Converting Spyderprint 225 patch target to i1Profiler

[Ink stained Fingers]

The printer just takes RGB data, not gray or ligh cyan or....the split of the RGB data into the additional ink colors gray - violett, light cyan,...... happens internally in the driver, and the user does not have any control over it. Gray is a color like any other for the printer. You would need to go for a RIP which would let you use more ink colors in the profile generation process, I think i1Profiler would support that. The user cannot even control the amount of black, you would need software like a RIP which bypasses the driver internal color translations, Printfab would let you do that with black - UCR - Undercolor removal.

I did another test with the 720 or 702 patchsheets as reported above - running the test on the Netbit paper - on a Epson T3100X 24 inch printer which runs with 4 colors CMYK and with pigment inks .

I'm getting a total average of DeltaE's of just 1,33 over the 720 patches, and the value for the largest outlier is just 4.16. None of the gray patches are in the upper 10% of delta values. This means that both profiles are practically identical.

 

[crenedecotret]

@crenedecotret : I have made an i1Profiler target that is based on the high quality target + grays of Spyderprint, that originally was based on RGB: 5x5x5, but I have changed it to a better RGB: 7x7x7 and including an bunch of near neutral patches (automatically generated by the i1Profiler target) and combining it with the extra patches from the Spyderprint 225 high quality target (minus the RGB 5x5x5 patches) + the neutralising gray patches + 2 extra patches (orange and violet) to fill up the target to 702 patches witch is divided into 26 rows, each including 27 patches (26x27=702). You might to try to convert it to the ArgyllCMS format with this: much appreciated.

My apologies for the late reply. I was out for a vacation and it got a little hectic at work when I got back

Here is the ArgyllCMS version of the chart. Letter sized. It's "as is". I didn't add any additional patches to this one so it will not completely fill the 2nd page.

 

[pharmacist]

Many thanks @crenedecotret . I did my part of the job and manipulated the target to fit exactly on a single sheet of A4 and optimized for the i1Pro2 scanning table.

 

[buxiaozizai.com]

Another AI creation that will be useful for users of ArgyllCMS. This python script will take a txt file with R G B values and output to a ti1 file that include the necessary data (patch number, RGB on a 0-100 scale and XYZ values). The TI1 file can then be used with printtarg to print for whatever device you are measuring for.

The input format is 1 patch per line, spaces between R G B values, R G B values 0-255

With this, we should be able to reproduce practically any popular profiling chart to argyllcms and see what kind of results we get. As soon as i'm done with my "datacolor replicas", I might go for the old TC9.18 since those RGB values are known.

Thanks for your share.
I found a small error in this script.The argyllCMS .ti1 file XYZ scale is 0~100,but the script generated XYZ scale is 0~1；This may affect chartread .The modified version of the script is attached，and i add a variable "number_start" to change the generated .ti1 SAMPLE_ID.

 

[crenedecotret]

Thanks for your share.
I found a small error in this script.The argyllCMS .ti1 file XYZ scale is 0~100,but the script generated XYZ scale is 0~1；This may affect chartread .The modified version of the script is attached，and i add a variable "number_start" to change the generated .ti1 SAMPLE_ID.

Hi.. this is totally normal for argyllcms. The original script works correctly. You can use r g b values, convert them to TI1 with my script. If you then use printtarg, the tiff files will have the correctt r g b values if you open the in argyllcms

 

[crenedecotret]

@buxiaozizai.com Looking at some documentation you may be right. That's bizarre. I ended up with the 0-1 file because printtarg wouldn't accept the ti1 from my first version which was like yours. When I compared with some existing ti1 files I had, I changed the scale.

Have to tried to read back a chart?

 

[knowing]

Thanks for your share.
I found a small error in this script.The argyllCMS .ti1 file XYZ scale is 0~100,but the script generated XYZ scale is 0~1；This may affect chartread .The modified version of the script is attached，and i add a variable "number_start" to change the generated .ti1 SAMPLE_ID.

Hi all,
I wondered how to use this script correctly, so I used chatGPT to make a markdown document to document its use. Here it is for anyone who is interested.

