Now on ScienceBlogs: "Investigative science journalism" and books I like to read [All of My Faults Are Stress Related]
Evolution, Life Sciences, Science Education, Human Evolution, and Stuff
Welcome to Greg Laden's Blog.
« Get your science news in 3D | Main | The video about the prop 8 video meme. Pass it on. »
Category: Linux
Posted on: December 7, 2008 10:16 PM, by Greg Laden
And the results are at the same time boring and astonishing. I mean, look at this graph:
Never mind the details of the graph. The point is that the two distros, running on an almost identical Linux kernel and using identical desktops and an almost identical X system, are almost identical when run on an identical computer.
TrackBack URL for this entry: http://scienceblogs.com/mt/pings/87331
PZ Myers 07.05.2009
PZ Myers 07.01.2009
Orac 07.04.2009
Greg Laden 07.05.2009
revere 07.05.2009
Latest science stories | More at nytimes.com
1:14 AM, 07.06.2009
12:40 AM, 07.04.2009
2:24 AM, 07.03.2009
6:21 AM, 07.04.2009
2:50 PM, 06.30.2009
Comments
How would a Windows machine fare in the same test? Both right after build and after 6 months of use?
Posted by: george.w | December 7, 2008 11:25 PM
I would have expected significant differences depending on the configuration of Compiz and the desktop environment. Sometimes it even matters if you configure the clock in the task bar to have a blinking colon or display seconds; or to scan for online updates, hotspots, Bluetooth devices; or to double-buffer window updates; or to have an xeyes/xsnow clone running; or to tweak power mgmt; sound effects; desktop search; or to have compiled one of the tweaks listed on lesswatts.org (xterm polling, tickless idling etc).
I am honestly surprised that the two distros are that on par with each other.
Posted by: marko | December 8, 2008 9:52 AM
So am I.But I reckon Fedora would cark it on older systems,or slower processors.
Posted by: clinteas | December 8, 2008 10:12 AM
When you try to do apples to apples testing, then you'd have them configured as near the same as possible, otherwise the results are irrelevant. Since the hardware pushing pixels is the same, and they're using the same driver, and very nearly the same kernel, I'd be more surprised if there was any meaningful variance.
Posted by: Fargo | December 8, 2008 12:06 PM