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

The GNOME View On Compositor Benchmarking

GNOME

Published on 13 June 2011 09:46 PM EDT
Written by Michael Larabel in GNOME
11 Comments

Red Hat's Owen Taylor who is largely responsible for the work on the GNOME Shell and Mutter has written his response to the recent article about How Unity, Compiz, GNOME Shell and KWin Affect Performance.

Owen's comments on his blog are pretty much as expected and come down to attributing the shortcomings of GNOME Shell / Mutter with not handling undirected full-screen rendering at this point.
Adel Gadllah has had patches for Mutter and GNOME Shell to add unredirection for over a year, but I’ve dragged my feet on landing them, because there were some questions about when it’s appropriate to unredirect a window and when not that I wasn’t sure we had fully answered. We want to unredirect fullscreen 3D games, but not necessarily all fullscreen windows. For example, a fullscreen Firefox window is much like any other window and can have separate dialog windows floating above it that need compositing manager interaction to draw properly.

We should land some sort of unredirection soon to benefit 3D gamers, but really, I’m much more interested in compositing manager performance in situations where the compositing manager actually has to composite.

This is where KDE's Kwin now does suspended compositing. Coming as a result of Owen's investigation though is a new benchmark to measure the effect of the compositing manager on the application re-draw performance. In Owen's blog he goes into more details about the application re-draw process and what his benchmark is doing. He also has some initial results comparing KWin, Compiz, and Mutter. He also looks at the affects of not compositing and off-screen rendering.
My goal here is that this is a benchmark that we can all use to figure out the right timing algorithms and get them implemented across compositors. At that point, I’d expect to see only minimal differences, because the basic work that every compositor has to do is the same: just copy the area that the application updated to the screen and let the application start drawing the next frame.

A Phoronix Test Suite test profile for this xcompbench program is likely to come.

Latest Linux Hardware Reviews
  1. Overclocking The AMD AM1 Athlon & Sempron APUs
  2. AMD Athlon 5350 / 5150 & Sempron 3850 / 2650
  3. Upgraded Kernel & Mesa Yield A Big Boost For Athlon R3 Graphics
  4. AMD Athlon 5350 APU On Linux
Latest Linux Articles
  1. A Quick Look At GCC 4.9 vs. LLVM Clang 3.5
  2. Are AMD Athlon/Sempron APUs Fast Enough For Steam On Linux?
  3. AMD Athlon's R3 Graphics: RadeonSI Gallium3D vs. Catalyst
  4. GCC 4.9 Compiler Optimization Benchmarks For Faster Binaries
Latest Linux News
  1. Trying Out Radeon R9 290 Graphics On Open-Source
  2. Intel Broadwell GT3 Graphics Have Dual BSD Rings
  3. Early Linux 3.15 Benchmarks Of Intel Core i7 + Radeon
  4. Red Hat Releases Its RHEL 7 Release Candidate
  5. New Features Coming To Xubuntu 14.04 LTS
  6. NVIDIA Officially Releases CUDA 6
  7. Google Releases An AutoFDO Converter For Perf In LLVM
  8. Fedora 21 To Evaluate Remote Journal Logging, 64-bit ARM Emulation
  9. Star Citizen Will Be Coming To Linux
  10. Ubuntu 14.10 Convergence To Focus On Replacing Core Apps
  11. The Results Of Optimizing Radeon's VRAM Behavior
  12. Kernel Developers Discuss Improving Kernel Configurations
Latest Forum Discussions
  1. Linux Kernel Developers Fed Up With Ridiculous Bugs In Systemd
  2. The GNOME Foundation Is Running Short On Money
  3. Bye bye BSD, Hello Linux: A Sys Admin's Story
  4. New tool for undervolt/overclock AMD K8L and K10 processors
  5. How to enable opengl 3.3 on r9 270?
  6. R290x sound problems
  7. radeon-profile: tool for changing profiles and monitoring some GPU parameters
  8. Torvalds Is Unconvinced By LTO'ing A Linux Kernel