Re: Is your ConTeXt too slow?

[Jan Hlavacek <jhlavacek@sf.edu>]

On Fri, Nov 08, 2002 at 11:03:48AM +0100, Jens-Uwe Morawski wrote:
> On Fri, 08 Nov 2002 09:30:56 +0100
> Hans Hagen <pragma@wxs.nl> wrote:
> 
> > At 09:36 PM 11/7/2002 +0100, Jens-Uwe Morawski wrote:
> > 
> > >Hi all (or mainly the users of Unix-like systems)
> > >
> > >after i've bought a new notebook and installed the recent
> > >Redhat Linux system i found that an updated ConTeXt runs
> > >very slow. First i thought that Hans had too much time
> > >and implemented a few thousand new features that need
> > >their time, but ...
> > >
> > >... it is a queerer story. Here are the run times of
> > >an example document (many inline MP graphics, 18 pages,
> > >many terminal messages !!) on different terminal programs
> > >for X-Windows:
> > >
> > >gnome-terminal:         2 min 18 sec
> > >konsole (KDE-terminal): 1 min 30 sec
> > >xterm:                  1 min  5 sec
> > >directly on the console
> > >         (no X-Windows):       58 sec
> > >
> > >Normally i use the first. Hmm... ;-(
> > >
> > >Don't ask me for reasons, but if your ConTeXt is too
> > >slow, ask yourself if you run it on the right terminal -)
> > 
> > Can you check if kpsewhich runs ok? For instance
> > 
> >    kpsewhich colo-ini.tex
> > 
> >    kpsewhich cmr10.tfm
> > 
> > maybe no ls-R file is found and consulted, resulting in a full scan of the 
> > texmf file structure.
> 
> No, no. TeX finds its files very fast. I could improve that with a smaller
> texmf-tree, but 1+ min for the job is IMO okay; it runs 3xConTeXt,
> 2xTeXUtil and 2xMetaPost. The problem is the terminal program. It seems
> that the drawing of those many messages on the terminal needs much time.
> If i use a slow terminal, i.e. gnome-terminal, then the runtime increases.
> 
> Can i switch off the messages generated by ConTeXt, this would help to
> compare.


Did you try to run context with output redirected to a file?  How long
does that take? 

try texexec yourfile.tex > logfile

then look at the file called logfile to see the messages.

-- 
Jan Hlavacek                                            (219) 434-7566
Department of Mathematics                             Jhlavacek@sf.edu
University of Saint Francis               http://www.sf.edu/jhlavacek/