[scribus] ?==?utf-8?q? Does scribus use only 1 of 4 cores?

ale rimoldi ale.comp_06 at xox.ch
Wed Jul 25 09:16:39 UTC 2018


hi ralf

> From my recent code browsing I'd say that the main reason for Scribus'
> devastating performance break-down for even moderatly sized story
> texts is the underlying data model ... :-/

well, if you found a way to improve the data model and solve this, i
think that people will be glad to hear your suggestions : - )

from what i know, the issue is rather that it is hard to know what
should be recalculated when a text frame is edited.
in doubt, scribus recalculates (and eventually redraws) everything that
could have changed, instead of trying to figure out what could have
been affected.

i think that contributions improving the reflowing of the text would be
very welcome!
such work is not trivial, but might be doable!

ciao
a.l.e

p.s.: you're not completely wrong in your analysis, the first step in
such an optimization would probably be: putting in the data model on
which page  item is placed... and which part of text belongs to which
frame... but i think that it's not the hardest part...
another welcome improvement would be to make it easy to move the text
and images back and forth from disk to memory... also a related issue...



More information about the scribus mailing list