This page is devoted to displaying benchmarks comparing an iMac running OSX 10.5.8 to the same machine running Ubuntu 9.10.

====== Network Speed Tests ======

Tests were performed by copying a 16.14GB file from a Linux server to a workstation's hard drive by a variety of methods: SFTP (Cyberduck|Nautilus, scp) and NFS (Finder|Nautilus, cp).

^ Operating System ^ Program ^ Protocol ^ Data Size ^ Time ^
| OSX | Cyberduck | SFTP | 16.14GB | 294 minutes (4.9 hours) |
| OSX | Finder | NFS | 16.14GB | 57 minutes |
| OSX | scp | SFTP | 16.14GB | 28 minutes |
| OSX | cp | NFS | 16.14GB | 19 minutes |
| Linux | Nautilus | SFTP | 16.14GB | 12 minutes |
| Linux | scp | SFTP | 16.14GB | 9 minutes |
| Linux | Nautilus | NFS | 16.14GB | 8 minutes |
| Linux | cp | NFS | 16.14GB | 6 minutes |

====== Real-World Network Speeds ====== 

Network usage under normal daily conditions:

^ OSX ^
| {{:linux-advocacy-bandwidthd-osx.png|}} \\ User #1|

^ Linux ^
| {{:linux-advocacy-bandwidthd-linux.png|}} \\ User #2 | 

**Disclaimer:** Both machines in the graphs above use NFS-automounting to provide the console user's home directory from a central server.  Every I/O the user does to their home area must go over the network.  However, these graphs are of two different machines with different users logged in on the console over the same time period.  More data is needed to discount discrepancies caused by different users and their working styles.  Unfortunately, at this time there is only enough data to cover a single month of Linux usage.

^ Server Bandwidth Use ^
| {{:linux-advocacy-bandwidthd-server.png|}} \\ Server Bandwidth Usage |

This third graph shows the bandwidth use of an NFS server over the last seven months.  Note the increase in graph density that begins around mid-February, this is around the time that we began migrating users over to Linux.  This server serves user home areas to 26 OSX workstations, of which 9 were in constant daily use by staff.

**Conclusion:** Despite the fact that Apple's iMac network card supports gigabit speeds, throughput on daily use never exceeded the speeds available from a typical 100base-T connection.  Even when copying files over the network it was difficult to exceed 100base-T speeds. After performing these network file copy tests under OSX it was thought that perhaps OSX itself was the bottleneck.  The data above seems to support this theory, as Linux, running on the same hardware, provides a significant increase in performance.

====== Benchmarks ======

Hardware: iMac8,1
  * CPU: Intel Core 2 Duo CPU E8235 @ 2.8GHz (Total Cores: 2)
  * Motherboard: Apple Mac-F227BEC8
  * Chipset: Intel Mobile PM965/GM965/GL960 + ICH8M
  * Memory: 2 x 2GB DDR2 800MHz

Operating System:
  * Mac OS X 10.5.8, Kernel: 9.8.0 (i386)
  * Ubuntu 9.10, Kernel: 2.6.31-19-generic (x86-64), Xorg: open-source video driver

Tests: Benchmarking software used is Phoronix Test Suite 2.4.1.

