Hands-On Report: Mac OS X 10.4.4--How Fast Is It?

Last week during Macworld, Apple released Mac OS X 10.4.4 Update to Tiger and the server edition too. As frequent readers know, we have benchmarked virtually every release of Mac OS X from the original 10.0 release through 10.4.3 to keep track of the improvements in its speed, a major complaint initially about Mac OS X.

Mac OS X runs very well these days and its speed enhancements have largely focussed on graphics. The improvements in 10.4.4 included Core Graphics, Core Image and updated ATI and NVIDIA graphics drivers so attention to graphics perfomance continues. Apple's new-to-Tiger Quartz 2D Extreme technology (in addition to and not to be confused with Quartz Extreme) is not implemented by default. One has to use a developer tool to turn it on. With the release of 10.4.3, Apple stated that it is permanantly disabled. This is curious because we have shown that when enabled it provides a huge performance boost to the User Interface actions and Quartz Graphics benchmarks using the Xbench speed benchmark. But one could still enable it with the developer tool Quartz Debug in 10.4.3 and the same is true for 10.4.4.

So how fast is it? Did 10.4.4 speed anything up or slow anything down? To determine this we ran our usual suite of speed tests on three test-bed Macs. The speed tests are composed of a series of "canned" benchmark applications and several so-called "real world" tests that anyone can repeat on their own Mac for comparison by following the steps that we perform.

Our long-term test Mac (since Mac OS X 10.0) has been a mostly-stock, year 2000-vintage 450 MHz Power Mac G4 Cube with 704MB RAM. With Mac OS X 10.3, Apple began including improvements that are aimed mainly at the G5 CPU and graphics cards with 32MB of video RAM or more (the Cube's has only 16MB) so we also began using our dual 2.5GHz Power Mac G5 with 1.5GB RAM and an Nvidia GeForce 6800 Ultra DDL graphics card, an upgrade, in the speed tests. We recently replaced the dual 2.5 GHz Power Mac G5 with a Power Mac G5 Quad (compared here) so we used the Quad that is stock except for an addition of 1GB RAM to 1.5GB. We also used our 20" 2 GHz iMac G5 (Rev. B), stock in every way, as a recent mid-range Mac.

Here are the results for the G4 Cube using Mac OS X 10.4.4 and Mac OS X 10.4.3, 10.4.2, 10.4 (Tiger) plus the last update of 10.3.9 (Panther). Other versions back to 10.0 can be found in our previous articles:

450 MHz G4 Cube Speed Tests (average* scores or secs).
Test 450 MHz G4 Cube
Mac OS X Version:
10.3.9
10.4
10.4.2
10.4.3
10.4.4
Benchmarks:
Altivec Fractal (GFLOPS)
1.5
1.5
1.5
1.5
1.5
Cinebench 2003 - Rendering
48
48
48
49
48
- OpenGL Hardware Lighting
91
90
91
92
91
Let1kWindowsBloom 1.0 (sec)
53
30
30
31
32
Xbench - CPU
54
54
54
53
53
- Threading
42
42
42
42
42
- Memory
76
77
79
80
79
- Quartz Graphics
74
73
77
77
75
- OpenGL Graphics
80
88
88
88
87
- User Interface Graphics
102
150
148
147
146
- Hard Disk
54
56
54
51
50
"Real World" Tests:
Startup (sec)
52
53
49
52
52
Shutdown (sec)
15
16
15
16
15
Launch Classic Mode (sec)
25
36
30
31
30
Dup. 700 MB Folder 3600 Files (sec)
152
128
130
133
132
Scroll 142p Acrobat Reader File (sec)
55
55
54
55
54
Encode MP3 in iTunes (sec)
105
104
104
104
104
Export QT FQDV Movie in iMovie (sec)
36
38
39
38
41
Quake III Arena 1.32 (frames/sec)
34
32
32
34
34
*Average of three runs.Red is much faster. Blue is much slower.

For this now low-end Mac, there are no improvements in 10.4.4. The only apparent loss in performance when compared with 10.4.3 is in exporting a QuickTime full quality DV clip in iMovie 3.0.3. The latter could be due QuickTime 7.0.4 released the same week as Mac OS X 10.4.4.

