Linked by Thom Holwerda on Tue 16th Sep 2008 12:50 UTC, submitted by Michael Larabel
PHoronix has taken a look at several Linux distributions running on the Eee PC 901. "Late last month we published our preview of the ASUS Eee PC 901 and we shared our plans for a number of benchmarks using this netbook with Intel's Atom processor. Following our Linux desktop encryption benchmarks of the ASUS Eee PC 901 and Intel Atom N270 CPU we have a performance comparison of Xandros, Fedora, Ubuntu, and Mandriva on this low-cost netbook PC."
Thread beginning with comment 330420
To read all comments associated with this story, please click here.
To read all comments associated with this story, please click here.
Member since:2005-07-06
especially on a system which is likely to be used for fairly light tasks
Likely, likely.. This piss me off: you do realise that many people don't have the money to buy several computers?
Or more precisely don't want to spend more money than they need on computers, so they'll buy a laptop for the mobility and use it for *everything*.
Member since:2006-04-18
Recent Original Stories
Recent Comments
Headlines
Random Comments
Random Stories
Random OS Link
Member since:
2005-07-06
...that Mandriva 2009 final will use kernel 2.6.27, not 2.6.26 as was in the beta tested.
I'm still not all that convinced by this kind of performance comparison, especially on a system which is likely to be used for fairly light tasks. I'd find a comparison of overall functionality under the given OS, including a subjective impression of performance if necessary, most interesting.
Some of the results are, I think, just silly, like the ImageMagick compile time. I can't think of any reason on Earth it would somehow magically be twice as fast on Fedora as every other distro tested. I find it far more likely that the default configuration of the distros led to something being disabled in the compilation on Fedora that was enabled in the compilation on the other distros, leading to the discrepancy. Compile times are a fairly simple function of CPU speed and having enough memory, and I just can't see anything a distro can do that would affect the performance of a compilation process to that extent.
To be honest, I don't think the integrity of this benchmark has been sufficiently demonstrated yet...it's a good effort on Phoronix's part to make it, but I'm not entirely sure it's yet reliable enough to draw much in the way of conclusions.