files there which will remain mostly useless because I want to use Amarok in a Gnome environment but if I want to remove even one, it'll throw a hissy fit. Way to go...a small media player with 180MB of dependencies...

    --
    I only please one person per day. Today is not your day. Tomorrow isn't looking good either. - Scott Adams
  120. Re:Performance isn't its raison detre by dotancohen · · Score: 1, Insightful

    It's nice to see that the Ubuntu fanboys have moved so quickly to 'shut up and like it'.

    It took Windows fanboys a decade to get there...

    That's because today's Ubuntu fanbois are yesterday's Windows fanbois.

    --
    It is dangerous to be right when the government is wrong.
  121. Re:Flexibility and freedom are its raison d'Ã by Computershack · · Score: 2, Insightful

    Linux (and Unix for the most part) tends to be a lot more modular than Windows. Windows does provide options. But not to the same degree. If you want to really dig in to a Windows system, it takes a lot more shennanigans than it does with a Linux distro

    Indeed. I might have to go install something like XPLite or create my own installation media with nlite/vlite. It's really taxing firing up a GUI and unticking a few boxes.

    --
    I only please one person per day. Today is not your day. Tomorrow isn't looking good either. - Scott Adams
  122. Re:good distro to balance speed/bloat vs. ease? by crazybilly · · Score: 1
    Can somebody recommend a good distro, then, that's a good balance of ease of use vs. speed/performance? Ubuntu has always felt bloated and unsnappy to me, no matter what hardware I run it on, but I'm too much of a noob (and too lazy) to do anything too different--it does make a lot of things that are difficult on other machines really easy.

    Anybody got any suggestions on a different distro to try?

  123. Re:Performance isn't its raison detre by LWATCDR · · Score: 1

    Performance should never go down.
    Yes as you expect a PC to do more and more their will be more overhead but the over all performance should go up as the code gets optimized.

    The problem here is we don't yet know why these benchmarks are running slower.
    Is it extra checks in the kernel for security?
    Is it fluff that isn't needed.
    Or is it bloat that needs to be optimized.
    So no people shouldn't just shut up. The cause needs to be found if possible fixed.

    --
    See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
  124. Re:Performance isn't its raison detre by Anonymous Coward · · Score: 0

    Lets leave Apple out of this!

  125. Re:Ubuntu? No way. by apoc.famine · · Score: 1

    I was in the same boat as you. Used Gentoo for a few years, and realized that I was spending way too much time screwing with it. I also switched to Ubuntu, but I have to say, my Gentoo machine was always faster and had more features. For that reason I went back to Gentoo on my desktop, with one major change: I stopped trying to friggin play games on the thing. Turns out that when you get it to a usable state, and stop screwing with it, Gentoo works a lot like Ubuntu. That is, it just works.

    Every time I have the urge to tinker with things, I try to make sure that it will really make a difference, and be worth the time. If not, I just go use Ubuntu. Now I have a fast desktop that just works. And on the off chance it doesn't the Ubuntu distros on the backup desktop or the Asus EEE Just Work.

    --
    Velociraptor = Distiraptor / Timeraptor
  126. Re:Performance isn't its raison detre by jonfr · · Score: 1

    Like running a system (Windows) that crashes even if it not doing anything special.

  127. Re:Performance isn't its raison detre by Anonymous Coward · · Score: 0

    Apple geeks at least have a chance of talking to a girl.

  128. Re:Flexibility and freedom are its raison d'Ã by mweather · · Score: 1

    It's still smaller than Windows Media Player, even when installed in the least efficient way possible.

  129. Re:Flexibility and freedom are its raison d'Ã by KasperMeerts · · Score: 1

    Even the most basic programs need more than a gigabyte of backend libraries in Windows Vista.
    You can probably program the complete UI of Amarok in one program. But that would probably be tens of megabytes itself, would bloat the program completely and would make reuse of code really hard.

    It seems you don't grasp the concept of libraries.

    --
    As long as there are slaughterhouses, there will be battlefields.
  130. Re:Performance isn't its raison detre by burroughsj1 · · Score: 1

    that's OK, I'm sure the next version, Moronic Monicker, is going to be a LOT faster.

    F, G, H, I,...M? Alphabet fail.

    --
    Suse vivo vixi victum reduco is ea id creatura absit decessus a facultas Linux! Dev root, dev root!
  131. Re:Flexibility and freedom are its raison d'Ã by Markspark · · Score: 1

    i have the same problem, but this is a gnome/kde problem, and has nothing to do with the speed of the computer, which has been constantly deteriorating, or rather, applications seemed to execute just as fast back in 99 as they do now, and even though the hardware is a lot faster, the software is a lot slower.

    --
    i find your lack of faith in science disturbing!
  132. Re:Performance isn't its raison detre by AvitarX · · Score: 1

    It is one of the stated goals of Jazzy Jackalope FWIW.

    Intrepid Ibex was mobility for example.

    Don't recall what Hardy Heron was. Gutsy Gibbon was about laying bases if I am not mistaken.

    --
    Wow, sent an e-mail as suggested when clicking on "use classic" banner, and got a fast response that addressed my msg
  133. telinit 3 by codepunk · · Score: 1

    telinit 3

    That is a pretty good cure for a slow running machine.

    --


    Got Code?
  134. No wonder you posted AC by DesScorp · · Score: 3, Insightful

    "If you really want performance, run FreeDOS. Otherwise, shut up and get used to progress."

    Jeez you're an idiot. I wouldn't have posted that under a registered nick either.

    So people should just settle for bloat simply because of the advance of technology? Apple manages to make OS X faster than older versions. Other Linux distros do. Bad software isn't "progress".

    --
    Life is hard, and the world is cruel
    1. Re:No wonder you posted AC by initdeep · · Score: 2, Insightful

      let me fix that for you.

      "Apple manages to make OSX faster on new hardware"

      i seriously doubt anyone will say that OSX 10.0 runs slower on a 4 year old 1gb ram intel chip than OSX 1.5.whatever.

      it's callled progress.

      More features are added because people want them.

      thus after a while, the original hardware is no longer the best solution to run the latest version of the OS.

    2. Re:No wonder you posted AC by TheRaven64 · · Score: 3, Informative

      i seriously doubt anyone will say that OSX 10.0 runs slower on a 4 year old 1gb ram intel chip than OSX 1.5.whatever.

      Try it. Seriously. The 10.0 kernel had a significantly inferior VM subsystem (10.5 improved it a lot). 10.3 and 10.4 introduced more GPU-offloading in to the windowing system. Each version of OS X has been faster on the same hardware, although 10.4 and 10.5 have been more RAM-intensive.

      --
      I am TheRaven on Soylent News
    3. Re:No wonder you posted AC by Anonymous Coward · · Score: 2, Insightful

      Oh, so "RAM Intensive" is Apple fan boi speak for "Slower" than, right?

      Turtle Necked Gentleman: Whoa, the new version of Mac OSX is totally RAM Intensive!
      Turtle Necked Gentleman's Companion: Why are you surprised? Apple is always working hard on adding new features!
      Turtle Necked Gentleman: No joke! The new version's Ram intensivity is radical!

    4. Re:No wonder you posted AC by TheRaven64 · · Score: 2, Informative

      No, RAM intensive means that it needs more RAM. If you have 1GB of RAM then 10.5 will be faster than 10.0. If you have 256MB of RAM then 10.0 will be faster. Try doing a Spotlight search in 10.5 and again in 10.4. The version in 10.4 will be slower and the indexing will be more CPU-intensive while it runs. 10.5 actually does slightly better in constrained memory situations than 10.4, because of the rewritten VM subsystem. If you have enough RAM for both to run without swapping then the newer version will be faster. If you have enough RAM for only the older version to run without swapping then the older one will be faster. If you have enough RAM for neither to run without swapping then 10.5 will be faster if it's the newer one, otherwise the older one will be faster.

      --
      I am TheRaven on Soylent News
    5. Re:No wonder you posted AC by Raenex · · Score: 1

      Jeez you're an idiot. I wouldn't have posted that under a registered nick either.

      The Anonymous Coward was making fun of the post above him, which was made under a registered nick.

    6. Re:No wonder you posted AC by toddestan · · Score: 1

      Just because you can find a few examples of where Apple has optimized things doesn't mean that the new release is faster overall. OSX 10.5 is slower than the previous releases, if you don't believe me try comparing it 10.2 or 10.3 on something like a 400Mhz G3. You'll find 10.5 sluggish no matter the amount of ram, whereas older versions of OSX would run acceptably on the same hardware.

    7. Re:No wonder you posted AC by CommanderIsm · · Score: 1

      Ubuntu 8.10 in real time everyday use the slowest yet. - so slow i took it off my machine - so unimaginative i wondered what make me waste time to put it on there in the 1st instance. so dull - i think the well meanining individuals must have been asleep and let the distribution out without the 'must have' features installed. i heard so much about it i had to try it - what a let down - do not try it, it is a waste of time.

  135. Re:Ubuntu? No way. by AvitarX · · Score: 1

    You don;t keep your servers up to date then?

    --
    Wow, sent an e-mail as suggested when clicking on "use classic" banner, and got a fast response that addressed my msg
  136. Can we wait until they release? by Godji · · Score: 1

    Shouldn't we at least wait until the final release is available? The release candidate, by definition, is unfinished.

    1. Re:Can we wait until they release? by markild · · Score: 1

      The release candidate, by definition, is unfinished.

      Silly me, thinking release candidate, by definition, meant that it was thought of as ready for release.

      --
      Scully: Should we arrest David Copperfield?
      Mulder: Yes we should, but not for this.
    2. Re:Can we wait until they release? by Godji · · Score: 1

      That used to be the case, but these days the .0 release is the RC, the RC is the new beta, and beta is the new alpha. I don't know what that makes alpha, but it's pretty much the same as beta.

  137. Ubuntu swapping by nostriluu · · Score: 1

    I have had a few cases where Ubuntu has run out of real memory and goes into a swapping loop. Sometimes the whole system will be unusable - cursor updates take minutes. The only solution is to reboot. I haven't seen this kind of behaviour, where essentially one process is taking down the entire computer, on an OS for a long time. It may be related to running Ubuntu 64bit (don't ask), but I've seen a number of other people with this problem, and there are long standing bug reports.

    1. Re:Ubuntu swapping by quantumphaze · · Score: 1

      I have experienced the same thing. I was playing with Imagemagick (convert command) to convert some scans to pdf. On my 1GB RAM laptop it soon hit swap and made the system unusable. Luckily I could hit Crtl+C on the terminal it was running in before a reboot was needed.

      Simply running convert *.png test.pdf for a large enough amount of files killed it. It ran fine on my desktop with 4GB RAM.

  138. Re:Performance isn't its raison detre by Anonymous Coward · · Score: 0

    But not like a truck?

    No, that's love.

  139. Re:Ubuntu? No way. by Bert64 · · Score: 1

    Most gentoo users have relatively conservative CFLAGS, equivalent to the flags used to build any other distro, but targeting your particular cpu, eg "-O2 -march=core2"...
    What this means is that the compiler can use features only present on this cpu, for many apps this doesn't help but also doesn't hinder, for some things it makes a significant difference (eg media encoding, modern cpus have features specifically for media encoding)

    But what really matters about gentoo, is not the ability to set your cflags, it's the ability to set USE flags, whereby you can compile a program using different parameters or options.

    --
    http://spamdecoy.net - free throwaway anonymous email - avoid spam!
  140. Re:Ubuntu? No way. by ethana2 · · Score: 1
  141. Re:Flexibility and freedom are its raison d'Ã by ushimitsudoki · · Score: 1

    Indeed. And Microsoft will be happy to support an nlite "custom" installation. That is the power of closed source proprietary software: the customer can select only those parts they need and the company is glad to support consumer freedom! Match that open source!

    --
    Me and U(buntu) - my blog about Ubun
  142. Ubuntu? yes way. by Cope57 · · Score: 0, Troll

    Ubuntu = "Linux for Dummies"

    --
    http://www.accountkiller.com/removal-requested
  143. Perhaps Dilbert can provide some insight ... by Mostly+a+lurker · · Score: 1
  144. Crapware invasion ? by Anonymous Coward · · Score: 0

    Probably TLB issue due to large amount of useless crap running in background. This is very windowish :[

  145. Re:Performance isn't its raison detre by LingNoi · · Score: 1

    Microsoft isn't in anyway related to mono, stop trolling.

  146. Re:Ubuntu? No way. by thatskinnyguy · · Score: 1

    "Set and forget" != "Leave bone stock install in place and pray there are no vulnerabilities"

    --
    The game.
  147. Ubuntu != Canonical != Everything in Ubuntu by ciroknight · · Score: 1

    First, Canonical chooses to pay some of the developers that work on Ubuntu, but they don't draw any lines with respect to what Ubuntu packages.

    Secondly, Ubuntu is made up of packages that are their own, and packages that are Debian's. Ubuntu picks the important stuff (Linux, GCC, Xorg, GNOME versions), but the rest (50+%) of the software is simply imported from Debian.

    Testing the performance between Debian builds and Ubuntu builds would be an interesting thing to do here, since it would point us in the correct direction as to whether it is GCC's fault for poorly optimizing code, as to whether it's the kernel's fault for being more aggressive about power management and/or I/O performance latency, whether the graphic slowdowns are due to commercial closed source drivers (which we can do nothing about) or due to Xorg changes, etc. etc.

    --
    "Victory means exit strategy, and it's important for the President to explain to us what the exit strategy is." G.W.Bush
    1. Re:Ubuntu != Canonical != Everything in Ubuntu by mollymoo · · Score: 1

      Secondly, Ubuntu is made up of packages that are their own, and packages that are Debian's. Ubuntu picks the important stuff (Linux, GCC, Xorg, GNOME versions), but the rest (50+%) of the software is simply imported from Debian.

      The fact that they choose to simply import a load of Debian packages is still their choice. I'm not saying it's a bad choice, but it's their choice, their decision, their responsibility. You can't pass the blame upstream when it's Ubuntu who decides what they are going to use. Nobody held a gun to their heads and forced them to select packages that way. Ubuntu consists of precisely what they decided to include in the distribution, even if they don't take a great deal of care in selecting some of the packages.

      You seem to think they aren't responsible for what goes into their distribution and that's just not the case. If Ubuntu is getting worse it's because Ubuntu chose to include worse software. It's certainly nobody else's fault.

      --
      Chernobyl 'not a wildlife haven' - BBC News
  148. Re:Performance isn't its raison detre by kv9 · · Score: 1

    F, G, H, I,...M? Alphabet fail.

    I'm sorry, I didn't realize the Noobuntu naming scheme has any meaning. let me try again: Jumbled Jalopy, Kamikaze Kyte, Lame Labyrinth. I'm sure by the time they reach Moronic Monicker shit will be flying allover the place, performance wise.

  149. Re:Ubuntu? No way. by Sancho · · Score: 1

    At the risk of soundling like a 1990s AOLer, me too.

    And for most of the same reasons. Originally, I tried Gentoo out for the promise of more speed. Once I bought a new notebook, I realized that speed was not going to be an issue. I benchmarked Gentoo and Ubuntu, and found that the differences were negligible in most applications. I stuck with Gentoo for a while, using -Os instead of -O3, just because I was somewhat stubborn and already had the notebook set up. Finally, I think after the third or fourth day in a row of rebuilding Open Office, I realized what an absurd pattern I was getting into, and I scrapped the whole thing for Ubuntu.

    The nice thing about Ubuntu, in my opinion, is that you can still configure and compile things to your specifications if you like. I always disable Beagle/Tracker. I tend to run my own kernel. I disable a lot of the useless startup scripts. It's very nearly as good as Gentoo, without the drawbacks.

  150. Re:Performance isn't its raison detre by Anonymous Coward · · Score: 0

    No, because then his 2-incher get laughed at.

  151. Why a laptop? What about a desktop? by AliasMarlowe · · Score: 1

    The support for power management (including cpu scaling) has improved in the successive versions of Ubuntu, largely due to kernel improvements. I've certainly noticed this on my laptop, which has had Ubuntu since the beta of Breezy. The result has been better support for hibernation and so forth, but not any particular change in application performance.

    TFA said that speedstep was disabled for the tests, but were there any other laptop power saving features which would slow down user processes?

    I'd like to see the same set of benchmarks run on a desktop.
    I've been running Xubuntu on a 10-year-old Dell (450MHz P3, upgraded to 384MB) since Dapper, and have not noticed any degradation in its admittedly modest performance.
    I've also been running Ubuntu since Feisty on a 5 year old Dell (2GHz, 1GB), also without any particular performance changes.

    --
    Those who can make you believe absurdities can make you commit atrocities. - Voltaire
    1. Re:Why a laptop? What about a desktop? by gunnk · · Score: 3, Insightful

      I'll hitch on to your post and add a couple of thoughts.

      First, I use ThinkPads regularly (due to deep discount available through a workplace contract with Lenovo). The closed-source ATI video drivers are constantly a little off (for example, compiz users will find that videos tend to flicker). I'd really want to exclude that can of worms if I could. Your question about other power-reducing features is also a good one. In any case, though...

      The real thing the tests appear to show is that Ubuntu has evolved in such a way that any single process may be slower. However, we rarely use our computers as single-process systems. We want it to be doing multiple simultaneous tasks without allowing any single task to dramatically reduce performance in other areas.

      I *believe* the new resource allocation methods in Ubuntu (and the kernel and elsewhere) have improved exactly this sort of performance. However, by not allowing single tasks to hog resources in a way that would degrade the user interface and other running software any benchmark run against that single task would appear to indicate that the system is slower.

      The SYSTEM is not slower. The TASK is slower.

      That's a trade-off. Do you want single tasks to be faster or the overall responsiveness of the system to be greater?

      --
      Life is short: void the warranty.
    2. Re:Why a laptop? What about a desktop? by PastaLover · · Score: 1

      The SYSTEM is not slower. The TASK is slower.

      The system is slower in terms of throughput, but not in terms of latency (which is arguably what matters to consumers).

      Still, I find ubuntu a bit sluggish lately, wouldn't hurt to do some profiling. The article is one big advertisement for their "phoronix test suite" though.

  152. Re:Ubuntu? No way. by R2.0 · · Score: 2, Funny

    "This may come as a revelation, but Linux is about choice."

    But only if you choose correctly.

    --
    "As God is my witness, I thought turkeys could fly." A. Carlson
  153. Kyle by Anonymous Coward · · Score: 0

    n00buntu. If you want performance go with Gentoo :]

  154. Linuxhater FTW! by stream0 · · Score: 0

    linuxhaters.blogspot.com FTW! Read the archive.

  155. Re:Performance isn't its raison detre by doti · · Score: 1

    I would love to see Phoronix do a retest with some of the major patchsets removed and see if they can find the one or ones that cause performance decreases.

    git-bisect is your friend.

    --
    factor 966971: 966971
  156. Re:Ubuntu? No way. by Anonymous Coward · · Score: 0

    ya but they're making the wrong choice!

  157. I don't see the slowdown by motang · · Score: 1

    I have upgrade since Ubuntu 6.06 (as the each release get to Beta release) on the same laptop HP dv5000z and let me tell you Ubuntu 8.10 has been the fastest and the most responsive yet. Much faster than XP Pro (as I booted into the partition last week to update the software I had installed it and it was crawling, mind you that I hadn't booted into the XP partition since I set up that partition two and half years ago).

  158. Re:Flexibility and freedom are its raison d'Ã by initdeep · · Score: 1

    except that anybody who wanted to, could remake their install using something like nlite.

    and have only those items/services they desired.

    it's called knowing what you are doing.

    it's not something the masses would do, just like they wouldn't roll their own linux install.

