[OpenSCAD] Jayesh: I goofed, was Re: Does PythonOCC work better than CGAL in Cadmium...?

Dan Zuras 3D threedee at nonabelian.com
Sun Jun 12 06:35:16 CEST 2011

> Date: Sat, 11 Jun 2011 22:22:01 -0500
> Subject: Re: [OpenSCAD] Does PythonOCC work better than CGAL in Cadmium...?
> From: Bryan Bishop <kanzure at gmail.com>
> To: Dan Zuras 3D <threedee at nonabelian.com>, openscad at rocklinux.org, 
> 	Bryan Bishop <kanzure at gmail.com>
> On Sat, Jun 11, 2011 at 6:27 PM, Dan Zuras 3D <threedee at nonabelian.com>wrote:
> > If you like.
> > Here is something simple.
> > It should be fairly easy to translate.
> >
> Have you considered just using pythonOCC directly? What gives? I see you
> posting about stuff that may or may not be resource intensive.. so going
> more direct would seem to be the solution, right?
> - Bryan
> http://heybryan.org/
> 1 512 203 0507

	You are quite correct.  And if I think about doing it again
	I will certainly consider that.  But, as I said, the source
	I gave Jayesh was one I worked on about a year ago when I
	was learning OpenSCAD.  I have yet to learn pythonOCC.  If
	Jayesh's experiment turns out well then I guess I will find
	myself motivated to learn it.  Do you think I should anyway?  - Dan

	P.S. - Jayesh: I made a mistake in the times I reported.
	In the year since I wrote the dragon curve code I have gotten
	into the habit of using f5 (with OpenCSG) rather than f6 (with
	CGAL).  It is with f5 that it takes 12 to 14 seconds & 1.3GB
	of memory to render dragon8192 (with 24576 elements).  With f6,
	it takes about the same amount of memory but I shot it after
	10 minutes.  So maybe not that much has changed in the last
	year after all.

More information about the OpenSCAD mailing list