^ System, Processor, and Disk Benchmarks ^^^
^ Test ^ OSX ^ Linux ^
| Bork file encryption \\ (seconds, fewer is better) | 68.06 | 59.61 (winner) |
| Apache compilation \\ (seconds, fewer is better) | 49.18 | 45.68 (winner) |
| ImageMagick compilation \\ (seconds, fewer is better) | 270.87 | 143.49 (winner) |
| Mplayer compilation \\ (seconds, fewer is better) | 132.98 | 129.43 (winner) |
| MySQL compilation \\ (seconds, fewer is better) | 390.16 | 219.47 (winner) |
| PHP compilation \\ (seconds, fewer is better) | 145.45 | 67.59 (winner) |
| Bullet Physics Engine 2.75 - 3000 Fall \\ (seconds, fewer is better) | 30.73 | 23.31 (winner) |
| Bullet Physics Engine 2.75 - 1000 Stack \\ (seconds, fewer is better) | 34.79 | 23.34 (winner) |
| Bullet Physics Engine 2.75 - 136 Ragdolls \\ (seconds, fewer is better) | 14.99 | 10.11 (winner) |
| Bullet Physics Engine 2.75 - 1000 Convex \\ (seconds, fewer is better) | 22.81 | 14.68 (winner) |
| Bullet Physics Engine 2.75 - Prim Trimesh \\ (seconds, fewer is better) | 25.32 | 15.92 (winner) |
| Bullet Physics Engine 2.75 - Convex Trimesh \\ (seconds, fewer is better) | 41.62 | 28.11 (winner) |
| Bullet Physics Engine 2.75 - Raytests \\ (seconds, fewer is better) | 49.63 | 32.68 (winner) |
| SciMark 2.0 - Dhrystone 2 \\ (LPS, more is better) | 8,418,988.80 | 14,498,089.70 (winner) |
| SciMark 2.0 - Register Arithmetic \\ (LPS, more is better) | 2,476,830.30 | 2,687,692.70 (winner) |
| SciMark 2.0 - Integer Arithmetic \\ (LPS, more is better) | 2,475,616.10 | 2,691,065.30 (winner) |
| SciMark 2.0 - Floating-Point Arithmetic \\ (LPS, more is better) | 2,250,090.20 | 2,260,545.00 (winner) |
| CompileBench - Initial Create \\ (MB/s, more is better) | 1.17 | 62.32 (winner) |
| CompileBench - Compile \\ (MB/s, more is better) | 7.26 | 65.99 (winner) |
| CompileBench - Compiled Tree \\ (MB/s, more is better) | 10.38 | 350.25 (winner) |
| 7zip \\ (MIPS, more is better) | 3279 | 4157 (winner) |
| gzip \\ (seconds, fewer is better) | 22.19 | 16.78 (winner) |
| Crafty \\ (seconds, fewer is better) | 80.15 | 34.24 (winner) |
| dcraw - RAW to PPM Image Conversion \\ (seconds, fewer is better) | 48.06 | 33.61 (winner) |
| LAME MP3 Encoding v3.98.2 \\ (seconds, fewer is better) | 37.51 | 31.32 (winner) |
| WavPack Audio Encoding v4.41.0 \\ (seconds, fewer is better) | 21.77 | 13.80 (winner) |
| FFmpeg v0.5 \\ (seconds, fewer is better) | 21.15 | 17.32 (winner) |
| Gcrypt Library v1.4.4 \\ (microseconds, fewer is better) | 3967 | 3293 (winner) |
| Java SciMark v2.0 - Composite \\ (Mflops, more is better) | 383.23 | 894.52 (winner) |
| Java SciMark v2.0 - Fast Fourier Transform \\ (Mflops, more is better) | 153.65 | 531.70 (winner) |
| Java SciMark v2.0 - Jacobi Successive Over-Relaxation \\ (Mflops, more is better) | 790.62 | 1112.23 (winner) |
| Java SciMark v2.0 - Monte Carlo \\ (Mflops, more is better) | 34.44 | 352.58 (winner) |
| Java SciMark v2.0 - Sparse Matrix Multiply \\ (Mflops, more is better) | 358.39 | 776.00 (winner) |
| Java SciMark v2.0 - Dense LU Matrix Factorization \\ (Mflops, more is better) | 584.10 | 1682.77 (winner) |
| John The Ripper v1.7.3.1 - Traditional DES \\ (Real C/S, more is better) | 2,617,333 | 2,653,333 (winner) |
| John The Ripper v1.7.3.1 - MD5 \\ (Real C/S, more is better) | 11,203 | 12,098 (winner) |
| John The Ripper v1.7.3.1 - Blowfish \\ (Real C/S, more is better) | 640 | 729 (winner) |
| Timed MAFFT Alignment v6.706 - Multiple Sequence Alignment \\ (seconds, fewer is better) | 40.57 | 32.63 (winner) |
| GROMACS MD Benchmark v4.0.5 \\ Molecule: Villin, Run-Type: Single Node, Precision: Single \\ (Gflops, more is better) | 3.28 | 3.59 (winner) |
| GROMACS MD Benchmark v4.0.5 \\ Molecule: Villin, Run-Type: Single Node, Precision: Double \\ (Gflops, more is better) | 2.12 | 2.24 (winner) |
