Can CMY cover all colors ? If not what does it ?

Mustafa Umut Sarac · Jul 8, 2015

Can CMY filters create all colors or what kind of filter set can cover all spectrums ?

Umut

Mainecoonmaniac · Jul 8, 2015

I think CMY is for printing. But for offset and inkjet printing you need some black also know as "K" or "keyline" to make the blacks black. Otherwise you'll get muddy brownish blacks.

gzinsel · Jul 8, 2015

I think the question should. . . Can red, green, blue create all colors? yours being the subtractive process, while the latter is being the additive. in addition, I think this is a philosophical question rather than a scientific question.

If you are starting with white light and subtracting ( thue filter set {cmy}) depending on what your "material, or subject, or negative " color has, it could. . . in theory. but we are starting to get into minutia. I really don't like going there. All FILM is an approximation of the real. The visual spectrum is a theorical limit, set by our numbers and our own way of measuring it. paper has its limits. its suspect to look for truth in THIS route. I advise against it. It's a fouls errand.

Mustafa Umut Sarac · Jul 8, 2015

gzinsel said:
I think the question should. . . Can red, green, blue create all colors? yours being the subtractive process, while the latter is being the additive. in addition, I think this is a philosophical question rather than a scientific question.

If you are starting with white light and subtracting ( thue filter set {cmy}) depending on what your "material, or subject, or negative " color has, it could. . . in theory. but we are starting to get into minutia. I really don't like going there. All FILM is an approximation of the real. The visual spectrum is a theorical limit, set by our numbers and our own way of measuring it. paper has its limits. its suspect to look for truth in THIS route. I advise against it. It's a fouls errand.

Hello , I am after to create perfect autochrome dye colors one by one with cmy filters ????

tony lockerbie · Jul 8, 2015

CMY will cover all colours in the photographic printing process, and of course RGB can be created from this combination. Mixing all three in equal amounts gives you a neutral density of course.
Don't know much about Autochrome butI'm sure that your CMY filters will work.

DREW WILEY · Jul 8, 2015

In theory, yes. In real life, never perfectly. That's because there is no such thing as a perfect dye or pigment set.

PinRegistered · Jul 8, 2015

This is a very interesting topic. I did a search and the author of one article states that 5 to 10 micron potato starch grains were dyed with either red-orange, green, or blue-violet dyes and between grain treatment was lamp-black. Good luck!

Sent from my SM-G900V using Tapatalk

wiltw · Jul 8, 2015

Both CYM and RGB can 'make all colors'... one is subtractive, the other is additive. Whether EITHER can 'make all colors' is really based simply upon the NUMBER OF BITS that each digital system is assigned to make all the possible colors.

If either use 8 bits per color (C-Y-M, or R-G-B) one is limited to 16.78 Million colors
If either use 16 bits per color, one is limited to 281,474976.71 Million colors
If either use 24 bits per color, one is limited to 4,722,366,482,869,650 Million colors

...in the representative of values of colors by each system's digital color space!

Using any analog (not digital) means of creating C-Y-M or R-G-B does not have the limit of number of bits.

Using C-Y-M vs. R-G-B is simply limited by the chemical means that are used to recreate each of the colors. An ink pigment process has its limits, a dye process has its limits.
- One dye process has different limits than a second dye process; we have seen this exemplified in the differences of reproduction
- conventional reflective reversal color prints vs. the Cibachrome ('Ilfochrome') azo dye prints, and also in
- the conventional reflective color prints vs. dye transfer color prints (of the late 20th century) vs. Carbo prints (of the 1930s)

MartinP · Jul 8, 2015

wiltw said:
Both CYM and RGB can 'make all colors'... one is subtractive, the other is additive. Whether EITHER can 'make all colors' is really based simply upon the NUMBER OF BITS that each system is assigned to make all the possible colors.

If either use 8 bits per color (C-Y-M, or R-G-B) one is limited to 16.78 Million colors

If either use 16 bits per color, one is limited to 281,474976.71 Million colors

If either use 24 bits per color, one is limited to 4,722,366,482,869,650 Million colors

...in the representative of values of colors by each system's color space!

In the real world there are an infinite number of colours. Thinking solely of some sort of digital approximation has no bearing on the OPs question, apart from an abstract philosophical consideration of how close is 'good enough' for some representation of a real colour.

