Re: [Gimp-user] Path Tool Problems (fwd)

[John Meyer <johnmeyer pueblocomputing com>]

Slightly OT, but that's why I'm glad Thunderbird has the reply list button.


On 10/4/2013 4:44 PM, Simon Budig wrote:
> Oops, sorry - forgot to have the list in cc as well.
>
> ----- Forwarded message from Simon Budig <simon budig de> -----
>
> Subject: Re: [Gimp-user] Path Tool Problems
> Date: Fri, 4 Oct 2013 19:38:07 +0200
> From: Simon Budig <simon budig de>
> To: Akovia <akovia1 eml cc>
>
> Akovia (akovia1 eml cc) wrote:
> > If I hover my mouse above a certain segments holding Control, some segments
> > will allow to add a node, while other segments won't. This also effects
> > deleting line segments.
> Weird.
>
> > Now that I think about it I have the last instance on video from a tutorial
> > I made. It's kinda perfect since I have it showing my key-presses as well.
> > https://dl.dropboxusercontent.com/u/93550827/Gimp%20Tutorial%20Series/he-is-my-master_full.mkv
> > 6:35 to around 8:30 is the problem. I also realized that this also was a
> > copied and pasted path.
> Really weird. I cannot reproduce this here though. *shrug*
>
> Where did you copy/paste the path from? Inkscape?
>
> > > > 3. Path strokes.
> > > >
> > > > Quite often on big complicated paths  that I try to stroke, there will
> > > > be a few sharp angles that won't protrude out to a point and will just
> > > > look like it was cut off.
> > > Do you stroke with a tool? Or by "stroke line"? with which options? For
> > > the Mitre-Join the Mitre-Limit determines the amount of "allowed
> > > protrusion".
> > >
> > > Can you provide screenshots of the problematic results?
> > Stroke line mostly unless I'm after some sort of effect. Playing with the
> > mitre setting does have some effect on the problem. Maybe I just can't get
> > the expected results I want in having an even stroke all the way around.
> > Have a look at this and maybe you can explain if there is another way. For
> > now I have just been patching some paths together to get what I need..
> > https://dl.dropboxusercontent.com/u/93550827/temp/path-stroke_mitre.mkv
> Ok, I think what you're seeing there is expected behaviour. The problem
> is, that when you move two tangents closely together (i.e. you have a very
> sharp corner) the end point of the stroke moves towards infinity, i.e.
> you need to cap this at some point. The parameter controlling this is
> the mitre limit. We use cairo for rendering and the behaviour is
> described as follows in the cairo documentation:
>
>      If the current line join style is set to CAIRO_LINE_JOIN_MITER (see
>      cairo_set_line_join()), the miter limit is used to determine whether
>      the lines should be joined with a bevel instead of a miter. Cairo
>      divides the length of the miter by the line width. If the result is
>      greater than the miter limit, the style is converted to a bevel.
>
>      [...]
>
>      The default miter limit value is 10.0, which will convert joins with
>      interior angles less than 11 degrees to bevels instead of miters.
>      For reference, a miter limit of 2.0 makes the miter cutoff at 60
>      degrees, and a miter limit of 1.414 makes the cutoff at 90 degrees.
>
>      A miter limit for a desired angle can be computed as: miter limit =
>      1/sin(angle/2)
>
> There is no way to completely disable the miter limit, but maybe the
> range of the limit in the UI is not sufficient.
>
> I hope this helps,
>          Simon