  159. Yes by jimpop · · Score: 1

    Yes, Ubuntu is getting slower. Not only slower, but also more bloated. And not only more bloated, but more Window's like (all sorts of sharing services running by default). Not only all that, but consider that Ubuntu Server installs wifi networking, bluetooth, avahi-daemon, parts of evolution, etc. Bloat, bloat, bloat. But, if you spend 2 days cleaning up the default install, you can have a nice decently running system.

  160. Too all the people hating on the performance test by ph1nn · · Score: 1

    It obviously wasn't the goal of their article to isolate to performance issues to help find the problems.

    It WAS the goal to install each version separately and test different areas of the performance.

    And the fact is that in most areas Linux performance (not just Ubuntu, it's not like they do anything others don't) is going down. This isn't the only benchmark that shows it. Whether it be in new kernel features like CFS, the file system, i/o changes, bad drivers, gcc changes, gtk issues and so on. Linux in general isn't getting any faster, and this is a huge problem that should be looked into by ALLLL distros. Ubuntu, for example, could put an entire (6 month) release cycle, or even 2 release cycles (Apple is doing this with the next version of Mac OS) into performance optimizations but it wouldn't do a lot of good because they would have to be looking into Gnome, GTK, and all the other applications that go into it.

    Even for me I have noticed issues, windows resize much slower in the last few years, I can't even use NVIDIA drivers with Firefox as the Ajax performance goes to hell, even writing to USB flash drives is noticeably slower than windows. And it's unfortunate because Linux is all I want to use.

  161. So Xorg, Linux, GCC == Ubuntu? by ciroknight · · Score: 2, Insightful

    In this case, Ubuntu is the sum of the software it packages. But if one piece of software is slower, then Ubuntu's not slower, that piece of software is slower. 1 + 1 + 1 = 3, 3 != 1.

    This would remain true if Ubuntu were replaced with $DISTRO.

    --
    "Victory means exit strategy, and it's important for the President to explain to us what the exit strategy is." G.W.Bush
    1. Re:So Xorg, Linux, GCC == Ubuntu? by lennier · · Score: 2, Insightful

      "In this case, Ubuntu is the sum of the software it packages. But if one piece of software is slower, then Ubuntu's not slower, that piece of software is slower. 1 + 1 + 1 = 3, 3 != 1. "

      I find your definition of 'sum' interesting.

      1 + 1 + 1 = 3
      1 + 1 + 2 = 4

      If one piece of software gets bigger, sure seems to me like the sum gets bigger too...

      --
      You are not a brain: http://books.google.com/books?id=2oV61CeDx-YC
  162. Re:Flexibility and freedom are its raison d'Ã by wbo · · Score: 1

    Windows Fundamentals (based on XP Embedded) is included in included in my Volume License Agreement from Microsoft. Our agreement is fairly basis so I suspect it is included in most (if not all) volume license agreements that include Client operating systems (Vista and XP).

    I currently have it installed on several older PCs (late Pentium II's and early Pentium IIIs) with impressive results. It performs much faster than even Windows 2000 on the same hardware. Those PCs are currently being used by students in an open lab.

  163. Re:Ubuntu? No way. by AvitarX · · Score: 1

    yeah, but compiling is still compiling.

    I would think some server type applications take a while. Especially if it is a library that is going to require a lot of other stuff to be recompiled.

    --
    Wow, sent an e-mail as suggested when clicking on "use classic" banner, and got a fast response that addressed my msg
  164. Re:Performance isn't its raison detre by Anonymous Coward · · Score: 0

    FYI, actually having sex is usually more fun than declining it.

    Having the position to be selective and not just lose yourself because a vagina walks by, is more fun then ending up with the fat girl in a dark corner, because she has a pretty personality :)

  165. Re:Cars! by ISoldat53 · · Score: 1

    Victory is one point away from a tie, not necessarily failure.

  166. OS X faster with new releases by Anonymous Coward · · Score: 0

    As software gets more useful (and Ubuntu has, Vista not so much) it gets bigger and thus gets slower on the same hardware.

    The last few releases of Mac OS X were faster than the previous ones on the same hardware.

    (Of course some people remark that this simply shows how "bad" the earlier ones were.)

  167. Re:Ubuntu? No way. by onefriedrice · · Score: 1

    All of you former Gentoo users who found yourselves tweaking more than using your computer were obviously not using Gentoo correctly. I guess it's a matter of balance between power and self control. Just because you can easily tweak anything doesn't mean you should or need to. The principle of diminishing applies here just as well as anywhere. Smart users should know that balance.

    Here's another way to look at it. You could use Ubuntu the same as you used to use Gentoo, but of course it wouldn't be as easy. Apparently you need the barriers of an OS that is more difficult to tweak in order to keep you on task. That's fine, but it seems unneccesary to me. It's probably the same reason that many people prefer Windows to linux (and let's be honest, there are many people who do prefer Windows to linux).

    --
    This author takes full ownership and responsibility for the unpopular opinions outlined above.
  168. Re:Flexibility and freedom are its raison d'Ã by Dystopian+Rebel · · Score: 4, Insightful

    Not even talking about the advantage of being free:

    - for optimization, I can select the window manager
    - for optimization, I can select the desktop environment
    - I get full compatibility with the open-source ecosystem
    - I can install programs from the huge apt-get application universe (including programming languages and tools)
    - I run it all from a very short script, unattended
    - it is fully supported, with automatic updates and no nonsense

    --
    Rich And Stupid is not so bad as Working For Rich And Stupid.
  169. Re:Flexibility and freedom are its raison d'Ã by Dystopian+Rebel · · Score: 1

    You might want to check out nLite [nliteos.com]

    Thanks for the suggestion. I really see no advantage to using MS Windows.

    --
    Rich And Stupid is not so bad as Working For Rich And Stupid.
  170. Re:Performance isn't its raison detre by ilikenwf · · Score: 1

    Archlinux FTW, if you understand Linux well enough. If you're above that understanding, go for Gentoo, if you're totally a Linux master, go for slack (you're nuts!). Ubuntu has always seemed like a stepping stone. I learned Linux with it, and moved on to greater distros (Arch). I'm too lazy to use gentoo or slack, but I'm sure I could if I had the many hours to invest that such action would require. Ubuntu is a great OS balance between being windows-y enough and somewhat faster than windows, which makes it good for indoctrinating new users...outside of that, once they learn more, another distro would probably suit them better if they want more speed, stability, and control.

  171. maybe i know what's the problem by someone1234 · · Score: 0

    They add more and more stuff in the default install which get their 'fair' share from CPU time and memory. You probably don't need half of them, but they eat time. Somehow this reminds me of Windows...
    In earlier distros i cared about what's running in the box, nowadays i'm more and more lazy, i don't really notice any performance increase despite i buy a stronger machine every 2 years.
    Probably, if i start killing some of these bloatware, i would get back a faster system.
    It's not really the distro's problem. The more they try to please everyone, the more code they gotta stuff in by default.

    --
    Patents Drive Free Software as Hurricanes Drive Construction Industry
  172. Re:Performance isn't its raison detre by burroughsj1 · · Score: 1

    Much better! I particularly like Jumbled Jalopy.

    --
    Suse vivo vixi victum reduco is ea id creatura absit decessus a facultas Linux! Dev root, dev root!
  173. Re:Performance isn't its raison detre by Hordeking · · Score: 0

    Actually, on that note, has anyone noticed "Linux" and "Windows" are antonyms?

    --
    Disclaimer: The opinions and actions of the US Gov't are in no way representative of those held by this author or its ci
  174. Re:Cars! by eggnet · · Score: 1

    If your goal is winning, a tie is failure.

  175. Re:Performance isn't its raison detre by Count+Fenring · · Score: 2, Insightful

    Wow. +20 jerk-off points for using other people (women, in this case) as props for your own ego.

    I mean, misogyny may be the rule at old Slash U more days than not, but you, sir, are a grade-a special asshole.

  176. Re:Flexibility and freedom are its raison d'Ã by jcookeman · · Score: 1

    Yes this is the part that some people don't seem to understand, but, I guess it is fair to do comparisons on default installs. It's also fair to say that it's obvious as time goes on and features are added that the default install will become slower. One thing that really got under my skin was the damn indexer in Hardy though. Other than that it's ok. I have big expectations for Intrepid. Keep up the good work Ubuntu!

  177. Re:Ubuntu? No way. by Fallingcow · · Score: 1

    Yeah, I got sick of the recompiling. That was always incidental to me, anyway--I liked Gentoo because the "just works" options of the time (Mandrake and the like) rarely "just worked" and were full of annoying crap, while my favorite techy-oriented distro--namely Debian--had, relative to Gentoo (and IMHO, of course) an inferior package management system and way, WAY fewer packages. The whole, "it's faster because we compile EVERYTHING!" deal never really appealed to me.

    I still miss the breadth of Gentoo's package selection, and I still prefer Portage to Dpkg+apt, but Ubuntu saves me the compiling, finally delivers on the "just works" promise (mostly), and gives me automation where I want it (discs automount, network config is hassle free, hardware autodetection, etc.) without bogging my system down. I switched when I realized that Ubuntu was almost exactly the system I'd like to custom-build my Gentoo to be.

  178. Re:Performance isn't its raison detre by Count+Fenring · · Score: 1, Insightful

    Yeah, cause women aren't people, awesome.

    Seriously, get the fuck out of my gender, you waste of sperm.

    Anyone who's not a fucked-up case of arrested development eventually realizes that women ARE PEOPLE. It's not cool or a sign of your manliness to reduce them to just sexual objects; what it is is a sign that you are going to have one-sided, unfulfilling relationships, that you're going to be a bad lover (probably physically, but certainly over any span of more than a couple sexual encounters), and that you're going to hurt people who don't deserve it. Making you an asshole.