| GROMACS MD Benchmark v4.0.5 \\ Molecule: Villin, Run-Type: Parallel, Precision: Single \\ (Gflops, more is better) | 3.89 (winner) | 3.79 |
| GROMACS MD Benchmark v4.0.5 \\ Molecule: Villin, Run-Type: Parallel, Precision: Double \\ (Gflops, more is better) | 2.66 (winner) | 2.35 |
| GROMACS MD Benchmark v4.0.5 \\ Molecule: DPPC, Run-Type: Single Node, Precision: Single \\ (Gflops, more is better) | 2.04 | 2.34 (winner) |
| GROMACS MD Benchmark v4.0.5 \\ Molecule: DPPC, Run-Type: Single Node, Precision: Double \\ (Gflops, more is better) | 1.38 | 1.56 (winner) |
| GROMACS MD Benchmark v4.0.5 \\ Molecule: DPPC, Run-Type: Parallel, Precision: Single \\ (Gflops, more is better) | 3.26 | 4.52 (winner) |
| GROMACS MD Benchmark v4.0.5 \\ Molecule: DPPC, Run-Type: Parallel, Precision: Double \\ (Gflops, more is better) | 2.18 | 2.58 (winner) |
| GROMACS MD Benchmark v4.0.5 \\ Molecule: Lysozyme, Run-Type: Single Node, Precision: Single \\ (Gflops, more is better) | 2.53 | 2.86 (winner) |
| GROMACS MD Benchmark v4.0.5 \\ Molecule: Lysozyme, Run-Type: Single Node, Precision: Double \\ (Gflops, more is better) | 1.74 | 1.91 (winner) |
| GROMACS MD Benchmark v4.0.5 \\ Molecule: Lysozyme, Run-Type: Parallel, Precision: Single \\ (Gflops, more is better) | 3.85 | 4.55 (winner) |
| GROMACS MD Benchmark v4.0.5 \\ Molecule: Lysozyme, Run-Type: Parallel, Precision: Double \\ (Gflops, more is better) | 2.69 | 3.11 (winner) |
| GROMACS MD Benchmark v4.0.5 \\ Molecule: Poly-CH2, Run-Type: Single Node, Precision: Single \\ (Gflops, more is better) | 1.35 | 1.63 (winner) |
| GROMACS MD Benchmark v4.0.5 \\ Molecule: Poly-CH2, Run-Type: Single Node, Precision: Double \\ (Gflops, more is better) | 1.00 | 1.24 (winner) |
| GROMACS MD Benchmark v4.0.5 \\ Molecule: Poly-CH2, Run-Type: Parallel, Precision: Single \\ (Gflops, more is better) | 1.44 | 1.50 (winner) |
| GROMACS MD Benchmark v4.0.5 \\ Molecule: Poly-CH2, Run-Type: Parallel, Precision: Double \\ (Gflops, more is better) | 1.17 (tie) | 1.17 (tie) |
| Mencoder v2009-06-04 - AVI to LAVC \\ (seconds, fewer is better) | 35.55 (winner) | 39.34 |
| OpenSSL v0.9.8k \\ (Signs/s, more is better) | 7.55 | 79.58 (winner) |
| PostMark v1.51 \\ (TPS, more is better) | 70 | 2500 (winner) |
| PyBench v2008-08-14 \\ (Miliseconds, fewer is better) | 5632 | 3523 (winner) |
| Sample Pi Program \\ (seconds, fewer is better) | 3.33 (winner) | 7.27 |
| SciMark v2.0 - Composite \\ (Mflops, more is better) | 535.16 | 549.78 (winner) |
| SciMark v2.0 - Fast Fourier Transform \\ (Mflops, more is better) | 105.27 (winner) | 85.43 |
| SciMark v2.0 - Jacobi Successive Over-Relaxation \\ (Mflops, more is better) | 812.79 | 821.95 (winner) |
| SciMark v2.0 - Monte Carlo \\ (Mflops, more is better) | 175.05 | 211.37 (winner) |
| SciMark v2.0 - Sparse Matrix Multiply \\ (Mflops, more is better) | 633.07 (winner) | 627.78 |
| SciMark v2.0 - Dense LU Matrix Factorization \\ (MFlops, more is better) | 942.14 | 1001.58 (winner) |
| Tachyon v0.98.7 \\ (seconds, fewer is better) | 97.13 | 91.86 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Write, Size/Thread: 32MB, Count: 4 \\ (MB/s, more is better) | 53.67 | 62.49 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Write, Size/Thread: 32MB, Count: 8 \\ (MB/s, more is better) | 50.59 | 63.48 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Write, Size/Thread: 32MB, Count: 16 \\ (MB/s, more is better) | 48.93 | 64.12 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Write, Size/Thread: 32MB, Count: 32 \\ (MB/s, more is better) | 46.41 | 67.18 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Write, Size/Thread: 64MB, Count: 4 \\ (MB/s, more is better) | 52.48 | 61.64 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Write, Size/Thread: 64MB, Count: 8 \\ (MB/s, more is better) | 48.89 | 64.18 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Write, Size/Thread: 