[scribus] Scribus Limits, Present and Future

[John Morris]

John Beardmore wrote
>John Morris wrote:
>
>>1. It sounds to me from a.l.e's and John Beardmore's comments that 
>>I should not expect Scribus to smoothly handle any single file of 
>>more than about 50 or 60 pages, regardless of the linkages between 
>>text boxes. Is that correct?
>
>Actually I didn't even notice Scribus slowing down (apart from load 
>time) for 50 to 60 page documents, but my runs of link text frames 
>never run to more than a few pages, and I generally have a lot of 
>graphs and photographs.

   Thanks for the feedback. This is clearly an avenue I'll have to 
explore when I have the time. This does remind me that I forgot to 
ask about the nature of the limitation. This seems to be saying that 
the program is expected to slow down but continue to do the work as 
the document gets larger. Do the limitations manifest themselves in 
other ways, such as crashing, simply refusing to accept more pages or 
links, or something else?

>I'm left with the impression that at least on a 2.4 Ghz machine with 
>4 gig of RAM, it could do bigger docs.  It was noticeably poorer in 
>its performance on a 1.8 GHz box with 2 gig of RAM, but that was 
>back in the days of 1.3.3.9 so things may have got better, and I've 
>seen it work  on P/III boxes at 800HMz too, though I've not played 
>much in that environment.

   This is good to hear. Most of the time, I'll be running Scribus on 
a 2.6GHz machine with 2GB of RAM, but my backup machine (which I'm 
using right now) is 1.86GHz with 2GB of RAM. I had noticed that 
Scribus did not seem particularly response, but I had hoped to 
discover the cause and remedy it for real production work.

>Editing within Scribus has been clunky but usable since before I 
>started using it, but has improved steadily.  Groups of linked text 
>frames can be a bit grumpy to edit, but I've never had significant 
>problems.

   This is good to know. I don't generally expect to do much editing 
in my layout application as I find that easier to do in a program 
focused on that type of work. However, there's always a few 
corrections to make just before the final output.

>It would be interesting to try importing some really long passages 
>into Scribus linked text frames, to try and tease out if the thing 
>that slows it down is number of pages, density of characters, or 
>number of linked frames etc.

   This is certainly something I will try when I have the time. I will 
report my results to this group if I have not seen others cover the 
subject sooner.

Best,
John