    Seriously, grow the fuck up. You are pathetic, and there's no excuse for this kind of bullshit, anywhere.

  179. -1, Off topic and troll by Anonymous Coward · · Score: 0

    More bullshit slashdot "moderating" here.

  180. Re:Performance isn't its raison detre by Bloodoflethe · · Score: 1

    Continued fail. {adjective} {animal (including fictional animals)}

    Game over. Insert coin(s) to continue.

    --
    "Little is much when little you need."
  181. Re:Flexibility and freedom are its raison d'Ã by _Sprocket_ · · Score: 1

    Windows Fundamentals (based on XP Embedded) is included in included in my Volume License Agreement from Microsoft. Our agreement is fairly basis so I suspect it is included in most (if not all) volume license agreements that include Client operating systems (Vista and XP).

    Great. If you happen to have a Volume License Agreement. And if you have a WinXP disk? What do you do then? With an Ubuntu install... any Ubuntu install... the option exists.

  182. Re:Flexibility and freedom are its raison d'Ã by Count+Fenring · · Score: 1

    Mod parent +1 ironic, I believe.

  183. -1, Off topic by Anonymous Coward · · Score: 0

    What in the fuck is going on with you retards doing the moderating here? This is BULLSHIT OFF TOPIC RANTING, not "insightful" in any way.

  184. Re:Ubuntu? No way. by mollymoo · · Score: 1

    There isn't much choice but to spend a reasonable amount of time tweaking Gentoo, or at least keeping it up to date (which does sometimes require tweaking). If you don't update regularly you end up having to spend hours fixing portage yourself, at least that was my experience when I last used it a couple of years ago. Perhaps it's better now and a box you've not powered up for six months can be brought up to date with less than two days compiling, interspersed with a couple of hours of fixing portage.

    --
    Chernobyl 'not a wildlife haven' - BBC News
  185. I'm glad that somebody else is noticing that, too! by cichon · · Score: 1

    Back in the days when processors were slow, and memory was scarce, it took 16 MB (sic) to run a full desktop. Like a word processor, a spreadsheet and a browser. 4 MB was a little tight, and 64 MB was plenty.

    I used to have a notebook (a Sony Vaio) with 64 MB RAM, a 300 MHz CPU, and it ran Linux SO DARN FAST! With emacs, Applix Office, 20 windows of Netscape 4, while listening to MP3 files, a compile in the background was barely noticeable.

    Today, I'm sitting in front of a multi-GB multi-GHz machine, and it's slow! This is some kind of sad, since I'd prefer a less feature rich system if I could run it on these slower machines that don't have a fan. E.g. on a NSLU2, you can't even open a web-browser. This thing has 32 MB memory while Firefox wants 250 MB. That's just too much for displaying a single web page, IMHO.

    The reason for this slowdown is as follows:

    1. the kernel

    In the recent time, everybody is optimizing for the big-iron servers (EMC, google, whatever), and nobody is looking any more for memory consumption and latency. And while the benchmark may still look great, the system becomes clumsy, bloated and un-resonsive.

    2. GCC / STL

    While the original idea of C++ was to make an OO language that's backward compatible with C, the older C++ program's can't even be compiled with the newer compilers! The templates, especially those of STL and boost, go totally out of whack when it comes down to memory footprint. It might give you a few percent of speed advantage to have a custom version of each single data type, but only if you don't use it much. Otherwise it's just totally trashing the cache. Not even to start talking about XML and consorts.

    And if you compile those programs, the compiler needs to compile 100eds of customized template versions in 100eds of source files, just to throw away most of this redundant work during linking. This is just a waste and it does not scale.

  186. Re:Flexibility and freedom are its raison d'Ã by 3p1ph4ny · · Score: 1

    Amarok is a KDE application. For those of us with KDE libraries already loaded into memory, it is small. If you'd stick to one DE the media player would be small; if you don't, you'll have to compromise, that's the nature of computing.

  187. Re:Performance isn't its raison detre by PitaBred · · Score: 2, Interesting

    I'd bet almost all of it is the effect of the scheduler. The benchmarks all show single tasks taking longer, but that's not taking into account multi-process performance. Is the desktop still responsive now even with a high-intensity background task running? I'd take that over the task finishing 5% faster any day.

  188. Re:Flexibility and freedom are its raison d'Ã by amRadioHed · · Score: 1

    Their are plenty of small media players out there for Linux. I would not classify Amarok as one of them.

    --
    We hope your rules and wisdom choke you / Now we are one in everlasting peace
  189. Re:Ubuntu? No way. by mickwd · · Score: 1

    I just want something that is fast, easy to use and gives me as little of a headache as possible.

    :

    Mod the parent up, I'm in the exact same boat.

    :

    I was pretty much exactly the same.

    Then why the hell were any of you running Gentoo in the first place ?

    Gentoo is all about tinkering and customising, and being offered the flexibility to do so, not ease of setting up.

  190. Re:Flexibility and freedom are its raison d'Ã by _Sprocket_ · · Score: 2, Interesting

    Indeed. I might have to go install something like XPLite or create my own installation media with nlite/vlite. It's really taxing firing up a GUI and unticking a few boxes.

    Yes, unticking boxes is easy. So's deleting a file. But what happens after the fact?

    Those utilities do look like a step in the right direction. Pity they're from a third party and a complete hack (albiet a very cool looking one). What happens when Microsoft releases a service pack? What happens if you change your mind and want to install a component?

    I know with Ubuntu I'm removing a package using the very same tools provided by the base distro. When I update, I only update whats installed. And I know I can always install / re-install anything missing at a later time.

  191. Re:Performance isn't its raison detre by LWATCDR · · Score: 1

    Maybe but that will take a lot more testing to find out.
    The think is that an ideal scheduler would detect that the UI is mostly static and not take a lot of CPU time unless you where using the desktop. Of course nothing is ideal.
    I also doubt that there is one answer.
    1. GCC compile times. Did they use the same version of GCC for all the tests? If not it could be extra optimization in the compiler taking up cycles. That is something I can live with.
    Did they use the same file system in all the tests?
    Just a lot of variables to sort through BUT Mono isn't a big problem in any of the tests I have listed.

    --
    See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
  192. Re:Performance isn't its raison detre by danieltdp · · Score: 1

    What the smiley has to do with it?! Oh, wait!

    --
    -- dnl
  193. Re:Performance isn't its raison detre by Anonymous Coward · · Score: 0

    You know, sucessfull people usually don't brag that much

  194. Re:Flexibility and freedom are its raison d'Ã by danieltdp · · Score: 2, Funny

    The user's, of course

    --
    -- dnl
  195. It really is getting slower by cecom · · Score: 1

    The truth is it is getting slower, which is causing serious problems, and is annoying the sh*t out of me.

    I recently upgraded my laptop to Kubuntu 8.4. I am not imagining it - it really is noticeably slower than before. Applications start slower, there is more disk activity and the disk itself feels slower. The laptop in question has Intel Core Duo and 2 Gigs of RAM, which should be plenty.

    I haven't investigated in detail why it seems so slow, but it is not something obvious: the disk UDMA is enabled, there is available RAM, there isn't a background process running and taking all CPU time.

    A co-worker's computer, which is somewhat less powerful than my laptop, became practically unusable after the upgrade to Kubuntu 8.10.

    So beyond mere annoyance this is creating real problems. We are seriously considering migrating all our [K]Ubuntu machines to Debian. Besides the speed, it has the important advantage that you don't need to upgrade your OS every 6 months (which is utter idiocy: I don't have time to upgrade my OS twice in a year, I have work to do. If you ask why I don't use the LTS - the previous LTS is too outdated and the new one is too slow).

  196. Re:Performance isn't its raison detre by Chutulu · · Score: 1

    then quit bitching about Vista being slower than XP you freetards

  197. Re:Performance isn't its raison detre by recoiledsnake · · Score: 1

    Really? Mono is related to Microsoft in every way except having MS code in it. All the specifications that Mono is built to are developed at Microsoft.

    --
    This space for rent.
  198. Re:Performance isn't its raison detre by recoiledsnake · · Score: 1

    Like running a system (Windows) that crashes even if it not doing anything special.

    The 90s called, they want your trolling back.

    --
    This space for rent.
  199. CK scheduler? by BlackSabbath · · Score: 1

    Out on a limb here...
    I presume the tests ran single tasks. As other posters have guessed, I also think its VM related. This would seem to be a result of VM advances that favour multi-process throughput/responsiveness over single-tasking speed. Are most of the recent VM changes geared at server or desktop use?

    I would be interested to know the results if the CK scheduler was used. Con always focused on typical "desktop user" workloads in tuning his scheduler.

    Just a thought...

  200. I am currently running Ibex and... by shellster_dude · · Score: 1

    Thus far it has been starting up much faster than its predecessor. Also, it starts up much more smoothly. It almost always takes the same about of time from boot to fully loaded. Previous versions of Ubunutu tended to be all over the charts on my system. While, I certainly am a fan of faster boot times, it isn't a make or break deal for me. It definitely should be a consideration, and I hope the Ubuntu devs devote more time to it in the future. However, Ubuntu would need to approach Windows XP start up times before I started to get seriously pissed off.

  201. get freebsd by Alvaro+Martinez · · Score: 1

    i use freebsd on a laptop with a 500Mhz amd k6, 128M on RAM. it runs gnome, vlc and can even run winXP on qemu pretty acceptable

  202. Re:Performance isn't its raison detre by Anonymous Coward · · Score: 0

    shut up and get used to progress.

    Progress towards what? Software quality can be divided into multiple dimensions, a few of which are:

    • parsimony: how small amounts of disk/RAM space is used
    • performance: how long between the user has given the command until it's done
    • feature-richness: many commands
    • ease of use: they are easy to give
    • efficiency of use: they do a whole lot, i.e. perl golf
    • security: only you can give them
    • safety: you can ungive them

    I'm simplifying a bit here, and leaving some out.

