Probably a good start. I can't tell you how much time I spent and how much
media I used to maximize my old printer and RIP, but I did have results
that other people were more than happy to see their images printed by me.
I've lost a lot of that knowledge in the past decade+ and certainly not in
a position to pick things back to that level.
On Fri, Dec 20, 2019, 20:24 <graxx@xxxxxxxxxxxx> wrote:
I’m experimenting with the -s and the -g targen parameters.
This is my current targen command line:
targen -v -d3 -f1000 -s20 -g25 RGB1000
The -s parameter is for each channel “extra” points. I’ve used -s20.
The -g parameter sprinkle the target with extra “neutral” points.
I’ve specified -g25. The extra points do appear along the “gray axis” as
can be seen in the 3D projection (I use BabelColor PatchTool on the ti1
file…).
But there is nothing that says that the presence of those “extra” points
is going to result in a profile having more “neutrality” than one
calculated without... I guess, it’s up to me to experiment this. Argyll
profiler probably does not need those extra points to “predict” neutrals…
I experimented with the -V parameter, using the maximum 4.0 value with
this targen command line:
targen -v -d3 -f1000 -V4 RGB1000
I would have expected the additional patches to show up (cluster) around
the “black point”, in 3D, but I am not seeing this. I probably need to
increase the total number of patches?
/ Roger
*From:* argyllcms-bounce@xxxxxxxxxxxxx <argyllcms-bounce@xxxxxxxxxxxxx> *On
Behalf Of *Greg E
*Sent:* December 20, 2019 7:55 PM
*To:* argyllcms@xxxxxxxxxxxxx
*Subject:* [argyllcms] Re: Custom "non-Argyll" charts
How much does paper and ink cost? I'd try both ways and see how each
works. But I'd also want to use a good spectro to measure everything. i1 or
Pulse would be my minimum. Wish my old Spectroscan still worked. You may
also want to use a black backing behind your paper, but I'd try measuring
with both black and white backing.
On Fri, Dec 20, 2019, 19:47 <graxx@xxxxxxxxxxxx> wrote:
So, Greg, not a good idea?
/ Roger
*From:* argyllcms-bounce@xxxxxxxxxxxxx <argyllcms-bounce@xxxxxxxxxxxxx> *On
Behalf Of *Greg E
*Sent:* December 20, 2019 7:43 PM
*To:* argyllcms@xxxxxxxxxxxxx
*Subject:* [argyllcms] Re: Custom "non-Argyll" charts
Talk about this topic from 15 years ago was to create a large number of
non-neutral patches near gray and a decent number of steps between black
and white. Throwing tons of "neutral" patches at it doesn't allow the
routines to calculate the fine changes that may be required to shift to
neutral. If I was printing black and white prints with a color inkset I
would give this a try. Even with multi black printers this helps unless you
are using the monotone presets for your printer.
On Fri, Dec 20, 2019, 13:43 <graxx@xxxxxxxxxxxx> wrote:
Suppose I want to « hand-optimize” charts, for some reason. Suppose I want
to throw in ‘extra’ color patches, here and there, like in the shadows,
will this “crash” Argyll in any way? Suppose I start with some fresh ti1
file, that I edit to my heart’s content, leaving the header “alone”,
except, the number of “sets”, such as, instead of, say, the base number of
sets of, say, 100, I type in, NUMBER_OF_SET 135, to reflect the 35 ‘extra’
patches I wanted to add in :
DESCRIPTOR "Argyll Calibration Target chart information 1"
ORIGINATOR "Argyll targen"
CREATED "Tue Dec 17 14:40:00 2019"
APPROX_WHITE_POINT "95.106486 100.000000 108.844025"
COLOR_REP "RGB"
WHITE_COLOR_PATCHES "4"
BLACK_COLOR_PATCHES "4"
COMP_GREY_STEPS "10"
OFPS_PATCHES "84"
NUMBER_OF_FIELDS 7
BEGIN_DATA_FORMAT
SAMPLE_ID RGB_R RGB_G RGB_B XYZ_X XYZ_Y XYZ_Z
END_DATA_FORMAT
NUMBER_OF_SETS 135
Would the edited ti1 file still be “honored” by printarg? And readchart?
And would I still be able to build a profile out this unique SET
combination?
Would I then have to supply the “predicted” XYZs for my ‘extra’ patches?
Or would printarg and readchart be able to “ignore” the missing
information? Or be able to “fill in the blanks” on their own?
I know, it may sound crazy but I’m experimenting with “custom charts
definitions” and I’m still not 100% clear on all targen options. I’m
probably reinventing the wheel, here? But just curious, in general, what’s
the underlying Argyll’s profiling capabilities?
With ProfileMakerPro, for example, it’s possible to submit pretty much any
kinds of charts for which we have device and ‘CIE’ values and still end up
with an ICC profile.
Best / Roger
