Possibility to convert csv-file with only L*A*B* values into a ArgyllCMS ti3 file

[Epatcola]

Hi Micronica: there is some cheating involved: I use photoshop to stretch and squeeze the whole thing for the patches to become square: this is not possible with printtarg only: you will end with elongated patches as always.

This
targen -v -d2 -g 16 -f 486 test
printtarg -r -v -i 22 -p A4 -a 1.2506 -T914.4 test

produces 486 x 10mm square patches on an A4 sheet:

The instrument type parameter seems to control the shape of the patches and peripheral annotations. The -a parameter scales the patch size. I didn't look at the output for other instrument types. There may be a more efficient option.

With the rule I printed I can read about 1 patch per second and about 5 seconds per column for rule alignment. So reading this would take just under 10 minutes.

This
targen -v -d2 -g 16 -f 216 testa5
printtarg.exe -r -v -i 22 -p 148.5x210 -a 1.2506 -T914.4 testa5

produces 216 10mm square patches on an A5 sheet.

The sheets need to be printed at 360 dpi. The -T parameter to printarg is supposed to set DPI (in tif terms resolution) but seems to need a value of dots per 2.54 inches to get dots per inch in the file.

 

[pharmacist]

That is some really nice target. I might 3D print your aligment ruler this weekend and give it a try. However not sure if @x64 already managed to enable the possibility of randomizing the patch configuration in order to match the csv data with a scrambeld ti2 file in order to generate the ti3 file. You can see the 4 white and 4 black patches in the beginning of the target are continous. In a non-randomized layout I like to put a tiny space bare (-A0.45) to separate each patch visually from another. Especially in scanning without a alignment ruler it becomes difficult to determine visually where the next patch will begin, when you use a target with let's say 32 grey steps as the gradient is gradually increasing in darkness. Scrambling patches helps to maximize contrast (Delta E) difference between adjacent patches and also forces the printer to change firing ink from different nozzles randomly during the whole printing process instead of just gradually intensifying the amount of ink pushed from the nozzles when moving from very light grey towards very dark grey, thus potentially improving profile quality.

 

[pharmacist]

For those interested in using this colorimeter to create printer profiles, be sure to get the CR30 and not the CR10 or the CR20 as the CR30 is the most accurate device (according to the specs).

 

[Epatcola]

That is some really nice target. I might 3D print your aligment ruler this weekend and give it a try.

The target is straight out of printtarg. No need to edit. All you need is something that can print without color management and to scale obeying the 360DPI file resolution. I use Qimage.

I played around some more with printtarg and -i22 is the best instrument setting for square patches. If you reduce the margins you can squeeze 260 10mm patches onto an A5 sheet, but, you need manual control of cropping so you can crop stuff you don't need instead of the bottom row.

With a revised rule you could probably get away with 8mm patches and that would put 384 on an A5 sheet.

A minor update attached. I moved the pips on the collar by 5mm so the rule is now aligned with notches on the edge not middle of the patches. I think that is a little easier to use. I also added a slight recess on the collar opening which makes it easier to fit/remove the calibration cap. The rule is unchanged.

It is a 40 minute and 29g of PLA print on my BBL X1.

 

[pharmacist]

@Epatcola : how is experience with the profiles generated with the CR30 to have a second opinion who can/cannot confirm my findings about using this cheap colorimeter for generating printer profiles. Please report back what you think about the printer targets generated using this device.