[argyllcms] Re: relativ volume of gamut -- vrml comparison?

  • From: Gerhard Fuernkranz <nospam456@xxxxxx>
  • To: argyllcms@xxxxxxxxxxxxx
  • Date: Fri, 04 Jan 2008 14:47:25 +0100

Klaus Karcher wrote:
> Klaus Karcher wrote:
>> Do you think it's worth the effort to add a volume intersection
>> report to viewgam? or the gamut volume to tiffgamut -v? (the latter
>> should be no problem, provided that iccgamut and tiffgamut use the
>> same gamut object)
>
> I just aded 3 lines to tifgamut:
>
> 557 if (verb) {
> 558   printf("Total volume of gamut is %f cubic colorspace
>     units\n",gam->volume(gam));
> 559 }
>
> and "tiffgamut_vol -v -ir ... timage_t_200_sRGB_inters_iC2_r.tif"
> returned: "Total volume of gamut is 398107.430337 cubic colorspace units"
>
> there's indeed a considerable difference (>5%) to qhull's result:
>
>> qhull s FA <timage_t_200_sRGB_inters_iC2_r.points [...]  Total
>> volume:       421367.74
>
> so much for "impressive precission" ;-)

My understanding is that the volume of a triangulated convex hull can be
computed analytically and exactly (i.e. limited only by machine
precision). But gamut hulls are not necessarily convex...

Regards,
Gerhard


Other related posts: