Re: make the cvs build zips smaller?

2006-05-04 Thread Linus Nielsen Feltzing
Jonathan Gordon wrote: can he lng's be moved to a seperate part of the dowload page? (have a drop list with the available languages, and another for the targets, then a button and it will upload the correct file to u..?) That would make it a pain for the user, having to download both the

Re: make the cvs build zips smaller?

2006-05-03 Thread Daniel Stenberg
On Wed, 3 May 2006, Jonathan Gordon wrote: ive setup my computer with linux and moved it downstairs so it can stay on 24/7 so youve got another build server if u want it (celly 2.4ghz, 512mb ram), the only downside is that my uploads are capped to about 20KB/s so uploading the 2mb zip

Re: make the cvs build zips smaller?

2006-05-03 Thread Manuel Dejonghe
On 5/3/06, Daniel Stenberg [EMAIL PROTECTED] wrote: But I don't think having the cvs builds do light packages is a solution to anything as there will still be people who want to download a full version and then we need to offer such ones. Doing full/light packages will help the users who

Re: make the cvs build zips smaller?

2006-05-03 Thread Bluechip
I can highly recommend 7-zip to make things smaller still. It worked wonders for The DevKit ...try it, I think you'll be pleasantly surprised :) And to boot it's all gpl open-source etc. :)) ive setup my computer with linux and moved it downstairs so it can stay on 24/7 so youve got another

Re: make the cvs build zips smaller?

2006-05-03 Thread Manuel Dejonghe
On 5/3/06, Daniel Stenberg [EMAIL PROTECTED] wrote: On Wed, 3 May 2006, Bluechip wrote: I can highly recommend 7-zip to make things smaller still. It worked wonders for The DevKit ...try it, I think you'll be pleasantly surprised :) And to boot it's all gpl open-source etc. :)) ... and we've

Re: make the cvs build zips smaller?

2006-05-03 Thread Bluechip
At 14:39 03/05/2006, you wrote: On 5/3/06, Daniel Stenberg [EMAIL PROTECTED] wrote: On Wed, 3 May 2006, Bluechip wrote: I can highly recommend 7-zip to make things smaller still. It worked wonders for The DevKit ...try it, I think you'll be pleasantly surprised :) And to boot it's all gpl

Re: make the cvs build zips smaller?

2006-05-03 Thread Erik Slagter
On Wed, 2006-05-03 at 17:14 +0200, Manuel Dejonghe wrote: On 5/3/06, Bluechip [EMAIL PROTECTED] wrote: But I found that even with a self-extractor stuck on, the .7z.exe file was still smaller than a gzip file [whether created by 7zip or gzip] ...I would suggest this be checked for a Rockbox

Re: make the cvs build zips smaller?

2006-05-03 Thread Dominik Riebeling
When talking about a platform specific solution, how about a web-aware windows *updater*? I'm thinking of a small wizard that - can be placed on the filesystem of the dap itself - can look up its settings from a config file on the player (say, /.rockbox/wizard.cfg) which stores player type,

Re: make the cvs build zips smaller?

2006-05-03 Thread Bluechip
At 16:14 03/05/2006, you wrote: On 5/3/06, Bluechip [EMAIL PROTECTED] wrote: But I found that even with a self-extractor stuck on, the .7z.exe file was still smaller than a gzip file [whether created by 7zip or gzip] ...I would suggest this be checked for a Rockbox distro as it is (by nature)

Re: make the cvs build zips smaller?

2006-05-03 Thread bk
Compression algorithms like 7zip, RAR, bzip2, etc tend not to be wins (except for very large files) since they use *much* more CPU time for relatively minor size savings. If you're concerned about server load, 7zip is a bad idea and best case scenario is that you'll only see 15-30% filesize

Re: make the cvs build zips smaller?

2006-05-03 Thread Manuel Dejonghe
On 5/4/06, bk [EMAIL PROTECTED] wrote: Compression algorithms like 7zip, RAR, bzip2, etc tend not to be wins (except for very large files) since they use *much* more CPU time for relatively minor size savings. If you're concerned about server load, 7zip is a bad idea and best case scenario is

Re: make the cvs build zips smaller?

2006-05-03 Thread Daniel Stenberg
On Thu, 4 May 2006, Manuel Dejonghe wrote: 7zip is a bad idea and best case scenario is that you'll only see 15-30% filesize savings. I don't see the point, as the original concern was cpu power bandwidth Yeps. In my quick and naive zip/7zip compression tests I did a number of months ago,