Hands-On Report--Xbench 1.3 (Universal)
As far as we knew, Xbench 1.2 had only one glitch--it produced invalid scores for the User Interface test when running on Intel-based Macs. But according to the developer, there are more fixes in v1.3 vs v1.2:
- Corrected a mistake that caused the altivec test to be turned off on PowerPC machines
- Turned off coalesced graphics updates for all platforms on Mac OS 10.4.4 and higher
- Switched compiler to GCC 4.0 on PowerPC. This provides some boost to floating point and AltiVec scores, and these have been recalibrated accordingly. This also raises Xbench's system requirements to 10.3.9 or higher.
- Revised machine database to include the MacBook, Intel iMac and several other models
- Added code to dynamically load machine database on launch from the Xbench website
- Added support for temporarily turning off beam sync on Tiger while running graphics tests
- Fixed an issue that causes Xbench to fail to launch on Leopard
- Built with Xcode 2.4
The list says nothing specific about the User Interface glitch but it is fixed as you will see below.
We ran Xbench 1.2 and 1.3 on our Intel-based 15" 2.16GHz MacBook Pro and our upgraded PowerPC-based 12" 1.4GHz PowerBook G4 to determine where the main differences are. Here are the results:
| Test | Mac OS X = 10.4.7 | ||
|---|---|---|---|
|
Xbench 1.2
|
Xbench 1.3
|
Perf. Chg. %
|
|
| Xbench Test | |||
| - CPU |
80
|
79
|
-1
|
| - Threading |
212
|
205
|
-3
|
| - Memory |
122
|
118
|
-3
|
| - Quartz Graphics |
61
|
103
|
+69
|
| - OpenGL Graphics |
154
|
156
|
+1
|
| - User Interface Graphics |
26
|
218
|
+838
|
| - Hard Disk |
36
|
32
|
-11
|
For our Intel-based Mac, Xbench 1.3 gives much higher scores in Quartz Graphics and in User Interface Graphics. These are two areas where Intel-based Macs appeared to lag PowerPC-based Macs (c.f. our Mac Pro vs Power Mac G5 comparison). The User Interface score of 218 for our MacBook Pro compares well with our former Power Mac G5 Quad at 236, consistent with our experience qualitatively. Xbench 1.3 seems to give more reasonable results comparing across the G5 and Core Duo CPUs.
| Test | Mac OS X = 10.4.7 | ||
|---|---|---|---|
|
Xbench 1.2
|
Xbench 1.3
|
Perf. Chg. %
|
|
| Xbench Test | |||
| - CPU |
56
|
68
|
+21
|
| - Threading |
59
|
60
|
+2
|
| - Memory |
23
|
28
|
+22
|
| - Quartz Graphics |
54
|
57
|
+6
|
| - OpenGL Graphics |
74
|
73
|
-1
|
| - User Interface Graphics |
24
|
29
|
+21
|
| - Hard Disk |
31
|
37
|
+19
|
For the PowerBook G4, a higher CPU score is expected due to a fix in the application's Altivec test. But three other scores were also higher at nearly 20 percent or more.
It is clear that Xbench 1.3 will give better and more comparable scores than Xbench 1.2 as we use it in future speed tests. [Bill Fox & Dana Baggett]
Saw something? Send a tip
The archive ran on reader tips. What did you see, where, and do you want the credit?