How does a RIP know a dot should lay down on another dot vs next to it?

[Gilligan]

Example, you print a halftone under base and follow up with a top color.  Then the top color needs to sit RIGHT on top of the underbase dots.

BUT, if you want to make say green (just for easy to follow example), you want to have Yellow lay NEXT to Blue to make it "look" like green from a distance.

First example is 50% pink with a 50% underbase on say a black shirt.  Second example would be 50% yellow and 50% blue on a white shirt.  RIP doesn't know what colors or what.

Maybe I'm completely mistaken on how this would work out, so please enlighten me.

I was thinking of a different way to work some art we printed the other day, just for future purposes sake, and it crossed my mind.

[Rockers]

Interlocking halftones. Send Jeff, The FullSpectrumSeparator, a message. He knows that stuff. He`s done a few seps for us with perfectly interlocking halftones. Looks so good.
Here is a link to a similar question on a different forum where he answered.
http://www.t-shirtforums.com/graphics-design-help/t206497.html

[blue moon]

all the dots are on top of each other. The field of dots is set by the lpi and angle and the RIP just calculates how big to make the dot in each position. In some places it will be 0% and in others it will 100%.
When blending blue and yellow to make green they should print on top of each other and mix (just like you would make a green color by mixing blue and yellow).
In 4CP, the dots are printed in a rosette and rather (also a fixed pattern) and rather than physically mixing the inks, it is their proximity that gives the illusion of the green.

pierre

[Gilligan]

Interesting Rockers, thanks!

Pierre, in that link that Rockers posted, it mentions the net result... it would be a negative space of 50% (assuming I'm printing in a vacuum   ).  So I'd have to change to some other percentage to achieve what I wanted.

Do any RIPS do this automagically?

[Rockers]

Interesting Rockers, thanks!

Pierre, in that link that Rockers posted, it mentions the net result... it would be a negative space of 50% (assuming I'm printing in a vacuum   ).  So I'd have to change to some other percentage to achieve what I wanted.

Do any RIPS do this automagically?

Jeff told me the new version of Simple Seps raster comes with an interlock halftones feature. Should be available via his website.

[blue moon]

Interesting Rockers, thanks!

Pierre, in that link that Rockers posted, it mentions the net result... it would be a negative space of 50% (assuming I'm printing in a vacuum   ).  So I'd have to change to some other percentage to achieve what I wanted.

Do any RIPS do this automagically?

nah, good idea, but not really usable for several reasons. . .

pierre

[ol man]

back in the old days -- i used to setup my screens to interlock -seemed to make sense.
although i haven't in years, since the advent of computers and rips.

Pierre, please explain the reasons ...

[Dottonedan]

There seems to be a difference of understanding in what makes a green from blue and yellow.

As Giligan has defined by placing the dot side by side, that's not truly mixing. That can be done, but in that case, it's the eye,size of libe count, and your eye that blends. An example would be indexing. That is putting dots side by side using square dots. The higher the Rez, the smaller the dot. The smaller the dot,?the higher the mesh needed and also the lighter the ink is put down.

Stochastic dots are also square but are treated like traditional sim process where the colors and dots run under each other to "blend" or mix. This is what you want to do with traditional halftones to achieve the best color blends as Pierre had indicated.

Vector programs by nature, knock out of each other and place space and dots butt up against each other.
If you want to blend colors in vector, make a colpy of your blend and past undernieth. Then extend the yellow fade out further. Tell that yellow to overprint. Then tell the top copy to also overprint. Simple blends will come our better.

In photoshop, sepping manually, you do this yourself. This is another reason that ps seps are used often over vector. Sure, it can be done, but it's raster in photoshop.

[Gilligan]

Dan, since you have so much time on your hands these days.

Would you be able to throw together a quick sample of what you are talking about?  Just big ugly blocks would be adequate for me to understand an do some R&D with.

[Full-SpectrumSeparator]

When blending blue and yellow to make green they should print on top of each other and mix (just like you would make a green color by mixing blue and yellow).
In 4CP, the dots are printed in a rosette and rather (also a fixed pattern) and rather than physically mixing the inks, it is their proximity that gives the illusion of the green.

pierre

The most accurate way for 4-color process is actually to overprint all CMY dots the same angle except invert-lock the black with the other colors. 

The most accurate way for plastisol/opaque inks to blend is Interlocking all dots, index/stochastic FM or halftone AM.

It is called the scientific method, it is a wonderful thing for having a question about the physical world and then running experiments and tests and comparisons to draw conclusions that are accurate and reliable.   

People can share their opinions based upon their apparent lack of understanding of these subjects, but it doesn't help the actual scientific progress going on, and it gets taken as fact by otherwise uninformed readers, which can actually harm the progress of the industry. 

[Frog]

Jeff, I think that you at least owe Dan a thank you for the fuel for you to heat up this discussion! I understand that you're looking at some cold temps up there! 

[Full-SpectrumSeparator]

Jeff, I think that you at least owe Dan a thank you for the fuel for you to heat up this discussion! I understand that you're looking at some cold temps up there! 

Lol... I was restraining myself for a while with this one,  but you're right now I've gone and lit the fuse. 

https://www.youtube.com/watch?v=wmin5WkOuPw

[blue moon]

I went ahead and removed Jeff's post due to name calling. As mentioned before, we will not tolerate those.

Pierre

[SteveS]

Why not just clean it up and leave it posted?

[blue moon]

When blending blue and yellow to make green they should print on top of each other and mix (just like you would make a green color by mixing blue and yellow).
In 4CP, the dots are printed in a rosette and rather (also a fixed pattern) and rather than physically mixing the inks, it is their proximity that gives the illusion of the green.

pierre

The most accurate way for 4-color process is actually to overprint all CMY dots the same angle except invert-lock the black with the other colors. 

The most accurate way for plastisol/opaque inks to blend is Interlocking all dots, index/stochastic FM or halftone AM.

It is called the scientific method, it is a wonderful thing for having a question about the physical world and then running experiments and tests and comparisons to draw conclusions that are accurate and reliable.   

People can share their opinions based upon their apparent lack of understanding of these subjects, but it doesn't help the actual scientific progress going on, and it gets taken as fact by otherwise uninformed readers, which can actually harm the progress of the industry.

I might not have explained it very well, but the question was about the RIP and the dot placement not color theory. So an easier to understand answer would have been to say that if the lpi and frequency are the same, all the dots will print on top of each other (physically blending the ink). If one or both are different, it will create a pattern. Further more, If the lpi is the same the dots can be printed next to each other and will create an evenly repeating (non moire) pattern that our eye will see as the blended color (this used in the printing industry other than garments to produce the 4CP prints).

These are facts and as long as your RIP is working properly any scientific testing will confirm it. So pleas do not take my answers out of context and twist them into something else.

Pierre

P.s. This pertains to the halftone dote that are amplitude modulated as they are the predominant type used in screenprinting