- Dec 29, 2005
- 6,271
- 2,238
- 136
I recently put a dent in my bank account and bought a macbook with a 2.4 GHz penryn Core 2 duo. It only has 3 MB level 2 cache vs 6 MB in the desktop penryn core 2 duos. Anyway, I decided to see how it compares to my 2.4 GHz Q6600 on a per core basis on the boinc poem@home project. Both computers have the same RAM speed (667MHz). The macbook is running macos leopard and the Q6600 is running Windows Vista Home. I also gave it a quick test on FAH SMP.
Boinc cpu benchmarks: (I know this is not terribly predictive)
macbook:
GenuineIntel
Intel(R) Core(TM)2 Duo CPU T8300 @ 2.40GHz [x86 Family 6 Model 23 Stepping 6]
Measured floating point speed 2689.34 million ops/sec
Measured integer speed 6272.78 million ops/sec
Q6600@2.4GHz:
GenuineIntel
Intel(R) Core(TM)2 Quad CPU @ 2.40GHz [x86 Family 6 Model 15 Stepping 7]
Measured floating point speed 2329.11 million ops/sec
Measured integer speed 5130.88 million ops/sec
Poem@home WUs:
macbook:
cpu time (sec)__claimed credit___granted credit
1,116.12_________5.79___________9.45
1,071.84_________5.56___________9.45
1,141.00_________5.92___________9.45
Q6600:
cpu time (sec)__claimed credit___granted credit
1,383.28_________5.97___________9.45
1,370.30_________5.92___________9.45
1,310.89_________5.66___________9.45
The caveat here is the 2 different OSes but the mobile penryn does seem to do quite well.
I also tested it on FAH SMP:
2605 WU (1760 points): 17 min step time --> 1490 ppd
The mobile penryn has a smaller L2 cache (3 MB vs 4MB). I guess I'm a bit disappointed in the 1490 ppd but not sure since I've never folded on a core2duo.
the Q6600 does 2200 ppd on 1xSMP and 2800 ppd on 2xsmp with affinity changer.
Cheers!:beer:
Boinc cpu benchmarks: (I know this is not terribly predictive)
macbook:
GenuineIntel
Intel(R) Core(TM)2 Duo CPU T8300 @ 2.40GHz [x86 Family 6 Model 23 Stepping 6]
Measured floating point speed 2689.34 million ops/sec
Measured integer speed 6272.78 million ops/sec
Q6600@2.4GHz:
GenuineIntel
Intel(R) Core(TM)2 Quad CPU @ 2.40GHz [x86 Family 6 Model 15 Stepping 7]
Measured floating point speed 2329.11 million ops/sec
Measured integer speed 5130.88 million ops/sec
Poem@home WUs:
macbook:
cpu time (sec)__claimed credit___granted credit
1,116.12_________5.79___________9.45
1,071.84_________5.56___________9.45
1,141.00_________5.92___________9.45
Q6600:
cpu time (sec)__claimed credit___granted credit
1,383.28_________5.97___________9.45
1,370.30_________5.92___________9.45
1,310.89_________5.66___________9.45
The caveat here is the 2 different OSes but the mobile penryn does seem to do quite well.
I also tested it on FAH SMP:
2605 WU (1760 points): 17 min step time --> 1490 ppd
The mobile penryn has a smaller L2 cache (3 MB vs 4MB). I guess I'm a bit disappointed in the 1490 ppd but not sure since I've never folded on a core2duo.
the Q6600 does 2200 ppd on 1xSMP and 2800 ppd on 2xsmp with affinity changer.
Cheers!:beer: