Ubuntu 7.04 to 8.10 Benchmarks: Is Ubuntu Getting Slower?

[glasen]

Hi,

A user at the german forum "www.ubuntuusers.de" did some tests with his Lenovo T60 (same hardware as the one in the Phoronix-test) on Ubuntu 8.10. His scores are dramatically better than the ones from the Phoronix test :

LAME-MP3 : 60.62s

OGG : 35.05s

FLAC : 28.82s

WAVPACK : 31.22s

FFMPEG : 33.76s

SQLITE : 24.21s

SCIMARK (composite) : 290.28 Mflops

Tandem-XML (write) : 54.22s

How can it be that two systems with the same hard -and software got so different test results?

[Michael]

The kernel used was SMP supported.

[deanjo]

it doesn't matter what pts is written for, when the kernel is UP.

Is the default ubuntu kernel not SMP? Thought that was pretty much default on all distro's now.

[energyman]

it doesn't matter what pts is written for, when the kernel is UP.

[deanjo]

You are right. I thought I read that the Core Duo wasn't a true dual core processor some time ago. This still does not mean the benchmarks used, took advantage of multiprocessing that is available.

https://help.ubuntu.com/8.04/installation-guide/powerpc/hardware-supported.html

The article makes no mention if the benchmarks are taking advantage if multiprocessing is being used.

PTS by design is written to utilize multithreading for applications that do support it. ie openSSL, compile tests, etc

[joshuapurcell]

You are right. I thought I read that the Core Duo wasn't a true dual core processor some time ago. This still does not mean the benchmarks used, took advantage of multiprocessing that is available.

https://help.ubuntu.com/8.04/installation-guide/powerpc/hardware-supported.html

The article makes no mention if the benchmarks are taking advantage if multiprocessing is being used.

My kernel is SMP by default on my T61 with an 8.04 install:

joshua@laptop:~$ uname -a
Linux laptop 2.6.24-21-generic #1 SMP Mon Aug 25 17:32:09 UTC 2008 i686 GNU/Linux

[LinuxAffenMann]

I really think you should check your suite or repeat the test on a different machine.

Those LAME numbers can't be right and are not reproducible (plain encoding, not with the suite) on my machine.

[Michael]

Ubuntu RC released only at the midnight of 23rd that means 24th October, The report says they tested it for 100 Hrs, The test result publication time is 27th . How much ubuntu competitors paid for it.

Before coming up with such a statement, did you take the time to think that not all testing had to begin after the RC was released? The Ubuntu 7.04, 7.10, and 8.04 testing started earlier in the week and then when the 8.10 release candidate came out, that was tested.

[hwertz]

As said in the article, Compiz was disabled.

Ahh man, I have to read the article text too? Just kidding, but OK I did miss that part... anyway, that's really interesting then. I wonder if it is the scheduler as some have speculated, or if the compiler makes that big a difference, or what? Those time differences are pretty large for quite a few apps.. I knew from the title that 8.10 was probably slower, but I was expecting a few %, not a drop like that .

[panayara]

I cannot believe the report

Ubuntu RC released only at the midnight of 23rd that means 24th October, The report says they tested it for 100 Hrs, The test result publication time is 27th . How much ubuntu competitors paid for it.

see more in

http://shibuvarkala.blogspot.com