measures of colors and lines in the middle
below: 6 invisible rectangles (click to zoom!)
Thanks, now I see. Have you figured the % of area the lines cover to make the most transparent grid? And the % of square colored dots under the grid?
I forgot to check the Notify box.
BTW, don’t forget to click on the pictures, that allows you to zoom and see much details
Actually, if the colors are added in good proportions and measured well, it doesn’t matters, really.
Only a too loose grid would become visible, just because we’ll clearly see the lines (especially if they are much darker than the background)
Something like a range between 15% and 95% should be almost equal…….
In the picture of the “6 transparents”, every color combination is tested 2 times: at 25% and at 50% (it’s very precise, max 1% error… so between 24,75% & 25,25%, and 49,5/50,5%)
Ah yes, what I forgot to say about what you see, well: what you DON’T see is the fact that this color combination (the lines of paint) have together exactly the same tint, saturation and brightness as the printed background, and that’s why they disappear on it, like if you paint white on white…
That’s a lot of work. All the line colors would have to match the density color of the squares. I wondered how you choose the line colors.
I’m not sure I get you, what do you mean by “line colors” and “density color”?
The lines are over the small squares. The “density color” is of the total small square colors as seen in a small sized image. When the small squares are really not discernable as objects. This would be a smaller image than any shown. That averaged color woud be the “density or saturation color” that the lines on top of the squares would be to make the squares “invisible”.
I think I get the question now.
How did I choose the colors of the lines to match the background?
Well, it’s all about “metamerism,” the average color of the background is calculated after some samples I did of the colors made by the printer. In colorimetric terms, the average color of the background is close to:
in D65 / 2° observer
This measure is the actual average of 64 measures of this print with a spectrocolorimeter that has a field diameter of 1,8 cm
With that measure, and all the other measures that I did (the different paints that I proffed on the same print, 2×2 cm squares) I can calculate a formula to “match” the color of the background. I have a technique to predict the amounts and proportions of paints that will match this measure.
So at the end, the choice of the colors is a tricky question, since several groups of different paints will appear metameric, i.e., look the same and disappear on this background.
My preference went to RGB + B/W, could have been Yellow/Magenta/Cyan + B/W or Orange/Purple/Green + B/W, or Yellow/ Blue + B/W, etc. it would have disappeared in the same way.
I have to read this again tomorrow.
I re-read it and it makes perfect sense. Any correct triad will work. It stands to reason that split-complements would work also if you wanted to increase the number of colors. A split-complement has colors equally spaced away from the main triad.
One more question: to make an opaque paint, you always mix PR122 with Titanium white or for ex. Zinc White??
I’ll buy it tomorrow in Brussels, the PR122 Sennelier
Any white will do, Lead white is the strongest and probably the best but poisonous in the air, titanium white is not as white or as brittle as zinc white. White is tinting it a little but you can go more saturated than cobalt violet and the color is more a more accurate magenta. Adding black will do about the same as adding the complement.
ok, I’ll try
Fill in your details below or click an icon to log in:
You are commenting using your WordPress.com account. ( Log Out / Change )
You are commenting using your Twitter account. ( Log Out / Change )
You are commenting using your Facebook account. ( Log Out / Change )
You are commenting using your Google+ account. ( Log Out / Change )
Connecting to %s
Notify me of new comments via email.
Notify me of new posts via email.
Blog at WordPress.com.