My first claim (geForce 6200 versus Radeon 9200) was based on looking up scores in futuremarks compare section.peharri said:You have a link to anything to back that up? cube was responding to the Macintouch benchmarks which most certainly do not test the Core Duo Mini. There's also little reason to believe the Core Solo's hardware acceleration are remotely different from the Core Duo, unless far more is being done in software, even when, supposedly, using hardware acceleration, than should be being done.
I hope you're wrong and the Core Duo and Core Solo have equivalent "hardware" OpenGL speeds, because if you're not, it's worse than we thought.
You are right about your hardware reasoning, but it is really an artificial problem you are creating. You are saying in fact that the great OpenGL performance of the Core Duo *CPU* is irrelevant because it is not a GPU. Who cares which unit does the processing as long as it is fast.
And let me clarify these results for you. The hardware lighting test requires a graphics card that support this feature. The fact is that the GMA950 does not have hardware acceleration for this, *but* the drivers for the card make any program assume that it does. Any call for a hardware lighting instruction gets intercepted by the driver and it instead applies a highly optimized software routine. Now for the Core Solo this pretend game actually yielded lower performace, but for the Core Duo the drivers were apparrently so optimized they could take full advantage of the second core. The core is so powerfull that it actually performs better than a dedicated graphics chip; even better than the x600 on a G5 imac. So wouldn't you be glad if you can actually use both chips of your core duo to such an extent that you are better off than with a dedicated card?
edit:
RECTIFICATION: My explanation is pointless, because I misread "iMac Core Duo" as "Mini Core Duo". Seems we have to wait untill someone else posts cinebench core duo results.
edit 2:
OK Cinebench results were actually already posted. They just barely beat the G4 Mini in the hardware test. Forget what I said about being better than a G5 iMac with an x600. Summary:
Software OpenGL
730 // iMac G5 iSight
981 // iMac Core Duo
427 // PowerPC Mini
869 // Mini Solo
1075 // Mini Duo
Hardware OpenGL
1075 // iMac G5 iSight
1687 // iMac Core Duo
530 // PowerPC Mini
438 // Mini Solo
545 // Mini Duo
Well at least the Core Duo Mini's software rendering is twice as fast as the Mini G4's hardware rendering