Ever since spotting a major Linux kernel power regression and increased Linux power usage followed by automatically locating the power regressions, there's requests every so often from Phoronix readers interested in new power-use benchmarks of the kernel.
The reason why I haven't put out any new kernel benchmarks looking at the battery use of laptops/ultrabooks has been since recently the Linux kernel is on a nice track. For at least the many different mobile systems in my possession, I haven't seen any dramatic changes on recent Linux kernel releases. For my MacBook Air/Pro systems, Apple OS X still does a superior job on delivering the longest battery longevity, but overall for my Apple and PC laptops/ultrabooks I haven't seen any major changes recently from kernel release-to-release.
With the Linux 3.13 kernel getting into shape, I ran some new benchmarks this holiday weekend to see how an ASUS Ultrabook is performing with Intel Core i3 "Ivy Bridge" processor. When running Ubuntu 13.10 x86_64 and testing all major kernel releases from Linux 3.7 to Linux 3.13 Git (using the Ubuntu Mainline Kernel PPA), I wasn't able to find any major changes in the ultrabook's power-use across the past seven kernel releases benchmarked.
All of this Linux kernel power use data is hosted on OpenBenchmarking.org via 1311296-SO-INTELULTR82. With the Phoronix Test Suite you can automatically monitor your mobile Linux system's power-use for any test profile / workload by simply setting the MONITOR=sys.power environment variable prior to running the Phoronix Test Suite. Setting PERFORMANCE_PER_WATT=1 will also generate performance-per-Watt graphs for each benchmark run via the Phoronix Test Suite. It's that easy! The Phoronix Test Suite also supports reading power usage data from the USB-based WattsUp power meter and other devices.
The ASUS Ultrabook has fluctuated about a Watt or so between kernel releases, but nothing too major and nothing like the ASPM issue of 2011.
The only issue to point out really was the Linux 3.10 kernel performing noticeably worse on this Intel Core i3 system than the 3.9 and 3.11+ kernels.
So there's really nothing too exciting to talk about with the Linux 3.13 kernel when it comes to power usage, but I'm still carrying out performance benchmarks and power monitoring on other mobile and desktop systems so stay alert in case there's something else uncovered. For now you can analyze the rest of the results.