64MB, Count: 16 \\ (MB/s, more is better) | 47.58 | 66.12 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Write, Size/Thread: 64MB, Count: 32 \\ (MB/s, more is better) | 44.82 | 66.22 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Write, Size/Thread: 128MB, Count: 4 \\ (MB/s, more is better) | 50.70 | 60.57 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Write, Size/Thread: 128MB, Count: 8 \\ (MB/s, more is better) | 48.65 | 62.36 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Write, Size/Thread: 128MB, Count: 16 \\ (MB/s, more is better) | 46.98 | 64.34 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Write, Size/Thread: 128MB, Count: 32 \\ (MB/s, more is better) | 43.02 | 65.34 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Write, Size/Thread: 256MB, Count: 4 \\ (MB/s, more is better) | 49.75 | 61.34 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Write, Size/Thread: 256MB, Count: 8 \\ (MB/s, more is better) | 47.34 | 62.86 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Write, Size/Thread: 256MB, Count: 16 \\ (MB/s, more is better) | 45.12 | 61.42 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Write, Size/Thread: 256MB, Count: 32 \\ (MB/s, more is better) | 41.65 | 60.19 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Write, Size/Thread: 32MB, Count: 4 \\ (MB/s, more is better) | 3.50 | 5.23 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Write, Size/Thread: 32MB, Count: 8 \\ (MB/s, more is better) | 3.03 | 4.90 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Write, Size/Thread: 32MB, Count: 16 \\ (MB/s, more is better) | 2.57 | 4.76 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Write, Size/Thread: 32MB, Count: 32 \\ (MB/s, more is better) | 2.19 | 4.60 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Write, Size/Thread: 64MB, Count: 4 \\ (MB/s, more is better) | 2.35 | 3.35 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Write, Size/Thread: 64MB, Count: 8 \\ (MB/s, more is better) | 2.13 | 3.28 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Write, Size/Thread: 64MB, Count: 16 \\ (MB/s, more is better) | 1.89 | 3.20 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Write, Size/Thread: 64MB, Count: 32 \\ (MB/s, more is better) | 1.64 | 3.06 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Write, Size/Thread: 128MB, Count: 4 \\ (MB/s, more is better) | 1.66 | 2.43 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Write, Size/Thread: 128MB, Count: 8 \\ (MB/s, more is better) | 1.63 | 2.36 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Write, Size/Thread: 128MB, Count: 16 \\ (MB/s, more is better) | 1.52 | 2.28 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Write, Size/Thread: 128MB, Count: 32 \\ (MB/s, more is better) | 1.30 | 2.19 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Write, Size/Thread: 256MB, Count: 4 \\ (MB/s, more is better) |  1.44 | 1.89 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Write, Size/Thread: 256MB, Count: 8 \\ (MB/s, more is better) | 1.33 | 1.84 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Write, Size/Thread: 256MB, Count: 16 \\ (MB/s, more is better) | 1.20 | 1.78 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Write, Size/Thread: 256MB, Count: 32 \\ (MB/s, more is better) | 0.98 | 1.69 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Read, Size/Thread: 32MB, Count: 4 \\ (MB/s, more is better) | 1344.54 | 1535.30 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Read, Size/Thread: 32MB, Count: 8 \\ (MB/s, more is better) | 1238.47 | 2018.02 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Read, Size/Thread: 32MB, Count: 16 \\ (MB/s, more is better) | 1425.75 | 2032.65 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Read, Size/Thread: 32MB, Count: 32 \\ (MB/s, more is better) | 