Testing For The Latest Linux Kernel Power Regression

Written by Michael Larabel in Software on 29 August 2014 at 12:00 AM EDT. Page 3 of 5. 11 Comments.

The Haswell desktop system used for testing was my go-to rig with the Core i7 4790K Devil's Canyon CPU that sports Intel HD Graphics 4600. This system was running Ubuntu 14.04 and from there I tested the Linux 3.14 through Linux 3.17 kernels. The desktop power consumption was measured using a USB-based WattsUp Pro power meter via the Phoronix Test Suite.

When running the Xonotic first person shooter like with the Ivy Bridge laptop, the only change in performance was found between Linux 3.14 and 3.15... With Linux 3.14 the power consumption was higher than 3.15, so it was an actual improvement.

The Xonotic performance for this high-end Core i7 desktop was unchanged between the tested kernels.

For additional measure I also ran the Tesseract game where the performance also showed an improvement between Linux 3.14 and 3.15.


Related Articles