LinuxMusicians

by j_e_f_f_g » Sat Dec 08, 2012 8:25 pm

If someone has hopes of showing an ubuntu-based distro favorably fairing against another distro base, the last criteria you want to focus on is runtime performance (or stability). You'd have to deliberately hobble the competition with the most oppressive bloatware you can find. At all costs, avoid comparing to an arch, slackware, or gentoo system; especially one tuned for speed.

The only thing ubuntu does faster than any other distro is to retrieve Amazon's current price for Windows 8.

I get the fact that you hate ubuntu. That is pretty apparent. Seriously though come on If you have computer with a 8 year old dual core and 2 gigs of ram ubuntu still boots in under 20 seconds. I can't light a smoke and scratch my butt that quick. It opens any program and performs most tasks instantly. Maybe gentoo can boot up in 15 seconds and calulates pi in .007395 seconds faster. I don't stay up at night worring about it though. I got better things to do with my time (like waiting for ubuntu to boot).

What ubuntu does have going for it. It's easy to install it has the best, easiest, most strait forward installer by far.

It's designed to work. That's the reason I use linux in the first place. I don't use linux becuases it's faster and I don't use it because it's difficult to use or understand I use linux because it's simple and it performes any task I would want a computer to do simpler and more effeciantly than any other OS.

Every program designed for linux is packaged for ubuntu and on the software center, or has a ppa, or has a tarball so I don't have to be a programmer to get the software I need to work.

Ubuntu is just a simple distro to use with high compatabilty with everything. It's not like they took the terminal away you can still use it when you need it. It's as fast as anything and really compared to other OSes it's like a shot of lightning. It might not be the fastest distro out of the box but you can configure it for speed.

Don't get me wrong I like all the distros and I think it's a great thing that linux unifies us all under one banner but at the same times gives the users the freedom to use a distro that either suits there computing needs best or reflects there ideals about computing and software and freedom. It's just I like ubuntu the best. I have tried other distros and it's not that I'm to stupid to get them to work. I can get computers to work that don't have monitors and just flash lights at you or make beeps. Ubuntu just suits my needs the best.

@ Alwaysanewb
I literally LOL'd a few times reading your post. Completely agree, I'm an audio engineer, not a programer. I use linux because the JACK ecosystem is the best I've ever worked with in pro-audio. It's nice to have choices, and there are certain distros suited for certain tasks. I could make a case for any distro out there for some given task in the audio world as being better than another, I just like what works out of the box better!

Ubuntu is good for pro audio work? Here, let me Google that for you:

http://lmgtfy.com/?q=pulseaudio+sucks

It must be hard being such an uber-noob, and yet having such strong opinions.

I'm not trying to be a jerk or start a flamewar with anyone dude. I don't know why you would google pulse audio sucks when linux recording is based around jack2. It pretty much installs it self

I do know this I never have to worry about using an unstable distro where everythime I want to turn om my computer I don't have to worry about if my mobo will reconize this pendrive as a boot device.

The reason you never see anyone on here asking questions about studio 13 is not that no one has problems with it or that is doesn't have bugs and it makes everything so easy they simply know how to prefom any task with it from the shear fact of it's great design. Its that nobody uses it. I'm not sure why no one uses it. Maybe it's a great distro and just needs better marketing or maybe it just wasn't the best idea in the first place. That fact still remains no one uses it.

People use it. More people should use it. But I do have customers, thank you.

I don't even have to ask my cousin Larry for a fake testimonial, all of those quotes on my site are real.

j_e_f_f_g wrote: Now what we need is a debian based distro with Enlightenment, compiled for wayland, running a DAW that uses the E libraries. Oh, and not from Canonical.

I'm running debian/sid for more then 10 years now, it is easy to install from siduction , for example. All you need to set up your very own "studio distro" from it, is in the debian repository’s. You get mostly the latest stable versions from upstream. siduction comes with additional user based repository's, like ppa's. As well with a "fast fix repository" to fix upcoming bugs from upstream. It is far more stable in terms of usability then debian /testing.
What I would say with this, is, we didn't need a special "studio distro", what we really need to improve linux audio is a more community based thinking of the users and developers as well.

j_e_f_f_g wrote:Well, at this point ubuntu is really a fork. Canonical has made too many intrusive changes such that ubuntu now has its own debian-incompatible repositories..

