🚀 go-pugleaf

In article <C0CIut.2vw@news.cso.uiuc.edu> wmagro@uiuc.edu (William Magro) writes:
>By the way, the slowness of Pagestream redraws is due to poor algorithms  
>and/or coding, not the speed of the ST.  DTP programs for the mac redraw  
>quite satisfactorily under Spectre GCR on a stock ST.  Take a look at  
>Calamus if you'd like to see the result of good redraw algorithms.  If  
>pagestream is depressing you, don't upgrade your machine, pressure  
>SoftLogik to upgrade the program to make it usable.
>
>--
>William Magro                                       NeXTMail welcome
>wmagro@uiuc.edu

I'm not sure if you were just talking about screen redraws or printing
as well. Calamus is tons faster at printing than Pagestream, even to the
same printer. What the problem turns out to be is that Pagestream's
printer driver calls TOS to print, i.e. it calls Bconout to print each
and every character to the printer. That's fine, except that as every
Quick ST / Warp 9 user knows, the Bconout and other trap handlers in TOS
are extremely slow. While Pagestream is printing it spends about 90% of
its time in the Bconout handler! Someone should call the Codeheads to
add some code to Warp 9 to speed up the handling of Bconout to the printer
(since it already speeds up Bconout to the screen). I would guess that a
similar hack to Warp 9 could significantly speed up Pagestream printing,
limited only by the baud rate of the parallel port.

Calamus's printer driver on the other hand already writes directly to the
hardware so I suspect there's little that can be done to speed up its
already fast printing.

- Darek (the PC bigot)

-- 
/--------------------------------------------------------------------------\
| Darek Mihocka, BraSoft, 14150 NE 20th St., Unit 302, Bellevue, WA 98007. |
| Creator of Gemulator, Quick ST, and Xformer. Views expressed are my own. |
\--------------------------------------------------------------------------/