Gentoo vs. Ubuntu performance comparison

[RealNC]

"Time to install" has dropped immensely the last couple of years due to incredibly fast CPUs. It was no fun installing on a Pentium, Pentium 2 or even Pentium 3. It took days.

Today, even on my poor Core2 Duo, it's a matter of hours

[elsie]

a comparison of "time to install" would be fun

hahaha,

I remember my days of Gentoo, it was the first *nix OS I really used. Every other OS I tried (including windows) was slow as f$%# but Gentoo ran really really well...

Of course that computer was completely unusable most of the time because I was compiling this&that.

Gentoo taught me sooo much about linux...

[Ant P.]

a comparison of "time to install" would be fun

To make it a fair test, it'd have to include the time spent uninstalling all the unwanted crap that Ubuntu installs by default.

[mits]

a comparison of "time to install" would be fun

[kraftman]

What's not faster than Ubuntu? Possibly Slowaris, but I can't think of anything else.

slowlaris, bsd and probably Fedora, because it has some 'heavy' things enabled.

[chithanh]

@krazy: Phoronix already tested the performance of different kernels. Some tasks are faster and some slower, but the assumption that kernel version skewed the results is not consistent with the almost sweeping win of the 2.6.30 based system over 2.6.28.

Different X server and NVidia driver affect X11 and 3D performance (that is why I said that this part of the comparison is meaningless), but not media encoding, database or similar tasks.

Both systems had the same gcc version and the same filesystem, so I would guess that the compute and I/O results are mostly valid.

[nightmorph]

What's not faster than Ubuntu? Possibly Slowaris, but I can't think of anything else.

[NullHead]

I wanna see Archlinux thrown into the mix as well.

[StringCheesian]

With a little work the Gentoo system could have had the same versions of everything.

@krazy: actually they did match DEs - both Xfce4, but your point still stands.

[krazy]

Different versions of Kernel, X server, Nvidia driver and completely different DEs. What conclusions can possibly be drawn from such a "comparison"?