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

  • From: Klaus Karcher <lists@xxxxxxxxxxxxxxxxx>
  • To: argyllcms@xxxxxxxxxxxxx
  • Date: Sat, 05 Jan 2008 16:34:33 +0100

Preben Soeberg wrote:
So next I tried to use the precision flag -p in cctiff, and this gave
huge changes in the calculated gamut volumes, but only when the camera
profile was involved.

Strange. Did you compare the tiffs visually? what says cctiff -k? can you send me the camera profile?

I found that the volume differences are negligible (<0.02%) with ISOcoated_v2:

timage -t -r300 -x t300.tif
cctiff -p -ir sRGB.icc -ir ic2.icc t300.tif t300_p_int.tif
cctiff -ir sRGB.icc -ir ic2.icc t300.tif t300_fast_int.tif

cctiff -k -ir sRGB.icc -ir ic2.icc t300.tif t300_p_test.tif
Worst error = 3445 bits, average error = 41.128641 bits
Worst error = 5.256733%, average error = 0.062758%

tiffgamut_vol -v -ir ic2.icc t300_p_int.tif | grep volume
Total volume of gamut is 397418.506559 cubic colorspace units

tiffgamut_vol -v -ir ic2.icc t300_fast_int.tif | grep volume
Total volume of gamut is 397358.358854 cubic colorspace units

echo "scale=10; 397418.506559/397358.358854*100-100" | bc -l
.0151368900

Klaus



Other related posts: