Linked by Thom Holwerda on Wed 7th Mar 2007 18:03 UTC, submitted by Michael Larabel
Benchmarks "We have compared the 32-bit and 64-bit performance of Ubuntu and started a performance comparison of Ubuntu/Kubuntu/Xubuntu, but how does the performance of the upcoming Feisty Fawn release compare to that of Fedora 7? In this article we have enclosed benchmarks from Fedora Core 6, Ubuntu 6.10 Edgy Eft, Fedora 7 Test 2, and Ubuntu Feisty Fawn Herd 5. In gaming and desktop benchmarks, which of these Linux distributions is faster?"
Order by: Score:
mh
by SK8T (2.16) on Wed 7th Mar 2007 18:38 UTC
SK8T
Member since:
2006-06-01
Fans: 1

as I remeber we had an articel about Windows Vista (XP vs. Vista Perfomance) last week. And I said that my ideal OS is getting faster with every new version and not just slower than Vista.

And I think this artical shows that it is possible.

Wrong benchmarks
by Luis (3.48) on Wed 7th Mar 2007 18:40 UTC
Luis
Member since:
2006-04-28
Fans: 1

I can't see the point in these benchmarks and the previous ones comparing Ubuntu vs. Kubuntu vs. Xubuntu. The things they're benchmarking are mostly kernel and drivers related, and all the systems compared have those things in common.

They should compare time to boot, time to load each desktop, memory used on a default install, time to open applications,...

What they do would be interesting if they were comparing Linux vs. Solaris vs. *BSD, but Linux vs. Linux is a dumb comparison.

RE: Wrong benchmarks
by r3m0t (1.32) on Wed 7th Mar 2007 20:56 UTC in reply to "Wrong benchmarks"
r3m0t Member since:
2005-07-25
Fans: 0

But the distros do patch the kernel, and compile with different options. They also run background programs like yum-updatesd.

RE[2]: Wrong benchmarks
by superstoned (2.8) on Wed 7th Mar 2007 20:59 UTC in reply to "RE: Wrong benchmarks"
superstoned Member since:
2005-07-07
Fans: 3

Indeed, these benchmarks at least where a bit more usefull compared to the earlier ones which really where stupid... If they continue to improve, their second next benchmarks might even be very interesting...

RE: Wrong benchmarks
by re_re (3.88) on Thu 8th Mar 2007 09:48 UTC in reply to "Wrong benchmarks"
re_re Member since:
2005-07-06
Fans: 0

>They should compare time to boot, time to load each desktop, memory used on a default install, time to open applications,...<

for me personally i couldn't care less about most of that. I almost never reboot my computers and when i do i really don't care about the boot time..... if it takes 2 minutes....... so what.... it's not like i have to reboot every time i install something like you have to do in windows so often.

and ...... well....... the memory........... linux uses all your memory all the time.............. what does your memory do if you aren't using it?

i am a hell of a lot more concerned with with how fast an app r uns after the system has booted then i am with boot time. I would rather have a slow boot time and a secure fast user environment myself.

What's the point ...
by Robert Escue (2.04) on Wed 7th Mar 2007 18:48 UTC
Robert Escue
Member since:
2005-07-08
Fans: 5

Since these tests don't appear to be run more than once (or at least the article text doesn't indicate more than one run) the variation between the test results would be considered normal. I don't see the point of the tests at all.

Ummm....
by arctic (2.68) on Wed 7th Mar 2007 19:06 UTC
arctic
Member since:
2006-04-19
Fans: 1

for me, speed is not as important as stability. But then, I am not a gamer. ;)

Difference is minimal
by TaterSalad (2.84) on Wed 7th Mar 2007 19:23 UTC
TaterSalad
Member since:
2005-07-06
Fans: 3

Looking at the tests they performed it looks like the difference is minimal between the distrobutions. A few seconds and a few frame rates. I'm not sure if that makes a lot of difference to people.

RE: Difference is minimal
by spikeb (2.52) on Thu 8th Mar 2007 00:58 UTC in reply to "Difference is minimal"
spikeb Member since:
2006-01-18
Fans: 1

notice that almost every test had a different winner?

Fedora 7 is much faster...
by Kwitschibo (2.28) on Wed 7th Mar 2007 19:24 UTC
Kwitschibo
Member since:
2006-01-17
Fans: 0

In the news to Fedora 7 Test 2 the Developer said, the many Debug Stugg is active so that the system is slower than the final.

What about Ubuntu at this point?

Important metrics for me
by John Nilsson (2.52) on Wed 7th Mar 2007 19:28 UTC
John Nilsson
Member since:
2005-07-06
Fans: 0

Benchmarking for me should focus on
Time to fire up a browser window (epihpany)
Time to fire up a text editor (gedit sucks in this arena, lefpad isn't to shabby though)
Time to fire up a media player or document viewer
"Feel" (small things lika latency between mouse click, visual feedback and actual effect feedback)

RE: Important metrics for me
by Kwitschibo (2.28) on Wed 7th Mar 2007 19:37 UTC in reply to "Important metrics for me"
Kwitschibo Member since:
2006-01-17
Fans: 0

Browser open on click, gedit, a great editor, open on click, evince open on click and rhythembox need one second. OOo Writer or Calc also open on click.

compress, encod etc. are betetr benchmarks.

RE[2]: Important metrics for me
by miscz (3.68) on Wed 7th Mar 2007 20:46 UTC in reply to "RE: Important metrics for me"
miscz Member since:
2005-07-17
Fans: 0

Here, on Athlon 3500+ with 1GB of ram, it's way longer to open Firefox or Gedit. Subjective feel of speed is more based on responsiveness, I don't really care if encoding an album will take 10 or 12 minutes because I'm going to switch to another task anyway.

RE: Important metrics for me
by Ford Prefect (3.04) on Wed 7th Mar 2007 20:47 UTC in reply to "Important metrics for me"
Ford Prefect Member since:
2006-01-16
Fans: 6

My applications typically run for months (using software suspend). Why should I care about the startup time?

I wonder how many years it will take until the majority of users is freed from starting up their working environment again and again and again...

RE[2]: Important metrics for me
by superstoned (2.8) on Wed 7th Mar 2007 21:04 UTC in reply to "RE: Important metrics for me"
superstoned Member since:
2005-07-07
Fans: 3

yeah, many windows users aren't really used to multi-tasking, as it doesn't work very well under windows. I experience that every day at work, having freezes and crashes - other ppl don't. They tell me 'you want too much too fast'. Well, if I want to search through several pdf's, I'm used to opening them at the same time - preferably in tabs in konqueror (select 10 files, richtmouseclick -> open in tab). Don't even think about it in windows.

And if it wasn't for the performance issues, the window management sucks as well. If you have a word file open (working on it, you know...) and are looking for another - you open an explorer window with some files. Open the first word doc. No, not the right one. close. What do I have on top? The first word window, instead of the explorer window which was supposed to be there! WTF???

And what about shortcuts? Often, alt-tab doesn't work - I can't get out of a word window to the next window. Why? Aaaah - a floating toolbar!!! Brings me right back to the window I was working on! WTF?!?!?

Sorry, but until Microsoft gets the basics working, I can't comfortably work in it, so I prefer to work at home - in a decent Desktop Environment (and no - in this regard, even though I prefer KDE - Gnome kicks windows ass as well - anytime...)

RE[3]: Important metrics for me
by raver31 (4.68) on Wed 7th Mar 2007 23:37 UTC in reply to "RE[2]: Important metrics for me"
raver31 Member since:
2005-07-06
Fans: 13

Agreed, after having had to use Windows(tm) all day, when I get home, I fire up this Linux box and chill out using KDE.

It is the same feeling you get coming home from work, slouching on the sofa wearing your favourite slippers.


Comfy !

RE[4]: Important metrics for me
by superstoned (2.8) on Thu 8th Mar 2007 10:24 UTC in reply to "RE[3]: Important metrics for me"
superstoned Member since:
2005-07-07
Fans: 3

Clearly you know what I mean. Windows users here most likely don't even understand what I'm talking about, used as they are to working around all the quirks in windows. Many don't try to properly use the scrollwheel anymore (like, for scrolling - in windows, it only works in the active application. Not over tabs, not over the taskbar, not over the volume icon, not in non-focused apps, and often not even in the focused app - esp in dialogs, you often have to click the scrollbar first. WTF does microsoft think the scrollwheel is made for???)

RE[3]: Important metrics for me
by stare (1.88) on Thu 8th Mar 2007 00:55 UTC in reply to "RE[2]: Important metrics for me"
stare Member since:
2005-07-06
Fans: 1

Don't even think about it in windows.

What's problem in doing that in Windows? I usually have upto 30-50 windows opened, switching is instant.

