thanks for your kind answer, it look more or less clear now. I hope progressive reading development goes on and issue can be solved. good work. best regards, francesco
Il giorno sab, 11/11/2006 alle 13.04 +0800, Craig Ringer ha scritto: > francesco fantoni wrote: > > It seems there is a serious memory problem with large images or huge > > number of images in Scribus documents using 1.3. > >>From what i understand reading bug reports it seems it could be related > > to use of Qstring for storing images. > > Partly, but not entirely. That's part of it, as it doubles or quadruples > the in-memory size of the decompressed image. However, Scribus also load > sall images into RAM, filters them into another buffer, and then writes > that out, so you'll end up with more than just the source image in RAM. > > Suddenly 2GB of RAM doesn't look like so much anymore ;-) > > The goal is to progressively read, filter, and write images to the PDF > so that the whole image need not be kept in RAM at once (let alone two > or three decompressed copies of it as now!). And, of course, not to use > QStrings inappropriately. > > > I'm having serious troubles with both 1.3.3.x and 1.3.4cvs, that turn > > out to be practically useless for producing a document with a large > > number of images. Someone stated in a bug-report that use of Qstring for > > images will be abandoned in 1.4, is it true? > > Yes, as above. > > > is there already a clear vision of problem and possible solutions? > > Yes. > > There was some progress with progressive image reading in 1.3.4. > > -- > Craig Ringer > > _______________________________________________ > Scribus mailing list > Scribus at nashi.altmuehlnet.de > http://nashi.altmuehlnet.de/mailman/listinfo/scribus
