Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.
That's not a problem, I will get to it this weekend. Am I assuming correctly, that if I were to run the Trim script, and then test with Xbench, and then run an edited version of the Undo script with just the line to move back the VSP framework, then test again, that would achieve what we need?

With regards to the fonts, I might do some testing to see just how much of an impact they make. I mostly use IceWesael PPC and Leopard Webkit on 10.5 so I don't use vanilla TFF if at all these days - anything I do use has FoxPEP installed though, naturally!

I'll have a play around this weekend and get back to you with some results!
 
If you could post both results here or relay them back, then that is correct. Alternatively, you could simply copy over the removed VSP.framework back to its original system location, using the script directories as a guide. You must then manually repair the disk permissions if you do this, however.

Excellent, that would also be a big help and would put to rest whether the fonts truly are worth bothering with in the first place.

Vanilla TFF has actually gotten slightly faster in recent releases than what it used to be - perhaps you might be inclined to give FPR25 a try, if only briefly?

Either way, delighted to hear how foxPEP is serving you as designed. :)
 
So sorry for not getting back to you sooner. Life with a toddler and returning towork more often, I'm sure we all know how that goes!

I'm not sure how helpful these results will be to you, really, I ran the script as-is and ran Xbench, and then I I restored just VSP frameworks, and then ran Xbench once again. As you can see, the results aren't probably what you were to expect, but I have included some cropped screens of the "altivec" labelled parts of the benchmark results - not sure if this is any more or less helpful, but I thought I'd include to. Also, happy to send you the Xbench files as well.

In terms of non-scientific "it feels like X" commentary, I didn't notice much of a difference, either way, to be honest. I would say, however, that running the Accelerator script was one of the biggest boosts to my iBook's performance. And I also have SPotlight fully disabled, Dashboard disabled, and I have used Monolingual to remove all kinds of unneccessary bunf, too. So, I'm not too sure how much the trimming actually did for a late G4 system like mine.

I am using TFF, but I have yet to upgrade from FPR24, I was going to wait for FPR26, but I may as well as update, I'm just being lazy in that regard. But FPR24 feels the best it ever has done along with FoxPEP 2.0 running.

Feel free to ask me to run anything else, and I will do my best to help, but I am new to doing much more than the odd bit of Terminal tinkering here and there to OS X this old.
 

Attachments

  • Just VSP Restored 1.png
    Just VSP Restored 1.png
    14.7 KB · Views: 226
  • No VSP Restored.png
    No VSP Restored.png
    13.9 KB · Views: 199
  • Both.png
    Both.png
    302.4 KB · Views: 176
No worries. I can imagine how things get hectic pretty fast and believe me, I can relate!

Hmm... Would you mind running the Undo script, and then get a benchmark of the system after the trimming has been undone? I'm terribly sorry, but I forgot to tell you to get a baseline of the system first, before any changes have been made. This would paint a little clearer of a picture of the situation in its entirety, especially when compared with the results here.

In any case, I feel like I will simply need to cut back a lot of what the script originally set out to do for the sake of AltiVec performance and system stability. However, the heavy hitters will be kept, as long as they aren't the ones at fault.

Thank you very much for the help, it's all greatly appreciated. :)
 
Hi - yes no problem. In the UK we have a saying "there's always one..." and this time - it's me! I can't believe I didn't even think to take a baseline benchmark, that is some basic ass s**t, sincere apologies.

I will re-run the undo script wholesale - returning it to "pre-trimcelator" state, I would note that in every instance the accelerator enabling Beamsync will have been run - on both my G4 machines this has had the biggest impact in my experience.

I will right by basic rookie error and come back to you!
 
When you can (and to everyone else listening), could you please scan through Extensions under Software in System Profiler on every (non-Trimcelerated) PowerPC machine on 10.4.11 you have and look for an AppleUSBIrDA if present? Tell me if it's there.

Also, if anyone has an iMac G3 or maybe CRT Studio Display also on 10.4.11 and wants to help out, please run the Accelerator there and detail what happens.

Thank you.
 
When you can (and to everyone else listening), could you please scan through Extensions under Software in System Profiler on every (non-Trimcelerated) PowerPC machine on 10.4.11 you have and look for an AppleUSBIrDA if present? Tell me if it's there.

Just had a look at my 667 MHz TiBook w/10.4.11, ne'er Trimcelerated (nice word btw). It's not there.

FWIW, it's also not present on a 1.42 GHz iBook running 10.5.8. It is present on a 867 MHz TiBook running 10.2.8. I don't know if that info does you any good, but there it is. Gave me an excuse to play around with my machines.:)
 
  • Like
Reactions: z970
And we're all done. :)

AquaTrimcelerator has been completely redesigned to be 100% safe, 88% more effective, and 49% more comprehensive.

AquaTrim and AquaAccelerator have been combined into one script (with all the fat removed), plus an additional script is on offer to disable Spotlight (this made a big difference in both operation and browsing on my MDD), another script is included to optimize the system network transfer rates for higher speeds (thanks @eyoungren), Monolingual and ShadowKiller have been bundled as recommended follow-up utilities, and a new (to most of you) and improved AquaTips has been ported in from AquaSuite to provide pointers for where the user can further improve performance manually after all scripts have been run.

Everyone give it another try. All the kinks, bugs, and wrinkles have been ironed out, the difference this makes is pronounced, and it has every reason to be another mainstay for Tiger just as foxPEP is.
 
Last edited:
What news! Some fun for me to play around with over the weekend! I'll be giving this a try over the weekend and I'm sure you've worked hard on this for us all.

Much appreciated!
 
Gave a minor update to the manual font blacklist in AquaTips to be more comprehensive and to actually reflect what is displayed in Font Book.

And with that, this project has most likely reached its apex. Thank you for your participation thus far.
 
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.