Forums | MacLife

I beat the iMac on all but one test on my Dual 2.7 GHz G5 with a Radeon X800. (I'm assuming the tests were run at the standard 1024x768)

CINEBENCH 9.5
****************************************************

Tester           : akb825

Processor        : Dual 2.7 GHz G5
MHz              : 2.7 GHz
Number of CPUs   : 2
Operating System : Mac OS X 10.4.6

Graphics Card    : Radeon X800
Resolution       : 1024x768
Color Depth      : 32

****************************************************

Rendering (Single   CPU): 409 CB-CPU
Rendering (Multiple CPU): 744 CB-CPU

Multiprocessor Speedup: 1.82

Shading (CINEMA 4D)                : 406 CB-GFX
Shading (OpenGL Software Lighting) : 1273 CB-GFX
Shading (OpenGL Hardware Lighting) : 2795 CB-GFX

OpenGL Speedup: 6.89

****************************************************

BTW, the title should read "Cinebench just as fast on OS X as on WinXP" since Windows is usually faster with graphics. However, OS X just edged out in front, which I'm pleased to see. Hopefully this means that OS X's OpenGL isn't really any slower if you use it properly. (since I know certain things cause it to slow down considerably)

Last edited by akb825 (2006-04-11 1:59 pm)

CINEBENCH 9.5
****************************************************

Tester           : P

Processor        : Power Mac G5
MHz              : 1800MHz Dual CPU
Number of CPUs   : 2
Operating System : Mac OS 10.4.5

Graphics Card    : ATi X800 XT 256MB AGP 8x Pro
Resolution       : <fill this out>
Color Depth      : <fill this out>

****************************************************

Rendering (Single   CPU): 278 CB-CPU
Rendering (Multiple CPU): 457 CB-CPU

Multiprocessor Speedup: 1.64

Shading (CINEMA 4D)                : 277 CB-GFX
Shading (OpenGL Software Lighting) : 816 CB-GFX
Shading (OpenGL Hardware Lighting) : 1642 CB-GFX

OpenGL Speedup: 5.93

****************************************************

After with the latest version


CINEBENCH 9.5
****************************************************

Tester           : P

Processor        : Power Mac G5
MHz              : 1.8GHz  G5 Dual CPU
Number of CPUs   : 2
Operating System : OS X.4.6

Graphics Card    : ATi X800 XT
Resolution       : <fill this out>
Color Depth      : <fill this out>

****************************************************

Rendering (Single   CPU): 272 CB-CPU
Rendering (Multiple CPU): 473 CB-CPU

Multiprocessor Speedup: 1.74

Shading (CINEMA 4D)                : 282 CB-GFX
Shading (OpenGL Software Lighting) : 938 CB-GFX
Shading (OpenGL Hardware Lighting) : 2180 CB-GFX

OpenGL Speedup: 7.74

****************************************************

That's pretty darn good.  I really like to know what the hell they fixed cause that's MUCH better

I did this on my Quad


CINEBENCH 9.5
****************************************************

Tester           : P

Processor        : Power Mac G5
MHz              : 2500MHz Dual CPU
Number of CPUs   : 2
Operating System : Mac OS 10.4.6

Graphics Card    : GeForce 6600
Resolution       : <fill this out>
Color Depth      : <fill this out>

****************************************************

Rendering (Single   CPU): 381 CB-CPU
Rendering (Multiple CPU): 1163 CB-CPU

Multiprocessor Speedup: 3.05

Shading (CINEMA 4D)                : 397 CB-GFX
Shading (OpenGL Software Lighting) : 1366 CB-GFX
Shading (OpenGL Hardware Lighting) : 3164 CB-GFX

OpenGL Speedup: 7.98x

akb825 wrote:

MacBoy4139 wrote:

So the XP drivers are in beta, and on OS X they are good.  Cinebench then gets better scores.

Compare it to a similarly equipped PC maybe?

Actually, the PC would probably win. Not because of the drivers, but because the X1600 in Intel Macs is underclocked.

Then take it out of the equation.  There are tools out there to overclock the thing.  The drivers will definitely have an impact on performance - negative or otherwise.

akb825 wrote:

IIRC, they are based on ATI's other drivers. Methinks they just kept the meat of the drivers and put in the compatibility for their EFI interface.

To be honest, I don't know.  That would make sense though - I mean, why reinvent the wheel?