Hi Peter, Due to machine troubles I have not been able to get to the list for about a week.
regarding Freehand drawing you said:
Thanks for the info. How recent is recent? In particular, before or after 2004-11-22 05:15:51 UTC, when I committed a partial regression in the sophistication of the fitter that happens to improve things for our current pencil tool.
It was before 22/11, possibly around build 041118. However, I have just brought down build 0411221024 and have compared it with SP0.34. Firstly, Inkscape Fd is smoother now than it has been but it still falls far short of 0.34, at least on my system. I did a small comparison side-by-side comparison drawing a shape first on Inkscape and then on Sodi Podi. Then another shape etc. and noted nodes.
Shape Nodes (Inkscape) Nodes (SodiPodi) triangle 18 8 figure S 19 8 circle 23 6 square 19 8 random1 32 10 5 link chain 42 16
Not strictly scientific but there is some consistency there. I think some people have said they think SP is sluggish. Yes it is slower but certainly not sluggish (again on my machine). However, SP has a very nice feel to it and as a user I like it. In everything else I reckon Inkscape is streets ahead but in smoothness........
Again this may be a personal thing but when I am drawing maps I nearly always find it simpler to start from a simpler outline and add nodes where necessary for fine tuning. Removing nodes nearly always means having to do more fiddling with the ones that are left. The Inkscape method certainly seems to create more nodes than SP. Yes I know I can 'simplify' but I don't necessarily want to do that.
Anyway as Bulia said:
but overall worse than sodipodi 0.34.
No. As I have demonstrated, it's better for some uses but worse for others.
If worse equals less smooth then Yes. It is.
It's just different, let's not try to compare them in terms of better/worse.
I agree not better/worse but maybe let's not forget there is very real difference there which may have a significant impact on people coming to Inkscape for the first time.
Peter said:
My current intent is to leave freehand tool as it is for 0.40 (except for trying to address a rare assertion failure), and do more experimentation after 0.40 is released.
Sounds great to me and I look forward to the results.
vellum