1469.15 | 2051.40 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Read, Size/Thread: 64MB, Count: 4 \\ (MB/s, more is better) | 1292.74 | 1552.54 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Read, Size/Thread: 64MB, Count: 8 \\ (MB/s, more is better) | 1647.64 | 1840.78 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Read, Size/Thread: 64MB, Count: 16 \\ (MB/s, more is better) | 1657.52 | 2091.90 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Read, Size/Thread: 64MB, Count: 32 \\ (MB/s, more is better) | 1644.28 | 2118.80 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Read, Size/Thread: 128MB, Count: 4 \\ (MB/s, more is better) | 1641.96 | 1745.07 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Read, Size/Thread: 128MB, Count: 8 \\ (MB/s, more is better) | 1645.63 | 1931.63 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Read, Size/Thread: 128MB, Count: 16 \\ (MB/s, more is better) | 1612.93 | 2143.48 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Read, Size/Thread: 128MB, Count: 32 \\ (MB/s, more is better) | 49.85 | 57.49 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Read, Size/Thread: 256MB, Count: 4 \\ (MB/s, more is better) | 1590.29 | 1613.43 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Read, Size/Thread: 256MB, Count: 8 \\ (MB/s, more is better) | 1651.21 | 2004.90 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Read, Size/Thread: 256MB, Count: 16 \\ (MB/s, more is better) | 57.54 | 98.08 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Read, Size/Thread: 256MB, Count: 32 \\ (MB/s, more is better) | 50.40 | 63.77 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Read, Size/Thread: 32MB, Count: 4 \\ (MB/s, more is better) | 1036.87 | 1312.89 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Read, Size/Thread: 32MB, Count: 8 \\ (MB/s, more is better) | 1042.30 | 1574.54 (winner) |
| Threaded I/O Teser v0.3.3 \\ Test: Random Read, Size/Thread: 32MB, Count: 16 \\ (MB/s, more is better) | 1084.21 | 1815.03 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Read, Size/Thread: 32MB, Count: 32 \\ (MB/s, more is better) | 1124.80 | 1752.13 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Read, Size/Thread: 64MB, Count: 4 \\ (MB/s, more is better) | 901.49 | 1294.89 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Read, Size/Thread: 64MB, Count: 8 \\ (MB/s, more is better) | 1067.93 | 1510.43 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Read, Size/Thread: 64MB, Count: 16 \\ (MB/s, more is better) | 1080.87 | 1807.89 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Read, Size/Thread: 64MB, Count: 32 \\ (MB/s, more is better) | 949.26 | 1790.50 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Read, Size/Thread: 128MB, Count: 4 \\ (MB/s, more is better) | 952.57 | 1370.93 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Read, Size/Thread: 128MB, Count: 8 \\ (MB/s, more is better) | 1006.37 | 1550.81 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Read, Size/Thread: 128MB, Count: 16 \\ (MB/s, more is better) | 1099.20 | 1798.61 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Read, Size/Thread: 128MB, Count: 32 \\ (MB/s, more is better) | 1.90 | 3.49 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Read, Size/Thread: 256MB, Count: 4 \\ (MB/s, more is better) | 925.42 | 1288.60 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Read, Size/Thread: 256MB, Count: 8 \\ (MB/s, more is better) | 768.38 | 1647.36 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Read, Size/Thread: 256MB, Count: 16 \\ (MB/s, more is better) | 1.88 | 3.25 (winner) |
| Threaded I/O Tester v0.3.3 \\ Test: Random Read, Size/Thread: 256MB, Count: 32 \\ (MB/s, more is better) | 0.86 | 0.87 (winner) |
| TSCP v1.81 - AI Chess Performance \\ (Nodes/s, More is better) | 244,486 | 273,556 (winner) |