If you have a word file open (working on it, you know...) and are looking for another - you open an explorer window with some files. Open the first word doc. No, not the right one. close. What do I have on top? The first word window, instead of the explorer window which was supposed to be there! WTF???

I wasn't able to reproduce that. After closing the document, explorer window is on top.

And what about shortcuts?

They work.

Often, alt-tab doesn't work - I can't get out of a word window to the next window. Why? Aaaah - a floating toolbar!!! Brings me right back to the window I was working on! WTF?!?!?

I wasn't able to reproduce that. Floating toolbar doesn't prevent switching from Word window in my case.

RE[4]: Important metrics for me
by superstoned (2.8) on Thu 8th Mar 2007 10:22 UTC in reply to "RE[3]: Important metrics for me"
superstoned Member since:
2005-07-07
Fans: 3

well, all of it is pretty reproducable here at work. Office 2003, MS Windows XP Media Center edition.

and having 30 windows open is of course horrible on a single desktop, but yes, that can be done. What IS horrible is that it takes word several minutes to become usable after you selected and opened 5 files at the same time... let alone 10 or more. Same with Adobe acrobat.

And the lack of Focus stealing prevention is horrible as well. If you open 5 word files, and word is 'working' on that in the background, it's pretty annoying to read a website - every time it has finished loading a word file, word pops up on the foreground, even when you're typing in firefox or whatever. At least KDE has proper window management - as long as you're actually WORKING in a window, another window doesn't pop up in front. But when you start an app, it DOES. Hard to describe to a windows user, I guess, let me just say it DOESN'T ANNOY THE HELL OUT OF ME. Like windows does.

No difference
by Liquidator (2.84) on Wed 7th Mar 2007 21:09 UTC
Liquidator
Member since:
2007-03-04
Fans: 2

Damn, that web site is full of ads. Anyway, the end user doesn't feel any difference between 1% faster or slower.

Wow
by Duffman (0.84) on Wed 7th Mar 2007 21:16 UTC
Duffman
Member since:
2005-11-23
Fans: 4

They have the same results.
What a surprise, it's not like they are sharing the same kernel, binaries, etc ...

RE: Wow
by Constantine XVI (1.84) on Thu 8th Mar 2007 13:25 UTC in reply to "Wow"
Constantine XVI Member since:
2006-11-02
Fans: 0

They aren't.
Fedora and Ubuntu use different kernel versions, and they have tweaked the stock kernel to fit their needs.
And, a lot of the software in the repos have distro-specific patches, which would change things some.
However, I still doubt there would be MAJOR differences. The one major difference I can think of in Ubuntu is Upstart, instead of sysVinit. They didnt test boot times, so that wouln't show up.

RE[2]: Wow
by Duffman (0.84) on Sat 10th Mar 2007 14:10 UTC in reply to "Wow"
Duffman Member since:
2005-11-23
Fans: 4

Fedora and Ubuntu use different kernel versions, and they have tweaked the stock kernel to fit their needs.

So it proves that 'tweaking' the kernel doesn't change anything and is more a geek stuff than real world optimizations.

at the very least
by djame (2.4) on Wed 7th Mar 2007 22:10 UTC
djame
Member since:
2005-07-08
Fans: 1

They should have run the unix byte benchmark +q3 test or whatever is in use now under these two distro using 5 variables and all with firefox, thunderbird and something like xmms or amarok playing an internet stream (let say a tipical session for anyone)

with gnome eyecandized
with kde
with a light windowsmanager like blackbox or e16
with x11 and an xterm
and on init 3 (without q3 or if it exist a text version, libcaca or something)



Djamé

Stability and applications
by Southern.Pride (-0.68) on Thu 8th Mar 2007 00:54 UTC
Southern.Pride
Member since:
2006-09-14
Fans: 0

None of these 'bench marks' mean anything to me I only care about the system running 24/7 and loading applications correctly without to many hassles.

This was geared for a Windows review with games who cares...

RE: Stability and applications
by spikeb (2.52) on Thu 8th Mar 2007 05:17 UTC in reply to "Stability and applications"
spikeb Member since:
2006-01-18
Fans: 1

people who use linux as something other than a server.

Want to bet?
by libray (2.28) on Fri 9th Mar 2007 19:51 UTC
libray
Member since:
2005-08-27
Fans: 1

... that the author probably ran the default desktops on each system, with xyz whizzbang?