Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

creoguy

macrumors newbie
Original poster
Apr 26, 2005
3
0
I just got done loading OS 10.4 on my iMac and thought I'd run Cinebench to see how it compared to my results from OS 10.3.9...

OS 10.3.9 running 2gb of ram

Processor: G5
MHz: 1.8 ghz
Number of CPUs: 1
Operating System: 10.3.9

Graphics Card: GeForce FX 5200
Resolution: 1680 x 1050
Color Depth: Millions

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

Rendering (Single CPU): 245 CB-CPU
Rendering (Multiple CPU): --- CB-CPU


Shading (CINEMA 4D): 232 CB-GFX
Shading (OpenGL Software Lighting): 638 CB-GFX
Shading (OpenGL Hardware Lighting): 922 CB-GFX

OpenGL Speedup: 3.97

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


OS 10.4.0 running 2gb of ram

Processor: G5
MHz: 1.8ghz
Number of CPUs: 1
Operating System: 10.4.0

Graphics Card: GeForce FX 5200
Resolution: 1680 x 1050
Color Depth: Millions

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

Rendering (Single CPU): 165 CB-CPU
Rendering (Multiple CPU): --- CB-CPU


Shading (CINEMA 4D): 192 CB-GFX
Shading (OpenGL Software Lighting): 563 CB-GFX
Shading (OpenGL Hardware Lighting): 865 CB-GFX

OpenGL Speedup: 4.51

****************************************************
 
Ok i know very little about Benchmarks..all i do know is that bigger usually means better...Is that the case here...cos then tiger didnt do to great.. :confused: :confused:
 
Did you remember to set the speed to highest and turn of HDD sleep after the upgrade ?
 
creoguy, are you sure your imac wasn't reset to "automatic" in the energy saver preferences pane after you installed tiger? It seems to me that tiger should NOT be worse than panther...
 
I thought those were better to be smaller? Where it says openGL speed up that is a bigger number. So i think he did get better results. As i recall those are times.
 
Both in automatic

Both tests were run with the automatic settings. I wanted to see what the day to day speeds would be because I run mine in automatic. I'll switch them over to highest and see what happens.

Results to come.
 
Updated Results

Here are the updated results with the highest setting and put HD to sleep disabled...

OS 10.3.9

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

Rendering (Single CPU): 244 CB-CPU
Rendering (Multiple CPU): --- CB-CPU


Shading (CINEMA 4D): 231 CB-GFX
Shading (OpenGL Software Lighting): 644 CB-GFX
Shading (OpenGL Hardware Lighting): 932 CB-GFX

OpenGL Speedup: 4.02

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



OS 10.4.0

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

Rendering (Single CPU): 177 CB-CPU
Rendering (Multiple CPU): --- CB-CPU


Shading (CINEMA 4D): 193 CB-GFX
Shading (OpenGL Software Lighting): 526 CB-GFX
Shading (OpenGL Hardware Lighting): 816 CB-GFX

OpenGL Speedup: 4.23

****************************************************
 
I have almost the exact same system without Tiger (as of Tuesday) and am looking forward to speed gains. I do believe the bigger number at the end means better/faster.

I think the above larger numbers are cycles and that more means slower.
 
clearly the "speed up" isn't enough to notice if the guy can't tell which one means "faster."
 
sorryiwasdreami said:
I have almost the exact same system without Tiger (as of Tuesday) and am looking forward to speed gains. I do believe the bigger number at the end means better/faster.

I think the above larger numbers are cycles and that more means slower.


From this results grid, it appears that larger is better, not the other way around. :(

Cinebench Results
 
This just doesn't make any sense. Something must be wrong with cinebench running in tiger. Can you try the same experiment using xbench on the 10.3.9 and the 10.4 and see if there is any similar speed loss?

What a disappointment :(
 
His results were lower on "Highest" setting under tiger than they were on "Automatic". I don't really take these benchmarks seriously.
 
Is it possible Spotlight was still indexing the volumes on Tiger? That would account for the numbers, I think.
 
I agree with daveL. I'm sure that would slow the machine down. You also might want to check what processes are running in the background to see what is running. If it's not system related, kill it.
 
It's clear that with either system, an occasional coffee break will be forced upon the user during complex rendering operations.
 
Wow that Geforce FX 5200 Ultra in the iMac really sucks , those scores are horrible.


My Athlon XP-M @2.3ghz
ATI Radeon 9600 128MB DDR(non pro)
1GB DDR3200 in Dual Channel

Rendering (Single CPU): 263 CB-CPU
Rendering (Multiple CPU): --- CB-CPU

Shading (CINEMA 4D): 300 CB-GFX
Shading (OpenGL Software Lighting): 1321 CB-GFX
Shading (OpenGL Hardware Lighting): 2584 CB-GFX

OpenGL Speedup: 8.61

This was done with me running iTunes , Firefox , Thunderbird , Zone Alarm , AVG-Antivirus , Weatherbug and Konfabulator with 2 widgets on screen.

No trying to shoot down the iMac cuz my CPU is OC'd after all from 1.8ghz-2.3ghz , but that GPU needs to go. Apple should atleast put an intergrated 9600XT.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.