I just spent the morning trying to make a profile strictly using
ArgyllCMS tools, usually I use DisplayCal. I went up to 2000 patches at
high quality, hours and hours. I use both an I1 Display Pro and an I1
Studio.
I didn't do these with the intent of getting the best "check" results.
My goal was to match as best as I can a paper I use for printing and now
at least my screen is as dark as my prints and it should be easy now to
make edits to bring both to what I want.
The other point I was hoping to solve is though my black and white image
show equal channel values in PS the actual tone I was seeing add some
color in it. Now it's better but still not perfect.
I tried a few option, I don't know if something is wrong or not but it
seems Lightroom doesn't like CLut profiles.
The first few where made using 2000 patches
colprof.exe -v -A"BENQ" -M"SW240" -D"-v -aX -qh -cpp -dmt" -qh -cpp -dmt
"SW240_2020_FEB_120cdm2_D5700"
Profile check complete, peak err = 0.978411, avg err = 0.260891, RMS =
0.294842
colprof.exe -v -A"BENQ" -M"SW240" -D"-v -aX -qh -cpp -dmt" -qh -cpp -dmt
"SW240_2020_FEB_120cdm2_D5700
Profile check complete, peak err = 1.196742, avg err = 0.275726, RMS =
0.326740
Profile check complete, peak err = 1.819702, avg err = 0.506932, RMS =
0.567717
500 patches:
targen -v -d3 -G -g50 -f500 SW240_2020_FEB_2PM
dispread -v -y8 -kSW240_2020_FEB_2PM.cal SW240_2020_FEB_2PM
colprof.exe -v -A"BENQ" -M"SW240" -D"SW240_2020_FEB_120cdm2_D5700" -as
-qm -cpp -dmt "SW240_2020_FEB_2PM"
Profile check complete, peak err = 1.696714, avg err = 0.455160, RMS =
0.526696
I wouldn't put to much faith on these numbers as I don't recall the
actual commands I use for targen and dispread except for the last one.
Regards,
Yves
On 2/3/2020 3:57 PM, Eric Brown wrote:
For those of us who can keep a monitor on all night, is there any benefit to doing a huge number of samples (thousands) when we're sleeping? I assume that the return on the investment isn't great, but if the investment is free (the cost of electricity), is there any downside?
-Eric
On Mon, Feb 3, 2020 at 2:14 PM <graxx@xxxxxxxxxxxx <mailto:graxx@xxxxxxxxxxxx>> wrote:
Decided to “Go for the gold!” and got 836 data points for my lowly
NEC PA271W monitor.
Here are the important statistics:
> peak err = 1.146645, avg err = 0.277301, RMS = 0.314956
For 300 data points, the statistics were:
> peak err = 0.888202, avg err = 0.244000, RMS = 0.285845
As Gerard pointed out, the statistics may have gone up but the
“splines” have more “real data” to “chew on” instead of leaving
whole swaf of color space uncharacterized.
What is the meaning of “Overdetermination” in the context of
fitting splines?
/ Roger
*From:* argyllcms-bounce@xxxxxxxxxxxxx
<mailto:argyllcms-bounce@xxxxxxxxxxxxx>
<argyllcms-bounce@xxxxxxxxxxxxx
<mailto:argyllcms-bounce@xxxxxxxxxxxxx>> *On Behalf Of *Gerhard
Fuernkranz
*Sent:* Sunday, February 2, 2020 6:03 PM
*To:* argyllcms@xxxxxxxxxxxxx <mailto:argyllcms@xxxxxxxxxxxxx>
*Subject:* [argyllcms] Re: colprof
Am 02.02.20 um 23:22 schrieb graxx@xxxxxxxxxxxx
<mailto:graxx@xxxxxxxxxxxx>:
Gerard,
The outFile300 -> XYZ Lux + Matrix profile ‘stats’ are as
follows :
> peak err = 0.888202, avg err = 0.244000, RMS = 0.285845
This is way better than the ‘-as’ Matrix+shaper profile stats:
> peak err = 2.220396, avg err = 0.642870, RMS = 0.747308
CLUT models can adapt even more "flexiby" to the data than
matrix/shaper profiles, having a typical number of 50...200
effective parameters, depending on the chosen amount of smoothing
(just a rough number from my experience). So 300 points may still
be too low in order to achive a significant overdetermination.
One comment…
A quick look at the TRC tags reveals an important difference
in the -ag compared to the -as encoding.
The -ag encoding starts at 0 where the -as encoding does not.
That’s an important difference.
A power function y=x^gamma always passes through the point (0,0).
The display's black level may not be zero, though. A potential
idea to might be to fit a gamma+offset model, and record the
resulting TRC as 1D LUT (but I think this would need to be
implemented first, and I can't say in advance whether it would
provide the expected value). Generally it is always a good idea to
use the "stiffest" model (with the fewest number of parameters)
which can describe the (noise-free) device characteristics
accurately enough.
Regards,
Gerhard
