On Sat, Jun 15, 2013 at 12:12 PM, Dmitri A. Sergatskov <address@hidden> wrote:
I wonder If the problem is due to bad OpenGL driver...
That's indeed something I have been wondering. For the record the way printing from FLTK/OpenGL works is as follows:
1) the OpenGL feedback buffer is turned on
2) octave makes the plot by sending all the required OpenGL commands
3) GL2PS retrieves the feedback buffer from OpenGL; this buffer contains the list of primitives representing the plot; those primitives are expressed screen coordinates (basically the buffer contains the 2D projection of the 3D scene)
4) GL2PS translates the primitives into equivalent PostScript commands
The point is that the content of the buffer in 3) is driver dependent. If you have spurious lines in the final postscript file, it can mean that:
1) GL2PS is buggy and misinterpret the content of the feedback buffer
2) the driver is buggy and put unwanted primitives in the feedback buffer