The practical problem is that the filters and the lens colours are imperfect to some degree and so the real colours are distorted in some way, at some point in the spectrum. So (and I now see this was mentioned already) the answer is "Yes, but . . ."

Somewhere I recall reading that the distribution of the coloured grains was critical to the appearance of the result, and that tests (by the writer of the description of the process) had not managed to achieve the reliable distribution of the filter-colours seen in original Autochrome plates -- so that will be another interesting problem within the implementation. How about scribed lines, or an inkjetted grid in place of starch grains?

wiltw · Jul 8, 2015

MartinP said:
In the real world there are an infinite number of colours. Thinking solely of some sort of digital approximation has no bearing on the OPs question, apart from an abstract philosophical consideration of how close is 'good enough' for some representation of a real colour.

The practical problem is that the filters and the lens colours are imperfect to some degree and so the real colours are distorted in some way, at some point in the spectrum. So (and I now see this was mentioned already) the answer is "Yes, but . . ."

Somewhere I recall reading that the distribution of the coloured grains was critical to the appearance of the result, and that tests (by the writer of the description of the process) had not managed to achieve the reliable distribution of the filter-colours seen in original Autochrome plates -- so that will be another interesting problem within the implementation. How about scribed lines, or an inkjetted grid in place of starch grains?

I already added to my post, compared to what you quoted!
As already stated, and as reflected by the second half (the edits to my first response), "In theory, yes. In real life, never perfectly. That's because there is no such thing as a perfect dye or pigment set. "

Bill Burk · Jul 8, 2015

https://en.wikipedia.org/wiki/Gamut

Look at the graphic titled "Comparison of some RGB and CMYK colour gamut on a CIE 1931 xy chromaticity diagram"

SWOP CMYK is the pentagonal shaped region. That would be good quality offset printing in four process colors.

Some of the RGB triangles are pretty large. I am impressed with what can be done.

Mustafa Umut Sarac · Jul 8, 2015

I learned that postcard printers printed with RYB inks , not rgb or cmyk.

Athiril · Jul 8, 2015

wiltw said:
Both CYM and RGB can 'make all colors'... one is subtractive, the other is additive. Whether EITHER can 'make all colors' is really based simply upon the NUMBER OF BITS that each digital system is assigned to make all the possible colors.

If either use 8 bits per color (C-Y-M, or R-G-B) one is limited to 16.78 Million colors

If either use 16 bits per color, one is limited to 281,474976.71 Million colors

If either use 24 bits per color, one is limited to 4,722,366,482,869,650 Million colors

...in the representative of values of colors by each system's digital color space!

Using any analog (not digital) means of creating C-Y-M or R-G-B does not have the limit of number of bits.

Using C-Y-M vs. R-G-B is simply limited by the chemical means that are used to recreate each of the colors. An ink pigment process has its limits, a dye process has its limits.

One dye process has different limits than a second dye process; we have seen this exemplified in the differences of reproduction

conventional reflective reversal color prints vs. the Cibachrome ('Ilfochrome') azo dye prints, and also in

the conventional reflective color prints vs. dye transfer color prints (of the late 20th century) vs. Carbo prints (of the 1930s)

That's only the fine ness of steps that can be recreated within the colour range that can be created by RGB/CMY etc.

RGB display screens can have an extremely broad colour range/gamut. Many can get basically 100% Adobe RGB (apple monitors only recently claimed 100% RGB, but they referred to sRGB or less than 70% Adobe RGB, before that were even worse, and many higher end laptop screens even approached 100% Adobe RGB predating their 100% sRGB by over 10 years).

Prof_Pixel · Jul 8, 2015

Mustafa Umut Sarac said:
I learned that postcard printers printed with RYB inks , not rgb or cmyk.

From: http://www.metropostcard.com/techniques8.html "Tricolor Lithograph: In attempts to render views through tricolor printing, dull looking images often resulted. While the RYB pallet was less than ideal to create natural color images it could still produce a bright gamut when reproducing illustration."

It has lots of examples that show all the shortcomings

removed account4 · Jul 8, 2015

hi professor pixel

tricolor/ is what kodachromes were, technicolor no ?
the russian trichromes from last century are stunning
as are modern ones .. its great to see post cards being made
with this technology !

Bill Burk · Jul 9, 2015

Mustafa Umut Sarac said:
I learned that postcard printers printed with RYB inks , not rgb or cmyk.

Yellow was always pure. Blue was Peacock Blue, so actually Cyan. Magenta was always a problem to make pure.

