MacBook Pro Disk Performance Problem

[jalb@jalb.net]

MacBook Pro Disk Performance Problem

I have had my new Mac Book Pro 2.16 for a week now and I have had problem playing movie files on the machine, it skips. I have decided to run Xbench. Compared to the ArsTechnica results it seems that I have a major problem with my harddrive:

Results 21.31
System Info
Xbench Version 1.2
System Version 10.4.6 (8I1121)
Physical RAM 2048 MB
Model MacBookPro1,1
Drive Type ST9100824AS
CPU Test 64.93
GCD Loop 202.44 10.67 Mops/sec
Floating Point Basic 71.90 1.71 Gflop/sec
vecLib FFT 37.94 1.25 Gflop/sec
Floating Point Library 60.98 10.62 Mops/sec
Thread Test 199.02
Computation 178.20 3.61 Mops/sec, 4 threads
Lock Contention 225.34 9.69 Mlocks/sec, 4 threads
Memory Test 109.55
System 102.26
Allocate 113.56 417.03 Kalloc/sec
Fill 102.72 4994.64 MB/sec
Copy 92.61 1912.92 MB/sec
Stream 117.96
Copy 113.79 2350.37 MB/sec
Scale 114.46 2364.74 MB/sec
Add 122.17 2602.50 MB/sec
Triad 121.96 2609.11 MB/sec
Quartz Graphics Test 53.75
Line 54.63 3.64 Klines/sec [50% alpha]
Rectangle 51.85 15.48 Krects/sec [50% alpha]
Circle 51.27 4.18 Kcircles/sec [50% alpha]
Bezier 77.80 1.96 Kbeziers/sec [50% alpha]
Text 43.34 2.71 Kchars/sec
OpenGL Graphics Test 122.16
Spinning Squares 122.16 154.97 frames/sec
User Interface Test 31.79
Elements 31.79 145.88 refresh/sec
Disk Test 4.16
Sequential 2.54
Uncached Write 26.69 16.39 MB/sec [4K blocks]
Uncached Write 19.72 11.16 MB/sec [256K blocks]
Uncached Read 0.71 0.21 MB/sec [4K blocks]
Uncached Read 12.61 6.34 MB/sec [256K blocks]
Random 11.39
Uncached Write 4.47 0.47 MB/sec [4K blocks]
Uncached Write 22.53 7.21 MB/sec [256K blocks]
Uncached Read 18.25 0.13 MB/sec [4K blocks]
Uncached Read 35.20 6.53 MB/sec [256K blocks]



How is it possible that my harddrive have such poor reading performance ?

Thank You

 

[Sun Baked]

You are not the only one, barefeats test of the 5400RPM Seagate showed poor performance at 29MB/Sec.

The drive should be running at 42MB/sec -- slightly below the 7200 Seagate, which tested fine in the barefeats tests.

Something is happening with these 5400RMP drives to make them goofy, since they are not supposed to perform this badly.

 

[DMPDX]

There have been major problems with Xbench and intel macs. You should wait until the bugs get sorted out before conducting an Xbench test again to get concrete results.
-dsm

 

[CanadaRAM]

Your first mistake is trusting that XBench is a valid test.

I can get 20% differences on XBench 10 minutes apart, on the same drive, same machine.

XBench IMO is fatally flawed and not to be trusted, even for relative, as opposed to absolute, measurements.