|
From: Petr M. <mi...@ph...> - 2007-12-17 13:56:54
|
I wonder that for "set hidden3d", the order of drawn surfaces vs lines is the same and not opposite for: splot ... with pm3d, ... with line splot ... with line, ... with pm3d In both cases, the pm3d Test: set term x11 1 splot (x+y)*0.1 with pm3d, \ x*x-y*y with lines lt 1 lc rgb "#000000" set term x11 2 set hidden3d replot set term x11 3 splot x*x-y*y with lines lt 1 lc rgb "#000000", \ (x+y)*0.1 with pm3d set term x11 4 splot (x+y)*0.1 with pm3d, \ x*x-y*y with lines lt 1 lc rgb "#000000" set term x11 10 splot (x+y)*0.1, \ x*x-y*y with lines lt 1 lc rgb "#000000" PS: David, you would get a "more expected" plot when "set hidden3d" is omitted. |
|
From: Ethan M. <merritt@u.washington.edu> - 2007-12-17 19:03:18
|
On Monday 17 December 2007 05:56, Petr Mikulik wrote:
> I wonder that for "set hidden3d", the order of drawn surfaces vs lines is
> the same and not opposite for:
> splot ... with pm3d, ... with line
> splot ... with line, ... with pm3d
You may recall that there was an extended discussion + disagreement about
whether 'set hidden3d' should affect pm3d plots. Part of this discussion
was captured in the comments attached to patchset #1077726.
As to your current set of examples..
The effect of "set hidden3d" is to sort all plot elements together into
one big list so that they can be tested for overlap on Z. Therefore it
makes no differenet what order the plots are listed on the command line.
The list of elements is then rendered first, before anything else on the
command line that is not part of hidden3d processing (pm3d surfaces,
contours, images).
If you want to provide an option for the hidden3d plot elements to
be drawn last, rather than first, one way to do this would be to add
{front|back} as an option to "set hidden3d". What do you think?
--
Ethan A Merritt
|
|
From: Ethan M. <merritt@u.washington.edu> - 2007-12-17 19:24:16
|
On Monday 17 December 2007 11:03, Ethan Merritt wrote:
> On Monday 17 December 2007 05:56, Petr Mikulik wrote:
> > I wonder that for "set hidden3d", the order of drawn surfaces vs lines is
> > the same and not opposite for:
> > splot ... with pm3d, ... with line
> > splot ... with line, ... with pm3d
>
> If you want to provide an option for the hidden3d plot elements to
> be drawn last, rather than first, one way to do this would be to add
> {front|back} as an option to "set hidden3d". What do you think?
It turns out quite nicely if the call to plot3d_hidden() is moved after
the normal loop over plot elements. It allows you to add a dummy
(invisible) hidden3d surface that is coincident with the pm3d surface:
set hidden3d
splot foo with pm3d, foo with lines lt -3, otherstuff with lines
The net result is a plot in which the lines are correctly occluded
by the pm3d surface, and vice versa.
--
Ethan A Merritt
|
|
From: Petr M. <mi...@ph...> - 2007-12-18 07:56:49
|
> > > I wonder that for "set hidden3d", the order of drawn surfaces vs lines is
> > > the same and not opposite for:
> > > splot ... with pm3d, ... with line
> > > splot ... with line, ... with pm3d
> >
> > If you want to provide an option for the hidden3d plot elements to
> > be drawn last, rather than first, one way to do this would be to add
> > {front|back} as an option to "set hidden3d". What do you think?
Good idea. Can you please provide a patch?
> set hidden3d
> splot foo with pm3d, foo with lines lt -3, otherstuff with lines
>
> The net result is a plot in which the lines are correctly occluded
> by the pm3d surface, and vice versa.
Should the "front" be the default option then?
---
PM
|
|
From: Ethan A M. <merritt@u.washington.edu> - 2007-12-20 10:07:12
|
On Monday 17 December 2007 23:56, Petr Mikulik wrote:
> > > > I wonder that for "set hidden3d", the order of drawn surfaces vs lines is
> > > > the same and not opposite for:
> > > > splot ... with pm3d, ... with line
> > > > splot ... with line, ... with pm3d
> > >
> > > If you want to provide an option for the hidden3d plot elements to
> > > be drawn last, rather than first, one way to do this would be to add
> > > {front|back} as an option to "set hidden3d". What do you think?
>
> Good idea. Can you please provide a patch?
> Should the "front" be the default option then?
Preliminary patch on SourceForge (#1854733)
https://sourceforge.net/tracker/index.php?func=detail&aid=1854733&group_id=2055&atid=302055
--
Ethan A Merritt
|
|
From: David B. <Dav...@mo...> - 2007-12-17 15:08:55
|
Petr Mikulik wrote: > I wonder that for "set hidden3d", the order of drawn surfaces vs lines is > the same and not opposite for: > splot ... with pm3d, ... with line > splot ... with line, ... with pm3d > In both cases, the pm3d > > Test: > > set term x11 1 > splot (x+y)*0.1 with pm3d, \ > x*x-y*y with lines lt 1 lc rgb "#000000" > > set term x11 2 > set hidden3d > replot > > set term x11 3 > splot x*x-y*y with lines lt 1 lc rgb "#000000", \ > (x+y)*0.1 with pm3d > > set term x11 4 > splot (x+y)*0.1 with pm3d, \ > x*x-y*y with lines lt 1 lc rgb "#000000" > > set term x11 10 > splot (x+y)*0.1, \ > x*x-y*y with lines lt 1 lc rgb "#000000" > > > PS: David, you would get a "more expected" plot when "set > hidden3d" is omitted. > > Ok, then I think I have a patch.. Thanks D. -- David Bateman Dav...@mo... Motorola Labs - Paris +33 1 69 35 48 04 (Ph) Parc Les Algorithmes, Commune de St Aubin +33 6 72 01 06 33 (Mob) 91193 Gif-Sur-Yvette FRANCE +33 1 69 35 77 01 (Fax) The information contained in this communication has been classified as: [x] General Business Information [ ] Motorola Internal Use Only [ ] Motorola Confidential Proprietary |