    Sometimes you can make software better in one dimension without making it worse in others. Sometimes you can't. When you can't, what exactly is progress? What is it progress towards? People value different things, and their values change in some contexts.

    Invalid token: progress.

    -- Jonas K

  203. Re:Performance isn't its raison detre by kv9 · · Score: 1

    you should look up sarchasm.

  204. Re:Performance isn't its raison detre by agrounds · · Score: 2, Insightful

    From time to time I download the newest Ubuntu and check it out to see what all the new fuss is. I am continually disappointed though, and it rarely makes it a week before being purged.

    It's not just the bugs and the long app-launch times. It's more than just the gratuitous resource consumption. It's the whole thing... It's bloat on top of eye-candy bloat with a hefty helping of fanboy zealotry.

    I love *NIX very much but it is precisely because I love it that I have to point out that something is rotten in the state of Denmark.

    When did the userbase of linux go so wrong? When did it start? Those of us that have been around for quite a while have watched the rockstar rise and subsequent plummet in the communities surrounding it. 10+ years ago you could jump into IRC type /join #[your distro] and be surrounded by people that truly loved their systems and would help you without being condescending, resorting to ad hominem attacks, or calling other *nix variants crap. A lot of those guys, myself included, were using Solaris, HP-UX, Tru64, AIX, etc. during the day and spending some of our evenings just helping out others that were interested in or 'testing the water' with linux. The slackware community was great. Even a little more recently when Gentoo first really got rolling between 1.0 and 1.2 the community on EFNET in #gentoo was a shining example of what a userbase should be. I spent many an hour in there helping people figure out their CFLAGS, configuring their XF86Config, and the like.

    With the influx of more and more folks that seem to be vastly more focused on hating Windows, Mac OSX, and even other UNIX variants, the face of linux has changed. People that used to use *NIX as a personal choice did so because they truly loved computers. Now it seems to be the equivalent of a battle standard. Your Operating System is your country, your flag, and your religion. Thousands of angry people focused on their hatred of anything that is unlike themselves. It's GNU/Xenophobia.

    The programming ramifications of this have become pervasive throughout many of the more popular distributions. In fact, the fundamental idea of the ideological Bazaar has been replaced by the Cathedral of intolerance. Instead of a focus on excellence and listening to the end-users, more and more developers are dismissive and prone to flame. More time is spent developing completely worthless and unrelated 'features' than in solidifying and optimizing the current code. Instead of, say fixing GNOME's inability to remember where I want my launch icons on the panel, we get wobbly windows that add absolutely nothing to the value of the desktop. Instead of writing just one really, really good IDE for C development, we get oodles of feature-incomplete environments that can't even compete with older Visual Studios or XCode; and this is supposed to be forte of *NIX. As children we are reinforced to eat our meat and vegetables before we get dessert. Yet more and more developers focus on the candy and leave the meat (optimizing) and vegetables (squashing bugs) virtually untouched. It's not as exciting of course, but it is necessary. In *buntu I struggle with yet another audio layer to cover the other layers to figure out why my sound card is doing a darn good impression of a french mime when I try to play some music. Meanwhile, a thousand fanboys upload yet another Youtube video of a spinning desktop cube with a Moby soundtrack.

    Perhaps it is the fate of those of us from the previous generation to make way for the new one, but as we do so there should be some guidance, some hope, and some direction given. Reading through the comments in this story really drive home how far this has gone and the need for a gentle hand to remind people not just of the Bazaar, but also that we need to eat our meat and vegetables before we get dessert. This reply has gotten much longer than I originally planned though, but perhaps this conversation can continue in another venue.

  205. Re:Performance isn't its raison detre by moreati · · Score: 1

    Perhaps: Software becomes bloated one bit at a time.

  206. Re:Flexibility and freedom are its raison d'Ã by Yfrwlf · · Score: 0, Flamebait

    I can install programs from the huge apt-get application universe

    Huge, but still fragmented and not huge enough, your repo should be all of Sourceforge and everything else. You shouldn't have to live in the walled garden of your distro's repository. In Windows and OS X, packages are always installable for the most part. After Alien is integrated into the existing common package managers; or the managers are made compatible with RPMs, DEBs, and others; or the formats are upgraded so they can be more easily adopted by the existing managers; or at least one new format is created that is easy for the managers to adopt, this will continue to be a problem for Linux users, the ones who don't want to spend all day looking for dependencies and compiling software just to run a program. Compatibility with the existing Linux software ecosystem, while being mandatory, doesn't mean much to normal computer users when they can't click to install software from some developer's website because the developer gave up trying to support the thousands of distros who thought they were doing everyone a favor by requiring different packages for every version they put out.

    Then, users will finally be able to update their system directly from the developers who put out those security updates instead of waiting for their distro to do it for them, because the devs will finally be able to support them directly because they're on GNU/Linux, and everyone will be much more free. The distro companies may not like that freedom very much, but tough for them, it's open source software so they will have to deal instead of trying to profit off that aspect of distro lock-in.

    --
    Promote true freedom - support standards and interoperability.
  207. Re:Performance isn't its raison detre by Bloodoflethe · · Score: 1

    I understand sarcasm quite well. I also know when to use it and when to indicate that I am using it.

    You see, the fact that you persisted in poor approximations actually makes you look less intelligent than if you were to produce your satirical responses in proper form and format (i.e. following the style and usage rules for the names but using derogatory adjectives) for a more relevant rejoinder with the poster you were attempting to razz.

    --
    "Little is much when little you need."
  208. Performance regressions by Anonymous Coward · · Score: 0

    I've seen a few performance regressions get flushed through the kernel. Some are just plain bugs and there are others certain scaling related improvements that sometimes trade off raw performance for throughput but ah looking at some of these benchmarks especially the memory read/write area has me thinking its got nothing to do with kernel bugs or bloat.

    I'm thinking whoever did the benchmark may have neglected to properly configure bus/power management or use the same cpu microcode revisions.

  209. Re:Flexibility and freedom are its raison d'Ã by Anonymous Coward · · Score: 0

    Please share your thinbuntu script? I have an old Thinkpad R30 and this would be nice to use. Much thanks.

  210. Re:Performance isn't its raison detre by Anonymous Coward · · Score: 0

    FYI, actually having sex is usually more fun than declining it.

    What is this "sex" thing you speak of?

  211. Re:Flexibility and freedom are its raison d'Ã by _Sprocket_ · · Score: 1

    I've never used nlite. Looks interesting - I'll have to check it out in more detail.

    Having said that, if I wanted to do a minimalistic Ubuntu install (as described by the parent), it doesn't take burning a special install disk. It's there. Included. And just as viable as the full-on all-defaults-enabled distro. No additional 3rd party apps and unexpected hacks required.

    Whether "the masses" would do this or not... possibly not. But I'm not sure how that's a factor.

  212. Is Ubuntu getting slower? by Clived · · Score: 1

    Well I converted from Mandriva 2008 to Ubuntu 8.04 about three months ago and I feel that Ubuntu is totally awesome. Fast friendly and easy to handle. I upgraded online to Ubuntu 8.10 and I can't say that I noticed any "slowing down" from 8.04.

    There again, maybe I'm just happy to have found a distro which runs well on this piece of junk I call a computer.

    My two bits

    --
    Clive DaSilva Email: clive.dasilva@gmail.com Ubuntu 18.10 Kernel 4.18
  213. Ummmm, using available tech? by fatalGlory · · Score: 1

    Yes. I would imagine ubuntu should get slower - on the same computer. Features in the OS get upgraded to make use of faster hardware as it becomes common. The criticism leveled at Vista is that it took too big of a leap in one go and the average computer didn't run it well enough.

    The great thing about Ubuntu over Vista is that it is so customizable. You can use Vista with less eye candy, but you can use Ubuntu with Xfce instead of Gnome, or Fluxbox, or no desktop environment at all if performance bothers you. Sure Hardy may run slower in tests than Feisty on the same machine. Change all the settings in Hardy so that it is essentially Feisty and this "issue" goes away.

    --
    Censorship is the opposite of education. If neo-darwinism were defensible, people would not need to try and censor ID.
  214. Re:Performance isn't its raison detre by petermgreen · · Score: 1

    software gets slow a percent or so at a time.

    Making something take one percent longer doesn't seem significant and probablly won't even be noticed in most testing. But do that a hundred times and your performance is worse than halved.

    --
    note: i'm known as plugwash most places but i screwd up registering that here somehow in the past and now can't register
  215. Re:Flexibility and freedom are its raison d'Ã by Pope+Raymond+Lama · · Score: 1

    You only forget that you are not supposed to be doing that. The very "license" (and I use parenthesis because it is so far from a license the y actually coined another term to call it) of windows say you can't hack it -
    SO, no neither XP, neither other proprietary things are meant to be tweaked , and the fact they can illegally tweaked don't make then better. They only make you a trespasser.

    --
    -><- no .sig is good sig.
  216. These are 32-bit kernel benchmarks by Anonymous Coward · · Score: 0

    Phoronix stated in the second paragraph: "For our testing we had used the final Intel 32-bit releases of the four most recent Ubuntu releases except for Ubuntu 8.10 "Intrepid Ibex" where we used the Intrepid release candidate."

    This benchmark would be more valuable if it compared how the ubuntu repository fared when compiled with the different available compilers:
    -Intel compiler
    http://www.ubuntugeek.com/howto-install-intel-c-compiler-10-on-ubuntu-feisty-fawn.html
    -GNU compiler
    -SunStudio compiler
    http://developers.sun.com/sunstudio/downloads/index.jsp
    -Portland group has unifying binary for both intel 64-bit and amd 64-bit
    http://www.pgroup.com/about/why_pgi.htm
    -Pathscale compiler
    http://www.pathscale.com/node/189

    It would also be nice to discuss benchmarks for:
    -AMD 32-bit Kernel
    -AMD 64-bit Kernel
    -Intel 64-bit Kernel

  217. Re:Ubuntu? No way. by Legion_SB · · Score: 1

    Mod me Redundant but the above post could've been written by me. When I was using Gentoo, the other options seemed to be either a too-out-of-date-for-the-desktop Debian install, or some other distro with overly tiny repositories.

