Hands-On Report: Mac OS X 10.4.6 (PPC)--How Fast Is It?

2004–09April 10, 2006Bill FoxReviews
Last Tuesday, Apple released Mac OS X 10.4.6 Update to Tiger for the Power PC CPU, Intel CPU and server editions (see this Apple Web page). As frequent readers know, we have benchmarked virtually every release of Mac OS X from the original 10.0 release through 10.4.5 to keep track of the improvements in its speed, a major complaint initially about Mac OS X. Fortunately, Mac OS X runs very well these days. Mac OS X 10.4 Tiger brought significant increases in speed to the Finder and OpenGL graphics. Subsequent updates through 10.4.5 brought stability and security improvements but no real speed changes of note that we could detect with our test bed Macs.

Now we have Mac OS X 10.4.6 (PPC), how fast is it? Did 10.4.6 speed anything up or slow anything down? To determine this we ran our usual suite of speed tests on two of our 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. For 10.4.6, however, we have begun using Universal Binary versions of some test applications that were not available for our tests prior to Mac OS X 10.4.5.

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. But time marches on so we have retired it for the rest of the "Tiger" updates. Doing so allows us to spend our time on the newer Intel-based Macs. We will use the G4 Cube again when Mac OS X 10.5 "Leopard" is released in order to gauge its effects on the performance of non-upgraded legacy G4 Macs.

In these tests, we used Apple's top-of-the-line Power Mac G5 Quad that is stock except for an additional of 2GB RAM to 2.5GB. Our second test Mac is a 20" 2 GHz iMac G5 (Rev. B), stock in every way, to represent a relatively recent mid-range PPC Mac. We will post the results from our Intel-based Macs at a later date.

Here are the results for the 20" 2 GHz iMac G5 comparing 10.4.6 with 10.4.5:

2.0GHz iMac G5 (Rev. B) Speed Tests (average* scores or secs).
Test+ Mac OS X
10.4.5
10.4.6
Chg. %
Benchmarks:
Altivec Fractal (GFLOPS)
4.8
5.1
+6
Cinebench 2003 (9.5)+ - Rendering
275
272
-0
- OpenGL Hardware Lighting
1226
1197
-2
Let1kWindowsBloom 1.0 (sec)
9
9
0
Xbench 1.2+ - CPU
84
88
+5
- Threading
55
60
+9
- Memory
64
64
0
- Quartz Graphics
88
88
0
- OpenGL Graphics
121
119
-0
- User Interface Graphics
69
65
-6
- Hard Disk
92
90
-2
"Real World" Tests:
Startup (sec)
40
42
+5
Shutdown (sec)
7
8
+14
Launch Classic Mode (sec)
13
13
0
Dup. 500 MB Folder of Files (sec)
62
61
-2
Scroll 144p Adobe 7.0.7 Reader File (sec)
39
39
0
Export QT FQDV Movie in iMovie HD 5.0.2 (sec)
21
22
+5
Quake III Arena 1.32 (frames/sec)
142
141
-0
Halo 1.5.2 (frames/sec)
19
20
+5
Call of Duty 1.4 (frames/sec)
56
56
0
Doom 3 1.3A+ MQ (frames/sec)
38
32
-16
*Average of 3 runs. +Universal binary version. Red is much faster. Blue is much slower.

The results for the 20" 2 GHz iMac G5 show, with one exception, no appreciable differences between 10.4.6 and 10.4.5. The frame rate for Doom 3 v1.3A dropped significantly to the rate for Doom 3 v1.3, i.e. before the Universal Binary patch. This also happened on our iMac Core Duo which we will be reporting on later.

Here are the Power Mac G5 Quad's results comparing 10.4.6 with 10.4.5:

Quad 2.5GHz Power Mac G5 Speed Tests (average* scores or secs).
Test Mac OS X
10.4.5
10.4.6
Chg. %
Benchmarks:
Altivec Fractal (GFLOPS)
24.6
25.3
+3
Cinebench 2003 (9.5)+ - Rendering
1194
1194
0
- OpenGL Hardware Lighting
2182
2219
+2
Let1kWindowsBloom 1.0 (sec)
6
6
0
Xbench 1.2+ - CPU
126
124
-2
- Threading
247
243
-2
- Memory
129
132
+2
- Quartz Graphics
147
149
+1
- OpenGL Graphics
137
136
-0
- User Interface Graphics
236
236
0
- Hard Disk
41
42
+2
"Real World" Tests:
Startup (sec)
44
41
-7
Shutdown (sec)
12
12
0
Launch Classic Mode (sec)
10
11
+10
Dup. 500 MB Folder of Files (sec)
35
34
-3
Scroll 144p Adobe 7.0.7 Reader File (sec)
32
32
0
Export QT FQDV Movie in iMovie HD 6.0.2 (sec)
44
44
0
Quake III Arena 1.32 (frames/sec)
414
417
+1
Halo 1.5.2 (frames/sec)
31
31
0
Call of Duty 1.4 (frames/sec)
85
84
-1
Doom 3 1.3A+ (frames/sec)
52
53
+2
*Average of 3 runs. +Universal binary version.Red is much faster. Blue is much slower.

For the current top-end Power Mac G5, the Quad 2.5 GHz, there are no appreciable differences between Mac OS X 10.4.6 and 10.4.5. We still get terrible QuickTime DV compression performance from iMovie HD 6.0.1. As we noted previously, it takes over twice as long for the new iLife '06 application to export a full quality DV movie from the test clips than does iMovie HD 5.0.2.

Mac OS X 10.4.6, like all the other 10.4.x updates, has mainly security, compatibility and stability improvements and no real speed improvements. The only speed effect seems to be that the performance increase in Doom 3 for at least some Macs with ATI Radeon cards brought by the v1.3A update, as we previously noted, has been eliminated for some reason.

The only problem that we've encountered with Mac OS X 10.4.6 was updating Mac OS X Server v10.4.5 on an external FireWire drive--after the second reboot it froze at the Mac OS X splash screen. After a restart it would mount but would not boot our G4 Cube. We had to reformat the drive after backing up the data and reinstall 10.4 Server. Then, we used the 10.4.6 Server combo updater and it worked fine. Otherwise, we have used Mac OS X 10.4.6 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