Is Arch Linux Really Faster Than Ubuntu?

[pingufunkybeat]

Everyone with those biases will view any other distro as bad because it isn't theirs.

but this is not necessary.

As a long-time SuSE and Debian user, I still like those distributions. It's just that gentoo is a better fit for me at this time. No need to bitch at the people who prefer a different distribution, if it is a better fit for them.

It is exactly this choice that is the main strength of Linux, like you say.

[V!NCENT]

Every distro will fill a niche in the community that people are looking for. Everyone will end up bias toward one distro or another over time. Everyone with those biases will view any other distro as bad because it isn't theirs.

Why do I always keep hearing this?

In the beginning person Y didn't have [object X].
Person Y then looked around and made a choice in what object to acquire based on what they thought was the best [object X].
People than favour their [object X] over others, not because they have [object X], but because they chose [object X] while they could have gotten something else.

It is not the other way around. Especially not when distro's are gratis...

[stargeizer]

Not a bad thing.

I mean, is good that users that doesn't care about tweaks and editing files have a good option that in most cases "Just works" like ubuntu.

Of course users that want/need to tweak everything still have a distro that allow them to do so.

Both kind of users wants the most of their systems and the fact that Linux right now can offer these features with no real difference between a full personalized distro and a distro that "Just Works" is good for everyone. The rest is a thing of personal taste.

Something that Microsoft/Apple can't offer right now.

Not bad at all.

[nanonyme]

yawn. Welcome to 2003.

This was actually my point.
To everyone else: stop using funroll-loops, it sometimes makes code run slower. You can't find out which one it is for the particular program without profiling before compiling. Don't use it for whole system. kthxbye

[pmorph]

I think the "problem" with source distro optimizations is this: the more you need the gains from compile time optimizations, the less convenient they will be to deploy.

That is, if you have a sucky computer, compiling the whole distro from sources will be a major pain in the ass. On the other hand, if your system is beefy enough to compile it swiftly, its also beefy enough to make the little optimization gains virtually impossible to detect in day-to-day use.

[nanonyme]

Yes, that's one of the two big problems. The other is that people are allowed to put stupid stuff in CFLAGS which actually just slow down execution and compile times.

[energyman]

I think the "problem" with source distro optimizations is this: the more you need the gains from compile time optimizations, the less convenient they will be to deploy.

That is, if you have a sucky computer, compiling the whole distro from sources will be a major pain in the ass. On the other hand, if your system is beefy enough to compile it swiftly, its also beefy enough to make the little optimization gains virtually impossible to detect in day-to-day use.

back in the glorious k6-2 400 days compiling some libs (including glibc) was the difference between 'watchable video' and 'dia show'. So it was very much worth the time spent.

[BenderRodriguez]

Why do i get the feeling that almost all tests were ultimately limited by the hardware (that's why such negligible differences among all those distros). While with those benchmarks we can say that pure power is the same we can't say if they are all responsive identically like starting programs, memory usage... Phoronix is starting to be a big joke making purely kernel/CPU bound tests.

[V!NCENT]

Why do i get the feeling that almost all tests were ultimately limited by the hardware (that's why such negligible differences among all those distros). While with those benchmarks we can say that pure power is the same we can't say if they are all responsive identically like starting programs, memory usage... Phoronix is starting to be a big joke making purely kernel/CPU bound tests.

I dissagree. Besides being the central place for all your interesting FLOSS news, Michael really listens to users and just wanted to benchmark Arch Vs. Ubuntu because there has been many situation where people said "Omg Arch is 100000 times faster than Ubuntu WTF!" in the forums.

[m_gol]

Even Ubuntu does not force Pulseaudio.

Buhahahaha. Removing PA in Ubuntu Lucid results in losing sound in many games (one has to manually configure OpenAL to use ALSA), sound applet doesn't work at all (one has to find PPAs for multimedia keys, sound applets etc.) etc. etc. I won't say a word about installing OSS4 as it's even far more difficult.

Sure, You can use ALSA. For the record, I use ALSA in Ubuntu 10.04! But don't tell me it's a piece of cake to set it properly.