Jump to content
8 posts in this topic

Recommended Posts

OK I have been doing a lot of install with 10.5.2. Probably 30+

 

Tried 10.5.1 Iatkos updated with kalyways 10.5.2 update

 

Tried Leo4all 10.5.2 - no x3100 graphics

 

Jas 10.5.2-- to confusing as nothing is labeled

 

& finally kalyway's 10.5.2 install

 

Now i did not test all of the above with benchmarks as some did not yield well to testing (jas and leo4all). The two I did test were Iatkos before 10.5.2 update and after and kalyway 10.5.2 install. In every case 10.5.1 outperforms 10.5.2 every time.

 

10.5.1 score 2354 with netkas 9.2.0 SS kernel

 

10.5.2 score 1694 with netkas 9.1.0 SS kernel

 

Why is this? Well i thought it was SMBios as this seems to be the biggest file tied to performance, that and the kernel. I have tried a lot of SMBios files from netkas, vanilla and mac.nub and all report my bus speed wrong and they dont seem to affect geekbench scores much.

 

Anyone have a clue why I'm seeing such a huge hit in performance in 10.5.2? And is it possible that geekbench is not correct?

 

Does SMBios really hold the cards to performance or is it all kernel? What if I went back to netkas 10.5.1 SS kernel?

 

Anyone else seeing this kinda hit in 10.5.2?

Link to comment
https://www.insanelymac.com/forum/topic/99559-1051-almost-2x-faster-then-1052/
Share on other sites

I don't see why is geekbench score so important. The real question would be how your computer reacts to real time tasks. If it shows that it's slower with 10.5.2 then you have a problem. Tests can lie :)

 

Well most of the time I would say who cares about the scores but after seeing my laptop go from really fast to just sorta fast I am concerned.

 

I just proved there is a problem definatetively. I went into vista home premium and got a geekbench score of 2134.

 

So that makes 10.5.2 the slowest of them all. Geekbench scores are

 

Windows Vista HP Score 2134

10.5.1 score 2354 with netkas 9.2.0 SS kernel

 

10.5.2 score 1694 with netkas 9.1.0 SS kernel

 

I also went online and found a laptop with lower specs then mine a C2D 1.66ghz, GMA950 graphics and 2gb of ram and its beats my 1694 with a score of 1780.

 

SO there is definately a hit with 10.5.2. WHY though. Whats causing it and can I do anything about this hit?

 

Hope someone has a clue and can shed some light on this.

 

THANKS

And i'll shed be the one to shed my own light I guess.

 

2 problems here:

 

1. Booting with slower FSB. Fixed with -g option at boot up. Huge speed increase.

 

2. Not having the right SATA controller driver loaded in setup. Still working on this. Trying like my 50th install of kalyway 10.5.2

Actually, I get better Xbench scores after 10.5.2 update (iATKOS install), a difference of almost 15, but then the system itself acts very slow everywhere. from the feel of the system and working with various apps, its clear it is slower than 10.5.1. Not to mention about long boot times (40sec - 1.5 min) when they were usually 20 sec with 10.5.1. Same goes with my PowerBook G4. It had Tiger intially. With Leopard install, it became very peppier (better than Tiger) and putting 10.5.2 made that machine a snail. It is bloody so slow that I have stopped working on it.

Actually, I get better Xbench scores after 10.5.2 update (iATKOS install), a difference of almost 15, but then the system itself acts very slow everywhere. from the feel of the system and working with various apps, its clear it is slower than 10.5.1. Not to mention about long boot times (40sec - 1.5 min) when they were usually 20 sec with 10.5.1. Same goes with my PowerBook G4. It had Tiger intially. With Leopard install, it became very peppier (better than Tiger) and putting 10.5.2 made that machine a snail. It is bloody so slow that I have stopped working on it.

 

 

Yeah all my issues with 10.5.2 are fixed now and its running on par if not better then 10.5.1.

 

I FSB issues is now fixed with hardcoded SMBIOS and the SATA controller issue is fixed with Dune's ICHx kext update.

 

When I first installed 10.5.2 I totally noticed the difference and thought that 10.5.1 was better. Now that I got this tweaked though this is better.

 

I get better geekbench scores in 10.5.2 then I do in 10.5.1 but its almost a dead heat.

This is more an observation and a suggestion. Why do you keep installing different releases over and over again? It doesn't really matter what release you use, what matters is how you tweak the install. You can make any release perform well by replacing what needs to be replaced to work best with your hardware. Doing constant reinstalls is counter productive and unnecessary.

This is more an observation and a suggestion. Why do you keep installing different releases over and over again? It doesn't really matter what release you use, what matters is how you tweak the install. You can make any release perform well by replacing what needs to be replaced to work best with your hardware. Doing constant reinstalls is counter productive and unnecessary.

 

Well I figured that out now. I was adding my own kexts to the installs and testing a lot of the options on the disk to see what they all did. Trying to find all that stuff in pacifist can be tricky. Plus that was a bit of an exaggeration anyway LOL.

×
×
  • Create New...