Cheers,
K

 

[knowing]

Hi,
I have a small question relating to converting the SpyderPrint targets to ArgyllCMS format reference files. Is it not possible to use the scanin command to read the color charts in order to produce ti3 files that may be used as reference data for when reading the SpyderPrint targets? Making a cht-file with patch location definition should alow scanin to read the colors in the tiff image, right? Then that reference can be used with profcheck later to verify the accuracy of the profile created with real measurements of the target?

Example:
scanin -v -dipn -c SpyderPrintTarget1.tif SpyderPrintTarget1.cht Scanner.icc OutputFile


Has anyone made such a cht file?

Thanks,
K

 

[buxiaozizai.com]

Hi,
I have a small question relating to converting the SpyderPrint targets to ArgyllCMS format reference files. Is it not possible to use the scanin command to read the color charts in order to produce ti3 files that may be used as reference data for when reading the SpyderPrint targets? Making a cht-file with patch location definition should alow scanin to read the colors in the tiff image, right? Then that reference can be used with profcheck later to verify the accuracy of the profile created with real measurements of the target?

Example:
scanin -v -dipn -c SpyderPrintTarget1.tif SpyderPrintTarget1.cht Scanner.icc OutputFile


Has anyone made such a cht file?

Thanks,
K

I speculate that what profcheck actually check the color difference between the Lab value obtained by patch’s RGB value through the printer ICC profile, and the measured Lab value of that same patch.

Verifying the printtarg-generated .tif target image against the XYZ values in the .ti2 file is likely meaningless — if the .ti2 file is script-generated, those XYZ values would simply be the RGB patch values converted to XYZ under the sRGB color space.

Using XYZ values under sRGB to validate the measured colors of printed patches would be both impractical and meaningless, because the printed gamut and the sRGB gamut have large non-overlapping regions.

Color management is to map colors between devices with different gamuts through a perceptual color space, preserving tonal gradation as much as possible and reproducing similar-looking colors.
I guess the values won't be similar.

 

[knowing]

I speculate that what profcheck actually check the color difference between the Lab value obtained by patch’s RGB value through the printer ICC profile, and the measured Lab value of that same patch.

Verifying the printtarg-generated .tif target image against the XYZ values in the .ti2 file is likely meaningless — if the .ti2 file is script-generated, those XYZ values would simply be the RGB patch values converted to XYZ under the sRGB color space.

Using XYZ values under sRGB to validate the measured colors of printed patches would be both impractical and meaningless, because the printed gamut and the sRGB gamut have large non-overlapping regions.

Color management is to map colors between devices with different gamuts through a perceptual color space, preserving tonal gradation as much as possible and reproducing similar-looking colors.
I guess the values won't be similar.

Thanks for your insights

My issue is that I would like to measure SpyderPrint targets, such as the "EZ 729 Colors Plus Grays" (9 pages), or "Expert Target Plus Grays" (4 pages). When I scan them with the SpyderPrint and export the measurement file, I only get Lab values. The measurement file before export also contain only lab values (actually integer values that must be divided with 100 to get the values that are exported). The attached Archive.zip contain examples of measurement file and exported data from SpyderPrint when scanning "EZ 729 Colors Plus Grays" target. I have also attached the targets from SpyderPrint.

So I wanted to confirm the sequence of the exported values, so that one could convert them into a ti1 or ti2 format usable with ArgyllCMS and create profiles with Argyll.

I started with making a python script to visualise the lab values to confirm the patch sequence, this script is attached also. This target uses a grid of 9 columns by 12 rows for the first 7 pages, and for the last two pages of grays has a grid of 10 columns by 12 rows. The grays start at line 756, so when I use the script attached

python3 plot_lab_values_as_grid.py "PrintTarget Measurement-_ex.txt" --rows 12 --cols 10 --patch_no_start 756

Then I can visualize the gray patches, or

python3 plot_lab_values_as_grid.py "PrintTarget Measurement-_ex.txt" --rows 12 --cols 9

to visualise the first page of the measured "EZ 729 Colors Plus Grays" target.

Now I realise that I am missing the RGB values from the original reference patches that SpyderPrint keeps hidden in its software. If I could read the reference colors from the images themselves, and generate the ti1 or ti2 file, similar to that targen and printtarg would create then I would have enough to make a profile in ArgyllCMS, right?

Is my thinking correct?

Has anyone in this forum attempted to do this before? And are there any scripts that can make the conversion of data from the images to RGB values in ti1 or ti2 form?

Thanks for all help...

Cheers,
K