There are two types of iMac graphics chips available for testing, both are tested.

Hardware: iMac8,1
  * CPU: Intel Core 2 Duo CPU E8235 @ 2.8GHz (Total Cores: 2)
  * Motherboard: Apple Mac-F227BEC8
  * Chipset: Intel Mobile PM965/GM965/GL960 + ICH8M
  * Memory: 2 x 2GB DDR2 800MHz

Hardware: iMac9,1
  * CPU: Intel Core 2 Duo @ 2.66GHz (Total Cores: 2)
  * Motherboard: Apple Mac-F2218FC8
  * Chipset: nVidia MCP79
  * Memory: 2 x 2GB DDR2 1067MHz

Operating System:
  * Mac OS X 10.5.8, Kernel: 9.8.0 (i386)
  * Ubuntu 9.10, Kernel: 2.6.31-19-generic (x86-64), Xorg: open-source video driver

^ Graphics Benchmarks ^^^^^
^ Test ^ OSX (ATI) ^ Linux (ATI) ^ OSX (nVidia) ^ Linux (nVidia) ^
| nexuiz-1.5.0 @ 800x600, HDR, Sound \\ (Frames Per Second, More is better) | 30.16 | 49.18 (winner) | 27.23 | 43.18 (winner) |
| nexuiz-1.5.0 @ 800x600, HDR, No Sound \\ (Frames Per Second, More is better) | 30.74 | 50.97 (winner) | 27.55 | 44.65 (winner) |
| nexuiz-1.5.0 @ 800x600, No HDR, Sound \\ (Frames Per Second, More is better) | 46.41 | 69.91 (winner) | 42.88 | 55.29 (winner) |
| nexuiz-1.5.0 @ 800x600, No HDR, No Sound \\ (Frames Per Second, More is better) | 48.02 | 68.08 (winner) | 43.18 | 56.44 (winner) |
| nexuiz-1.5.0 @ 1024x768, HDR, Sound \\ (Frames Per Second, More is better) | 27.28 | 39.27 (winner) | 23.36 | 33.03 (winner) |
| nexuiz-1.5.0 @ 1024x768, HDR, No Sound \\ (Frames Per Second, More is better) | 27.82 | 38.84 (winner) | 23.63 | 33.78 (winner) |
| nexuiz-1.5.0 @ 1024x768, No HDR, Sound \\ (Frames Per Second, More is better) | 38.51 | 54.18 (winner) | 33.69 | 40.09 (winner) |
| nexuiz-1.5.0 @ 1024x768, No HDR, No Sound \\ (Frames Per Second, More is better) | 38.99 | 53.64 (winner) | 33.87 | 40.61 (winner) |
| nexuiz-1.5.0 @ 1280x1024, HDR, Sound \\ (Frames Per Second, More is better) | 32.49 (winner) | 31.33 | 31.85 (winner) | 23.68 |
| nexuiz-1.5.0 @ 1280x1024, HDR, No Sound \\ (Frames Per Second, More is better) | 33.54 (winner) | 31.14 | 32.60 (winner) | 24.00 |
| nexuiz-1.5.0 @ 1280x1024, No HDR, Sound \\ (Frames Per Second, More is better) | 55.14 (winner) | 41.42 | 53.59 (winner) | 27.63 |
| nexuiz-1.5.0 @ 1280x1024, No HDR, No Sound \\ (Frames Per Second, More is better) | 57.63 (winner) | 40.32 | 54.88 (winner) | 27.73 |
| nexuiz-1.5.0 @ 1400x1050, HDR, Sound \\ (Frames Per Second, More is better) | 32.48 (winner) | 29.66 | 32.03 (winner) | 21.78 |
| nexuiz-1.5.0 @ 1400x1050, HDR, No Sound \\ (Frames Per Second, More is better) | 33.65 (winner) | 30.00 | 32.46 (winner) | 22.09 |
| nexuiz-1.5.0 @ 1400x1050, No HDR, Sound \\ (Frames Per Second, More is better) | 55.07 (winner) | 38.03 | 53.62 (winner) | 25.10 |
| nexuiz-1.5.0 @ 1400x1050, No HDR, No Sound \\ (Frames Per Second, More is better) | 57.61 (winner) | 38.18 | 55.24 (winner) | 25.25 |
| nexuiz-1.5.0 @ 1680x1050, HDR, Sound \\ (Frames Per Second, More is better) | 32.45 (winner) | 27.17 | 31.87 (winner) | 19.48 |
