1. Computers
  2. Display Drivers
  3. Graphics Cards
  4. Memory
  5. Motherboards
  6. Processors
  7. Software
  8. Storage
  9. Operating Systems


Facebook RSS Twitter Twitter Google Plus


Phoronix Test Suite

OpenBenchmarking.org

Chromium OS, Moblin, Ubuntu Netbook Remix Benchmarks

Michael Larabel

Published on 23 November 2009
Written by Michael Larabel
Page 5 of 7 - 25 Comments

Moving to the more interesting netbook performance tests of these much talked about Linux distributions, we have our results from the battery-power-usage test profile. All of the previous tests in this article were run when the Samsung NC10 netbook were running off of its AC power adapter, but with the remainder of the results in this article were run off of the netbook's six-cell battery. The battery-power-usage test profile monitors the battery power consumption via ACPI as the system idles for 60 seconds, then for another 60 seconds as the system continues to idle but with the display automatically shut off via DPMS and then lastly the display is turned back on via DPMS and then a sample video file is played once via MPlayer. The battery power consumption is plotted in Milliwatts in the graph below.

What is most apparent from the five lines is that Chromium OS is by far consuming the most power. Ubuntu Netbook Remix, Moblin, Fedora, and openSUSE all had an average power consumption rate of 11.3 Watts, but Chromium OS was consuming about 15% more power than the rest with its average at just over 13.0 Watts. Chromium OS also peaked at consuming 17 Watts where as the others topped out at 15~16.0 Watts. Likewise, Chromium OS bottomed out at 9 Watts when the display was turned off where all the others managed to hit 7 Watts on this Intel-powered netbook. Coming up in a set of results on the next page is a possible explanation why Chromium OS may be consuming so much more power in this Linux environment than the rest.

With the graph above and the rest of the graphs in this article are looking at various system vitals through the Phoronix Test Suite in a similar process to the battery-power-usage test profile, but this test case let the system idle, then ran the compress-7zip test profile, ffmpeg video encoding, and then lastly it fired up OpenArena and let that run for a round. The system temperatures for the Samsung NC10 between the five Linux distributions were rather close, except for Moblin 2.1 that ran warmer and then openSUSE 11.2 ran a bit cooler. The ambient room temperature was maintained during the testing process, but take these results as you want. It is rather surprising though that the average system temperature for Moblin 2.1 was over 6°C warmer than Chromium OS and 9°C for Ubuntu Netbook Remix. We may run this test on another netbook to see if this remains the case of Moblin 2.1 running much warmer than other Linux distributions, but our earlier results in this article did show it running generally slower than the other tested Linux environments. Also worth noting is that due to Chromium OS not being able to run OpenArena, its monitoring had ended after completing the ffmpeg test.

With this graph, we are looking at the battery power consumption again from the same idle, compress-7zip, ffmpeg, and openarena monitoring session. These results are quite similar to our earlier battery power usage testing from the idle/DPMS off/video playback testing, but with different numbers of course. Chromium OS again consumed the most power as you can see from the graph during the time in which the monitoring was active. OpenArena takes more power than compress-7zip/ffmpeg, so the average for Chromium OS is lower since it ended before OpenArena ran, due to compatibility issues, and thus only averaged the idle/compress-7zip/ffmpeg portion for Chromium OS where as the other distributions had their OpenArena results in the average. In our first battery test, openSUSE 11.2 had the lowest power consumption and this was the case again in this second battery test. In fact, openSUSE 11.2 with its KDE 4.3 desktop consumed noticeably less power than the others -- even Ubuntu Netbook Remix and Moblin! openSUSE 11.2 bottomed out at 8.8 Watts and peaked at 13 Watts where as the other four all bottomed out at 11~12 Watts and then peaked at 15~16 Watts. Novell is doing something right here when it comes to aggressive power management and consuming the battery life in a stock configuration.

Latest Linux Hardware Reviews
  1. Preview: AMD's FX-9590 Eight-Core At Up To 5.0GHz On Linux
  2. Intel Launches The Core i7 5960X, Mighty Powerful Haswell-E CPUs
  3. AMD Radeon R9 290: Gallium3D vs. Catalyst Drivers
  4. AMD Radeon R9 290 Open-Source Driver Works, But Has A Ways To Go
Latest Linux Articles
  1. Ondemand vs. Performance CPU Governing For AMD FX CPUs On Linux 3.17
  2. How Intel Graphics On Linux Compare To Open-Source AMD/NVIDIA Drivers
  3. The Fastest NVIDIA GPUs For Open-Source Nouveau With Steam Linux Gaming
  4. Testing For The Latest Linux Kernel Power Regression
Latest Linux News
  1. Marek Lands Radeon Gallium3D HyperZ Improvements
  2. Mozilla Firefox 32 Surfaces With HTML5, Developer Changes
  3. Nouveau X.Org Driver Released With DRI3+Present, Maxwell, GLAMOR
  4. Microsoft & AMD Release C++ AMP Compiler With Linux Support
  5. AMD, Wine & Valve Dominated August For Linux Users
  6. Linux 3.17-rc3 Kernel Released Back On Schedule
  7. Lennart Poettering Talks Up His New Linux Vision That Involves Btrfs
  8. Mesa 10.3 RC2 Arrives Via Its New Release Manager
  9. Ubuntu 14.10's Lack Of X.Org Server 1.16 Gets Blamed On AMD
  10. MSI Motherboard BIOS Updating Remains A Pain For Linux Users
Latest Forum Discussions
  1. AMD graphics doesn't work with AMD Catalyst drivers
  2. Lennart Poettering Talks Up His New Linux Vision That Involves Btrfs
  3. Best Radeon for a Power Mac G5?
  4. The dangers of Linux kernel development
  5. Updated and Optimized Ubuntu Free Graphics Drivers
  6. AMD Releases UVD Video Decode Support For R600 GPUs
  7. SSD seems slow
  8. Is laptop with Intel CPU and AMD dGPU worth buying considering especially AMD Enduro?