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

Xinerama v. TwinView

Michael Larabel

Published on 4 February 2006
Written by Michael Larabel
Page 1 of 7 - Comment On This Article

When it comes to a multi-headed environment under Linux, there are two popular options for consumers -- Xinerama and TwinView. TwinView was developed by NVIDIA for allowing multiple monitors to be powered by a single GPU with their array of GeForce graphics cards. NVIDIA TwinView is also compatible with a wide array of displays from DFP displays to traditional television sets, and support remains unified with both Microsoft Windows and Linux. On the other hand, Xinerama was originally developed by DEC (Digital Equipment Corporation) under the name of PanoramiX, and was later incorporated into the X Window System as Xinerama. Not only is Xinerama compatible with NVIDIA solutions, but also is universal to all graphics cards with supportive Linux drivers and the X Window System -- including the ATI Radeon and FireGL series as of their v8.18.6 release. With Xinerama and TwinView being two of the popular multi-headed options for Linux users, we have decided to study the frame-rate performance for both of these configurations, as well as a traditional single-head setup, under a variety of popular games. As the focus of this article today is strictly upon the solid frame-rate results, we will not be sharing the various features directly nor the various advantages and disadvantages for each, however, more information is available at the NVIDIA website and the X.Org Foundation. For our testing, we used two Acer AL1715 LCD displays at 1280 x 1024, while below is the complete hardware, and software, specifications.

Hardware Components
Processor: Intel Pentium D 820 @ 3.36GHz
Motherboard: Abit AW8-MAX v1.0 (i955X)
Memory: 2 x 512MB Crucial Ballistix DDR2-800
Graphics Card: Leadtek PX7800GTX 256MB
Hard Drives: Seagate 160GB SATA2
Cooling: Sytrin KuFormula VF1 Plus
Case: Chenming Full ATX
Power Supply: Enermax Whisper II 2.0 535W SLI
Software Components
Operating System: Fedora Core 4
Linux Kernel: 2.6.14-1.653_FC4smp (x86_64)
GCC - GNU Compiler: 4.0.0
Graphics Driver: NVIDIA 1.0-8178
X.Org: 6.8.2

Onto the X.Org configurations for multi-headed environments, NVIDIA supports the configuration of both TwinView and Xinerama through its nvidia-xconfig utility. Although the advanced-help area is able to share all of the specific flags for automatically configuring the multi-headed support, the automatic configuration alone may not be enough for proper control, and its likely some manual X.Org modifications will be needed. As is stated inside of NVIDIA's official Linux manual, the X TwinView options needed is the boolean TwinView as well as the MetaModes and SecondMonitorHorizSync/SecondMonitorVertRefresh or HorizSync/VertRefresh. In addition, other optional items include TwinViewOrientation, ConnectedMonitor, and UseEdidFreqs. As far as the Xinerama enabling goes, it is not as elaborate as NVIDIA's present TwinView implementation, but rather drives two separate X.Org screens. For knowing the specific multi-headed and single-headed configurations used during testing, as well as the specific parameters for those who have experienced problems, on the following pages is our xorg.conf files. Following these sample configurations are our benchmarking results from Xinerama, TwinView, and a single-head setup using the above-mentioned hardware and software selection. The gaming benchmarks ran in this article was Enemy Territory, Doom 3, and Quake 4. With each of these Linux-native games, the graphical settings were unique (as noted on the actual results) and are designed to display the differences between the two Linux multi-headed options under our microscope today. Two of our benchmarks focused upon the dual-headed capabilities, while Quake 4 was running on the primary display, we had mplayer playing a movie full-screen on the secondary display, and then in the other benchmark we simply had glxgears running on the second LCD to monitor how the performance was effected. In the event of the single head tests, we simply ran glxgears and mplayer behind Quake 4. On the following pages are our official findings.

Latest Linux Hardware Reviews
  1. Intel Xeon E5-1680 v3 & E5-2687W v3 Compared To The Core i7 5960X On Linux
  2. Intel 120GB 530 Series SSD Linux Performance
  3. Btrfs/EXT4/XFS/F2FS RAID 0/1/5/6/10 Linux Benchmarks On Four SSDs
  4. AMD's Windows Catalyst Driver Remains Largely Faster Than Linux Drivers
Latest Linux Articles
  1. NVIDIA vs. Nouveau Drivers With Linux 3.18 + Mesa 10.4-devel
  2. Is The Open-Source NVIDIA Driver Fast Enough For Steam On Linux Gaming?
  3. Linux 3.18 File-System Performance Minimally Changed But Possible Regressions
  4. AMD Radeon Gallium3D Is Catching Up & Sometimes Beating Catalyst On Linux
Latest Linux News
  1. Linux 3.18-rc6 Released, A Worrisome Regression Remains
  2. HandBrake 0.10 Brings H.265 & VP8 Encoders
  3. Gngr: A New Web Browser Focused On Privacy
  4. Linux 3.18 Kernel: Not Much Change With Intel Haswell Performance
  5. More File-System Tests Of The Linux 3.18 Kernel
  6. Using NVIDIA's NVENC On Linux With FFmpeg
  7. There's Talk Again About An "Open To The Core" Ubuntu Laptop
  8. PowerVR SGX Driver Code Gets Leaked
  9. V2 Of KDBUS Published For Linux Kernel Review
  10. VirtualBox 4.3.20 Arrives, Still No Sign Of VirtualBox 4.4
Latest Forum Discussions
  1. PulseAudio 6.0 Is Coming & Other Linux Audio Plans For The Future
  2. Debian Developer Resigns From The Systemd Maintainership Team
  3. Roadmap to Catalyst 14.10 ?
  4. Updated and Optimized Ubuntu Free Graphics Drivers
  5. Cant get working Kaveri APU - A10-7850k
  6. Script for Fan Speed Control
  7. Debian Init System Coupling Vote Results
  8. The Slides Announcing The New "AMDGPU" Kernel Driver