Interesting thread.... The first example on page one did not work on Tahrpup.
Best,
Slavvo67
system optimization techniques
- technosaurus
- Posts: 4853
- Joined: Mon 19 May 2008, 01:24
- Location: Blue Springs, MO
- Contact:
You mentioned some tips about portability in thisother thread. There were some good links posted in that thread too. Do you have a summary that could be posted in your subsection at top of this thread?technosaurus wrote:Portability
- potchan
- Posts: 109
- Joined: Sat 05 Apr 2008, 11:46
- Location: Pilots' Height Tel Aviv - Yaffo, Israel ×œ×™× ×•×§×¡_פותחן
- Contact:
System optimization techniques
Hello, I'm Timmo' from http://potchan.org.
My main s(q)fs file is a 3.0G. size CREAMED to 3.8G. to earn about 15 s' in mounting out of total 1.5 min' mounting depends on PC. I find this discussion interesting regarding BIHIHIHIG sqfs needs to be 'educated' in a more manner to ststistically mount the more essentials faster instead of just scanning it. In my humble POV busybox gives a partial-frozen solution to it. Two points to relate to...
1. 's(q)fs' making with new codes: to run the 'naked' basis "/" on desktop partition with some essentials such as network and browser (for instance). The system will acquire it in several plays statistically (best will be thousands times) and then we freeze it into sqfs or sfs which is the basis.
- a possible con of this may be the specific HW coder uses. So we must find the "average politically-correct standard agreed" APCS new negging standard for it, and we as community may compromise a lot doing it right... if "right" is a some absolute value out there...
2. to do the same on "savefile.4fs" or save folder and the user will 'teach' the system according to specific HW and habits on the run, "next time will be faster" in a better way.
Alternatively, we create several of "file.4fs" or save folder on shelf for various users and machines. Not a BarryK vision at all (and not mine at all).
Which one - or which combination - we do better or shorter way ?
My main s(q)fs file is a 3.0G. size CREAMED to 3.8G. to earn about 15 s' in mounting out of total 1.5 min' mounting depends on PC. I find this discussion interesting regarding BIHIHIHIG sqfs needs to be 'educated' in a more manner to ststistically mount the more essentials faster instead of just scanning it. In my humble POV busybox gives a partial-frozen solution to it. Two points to relate to...
1. 's(q)fs' making with new codes: to run the 'naked' basis "/" on desktop partition with some essentials such as network and browser (for instance). The system will acquire it in several plays statistically (best will be thousands times) and then we freeze it into sqfs or sfs which is the basis.
- a possible con of this may be the specific HW coder uses. So we must find the "average politically-correct standard agreed" APCS new negging standard for it, and we as community may compromise a lot doing it right... if "right" is a some absolute value out there...
2. to do the same on "savefile.4fs" or save folder and the user will 'teach' the system according to specific HW and habits on the run, "next time will be faster" in a better way.
Alternatively, we create several of "file.4fs" or save folder on shelf for various users and machines. Not a BarryK vision at all (and not mine at all).
Which one - or which combination - we do better or shorter way ?
- nosystemdthanks
- Posts: 703
- Joined: Thu 03 May 2018, 16:13
- Contact:
Re: System optimization techniques
it isnt like that at all. there is no standard. basically people do whatever they want, and a small portion of that filters up into official this or that.potchan wrote:So we must find the "average politically-correct standard agreed" APCS new negging standard for it, and we as community may compromise a lot doing it right... if "right" is a some absolute value out there...
afterwards, official this or that comes back down and is reused to make about 100 derivatives. maybe its 10. lets say its in that range.
for the most part, this works out. instead of traveling the world to find out what people are running, you can pretty much just download a copy of the setup they like most. its not completely accurate, because some people add features they dont use because "someone else might."
though basically every setup is different, and somewhere between 10 and a 100 users even offer you a copy of theirs for you to try. theres no standard. posix and woof help keep some things familiar. a very practical reinterpretation of your question is to start with your "what do people want" question and simply ask: what would you like to have? then stick that in a distro (you already have-- 12 virtual desktops! fun!) and offer it for people to try. whats missing? what would you like? smaller sfs files? thats doable.
[color=green]The freedom to NOT run the software, to be free to avoid vendor lock-in through appropriate modularization/encapsulation and minimized dependencies; meaning any free software can be replaced with a user’s preferred alternatives.[/color]