Path: utzoo!mnetor!uunet!cbmvax!grr
From: grr@cbmvax.UUCP (George Robbins)
Newsgroups: comp.sys.amiga
Subject: Re: Amiga serial questions
Message-ID: <3025@cbmvax.UUCP>
Date: 27 Dec 87 23:05:14 GMT
References: <8712050831.AA11465@cory.Berkeley.EDU> <5568@oberon.USC.EDU> <1253@sugar.UUCP>
Reply-To: grr@cbmvax.UUCP (George Robbins)
Organization: Commodore Technology, West Chester, PA
Lines: 45

In article <1253@sugar.UUCP> peter@sugar.UUCP (Peter da Silva) writes:
> In article <5568@oberon.USC.EDU>, papa@pollux.usc.edu (Marco Papa) writes:
> > >> will explain "sustained".  8 hours of continuous 19.2K transmission with no
> > >> ...
> > >> Scrolling is mandatory...
> > 
> > Matt responds:
> > >	Don't be silly.  Even Tek terminals have flow-control, and assuming
> > >all 55 Meg will get through without an error is ridiculous.
> 
> True.
> 
> > >The Amiga can
> > >read data at that speed, but not scroll at that speed.
> 
> Sure it can. It'd take some smarts, but it can.
> 
> DTerm already does multicharacter reads from the serial.device. Now all you
> need to do is multiline scrolling. You don't scroll every line, instead if you
> feel yourself getting behind you jump-scroll 3 or 4 lines at a time. It doesn't
> take any more time to scroll 4 lines than to scroll one (a blit is a blit,
> the world around), so if you can scroll at 9600 normally you can jump-scroll
> at 19.2 by skipping every other scroll when the heat's on.
> 
> The rest of the problems (like doing area fills at 19.2) are probably real,
> but this is a paper tiger.

This "can't scroll" is a little silly.  It depends somewhat on whether you
want your emulator to run on a screen or in a window.  If you're in a window,
then scrolling basically requires a fair amount of data movement.  You can
either optimize or cheat to obtain the level of performance you want, although
most of the emulator's I've tried aren't as fast as I'd like.  

If you use a screen, or take over the system, then scrolling is a different
matter entirely.  You can either play games with the display start pointers
or even implement a display list architecture using the copper to reload
the pointer for every line.   This quickly gets you back to the character
rendering time as being the major performance contstraint, and there you
can do the traditional simplifications involving byte-wide aligned characters
to speed things up.

-- 
George Robbins - now working for,	uucp: {uunet|ihnp4|rutgers}!cbmvax!grr
but no way officially representing	arpa: cbmvax!grr@uunet.uu.net
Commodore, Engineering Department	fone: 215-431-9255 (only by moonlite)