That isn't "fully" true any more, debian and ubuntu maintainers work more together then ever. It comes more to a partnership.
So ubuntu takes automatically all packages from debian testing and try to build it on there servers without any change in the /debian folders, well in a automated process. This way, debian maintainers get listed as ubuntu maintainers as well.
Those maintainers from debain and ubuntu have understand that it is important to "work together", regardless how difficult it seems.

Alwaysanewb wrote:ubuntu still boots in under 20 seconds.

Well, unless you need to reboot your pc during a gig, as others have point out, boot time is not much of a selling point to a musician. (And that's a good thing for ubuntu since it tends to be a slow distro, including in terms of boot time. And this will only get comparitively worse as other distros switch to systemd while canonical sticks with its much less efficient, more convoluted upstart.)

What is important is run-time performance (ie, speed) and stability, which are two areas where folks who have tested non-ubuntu distros discover that ubuntu lags. And that's understandable given that ubuntu/canonical's goal is not run-time performance (nor even stability per se), but rather trying to compete with the formidable commercial forces of Windows, Apple, and now google to come up with a general use system for people who surf the internet, use a word processor, and do stuff like that. To achieve this, canonical has to, for example, load up lots of background daemons to handle the myriad general use support a typical non-technical windows/apple user expects to be handled transparently. And canonical needs to ship new updates as soon as possible (not necessarily when stable or debugged) in order to keep pace with windows/osx. An example is how canonical/ubuntu shipped Pulse Audio well before it was really useable (and the nightmare of issues that caused).

Canonical/ubuntu's use case makes it a distro for which run-time performance and stability are secondary concerns. Primary concerns are enticing non-technical windows/osx/android users to use ubuntu (with canonical's online services and affiliate programs). When you understand that, it's easy to see why ubuntu can, and typically is, bested in run-time performance and stability by other distros.

Can someone replace ubuntu's stock kernel with a realtime kernel, and do things like get rid of the more esoteric daemons, to make it faster? Yes. Will such a setup best another distro in real-time performance? That's really iffy because, in order to make ubuntu as "lean and mean" as slackware, arch, or gentoo, or even a non-specialist distro like Debian, you have to rip out lots and lots of canonical/ubuntu bloatware. If one was going to that much trouble, he may as well start from a more apropos base.

So here we have someone showing a slackware based distro, as well as a debian based distro, stomping all over 3 ubuntu based distros optimized for realtime audio, in terms of runtime performance. Is it believeable? Absolutely, without a doubt. It fits in with what I, and other folks who have compared distros, have discovered about ubuntu versus other distros. Ubuntu is the bloatware of linux distros. You can even see the speed difference between Linux Mint Debian Edition versus Linux Mint's version of Xubuntu. I feel sorry for anyone who tries to recreate that sort of speed test, hoping that ubuntu will perform markedly different. You're either going to need to deliberately cripple the competition, or face a disheartening reality that things greatly valued by musicians (such as lowest latency and fastest speed) are just not ubuntu's forte.

Ubuntu is all about enticing non-technical windows user to use linux to purchase apps from the ubuntu store, purchase space on canonical's cloud service, and click on affiliate links via the unity dashboard. Given such goals, is anyone really surprised by a demonstration of other distros beating ubuntu in realtime performance?

it has the best, easiest, most strait forward installer by far.

That it does, because this fits in with ubuntu's intended use case.

But I don't spend much time installing linux. (Using a rolling release distro here). So that's not a selling point for people me.

On the other hand, if was a novice computer user who just wanted a free OS in order to purchase mp3's from Amazon, I'd still be using ubuntu.

It's designed to work. Ubuntu is just a simple distro to use with high compatabilty with everything.

In other words, it's designed to entice non-technical windows to... etc. That it does do. What it's not particularly well suited for is power users who intend to customize their software setup, nor esoteric use cases like anything needing demanding realtime performance.

As a musician who used to use ubuntu, but then tried other distros, I find that using ubuntu for music is like using BeOS for accounting. With enough effort it can be done. But there are better tools for the job. I don't say this to make someone feel bad about ubuntu (nor to spend $50 on a linux distro -- I didn't). I say it because I believe a musician can be as pleasantly surprised by being convinced to try non-ubuntu distros, as that person was apparently as pleasantly surprised by being convinced to try ubuntu (instead of just continuing to use windows/osx).