| nexuiz-1.5.0 @ 1680x1050, HDR, No Sound \\ (Frames Per Second, More is better) | 33.39 (winner) | 27.13 | 32.47 (winner) | 19.57 |
| nexuiz-1.5.0 @ 1680x1050, No HDR, Sound \\ (Frames Per Second, More is better) | 55.00 (winner) | 27.13 | 53.56 (winner) | 20.45 |
| nexuiz-1.5.0 @ 1680x1050, No HDR, No Sound \\ (Frames Per Second, More is better) | 57.50 (winner) | 34.00 | 55.29 (winner) | 20.14 |
| nexuiz-1.5.0 @ 1600x1200, HDR, Sound \\ (Frames Per Second, More is better) | 17.98 | 25.83 (winner) | n/a | n/a |
| nexuiz-1.5.0 @ 1600x1200, HDR, No Sound \\ (Frames Per Second, More is better) | 18.07 | 25.85 (winner) | n/a | n/a |
| nexuiz-1.5.0 @ 1600x1200, No HDR, Sound \\ (Frames Per Second, More is better) | 20.81 | 32.18 (winner) | n/a | n/a |
| nexuiz-1.5.0 @ 1600x1200, No HDR, No Sound \\ (Frames Per Second, More is better) | 20.80 | 31.92 (winner) | n/a | n/a |
| nexuiz-1.5.0 @ 1920x1080, HDR, Sound \\ (Frames Per Second, More is better) | 32.70 (winner) | 24.78 | 31.93 (winner) | 15.87 |
| nexuiz-1.5.0 @ 1920x1080, HDR, No Sound \\ (Frames Per Second, More is better) | 33.23 (winner) | 24.89 | 32.59 (winner) | 15.94 |
| nexuiz-1.5.0 @ 1920x1080, No HDR, Sound \\ (Frames Per Second, More is better) | 55.22 (winner) | 30.84 | 53.09 (winner) | 17.72 |
| nexuiz-1.5.0 @ 1920x1080, No HDR, No Sound \\ (Frames Per Second, More is better) | 57.75 (winner) | 30.50 | 55.33 (winner) | 17.86 |
| openarena-0.8.1 @ 800x600 \\ (Frames Per Second, More is better) | n/a | n/a | 169.43 | 199.63 (winner) |
| openarena-0.8.1 @ 1024x768 \\ (Frames Per Second, More is better) | n/a | n/a | 136.16 (winner) | 134.90 |
| openarena-0.8.1 @ 1280x1024 \\ (Frames Per Second, More is better) | n/a | n/a | 98.50 (winner) | 96.16 |
| openarena-0.8.1 @ 1440x1050 \\ (Frames Per Second, More is better) | n/a | n/a | 90.50 (winner) | 82.06 |
| openarena-0.8.1 @ 1680x1050 \\ (Frames Per Second, More is better) | n/a | n/a | 78.73 (winner) | 70.43 |
| openarena-0.8.1 @ 1920x1080 \\ (Frames Per Second, More is better) | n/a | n/a | 69.40 (winner) | 61.56 |
| warsow-0.5 @ 800x600 \\ (Frames Per Second, More is better) | n/a | n/a | 75.76 | 86.00 (winner) |
| warsow-0.5 @ 1024x768 \\ (Frames Per Second, More is better) | n/a | n/a | 64.23 | 69.23 (winner) |
| warsow-0.5 @ 1680x1050 \\ (Frames Per Second, More is better) | n/a | n/a | 44.00 | 44.60 (winner) |
| warsow-0.5 @ 1920x1080 \\ (Frames Per Second, More is better) | n/a | n/a | 40.60 (winner) | 39.66 |
| urbanterror-4.1 @ 800x600 \\ (Frames Per Second, More is better) | 99.05 (winner) | 89.60 | 91.60 (winner) | 84.90 |
| urbanterror-4.1 @ 1024x768 \\ (Frames Per Second, More is better) | n/a | n/a | 86.50 (winner) | 84.10 |
| urbanterror-4.1 @ 1400x1050 \\ (Frames Per Second, More is better) | 99.85 (winner) | 89.90 | 68.50 | 73.95 (winner) |
| urbanterror-4.1 @ 1680x1050 \\ (Frames Per Second, More is better) | 97.10 (winner) | 89.60 | 63.35 | 69.25 (winner) |
| urbanterror-4.1 @ 1920x1080 \\ (Frames Per Second, More is better) | 95.35 (winner) | 88.25 | n/a | n/a |