    Ubuntu came along, leveraging Debian's big software repository while providing up-to-date key packages in their own repos.

    I never looked back. Debian powers my servers and Ubuntu powers my desktops (along with OS X)

    --
    'a';DROP TABLE users; SELECT * FROM DATA WHERE name LIKE '%'... if you're reading this, it didn't work.
  218. lol by Anonymous Coward · · Score: 0

    Ubuntu = Vista of the linux world.

  219. Re:Cars! by Annymouse+Cowherd · · Score: 1

    What's failure then?

  220. Re:Cars! by TheSpoom · · Score: 1

    If your tie is win, you shall not fail.

    --
    It's better to vote for what you want and not get it than to vote for what you don't want and get it.
    - E. Debs
  221. Re:Performance isn't its raison detre by TheSpoom · · Score: 1

    +1, Epic Win.

    --
    It's better to vote for what you want and not get it than to vote for what you don't want and get it.
    - E. Debs
  222. Re:Flexibility and freedom are its raison d'Ã by setagllib · · Score: 1

    That's just retarded. KDE and GNOME libraries comprise similar functionality to the Windows foundation, in different ways with different strengths, and it just happens you are able to run both at once.

    There are plenty of "small" media players you can install that depend on as few libs as possible, including some that run via a command line interface. You have the full spectrum of features and requirements available to you. Try saying that about Windows, where a multi-megabyte beast like foobar2000 is considered "lightweight".

    --
    Sam ty sig.
  223. Re:Performance isn't its raison detre by Samah · · Score: 1

    http://en.wikiquote.org/wiki/Futurama#The_Problem_With_Popplers
    Fry: They're like sex, except I'm having them!

    --
    Homonyms are fun!
    You're driving your car, but they're riding their bikes there.
  224. Re:Performance isn't its raison detre by ypctx · · Score: 1

    You're so cool, may I be your fanboy?

  225. Re:Cars! by master5o1 · · Score: 1

    The outcome you don't want. Aim for failure, you can always be surprised when you win.

    --
    signature is pants
  226. I'd really like to see better benchmarks by m6ack · · Score: 1

    Yes, Ubuntu is getting slower in some ways & faster than others.

    I'd really like to see better benchmarks than this... A lot of this could be caused by the compiler -- as they are compiling the suite separately on each platform? Also, the is it something underneath Java that is a problem, or just the new Java version (which one?).

    I know and understand that GCC takes longer to compile code -- but that's because the GNU's are trying to put together an architecture that is easier to create optimizations in the chain. Looks like they have a long way to go to actually build better code... Unless the code intentionally tries to defeat GCC from it's attempts at optimizing.

    What I'd like to see is more focused benchmarks so that it's possible to track the problems down.

    1) Kernel benchmarks compiled on the same version and flags of the compiler in ANSI C.
    2) Java benchmarks per java version in each OS version.
    3) GTK Benchmarks written in C -- same binary.
    4) Mesa benchmarks written in C -- same binary.

  227. Re:Performance isn't its raison detre by Mista2 · · Score: 2, Insightful

    I agree about cars. I want to replace my 1990 MR2 Turbo which I bought in 2000 for NZ1313K. Now if I try and find a new car for $13,000 there is no way I can find any mod engined 200HP rice rockets. They are all old slow junkers or much more expensive than my budget allows.
    They are also loaded with extra features like ABS brakes, adaptive dampers, airbags, pretensioning seatbelits, CD changers, run flat tyres, and all manner of other goodies that just add weight, not performance.
    Just like new OS's I guess

  228. Scheduler Changes - Not necessarily an Increase by gbutler69 · · Score: 1

    If I recall correctly, the scheduler changes that were put into the Kernel were primarily to increase "Perceived Desktop Performance" at the expense of "Actual Overall Performance". What this means to me, is that there should be less latency switching between applications and the application I'm currently focused on should get scheduling preference, but, the overall performance, meaning actual work getting done by all applications, could be less.

    --
    Over-the-top Response Guy! Giving "Over-the-Top Responses" since 1970.
  229. Re:Performance isn't its raison detre by HiThere · · Score: 1

    You mean like "Jealous Jabberwock"?

    Personally my preference would be for something like "Jumpy Jackalope", but that's not the tenor that I think he was reaching for.

    --

    I think we've pushed this "anyone can grow up to be president" thing too far.
  230. Re:Arch Ubuntu by Whiteox · · Score: 1

    Hey! I liked my bubble!
    Now I'm going to spend the rest of the day installing Gentoo.

    --
    Don't be apathetic. Procrastinate!
  231. Re:Flexibility and freedom are its raison d'Ã by Anonymous Coward · · Score: 0

    "your repo should be all of Sourceforge and everything else." ./configure && make && make install. It's better that its Windows' counterpart, isn't it? (or is this a case of double measure? When will be Windows "repo" all of Sourceforge and everything else?)

    "You shouldn't have to live in the walled garden of your distro's repository."

    Why not?

    "In Windows and OS X, packages are always installable for the most part."

    Sure? Try installing "all of Sourceforge and everything else" on Windows or OS X. When you manage to install even a significant fraction of what you can install on Linux return, please.

    "After Alien is integrated into the existing common package managers"

    First you should know what are you talking about. You'd know then Alien doesn't add so much to the equation.

    "or the managers are made compatible with RPMs, DEBs, and others"

    Again, you don't know what are you talking about but still try to offer your opinion as one from a knowledgeable person.

    "this will continue to be a problem for Linux users, the ones who don't want to spend all day looking for dependencies and compiling software just to run a program."

    I simply can't remember when was the last time I compiled a program on my desktop, probably years. The plain truth is that people that "don't want to spend all day looking for dependencies and compiling software" simply don't do that.

    "Then, users will finally be able to update their system directly from the developers who put out those security updates instead of waiting for their distro to do it for them"

    That's so wrong from so many points of view, I simply don't know where to start. Anyway, I'll start by the easy part: how many of those developers you mention do indeed publish security (mind you: just *security*) updates? Answer: almost no one. So you are talking about an imposibility; you are offering an "update" for a non-existant landscape. And then, even on the land of "compile once run everwhere", Java, that's it, version X of foo only runs nice along version Y of bar (as long as the zoot toolkit is installed too, of course). Are those "end developers" so to say, the ones accepting the burden of integrating their software nicely with dozens of other related programs? Is my version of ssl going to run nice with the last version the people from glibc provides? Will work last Venema's Postfix with last version of Amavis? (but Amavis people is still working with last-but-one Postfix). How long will Venema support his last-but-one Postfix? I'm still not ready to integrate the changes from the last one on my environment...

    It's easy to offer almighty solutions to imagined problems... as long as it's not one himself the one to deal with all those nitty-gritty details (all those petty nuisances are not for us, the people with the great vision, are they?).

  232. Re:Cars! by Requiem18th · · Score: 1

    Epic thread if I have ever seen one is indeed, is epic.

    --
    But... the future refused to change.
  233. Re:Ubuntu? No way. by Anonymous Coward · · Score: 0

    Slackware -- 1. Anglo Saxon word, meaning "I'll keep looking for glory fighting a war that has already been won by not less than 10 other factions of Linux." 2. Opposite of Slack (Sub-Genious) 3. Some old bastard that couldn't cope with the change that made life easier so they could go on and cure cancer, aids, write great games in OpenGL.

    Fedora -- 1. French word for giving the man a hand litterly (reach around). Think French hooker working the corporate ladder but just enough of a slut to get stuck on wall street as the dirty secret wives find out about from chapter 11 filings.

    Ubuntu -- 1. Ancient Atlantian word for enlightenment. 2. The church of subgenious refers to this as Slack. (you can slack off and it still works for you). 3. Will brang world unitity. 4. The Anti-Christ Ph33r's it.

    Mandriva -- 1. German word for " Frankly France Fucking Sucks " 2. The sound of dribbling gravey out of your mouth on your mama's mashed potatos. 3. Man-Driver -- Something remotely sexual. Mandrake was a renown homosexual. That's why they went atomic when they found out the name was being used by a french whore selling tricks for x86.

    Debian -- Ghandi -- 1. An Incan word for a Hindu cow that when beat with a stick pops like a pinata with all sorts of delicious candies full of razor blades and rat poison. 2. A mistranslation of an Incan word for a Hindu cow that when beat witha stick pops like a pinata with a bunch of mexicans hiding out in hopes of finding a job in America. 3. An obsessed pervert who liked sed(x) so much he named his condom after his girlfriend.

    *I can keep going as this is fun

    Turbo-LInux -- A Japaneese word for a french whore who didn't use a condom. Nobody has heard much of her lately and it was assumed that she was a parasite who got absorbed into another parasitic whore named deb L i n dos (dallas). *say it really fast. ... hope I helped with the lexican

  234. Re:Flexibility and freedom are its raison d'Ã by mvdw · · Score: 1

    I currently have three Ubuntu-based systems. I have customized the GNU/Linux distribution for each system (...) using a script that I call "Thinbuntu". It has all the features I need.

    Can you share this script with the rest of us?

  235. Re:Performance isn't its raison detre by Vexorian · · Score: 1
    I love ubuntu, however Ubuntu is getting slower.

    But hey, since it is still faster than windows XP, I don't care.

    --

    Copyright infringement is "piracy" in the same way DRM is "consumer rape"
  236. Re:Performance isn't its raison detre by Anonymous Coward · · Score: 0

    Oh come on, thats funny shit! I havent laughed this hard in weeks!

  237. Yes by fortapocalypse · · Score: 1

    Almost all OSes get slower as they grow, but processing speed gets faster, and in the end, it will even out. You can count on this trend continuing to happen at least until AI gets to the point where computers can get smart enough to redesign and reimplement their OS to be more efficient. However, at that point SkyNet becomes self-aware, and you'll need to contact John Connor via a ham radio.