Seriously ask yourself -- do you really know you may be missing?

This message brought to you by "Folks who will never be invited to Shuttleworth's dinner party (should someday Canonical turn a profit)".

j_e_f_f_g wrote:What is important is run-time performance (ie, speed) and stability, which are two areas where folks who have tested non-ubuntu distros discover that ubuntu lags.

While it may be true, it's still a rather bold statement and it would be nice to back it up with some kind of easy to reproduce benchmarking suite that tests the stuff we care about (so you can easily test things like "ok, it might not perform well out-of-the-box, but what if you make optimizations X Y and Z?").

Can someone replace ubuntu's stock kernel with a realtime kernel, and do things like get rid of the more esoteric daemons, to make it faster? Yes.

Actually, not such a long time ago, when the -rt patches still made a huge difference, Ubuntu had official(-ish?) -rt kernel packages while Debian did not.

In general, intuitively, I sort of feel you're exaggerating the impact Ubuntu's focus has on the possibility of tweaking it to perform well. But that's just my intuition, to make real statements about it we should perform actual tests. (you mention Ubuntu-based distro's haven't had the upper hand in comparisons so far - that might be so, but I'm having a hard time believing this is somehow inherent to using Ubuntu as a basis).

(full disclosure: personally I do prefer Debian to Ubuntu, but not for performance reasons)

falkTX wrote:afaik comparing top distros is like comparing religions. It doesn't make much sense... (just a tiny bit of it )

I agree if it's "mine is bigger than yours"-comparisions.

However, I do think comparisons are useful when they can be used to track down what made 'foo' faster than 'bar', document it, and eventually also improve 'bar' to carry the same awesomeness.

It's all linux, so 1 software that works in Ubuntu should work in Arch or Fedora.

Exactly. So if it doesn't that's worth investigating.

What we need right now is more decent, high-quality apps and plugins. (...) trying hard to get our opensource apps better and better (bug/feature reports, documentation, tutorials, donations, etc)

Word!

j_e_f_f_g wrote:

Alwaysanewb wrote:ubuntu still boots in under 20 seconds.

Well, unless you need to reboot your pc during a gig, as others have point out, boot time is not much of a selling point to a musician. (And that's a good thing for ubuntu since it tends to be a slow distro, including in terms of boot time. And this will only get comparitively worse as other distros switch to systemd while canonical sticks with its much less efficient, more convoluted upstart.)

Well, this machine boots in about 5-10 minutes (given that the valves have to warm up before I hit that standby switch), but I still love to play it



And it suits well the color of my Ubuntu desktop. And I like it very much working in Ubuntu, and I don't find it slow at all and it's damn easy to install and connect things. I tried to make my own "bad luck number" studio just for kicks with Macpup, but I just couldn't get it going. So, anybody who wants to brag about having a fast studio in the back pocket of his pants should spit out 50 bucks But I still like Ubuntu better.

I guess this thread doesn't really have a topic anymore, so SEXY PIECE OF HARDWARE!

i2productions wrote:I guess this thread doesn't really have a topic anymore

the ghosts which you have called

raboof wrote:back it up with some kind of easy to reproduce benchmarking suite that tests the stuff we care about

Regardless of whether it was or wasn't easy to setup, that youtube comparison video is a perfectly valid benchmarking. He installed each contender in its intended "out of the box" configuration (ie, he didn't deliberately cripple a contender), run on the exact same hardware, using the same test app, with the same audio configuration (ie, jack with the same settings), and obtained results using the same software measurement tool (ie, jack's own latency and overrun diagnostics). Bottom line is that his methodology is proper. It's a valid, accurate test of audio performance (where he focused on that).

Apparently, the results shattered some illusions, thereby upsetting people who are inclined to be upset by such. But that's irrelevant to the test results, which show 3 ubuntu-based distros having demonstrably higher latency and more overruns than a debian-based and a slackware-based. That's reality.

