- This topic has 21 replies, 2 voices, and was last updated 2 years, 4 months ago by
.
-
Posts
-
Hello Vincent. When I checked the CCSS, I found that some CCSS can accommodate two sets of spectral data. What are the reasons for doing this? Is there any benefit? How many sets of such data can the CCSS file contain at most?
Argyll (and Xrite counterparts) will generate a custom matrix correction for i1d3 that minimizes squared error of n display samples in EDR/CCSS. ArgyllCMS code, i1d3.c (i1d3_comp_calmat), will show you more detail
Calibrite Display Pro HL on Amazon
Disclosure: As an Amazon Associate I earn from qualifying purchases.If you had a better minimization algorithm for n display samples you can place it there and recompile/test it.
I don’t quite understand, is it possible to test several monitors of the same model, and then write all their data in to increase fault tolerance? Or can the different color gamut modes of the monitor generate spectral corrections separately to improve fault tolerance?
Look at that ArgyllCMS function, it’s all there. Resulting n display sample correction is an “average” for all of them.
Whitepoint of them (relative heights of each channel) is not relevant, only “shape”. Corrections for n display shapes are “averaged”. If shapes (“shapes”, not “sizes”) are close, like in your example, individual displays correction will be very very close to “averaged”.Easy to test. Split that 2 display sample in 2 CCSS (need to modify overaal samle number and each row ID). Measure WP with each one, calculate difference between 2 and averaged.
Thank you for your answer. Compared with CCMX, CCSS has more information, so it has better versatility. Because CCSS can know the shape of the spectrum, CCMX can only know 7 points. As long as the data is different, ccmx cannot be used, because the same paragraph The spectra of the monitors are all similar, so a set of CCSS spectrum data is enough, right?
CCSS has half of the required information. The other half is stored on colorimeter firmware and you “trust” it to be accurate (sensivity curves). This is the reason they can be ported between different i1d3 and same display model or family (the other half information is provied by each colorimeter) . It’s all on argyllCMS code.
If you have a JETI or an EX1 you can avoid “trusting” i1d3 firmware and make a CCMX. It would make i1d3 match better the readings of your reference device… but it cannot be ported to other i1d3 => If you have a JETI use CCMX for you (it will be more accurate), but save a CCSS for others.
OTOH on a lesser device like an xrite spectro you may choose to trust i1d3 firmware if you suspect that even a 3nm reading is not a good match of afctual spectral power distribution (WLED PFS, myabe some widegamut CCFL). On less extreme spectral power distributions like common WLED (srGB only) CCSS and CCMX should bet a close match on average.
Hello Vincent, I tried to use the PR spectrometer to import the information to make a CCSS file, but it didn’t work. Can you help me see what went wrong?
Plotting SPD in Excel shows all is wrong. All are white samples of several WLED samples
You must generate 4 readings:
WHITE 255 255 255
RED 255 0 0
GREEN 0 255 0
BLUE 0 0 255
=> FOR EACH DISPLAYThen pack those 4 spectral samples data in a CCSS