  238. Re:Performance isn't its raison detre by ozphx · · Score: 1

    Making love is more like a series of tubes.

    --
    3laws: No freebies, no backsies, GTFO.
  239. Re:Performance isn't its raison detre by ozphx · · Score: 1

    I was in a similar situation. It was really awesome.

    Then I had a spiritual moment and got married.

    I recommend you don't do this.

    --
    3laws: No freebies, no backsies, GTFO.
  240. Re:Performance isn't its raison detre by phyrz · · Score: 1

    Jaunty Jackalope FWIW.

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

    "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."

    --
    Don't point that gun at him, he's an unpaid intern!
  241. Why? by Bootarn · · Score: 1

    A number of significant kernel changes had went on between these Ubuntu Linux releases including the Completely Fair Scheduler, the SLUB allocator, tickless kernel support

    These are kernel level modifications. Many distributions have switched to the SLUB allocator as the default allocator. CFS has proven it's worth (that's why it was included in the mainline kernel). I don't know whether the tickless kernel support is enabled for most stock kernels or not. I use Arch Linux, which upgrades the kernel if available every time I do a system upgrade. So far, I haven't experienced any degradation in performance. It's likely that the drivers for some chipset have been flawed, and this chipset is used in the machine where the recent Ubuntu was tested. Also, if the desktop effects feature got activated during upgrade of the text machine, that can explain a small amount of performance degradation, but not as much as a 2x-3x degradation in LAME performance.

    The point is, we must test Ubuntu on different machines to determine if it's Ubuntu, the kernel or some hardware specific issue that is causing this.

  242. Re:Flexibility and freedom are its raison d'Ã by Bert64 · · Score: 1

    That's just how KDE is designed...
    That said, if you have a distro which defaults to gnome it's unlikely to have included amarok as it's default media player, and if you want a lightweight media player there are several better choices.

    --
    http://spamdecoy.net - free throwaway anonymous email - avoid spam!
  243. Re:Performance isn't its raison detre by sg_oneill · · Score: 1

    See, there's this thing called an analogy. It's kinda like a car

    Will red ones get me a blowjob?

    --
    Excuse the Unicode crap in my posts. That's an apostrophe, and slashdot is busted.
  244. Re:Performance isn't its raison detre by Durkheim · · Score: 1

    Removing my mod.

  245. Re:Ubuntu? No way. by Abasher · · Score: 1

    Ehm, how many forks of Linux can you list? I know of none that have wide enough use to be confusing to anyone susceptible to confusion over the matter.

  246. Re:Performance isn't its raison detre by kv9 · · Score: 1

    hi there, new best friend!

  247. Re:Performance isn't its raison detre by kv9 · · Score: 1

    how does Butthurt Busybody work out for you?

  248. The 5% theory? by Sits · · Score: 1

    In an internal memo SGI's Tom Davis criticised some of what had come to pass in Irix 5.1. Within the memo he propsed that it was effectively a death by a thousand cuts (well 20 cuts I guess):

    What's most frightening about the 5.1 performance is that nobody knows exactly where it went. If you start asking around, you get plenty of finger-pointing and theories, but few facts. In the May report, I proposed a "5% theory", which states that each little thing we add (Motif, internationalization, drag-and-drop, DSOs, multiple fonts, and so on) costs roughly 5% of the machine. After 15 or 20 of these, most of the performance is gone.

    The memo is a great read and if you have the time it's worth a look (it's a shame no one will see this comment really). It's hard to tell whether Ubuntu has a similar situation - without correct measurement it can be terrifyingly hard to tell whether performance has really declined...

  249. Re:Flexibility and freedom are its raison d'Ã by Dystopian+Rebel · · Score: 1

    If you can run apt-get commands, you can make your own subset of Ubuntu. Grab and burn the Minimal installation CD that suits you:
    https://help.ubuntu.com/community/Installation/MinimalCD

    LTS gives you the most bang for the buck. (o;

    Boot; install the minimal command-line environment (CLI).

    Restart. Run apt-get update.

    Then run apt-get {pkg-list}

    where pkg-list is a space-separated list of the pieces that you want to install. Here's my list.

    xorg gnome-core gdm synaptic gnome-app-install gdebi ubuntu-gdm-themes gconf-editor gnome-volume-manager gparted conky alacarte xterm smbclient update-manager update-notifier pcmanfm file-roller nautilus-open-terminal evince cups-pdf system-config-printer-gnome arj genisoimage lha ncompress p7zip p7zip-full sharutils unace unrar build-essential
    firefox gnome-utils

    --
    Rich And Stupid is not so bad as Working For Rich And Stupid.
  250. Phoronix has done cross distro benchmarks before.. by Sits · · Score: 1

    For example you can read an Ubuntu/Mandriva/Fedora comparison. You can find other benchmarks by scouring the Phoronix articles page. I have a feeling the result is usually a wash with no distro coming out on top on all tests though...

  251. Re:Ubuntu? No way. by harry666t · · Score: 1

    One thing I do not yet understand is why all modern distros targeted at at least "mid-end" hardware (say, you won't probably try to run Ubuntu with less than 128MB of RAM) are compiled for >=486. As far as I recall, the original Pentium was 586 and Pentium Pro was the 686. How many of the <=586s have you seen "in the wild", running Ubuntu? Or at all, for that matter?

  252. Re:Phoronix has done cross distro benchmarks befor by LWATCDR · · Score: 1

    I have read those in the past. I was wondering if something had changed.
    If it is still a wash then the decrease in benchmark performance is a Linux issue and not an Ubuntu issue.
    Which to be honest what I would bet really is going on.

    --
    See my blog http://ilovecookes.blogspot.com/ for light hearted technical information.
  253. Re:Performance isn't its raison detre by Bloodoflethe · · Score: 1

    That was almost good, but your skill with adjective use needs to be redefined, and as normal, you are now attacking me instead of the distribution. Maybe I should cry? I would prefer to give you a more intelligently worded insult. I'll even use Busybody for your noun, since that was relatively clever.

    Maybe Bilge-sucking Busybody?
    Too piratical? What about Bobble-brained?

    If you feel you absolutely must go with the whole "gay" reference (I wonder why you would choose such a immature insult...), perhaps you would consider Sphincter-breached Slug.

    It may not be a "b" but it works much better from the standpoint of including an actual animal instead of an adjective that nominally serves as a noun.

    Or you could apply those to yourself and save me the trouble of finding an insult for you.

    --
    "Little is much when little you need."
  254. Re:Performance isn't its raison detre by Bloodoflethe · · Score: 1

    Hah, I wonder if you saw that I had mentioned the Jabberwock in an earlier reference to Ubuntu.

    Indeed Jealous Jabberwock could serve or for those that are tired of it all - Jaded Jabberwock?

    --
    "Little is much when little you need."
  255. Re:Performance isn't its raison detre by Anonymous Coward · · Score: 0

    You make a good point, but finding out that the girls are purely imaginary constructs of your broken ego is less fun.
    Probably best he stick to declining them.

  256. Re:Ubuntu? No way. by shutdown+-p+now · · Score: 1

    Same here. I started with Gentoo, and after two years had grown sick of having to compile everything. Moved on to Debian - was much better, but I eventually got tired of mucking with config files in vim. So, as far as Linux goes, it's Ubuntu for me now. I don't really miss anything, and my past experience with Debian still pays off every now and then (and, well, I just like aptitude to manage my packages).

  257. UBUNTU IS NOT GETTING SLOWER by teeteebahbah · · Score: 1

    I have Ubuntu 8.10 Intrepid Ibex Beta on the same PC, same H/D, that I have version 8.04 Hardy Heron. I timed the boot up time for both versions. Ibex actually boots up slightly faster than Hardy.

  258. Re:Performance isn't its raison detre by kv9 · · Score: 1

    I accept your apology.

  259. Re:Flexibility and freedom are its raison d'Ã by Anonymous Coward · · Score: 0

    True - but you still can't plug in a random wireless card from BestBuy and connect to the net.

    And you can't play 95% of the games people play.

  260. Re:Performance isn't its raison detre by Bloodoflethe · · Score: 1

    You made me apologize a lot. You are a tough person to please.

    --
    "Little is much when little you need."
  261. Re:Ubuntu? No way. by SanityInAnarchy · · Score: 1

    Turns out that when you get it to a usable state, and stop screwing with it, Gentoo works a lot like Ubuntu.

    I think Ubuntu provides more, but even assuming that's the case, I'll take Ubuntu any day for the simple reason that precompiled binaries install faster than source code.

    Sure, you can use binary tarballs on Gentoo, and you can speed things up with a shared binary cache, and with distcc and ccache (or whatever you kids are using nowdays), but that falls under "screwing with it" -- and if it brings me to something which works as well as Ubuntu, what was the point?

    I'm glad I used Gentoo -- it gave me a much deeper understanding of how the OS is put together. But I'm equally glad I no longer use Gentoo.

    --
    Don't thank God, thank a doctor!
  262. Re:Ubuntu? No way. by SanityInAnarchy · · Score: 1

    Just because you can easily tweak anything doesn't mean you should or need to.

    However, if I don't need to tweak more than what Ubuntu gives me, what's the point of Gentoo? At least with Ubuntu, I'm not compiling all the time.

    Apparently you need the barriers of an OS that is more difficult to tweak in order to keep you on task.

    I haven't found Ubuntu to be more difficult to tweak, for things that I care about. I could probably use Gentoo just fine.

    But again -- no point. Ubuntu just works, out of the box, no tweaking needed, no waiting for things to compile. It installs faster, and it updates faster.

    On top of that, it's popular, which means more packages out of the box, more tutorials, etc.

    It's probably the same reason that many people prefer Windows to linux

    I've found Windows to be harder to tweak in ways that I actually want to tweak it.