Europan · Jul 9, 2015

It might be a shock to most among us that violet is not reproduced by either RGB or YCM systems. No violet has ever been brought to a cinema screen, no violet was available with the Lumière Autochrome plates, violet doesnt exist on photo paper.

Not that it were impossible to display violet, yet it takes a separate channel so to speak. One would have to install an extra violet dye with printing presses. In the negative-positive process of colour photography and cinematography the films are not prepared for violet and there is no complementary colour to it.

Violet is a rather rare colour. Most people feel a certain discomfort when seeing pure violet.

gzinsel · Jul 9, 2015

mustafa, I hope you get your autochromes up and running. of all the processes, I would love to do is autochrome.

alanrockwood · Jul 10, 2015

There is no three color combination, either additive or subtractive, that can create all visible colors.

Bill Burk · Jul 10, 2015

alanrockwood said:
There is no three color combination, either additive or subtractive, that can create all visible colors.

Right. And I guess I was wrong about tricolor before my time, looks like they may have used red and blue.

Meanwhile, with all the wonderful non-APUG technology, I don't see why there isn't a push for seven-color processes which could really improve the gamut of color imagery.

AgX · Jul 10, 2015

Mustafa Umut Sarac said:
I am after to create perfect autochrome dye colors one by one with cmy filters ????

Autochrome is a additive process, CMY stands for substractive processes.

CMY filtering is in no way employed in Autochrome.

RobC · Jul 10, 2015

the operative word is "Filter" and thats what filters do. Any filter will only pass wavelengths that it is NOT blocking. So you should already know the answer to your own question if you have done your research and know what wavelengths each of your filters block and what wavelengths each of your dyes requires to activate/create them. Your question seems to indicate you haven't done that and its not what you've asked. Do you know what you're doing?

Mark Antony · Jul 10, 2015

alanrockwood said:
There is no three color combination, either additive or subtractive, that can create all visible colors.

That's interesting because most humans have trichromic vision, there are only RGB chemical receptors in the cones to define our idea of the visible spectrum.

So there are no RGB or CMY dyes that cover the whole visual spectrum? how do those films that EXCEED the human visual spectrum work?
We mapped that spectrum in 1931, so as long as we have a combination of dyes that are wider than the HVS the main problem being the purity of those dyes, so yes we can create emulsions that match or exceed the HVS
There are tri-pack emulsions that exceed HVS too!

chrisaisenbrey · Jul 10, 2015

Mark Antony said:
That's interesting because most humans have trichromic vision, there are only RGB chemical receptors in the cones to define our idea of the visible spectrum.

So there are no RGB or CMY dyes that cover the whole visual spectrum? how do those films that EXCEED the human visual spectrum work?
We mapped that spectrum in 1931, so as long as we have a combination of dyes that are wider than the HVS the main problem being the purity of those dyes, so yes we can create emulsions that match or exceed the HVS
There are tri-pack emulsions that exceed HVS too!

The perception of color by the human eye with three types of cones is the reason why the RGB of CMY system can work. Three colors are sufficient to give us the illusion to see the reality.

Then I guess pigments are not only optimized for the biggest hue, but also to discriminate the most important colors and to give a larger brightness range.

The next problem is of cause that real pigments are not ideal filters. Ive attached some diagram for some printing colors.

Mark Antony · Jul 10, 2015

The problem is there is no such thing as colour, its all just human perception, and our colour model for human vision is and can only be RGB, (OK there are a few known people in the world who are tetrachromic).

The argument is that no RGB dyes can cover the range of wavelengths that for the human perception of colour? well I don't think that is correct.

In the real world there are* many films that greatly exceed the range of colours we can perceive. The dye purity and problems with overlap are a wholly separate issue which is more akin to colour accuracy that is the actual colour number being viewed as a different one as some dyes have different metermeric values under variable light sources etc.

To say no there are no dyes that exceed human vision is false, and would be better to state that no models MATCH perfectly to human vision is a more correct way of putting it.
It is simple to create dyes in films that exceed human vision, but they won't look real.
Your attached file just seems to be a simple measure of individual dyes matched against perfect response, which isn't about the gamut they can define but rather the purity of dyes giving perfect colour matching (in perception) to source stimuli.

*EIR films exceed HVS

Can CMY cover all colors ? If not what does it ?

Attachments

Privacy & Transparency

Privacy & Transparency