Follow Up--Mac OS X 10.5.2, How Fast Is It?
| Test | 10.5.2 | 10.5.1 | 10.4.9 |
Performance Change (%) |
|---|---|---|---|---|
BENCHMARKS |
10.5.1vs10.4.9 | |||
Altivec Fractal 1.3 (GFLOPS) |
9.8 |
9.6 |
10.0 |
-4 |
Cinebench 9.5 Rendering (2x) |
646 |
707 |
738 |
-4 |
-OpenGL Hardware Lighting |
4122 |
4027 |
3903 |
+3 |
Let1kWindowsBloom 1.0+ |
11 |
11 |
8 |
-38 |
Xbench 1.3 - CPU |
150 |
137 |
122 |
+11 |
- Threading |
188 |
187 |
236 |
-21 |
- Memory |
158 |
159 |
135 |
+18 |
- Quartz Graphics |
192 |
201 |
175 |
+15 |
- OpenGL Graphics |
161 |
129 |
137 |
-6 |
-User Interface Graphics |
326 |
303 |
432 |
-30 |
"REAL WORLD" TESTS |
||||
Startup (sec) |
52 |
56 |
31 |
-80 |
Shutdown (sec) |
7 |
8 |
10 |
+20 |
Dup. 500 MB Folder (sec) |
63 |
62 |
61 |
-2 |
Quake III Arena (frames/sec) |
686 |
398 |
407 |
-2 |
Halo 2.0.2 (frames/sec) |
115 |
100 |
86 |
+16 |
In our original speed test that compared Tiger (10.4.9) to Leopard (10.5.1), Leopard ran substantially slower on four tests: Let1kwindowsbloom, Xbench 1.3 Threading, Xbench 1.3 User Interface Graphics and Startup time. The latest Leopard update 10.5.2 shows a slight improvement in two of them, Xbench 1.3 User Interface Graphics and Startup time.
Leopard 10.5.1 ran substantially faster than Tiger 10.4.9 in four tests: Xbench 1.3 Memory, Xbench 1.3 Quartz Graphics, Shutdown time and Halo frame rate. 10.5.2 shows a further improvement in two of them: Shutdown time and Halo frame rate.
Most notably, there are clear improvements in OpenGL performance in 10.5.2 as shown by Cinebench OpenGL Hardware Lighting (+2%), Xbench 1.3 OpenGL Graphics (+25%), Quake II Arena (+72%) and Halo (+15%). [Bill Fox]
Saw something? Send a tip
The archive ran on reader tips. What did you see, where, and do you want the credit?