Here are the results for the 20" 2 GHz iMac G5 comparing 10.4.4 with 10.4.3 and 10.4.1:

2.0GHz iMac G5 (Rev. B) Speed Tests (average* scores or secs).
Test 20" 2 GHz iMac G5 (Rev. B)
Mac OS X Version
10.4.1+
10.4.3
10.4.4
Benchmarks:
Altivec Fractal (GFLOPS)
4.9
4.8
4.7
Cinebench 2003 - Rendering
276
277
277
- OpenGL Hardware Lighting
1152
1142
1145
Let1kWindowsBloom 1.0 (sec)
10
9
9
Xbench - CPU
139
144
145
- Threading
111
112
113
- Memory
232
236
236
- Quartz Graphics
332+
232
231
- OpenGL Graphics
212
207
210
- User Interface Graphics
296
345
354
- Hard Disk
138
137
136
"Real World" Tests:
Startup (sec)
37
40
40
Shutdown (sec)
8
8
8
Launch Classic Mode (sec)
14
14
14
Dup. 700 MB Folder 3600 Files (sec)
53
49
48
Scroll 142p Acrobat Reader File (sec)
18
15
15
Encode MP3 in iTunes (sec)
56
52
52
Export QT FQDV Movie in iMovie HD (sec)
23
27
22
Quake III Arena 1.32 (frames/sec)
140
147
142
Halo 1.5.2 (frames/sec)
19
20
19
Call of Duty 1.4 (frames/sec)
56
56
57
Doom3 1.3 MQ (frames/sec)
-
30
31
*Average of 3 runs. +With Quartz 2D Extreme enabled.Red is much faster. Blue is much slower.

The results for the 20" 2 GHz iMac G5 are opposite that of the G4 Cube. There is one apparent performance improvement compared with 10.4.3 and that is in the export of a QuickTime full quality DV movie in iMovie HD 5.0.4. However, it is really no better than achieved in Mac OS X 10.4.1 and may also be due to QuickTime 7.0.4 changes.

Here are the Power Mac G5 Quad's results comparing 10.4.4 with 10.4.3:

Quad 2.5GHz Power Mac G5 Speed Tests (average* scores or secs).
Test Quad 2.5GHz Power Mac G5
Mac OS X Version:
10.4.3
10.4.4
Benchmarks:
Altivec Fractal (GFLOPS)
25.0
25.0
Cinebench 2003 - Rendering
1091
1101
- OpenGL Hardware Lighting
1854
1848
Let1kWindowsBloom 1.0 (sec)
6
6
Xbench - CPU
243
242
- Threading
516
520
- Memory
517
534
- Quartz Graphics
311
310
- OpenGL Graphics
237
241
- User Interface Graphics
541
549
- Hard Disk
111
123
"Real World" Tests:
Startup (sec)
44
46
Shutdown (sec)
13
25
Launch Classic Mode (sec)
10
11
Dup. 700 MB Folder 3600 Files (sec)
42
42
Scroll 142p Acrobat Reader File (sec)
11
11
Encode MP3 in iTunes (sec)
52
52
Export QT FQDV Movie in iMovie HD (sec)
19
19
Quake III Arena 1.32 (frames/sec)
416
417
Halo 1.5.2 (frames/sec)
31
31
Call of Duty 1.4 (frames/sec)
69
85
Doom3 1.3 (frames/sec)
52
52
*Average of 3 runs. Red is much faster. Blue is much slower.

For the current top-end Power Mac G5, the Quad 2.5 GHz, there are two differences of note between Mac OS X 10.4.4 and 10.4.3. First, the Quad is taking almost twice as long to shut down under 10.4.4 and much longer than either the iMac G5 or G4 Cube. We could not isolate a reason for this but it is probably related to one of several applications with menu items installed since we ran it with 10.4.3. If so, that would also account for the slightly longer startup time. The one application that really improved in performance is our favorite first person shooter, Call of Duty. It's frame rate is now equal to that of our former dual 2.5 GHz Power Mac G5 with a much faster Nvidia GeForce 6800 Ultra DDL graphics card. The Quad has a stock Nvidia GeForce 6600 graphics card while we wait for an after-market Nvidia GeForce 7800 GT or ATI Radeon X1800 to be released--we may not need either unless we get interested again in Halo or Halo 2 is released. Unlike the two other Macs, there was no change in the time to export a QuickTime DV movie in iMovie HD 5.0.4.

Mac OS X 10.4.4, like 10.4.3, 10.4.2 and 10.4.1, has mainly security, compatibility and stability improvements. We found opposite changes in exporting a QuickTime DV movie in two computers but no change in the third. We found a slight but not significant improvement in the Call of Duty frame rate on the iMac and a significant 23% improvement on the Quad G5. This may be because the iMac G5 has an ATI graphics (as does the Cube) card and the Quad G5 has an Nvidia graphics card and the improved Nvidia drivers are much more improved than ATI's. We are very happy with the result.

We have used Mac OS X 10.4.4 for a week on a number of Macs and PowerBooks without running into any new or significant problems. [Bill Fox]



Saw something? Send a tip

The archive ran on reader tips. What did you see, where, and do you want the credit?

Read by the editor. Never published without your say.

More in Reviews · This month in the archive