    Example: I had to use a Mac for a week or two, when my old laptop was dead. I was mousing all the time. The few things which had decent keyboard shortcuts, they were all different than what I had on Ubuntu, and pretty much none of them were customizable.

    Then I bought this new laptop, with Ubuntu preloaded -- wiped it and installed Kubuntu.

    Within the first few hours of use, I'd either set or discovered keyboard shortcuts for everything. Logout menu is win+backspace -- then alt+s to suspend, alt+t for turn off, etc. Win+arrowkeys packs the current window to the next edge -- makes it very easy to organize my windows, or pack things against the edge of the screen. When dragging or resizing windows, they snap to each other and the screen. Sloppy focus. Good package management.

    I could go on...

    These are all things that I was able to do either out of the box, or within the first hour or two of use. Most of them simply are not available, in any form, on Windows or OS X. The exceptions aren't customizable.

    Understand: These are things I can tweak quickly and easily, and directly improve my user experience. Changing a USE flag, or playing with -Os/O2/O3/etc, provides no obvious benefit, often no measurable benefit, over the defaults Ubuntu gives me.

    And that ignores sibling poster -- yes, Gentoo does require a lot of maintenance. But even if it didn't, I've simply found Ubuntu to be as good or better in almost every respect.

    --
    Don't thank God, thank a doctor!
  263. Re:Ubuntu? No way. by SanityInAnarchy · · Score: 1

    Then why the hell were any of you running Gentoo in the first place ?

    Because, at the time, it seemed logical -- after all, it would give me a faster computer (though mostly because I was more careful about which packages to install, not because I was so incredibly optimized).

    And it was easier to hack, in some respects.

    What happened was, I grew up, I got a job, and I now have significantly less time to spend fucking around in a terminal just to get my OS to work.

    And because of said job, I can now afford computers that are fast enough that I can't actually keep them busy. I used to be all about Fluxbox, -Os (not a lot of RAM or cache), and removing everything I possibly could from USE, killing as many daemons as I could, etc. Now I realize that I can load up a full KDE, and pile things on top of that (including several GNOME apps, even a Windows app), even install MySQL (which on Ubuntu means it's always running) and forget about it for months, and the only thing that ever feels slow on this machine is Flash.

    And this is a laptop.

    Gentoo is all about tinkering and customising, and being offered the flexibility to do so, not ease of setting up.

    The point is that I can tinker and customize Ubuntu, also -- and it's easy to setup.

    The things which Gentoo provides that Ubuntu doesn't -- mainly, the compile-time tweaks and optimizations -- ok, it's cool to play with those, but in the end, they don't do much for me. Maybe -- maybe, possibly, if I sacrifice a goat -- I could squeeze a few more frames per second out of a game. Now, I get ten or twenty more by rebooting to Windows.

    --
    Don't thank God, thank a doctor!
  264. Re:Ubuntu? No way. by apoc.famine · · Score: 1

    or whatever you kids are using nowdays

    Me personally? Installed via a live-cd, got the essentials working, and called it quits. Yes, more work than Ubuntu. A day or so more. But the speed is impressive, and it was worth the first few days of fighting with it.

    Make no mistake - I have a Ubuntu netbook and a backup desktop running Ubuntu for when I need something special that doesn't exist on the Gentoo machine. But for day-to-day stuff, it's far faster.

    I just have to treat it like a prima-dona, and never change things. To each their own, I guess.

    --
    Velociraptor = Distiraptor / Timeraptor
  265. Re:Ubuntu? No way. by try_anything · · Score: 1

    "Ubuntu" -- an African word, meaning "I'm sick of fucking with Linux in order to get it to do what I want but I really don't like the alternatives."

    Quite. Or, alternatively:

    "Ubuntu" -- an African word meaning, "I'm paid to make my code run efficiently on other people's computers, not to dick around making other people's code run efficiently on my computer."

  266. Re:Ubuntu? No way. by SanityInAnarchy · · Score: 1

    the speed is impressive, and it was worth the first few days of fighting with it.

    I think I've said it farther up in this thread, but...

    I found no difference in speed. The difference was largely what I chose to install, but frankly, I run a full Kubuntu/KDE desktop, it works well, and I can't actually make it feel slow. (Yes, I've tried.)

    About the most I could do to optimize at this point is to shave a few more seconds off boot time, but... it's login in 25 seconds, and usable maybe 3 seconds after login.

    Yes, I do have very good hardware. But you know what? Spending days fighting with something is no longer fun, and is also no longer economical. It doesn't take that many days of fighting with it to realize I could be working instead, and simply buy more speed.

    --
    Don't thank God, thank a doctor!
  267. Re:Performance isn't its raison detre by kv9 · · Score: 1

    that was a short post. I was expecting another long winded disertation failing to prove me wrong.

  268. Re:Performance isn't its raison detre by Bloodoflethe · · Score: 1

    My posting history is varied and interesting. You seem to have a history of short posts that miss the point entirely or are simply trolls :)

    By the way: I think you missed the part where I won every time. You think you are the one playing the game, but, if you noticed my posting history, I usually only respond to the people that sincerely believe their tripe.

    In this case, I've been using an obvious and well-documented troll and his half-way clever comments to massage my ego this whole time! There: I admitted it! And now that I have done so, I'll even let you get the last word.

    --
    "Little is much when little you need."
  269. Re:Flexibility and freedom are its raison d'Ã by JasterBobaMereel · · Score: 1

    XP Embedded - Not available to the general user - and 140MB .... for an embedded system!

    Try DSL - Fully GUI system, with everything you need for a basic system can boot from a USB key and will work on a 486DX with 16MB RAM - Can run totally in RAM with only 128MB memory - try that with windows ...

    --
    Puteulanus fenestra mortis
  270. Re:Performance isn't its raison detre by kv9 · · Score: 1

    here's 3 last words: unwarranted self importance.

  271. Re:Ubuntu? No way. by boshi · · Score: 0

    I too loved the compiler optimizations, however the process could be offloaded elsewhere if you just had a package management system that stored different architecture specific optimizations pre-built. The idea is already implemented in things these optimized Mac builds of firefox. What if we had an entire OS installer and package manager that downloaded binaries for our particular architecture?

    There would certainly be space costs on the server side, but I don't see why it's anything worse than hosting entire livedvds for obscure architectures when most of your users will be using x86 anyways.

    --
    Blog
  272. Re:Performance isn't its raison detre by TheCycoONE · · Score: 1

    That is now one of my favorite quotes. You should write a book so it would be easier to quote you.

  273. Ubuntu VS. Windows XP by lingiu · · Score: 1

    There are two OS in My ASUS Intel notebook. one is Win XP-sp3 another is Ubuntu8.04. when I am running ubuntu,the fan of the CPU makes little sound,but when it is xp, the fan is so noisy,just like a motor...

  274. Re:Flexibility and freedom are its raison d'Ã by dave420 · · Score: 1

    No, this is a version of windows specifically designed to be configured to make customised install discs. There is nothing illegal about any of this.

  275. unhappy with 8.10 by Anonymous Coward · · Score: 0

    I moved to Ubuntu from SuSE a couple of years ago. I used Debian for years, but wanted something a bit more up-to-date without having to run unstable. SuSE was quite good, but I was even happier with Ubuntu.

    So, yesterday I upgraded from 8.04 to 8.10.

    Now my system is slow. Paging down in Firefox isn't instantaneous like it used to be. Switching from one virtual desktop takes a couple of seconds.

    Text mode apps running in a terminal are still fast. Apache still serves up pages quickly. It's just GUI stuff that is slow. I've also tried switching from Gnome to KDE and that didn't help. (It actually made things worse. They seem to be trying to match the performance of Vista as well as the look with this version of KDE.)

    I'm probably going to have to move back to 8.04. :-(

  276. Re:Flexibility and freedom are its raison d'Ã by Yfrwlf · · Score: 1

    Yes, running binaries is soooooo insecure, because of course you go through every line of source code before you compile your programs to make sure there's nothing malicious in them. Sorry, but that isn't my idea of security even though that's one option. I can download and run binary packages if I want to, and there's nothing impossible about doing so. You can *already* do that, what I'm saying is they should be *packages* instead of binaries so you can easily remove and update them if you want to. There is nothing impossible about that, so stop living in the past and think about what new users want for a change.

    --
    Promote true freedom - support standards and interoperability.
  277. personal impressions by Anonymous Coward · · Score: 0

    xp and kubuntu intrepid ibex on dual boot here on a brand new laptop.
    Ibex is slow compared to debian. XP is slower than both or than ubuntu 8.04 32 bit: AV software to blame?

    Ibex is a lil unstable - though some updates today replace the crashing kde panel - but everything worked out of the box: graphic wifi network printers. windows started downloading jvm updates twice, and i had to track down the printers cds for the drivers.

  278. Re:Performance isn't its raison detre by mark0978 · · Score: 1

    A 20% loss in performance is NOT progress, but regression. We slam Vista because it provides very little MORE than XP did with a huge cost in speed, yet somehow when Ubuntu does the same thing we call it progress.

    Progress in an OS is providing more with less, hence the Kernel changes to provide better service. Sounds like maybe they aren't working out as well as expected. Would be interesting to see if the apps that performed better like GPG had their gains in performance from changes in their code vs changes in the kernel.

  279. Re:Flexibility and freedom are its raison d'Ã by Yfrwlf · · Score: 1

    Erm, no that wasn't a trolling comment, that's actually my honest opinion. I believe in total software freedom for Linux users, so sue me.

    --
    Promote true freedom - support standards and interoperability.
  280. not 180MB of dependencies by pxc · · Score: 1

    Amarok's dependencies aren't actually 180MB--that's just the way Ubuntu (or Debian) packages the KDE libraries. On Gentoo, for example, you can compile Amarok to total, with dependencies, less than iTunes' 60+MB.

    Finally, if your program actually *is* going to need 180MB of code, you can bet it's better to have it separated into a reusable library.