Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!utgpu!water!watnot!watmath!clyde!rutgers!mit-eddie!mit-amt!mit-atrp!ralph
From: ralph@mit-atrp.UUCP
Newsgroups: comp.sys.amiga
Subject: faster icons
Message-ID: <953@mit-amt.MEDIA.MIT.EDU>
Date: Fri, 10-Apr-87 15:39:25 EST
Article-I.D.: mit-amt.953
Posted: Fri Apr 10 15:39:25 1987
Date-Received: Sat, 11-Apr-87 18:08:49 EST
Sender: usenet@mit-amt.MEDIA.MIT.EDU
Reply-To: ralph@mit-atrp.UUCP (Amiga-Man)
Distribution: na
Organization: MIT Amiga Users Group
Lines: 35


Oops, in a previous posting (which I fear didn't get out on the net
since I got only one response to it) I mentioned the idea of rewriting
or replacing the "icon.library". I think it's actually called the
"info.library". This is the set of routines which let the amiga diddle
with the icons. Well, the crux of the idea is to have a single icon
file in each directory, and by rewriting the "info.library" have a
transparent use of that scheme instead of the old single .info file
for each file. This would speed things up quite nicely.

Questions:
 - this is written in assembly language, No ?
 - is this really hard, or just difficult ?
 - is anyone trying such an idea out ?
 - could it be made it work with both old and new schemes simultaneously,
   searching first for the grouped icons, and if not found resorting
   to the scattered .infos ? Then upon exit, grouping them automatically.
   A user would never have to worry, and his whole system would just
   keep speeding up (WOW!).

I have a buddy who wrote his own 3-D library to complement the amiga's own
2-D library. It works as smoothly as the 2-D does. He did it in assembly
language and it runs like a charm, autoloading just like it's supposed to
and going away when not used (and RAM is needed).

One person mentioned to me that it would then be a pain to manipulate the
individual icons, being forced to use a "tool" to work on them.
True. But the pain for the rare event of working on an icon is outweighed
by the speed of opening drawers which I spend a lot of time waiting for
and do much more often than making icons. We need to optimize the part of
the system that is used the most ! I can only hope that Commodore is
already (perhaps secretly) working on such stuff. Because I use the
icon environment 95% of the time, and I need that improvement.

Discussion is solicited....I may work on this as soon as I graduate (yea!).