I imagine an easy-to-use, real-world latency/overrun test app would be one that uses alsa to directly open the card's hardware interface (no dmix or plug, and with buffering set by the app). The enduser would physically connect his card's audio out to audio in (or better, SPDIF out to in), and MIDI In to Out. The software would then play audio and MIDI while detecting the lag/underruns happening with the resulting input, averaging the results over time. I'd have no problem writing such a tool, and would be willing to do so if someone is seriously committed to installing/testing several distros (from various bases). But I get the impression that the OP was disillusioned by that initial benchmark test, assumed the methodology was flawed, and thought that he could produce different results with his own test. Then when those assumptions turned out to be incorrect, the reaction changed to "I don't care my distro has worse audio performance. All that matters is that it's not Windows. And can't we all just stop saying that something has an advantage over ubuntu because somehow, someway, in my mind that's tantamount to declaring war on linux, and you don't want to be unpatriotic, do you?".

At this point, I'm not sure anyone (well, anyone using an ubuntu base) is interested in audio performance differences between distros. But if someone is serious about that, let me know and I'll give you a tool.

you're exaggerating the impact Ubuntu's focus has on the possibility of tweaking it to perform well.

Are you implying that the creators of those 3 ubuntu distros don't know how to tweak ubuntu so that it runs as well as other distro bases? Because the only other possibility is that the ubuntu base itself needs significantly more modifications (in order to "come up to speed") than even the maintainers of ubuntu-based audio distros have done. Be careful when arguing hypotheticals because it sometimes invites even more detrimental possibilities than whatever it is your hypothetical seeks to counteract.

Calm down. Your hatred for ubuntu has been made clear. And the tune never changed to,"I don't care if there's better distros.". Pretty sure that it comes more down to not being concerned with l0wl1fe and his test. In the end its all about the music. And if your distro gets you to the end result, than good for you. I don't doubt there are distros with better performance, and I don't disagree with the methodology used in the video that's out there. I just disagree with the person testing it. Hardly and objective test, and easily manipulated. If it were true and reproducible, the maker of that distro would let someone independently verify their claims. That's all I was ever thinking. I only wanted to save face for the creators of distros I know for a fact work better than what was "benchmarked" of them. Calm down, this isn't about ubuntu being better(if this was the case overall Linux Mint wouldn't have a reason to be in existence, and arguably the most popular distro.

Whoa!

I don't think there is any 'hatred' directed at Ubuntu merely the observation that it is not developed with the ideals of the Audio power user foremost in mind and requires some 'deconstruction' to perform better. If you are not a core Ubuntu developer then I don't think anyone needs to get their back up about j_e_f_f_g's observations. Even a core Ubuntu developer would likely say "Uhh yeah, we don't really care about that stuff"

As one of the 'distributors' in the benchmark video I thought the methodology was quite good and sound...EXCEPT, I think he could have tried a bit harder with KXStudio ISO, obviously something was amiss there, a genius like falkTX is not going to put up an ISO that doesn't boot into it's desktop. I suppose it wouldn't have hurt to annotate a few things as well like Ubuntu needing the user added to the 'Audio' group etc. But other than that like was already mentioned he tried them all on the same hardware and simply let them boot into their native Desktops and went from there...

The key here is that Studio 13.37 was booting with the 'toram' parameter and the others weren't so that alone accounts for a lot of speed difference. Really boot speed and JACK latency are only 2 facets of the user experience so I don't think anyone here needs to get too wounded especially if they're happy with whatever they're running.

In the future I'd like to see an expanded video with all participating OS's booting properly and maybe even all OS's being tried with a 'toram' option (although Unity and KDE would need a large amount of RAM available simply because they are bulkier DE's).

Oh and BTW our fellow forum members nick is l0wt3ch (not l0witch or l0wl1fe)... Sorry, I certainly don't agree with him all the time either but c'mon this kind of stuff is a bit Kindergarten. Nobody should have their username mangled like that...

j_e_f_f_g wrote:I believe a musician can be as pleasantly surprised by being convinced to try non-ubuntu distros, as that person was apparently as pleasantly surprised by being convinced to try ubuntu (instead of just continuing to use windows/osx). Seriously ask yourself -- do you really know you may be missing?

Just to let you know, I have wiped Ubuntu off my laptop, and have installed Debian sid. No more spyware, or Ubuntu One store, or Unity, or pop-up slider handles. No more nail-polish magenta, slow boot-up times, or Canonical. Feels good to be back!