[argyllcms] Re: Black Point Compensation and low dMax papers
- From: Graeme Gill <graeme@xxxxxxxxxxxxx>
- To: argyllcms@xxxxxxxxxxxxx
- Date: Wed, 14 Nov 2018 11:33:24 +1100
When building ICC profiles for uncoated papers and inkjet printers (let’s say
L* value is something like L24) I see almost an over-doing of the Black Point
Compensation. Photoshop values (0-255) that were 0,1, or 2, jump all the way
to 5 or up to
10 when soft-proofing (and printing) with these ICCs. There is a measurable
loss of dMax in the print.
Does anyone know of a way to lower the BPC “amount” using argyll or is this
is just inherent with various application’s use of BPC? I’m on Mac
Sounds like something in the soft proofing algorithm, and/or BPC being double
In the print workflow some form of BPC (either implemented in the CMM [ i.e.
BPC ] or pre-computed in the profile [ i.e. ArgyllCMS Perceptual ]) would be
desirable to avoid loss of shadow detail in the print when images are full
When soft proofed, an absolute soft rendition should have a black no
darker than the print is capable of (i.e. L24), but if BPC or a perceptual
intent is also applied to the soft proof transform, then you may end up with
a black lower than the print. This is sometimes seen as desirable,
given the different viewing conditions on a display, particularly
when seen side by side with images that use the full display black.
A workaround would be to take control of the softproof by doing it
manually as a 2-step. i.e. convert from source to print raster using the
desired print intent, and then use a print to display conversion that treats
the black point how you wish.
i.e. see <http://www.argyllcms.com/doc/Scenarios.html#LP2
but note that to preserve the print black point you would
need to choose a different intent, i.e. try -ir, but worth
experimenting with. (You can of course do similar things
faster using cctiff, but with less fine control.)
Other related posts: