Re: OpenGL-based framebuffer concepts

[Pavel Machek <pavel@ucw.cz>]

Hi!

> >>I just implemented text mode switch and tileblit ops into viafb
> >>(http://davesdomain.org.uk/viafb/index.php) and it is about four
> >>times faster than accelerated graphics mode and about eight times
> >>faster than unaccelerated graphics mode (both measured using cat
> >>largefile with ypan disabled). So textmode is meaningful
> >>alternative.
> >
> >I mean.... it is displaying text faster than refresh rate... so who
> >cares?
> >
> >You can only *display* so much text a second (and then, user is only
> >able to see *much* less text) and both text mode and frame buffers are
> >way past that limits. so.... who cares?
> 
> there are quite a few times when you have text output that you need to 
> scroll through, but you really don't need to read it as it goes by.
> 
> for example, accidently cating a large file, running a program with overly 
> verbose debugging output, etc.
> 
> yes, if you never make mistakes and know that these are problem cases 
> ahead of time you can redirect the output. but in the real world sysadmins 
> really do notice when they are on a console that is slower.
> 
> if reading speed was the limiting factor very few people would need 
> anything faster then a 9600 baud terminal.

I'm not talking about reading speed, I'm talking about displaying
speed. Once you display more than refresh rate times screen
size... you may as well cheat -- xterm-like. If xterm detects too much
stuff is being displayed, it simply stops displaying it, only
refreshing screen few times a second...
									Pavel
-- 
(english) http://www.livejournal.com/~pavelmachek
(cesky, pictures) http://atrey.karlin.mff.cuni.cz/~pavel/picture/horses/blog.html