Strange, my iMac seems to have scored lower with twice the amount of RAM?


System Info

Model Name: iMac
Model Identifier: iMac7,1
Processor Name: Intel Core 2 Duo
Processor Speed: 2.8 GHz
Number Of Processors: 1
Total Number Of Cores: 2
L2 Cache: 4 MB
Memory: 4 GB
Bus Speed: 800 MHz
Boot ROM Version: IM71.007A.B03
SMC Version: 1.21f4


Results

150.91 (previous 157.26)

Xbench Version 1.3
10.5.3

CPU Test 174.74 (176.48 )

GCD Loop 329.54—> 17.37 Mops/sec
Floating Point Basic 159.24 3.78 Gflop/sec
vecLib FFT 127.16 4.19 Gflop/sec
Floating Point Library 175.05 30.48 Mops/sec
AltiVec Basic n/a
Thread Test 224.83
Computation 223.91 4.54 Mops/sec, 4 threads
Lock Contention 225.75 9.71 Mlocks/sec, 4 threads


Memory Test 175.22 (172.01)

System 194.31
Allocate 310.15—> 1.14 Malloc/sec
Fill 160.49 7803.29 MB/sec
Copy 167.11 3451.49 MB/sec
Stream 159.55
Copy 148.02 3057.21 MB/sec
Scale 153.78 3177.13 MB/sec
Add 168.07 3580.22 MB/sec
Triad 170.58 3649.07 MB/sec

Quartz Graphics Test 227.75 (229.59)

Line 207.54 13.82 Klines/sec [50% alpha]
Rectangle 273.01 81.51 Krects/sec [50% alpha]
Circle 219.54 17.90 Kcircles/sec [50% alpha]
Bezier 217.96 5.50 Kbeziers/sec [50% alpha]
Text 230.98 14.45 Kchars/sec
OpenGL Graphics Test 202.84 (219.57)
Spinning Squares 1 frames/sec
User Interface Test 338.53—> (394.33)
Elements 338.53—> 1.55 Krefresh/sec (394.33)


Disk Test 54.84 (56.51)

Sequential 73.40 (78.21)
Uncached Write 78.62 48.27 MB/sec [4K blocks]
Uncached Write 123.58 69.92 MB/sec [256K blocks]
Uncached Read 37.83 11.07 MB/sec [4K blocks]
Uncached Read 137.84 69.28 MB/sec [256K blocks]

Random 43.78 (44.23)
Uncached Write 15.70 1.66 MB/sec [4K blocks]
Uncached Write 104.83 33.56 MB/sec [256K blocks]
Uncached Read 91.63 0.65 MB/sec [4K blocks]
Uncached Read 138.37 25.67 MB/sec [256K blocks]


EDIT: To clean up extraneous info (Previous scores in parenth.)

Added info:
Isn't "User Interface test" part of the Graphics test?
What they use to calculate it beats me.

GeekBench also showed a slight drop.
http://macosg.com/group/viewtopic.php?t=11543&start=15

You cannot expect to see consistent benchmark numbers until you positively eliminate all of the variables. Any processes that compete for system resources with Xbench will cause lower scores. It would be more accurate to average the scores of a sequence of Xbench runs, ensuring identical conditions for each test, BUT...

This is a synthetic benchmark that provides nothing beyond a relative comparison of the hardware of different machines. It has nothing to do with application performance, which is the gold standard of computing comparisons.

Chris,

Have you posted your results to the XBench site?

If so, under what name?

It is very convenient to be able to compare machines using the XBench site. You can compare to my new one and my original CD 2.0 if you do a search for "Macrick"

Your numbers appear to be line with mine. My iMac is a little quicker where it should be (memory and CPU).

The last test I did was fresh off of a re-boot with nothing else running. I also did a test using the "Guest" account that Leopard has but got dismal results. I have to go back and try to figure that out.

I realize scores may vary - this was the ave out of three tests right off of restart and the score's not really that much lower than previous ones.

I wasn't able to submit the previous ones because of connection problems, however, I did save a copy. http://macosg.com/group/viewtopic.php?t ... c&start=15 Albeit, posted on the wrong thread. Anyway, CPU, Memory and Graphics were way faster than the benchmark model and well within range of other similar macs.

This one is under "Chris's iMac" http://db.xbench.com/merge.xhtml?doc1=296521

All seems in order, except for the "User Interface Test".

I have yet to figure out what that test encompasses.

Isn't "User Interface test" part of the Graphics test?
What they use to calculate it beats me.