[argyllcms] Re: 2D view of icc's boundary is not smooth
- From: Graeme Gill <graeme@xxxxxxxxxxxxx>
- To: argyllcms@xxxxxxxxxxxxx
- Date: Mon, 11 Apr 2011 13:39:34 +1000
tmwu wrote:
I used a colormunki to make a profile for my epson 3800,
I then use the gamut viewer of PM 5.05 to view the profile.
In the 2D(ab) view, there are some “strange behavior” from L=25 to L=47.
As you can see from the attached sample1.png (a snapshot of L=39),
sample1
a portion of the boundary is not smooth.
And I wonder if this is normal or not?
Hi,
There's a trade-off between smoothness and degree of fit
to the sample points. You're using a relatively small number
of test points, and with the particular printer behaviour
there is some bumpiness near the black, due in part to
the nature of the black separation, and RGB to CMYK mapping
that the driver is using. Apart from the odd plot, are there
any actual practical problems though ?
If there are indeed some practical problems, then I'd suggest
using more points (say 1000 - 2000), and also try
increasing the colprof -r value to something like 1, 2 or even 4.
Graeme Gill.
Other related posts:
