Monad Studios wrote:
The relationships of colors to each other can vary depending on the spectrum of the light (and of the photographic material or sensor, and of the ink or monitor).

This is independent of white balance.  Even if everything is perfectly white balanced it is still the case.

digital Artform wrote:
Awesome. Now use that to introduce the concept of subtraction into the discussion.

Good idea!

Picture a brilliantly colored flower of a striking and unusual hue when seen in bright sunlight.  On a monitor, one can never get a color that does justice to that particular hue.  This does happen, right?  It's because the sunlight has a broad spectrum and, in particular, it has a certain spectral color that bounces well off the flower and stimulates our color receptors in proportions that the monitor can't manage.  In other words, the monitor produces very little of that particular wavelength of light; it's in a flat part of the monitor's spectrum, not on a peak.  One could say that in moving from sunlight to color monitor one has subtracted big chunks of the spectrum, as though one has used a special sort of filter.

EDIT:  You're awesome, too.  Now, you use "wraparound" to introduce multiplication into the discussion.

NewBoldPhoto wrote:
See kiddies... this is what happens when you spend too much time on them new fangled interwebs... now run along and play.

Seriously... I understand what you are getting at with the whole a-b not being equal to b-a.
But, you are aware that you are try to mix apples with orangutans to produce an inorganic compound.... right?

LOL  I wondered if anyone would point that out.

I'm surprised no one pointed out that what you are subtracting is *NOT* one filter from the other, but the wave lengths from the ORIGINAL source, so it's not a-b vs b-a, but c-a-b or c-b-a, where c is LARGER than either a or b, AND larger than a+b; where C is the original light source, presumably a WHITE light.

What filter 2 is subtracting is what passes THROUGH filter 1, *not* filter 1 itself.  So, the reason a 50% red followed by a 50% red = 25% red, is that the first filter removes 50% of the red from the original source, and the second filter removes 50% of the red from the 50% reduced red that made it through the first filter.

I love people arguing a flawed premise.

Filters are subtractive, because you are subtracting from WHITE light, or the ORIGINAL source, not each other.  A filter "removes" an amount of light - eg: SUBTRACTING those wave lengths from the original source - thus changing the "color" of the light emerging from the other side.  A filter can't "add" light   Stacking filters can't add more light. Light is alway subtracted until we reach "black" or "zero" transmitted light.

BodyartBabes wrote:

LOL  I wondered if anyone would point that out.

I'm surprised no one pointed out that what you are subtracting is *NOT* one filter from the other, but the wave lengths from the ORIGINAL source, so it's not a-b vs b-a, but c-a-b or c-b-a, where c is LARGER than either a or b, AND larger than a+b; where C is the original light source, presumably a WHITE light.

What filter 2 is subtracting is what passes THROUGH filter 1, *not* filter 1 itself.  So, the reason a 50% red followed by a 50% red = 25% red, is that the first filter removes 50% of the red from the original source, and the second filter removes 50% of the red from the 50% reduced red that made it through the first filter.

I love people arguing a flawed premise.

Filters are subtractive, because you are subtracting from WHITE light, or the ORIGINAL source, not each other.  A filter "removes" an amount of light - eg: SUBTRACTING those wave lengths from the original source - thus changing the "color" of the light emerging from the other side.  A filter can't "add" light   Stacking filters can't add more light. Light is alway subtracted until we reach "black" or "zero" transmitted light.

Maybe he really does think that it's purely a digital artform.

BodyartBabes wrote:
I'm surprised no one pointed out that what you are subtracting is *NOT* one filter from the other, but the wave lengths from the ORIGINAL source...

got it covered

Monad Studios wrote:
Subtractive mixing:  start with all colors (e.g. white) light and subtract some to make a new color.  If you "mix" in a yellow filter (i.e. a filter that subtracts blue light) you create the perception of yellow light. 

Then mix in a magenta filter (a filter that subtracts green photons) and we see red light.

Additive mixing: start with no light (e.g. black) and add some to make a new color.  If you add some red light and then add some green light you create the perception of yellow.

"Multiplicative mixing"  No, not really.  Now we're talking real, actual, physical photons vs. numbers in a computer.

digital Artform wrote:
That is the way many people understand it. Yes.

It's just false.

Please help me here!  Can you point me to where I went wrong?

HarveyT wrote:
Call it iguana...
Does the picture look good when your done?

To be fair: this whole thread started with a nit-picking complaint about terminology.  You can't really expect it to rise to art.

Alexander Image wrote:
As my understanding, a filter purpose is to blank off one color in the light, so the result is called “subtractive color”

That's my understanding, too.  I was just trying to think in my head how to say that, but you did it so nicely.

Le Beck Photography wrote:
From the physics department at Georgia State University:

Thanks!  Good illustrations and nice, clear explanation.

Monad Studios wrote:

Thanks!  Good illustrations and nice, clear explanation.

Yuppers.   Thanks, Beckster.

digital Artform wrote:
Change 'subtracted' to 'multiplied by a fraction between 0 and 100%' and you've got it

The relationship between multiplication and subtraction:

50%= 1-(1-50%)=1 –50%
20%=1-(1-20%)=1-80%
...

It is Math!

Ultra Magnus wrote:

Yuppers.   Thanks, Beckster.

That's Le Beckster thank you very much :-)
I'm a Frog! well... Norman French, Scots, Huguenot, and Dutch! All old guard Americans emigrated here during the 17th century! not German although there very well may be one of them hiding in the family tree too along with the Amerindians!

Alexander Image wrote:
A color in a light is totally different from a color in paints. In a light it is subtracted, and in paints it is added.

Color theory is based on 2 principles as addressed in the OP.

Basically, in additive color theory, you have 3 sources, Red, Green, and Blue. when you overlap the three color, you have a mixture. In the area where all 3 colors overlap, you have white light.

With subtractive color, think of using one light source and overlaping the colors Y,M,C. where the 3 overlap, you have black.

John

digital Artform wrote:
Change 'subtracted' to 'multiplied by a fraction between 0 and 100%' and you've got it

Alexander Image wrote:
The relationship between multiplication and subtraction:

50%= 1-(1-50%)=1 –50%
20%=1-(1-20%)=1-80%
...

It is Math!

A fella could get mixed up here between the math and the physiology and the physics.

Le Beck Photography wrote:
From the physics department at Georgia State University:

First there is no such thing as multiplicative color. Light is additive color. Dyes, pigments, and colored filters such as tricolor filters used to make color separation negatives, are subtractive color. I learned this in the third grade during art class at The Cabrillo School in Hawthorne California back in 1961.

Additive color mixing is the kind of mixing you get if you overlap spotlights in a dark room, as illustrated at left. The commonly used additive primary colors are red, green and blue, and if you overlap all three in effectively equal mixture, you get white light as shown at the center. Additive color mixing is conceptually simpler than the subtractive color mixing you get with paints and pigments since you are just adding light energy in different ranges of the visible spectrum.




Subtractive color mixing is the kind of mixing you get if you illuminate colored filters with white light from behind, as illustrated at left. The commonly used subtractive primary colors are cyan, magenta and yellow, and if you overlap all three in effectively equal mixture, all the light is subtracted giving black. Subtractive color mixing is more complex than the additive color mixing you get with colored spotlights.

Thanks for illustrating my point.

John

Hermesz Fine Art wrote:

Color theory is based on 2 principles as addressed in the OP.

Basically, in additive color theory, you have 3 sources, Red, Green, and Blue. when you overlap the three color, you have a mixture. In the area where all 3 colors overlap, you have white light.

With subtractive color, think of using one light source and overlaping the colors Y,M,C. where the 3 overlap, you have black.

John

It's really pretty straightforward, most of us seem to agree.  The hubbub here was just a confusion between the real world (where light gets subtracted) and the math (where numbers get multiplied),

Monad Studios wrote:

It's really pretty straightforward, most of us seem to agree.  The hubbub here was just a confusion between the real world (where light gets subtracted) and the math (where numbers get multiplied),

Oh My God... that was funny...
I understood that he was talking about the mathematical model that the computer uses but the explanation and examples were so far out there...  now my ribs hurt...

Monad Studios wrote:
The subtraction that people are talking about is the removal (by absorbtion) of physical light.

And that removal occurs by percentages. Not fixed amounts. We call such removal 'multiplication'

Chris Macan wrote:

Wrong.....

the CRT monitor is mixing RGB light to creat white,
The paper is reflecting white light.

The important difference is the reason that printers work in CMY and CRT Monitors work in RGB.

The CRT starts with a black tube
and through the introduction or red, green and blue light creates all the colors you see on the screen including white.
This is an example is additive color

Printers work in CMY because they are starting with a white base and they need to trick you eye into seeing all the colors it expects. You can do this with CMY because you can systematically subtract the light reflected to create the illusion of a full tone image. The reflective model that printers and photo printing paper uses works much like the subtractive theory of light.

Yeah. Like I said . . .

digital Artform wrote:
There is no important difference.

In the case of a monitor a wide spread of wavelengths, some from a red emitter, some from a green emitter, and some from a blue emitter come together in your brain. The wide variety of apparently equal strength wavelengths stimulates the perception of white.

On paper a wide spread of wavelengthsis reflected into your eyes where it comes together in your brain. The wide variety of apparently equal strength wavelengths stimulates the perception of white.

Now just replace 'subtraction' with 'multiplication' in your description of inks and printers and you'll be good to go.

Monad Studios wrote:

Monad Studios wrote:
The relationships of colors to each other can vary depending on the spectrum of the light (and of the photographic material or sensor, and of the ink or monitor).

This is independent of white balance.  Even if everything is perfectly white balanced it is still the case.

Good idea!

Picture a brilliantly colored flower of a striking and unusual hue when seen in bright sunlight.  On a monitor, one can never get a color that does justice to that particular hue.  This does happen, right?  It's because the sunlight has a broad spectrum and, in particular, it has a certain spectral color that bounces well off the flower and stimulates our color receptors in proportions that the monitor can't manage.  In other words, the monitor produces very little of that particular wavelength of light; it's in a flat part of the monitor's spectrum, not on a peak.  One could say that in moving from sunlight to color monitor one has subtracted big chunks of the spectrum, as though one has used a special sort of filter.

EDIT:  You're awesome, too.  Now, you use "wraparound" to introduce multiplication into the discussion.

You are right, an RGB monitor has a limited color gamut. But that is orthogonal to this discussion.

Monad Studios wrote:
EDIT:  You're awesome, too.  Now, you use "wraparound" to introduce multiplication into the discussion.

Conceptualizing a spectrum as a wheel is wrapping it around.

BodyartBabes wrote:
LOL  I wondered if anyone would point that out.

I'm surprised no one pointed out that what you are subtracting is *NOT* one filter from the other, but the wave lengths from the ORIGINAL source, so it's not a-b vs b-a, but c-a-b or c-b-a, where c is LARGER than either a or b, AND larger than a+b; where C is the original light source, presumably a WHITE light.

What filter 2 is subtracting is what passes THROUGH filter 1, *not* filter 1 itself.  So, the reason a 50% red followed by a 50% red = 25% red, is that the first filter removes 50% of the red from the original source, and the second filter removes 50% of the red from the 50% reduced red that made it through the first filter.

I love people arguing a flawed premise.

Filters are subtractive, because you are subtracting from WHITE light, or the ORIGINAL source, not each other.  A filter "removes" an amount of light - eg: SUBTRACTING those wave lengths from the original source - thus changing the "color" of the light emerging from the other side.  A filter can't "add" light   Stacking filters can't add more light. Light is alway subtracted until we reach "black" or "zero" transmitted light.

You are half right.

The part you are right about was that when I explained all this 6 ways to Sunday one of my ways was wrong. That was the A-B  B-A explanation.

Filters are multiplicative because you are multiplying by a percentage between 0% and 100%. A filter "removes" an amount of light - eg: MULTIPLYING those wave lengths from the original source - thus changing the "color" of the light emerging from the other side.  A filter can't "add" light   Stacking filters can't add more light. Light is always attenuated (multiplied downward) as we approach "black" or "zero" transmitted light.

Monad Studios wrote:

BodyartBabes wrote:
I'm surprised no one pointed out that what you are subtracting is *NOT* one filter from the other, but the wave lengths from the ORIGINAL source...

got it covered

Monad Studios wrote:
Subtractive mixing:  start with all colors (e.g. white) light and subtract some to make a new color.  If you "mix" in a yellow filter (i.e. a filter that subtracts blue light) you create the perception of yellow light. 

Then mix in a magenta filter (a filter that subtracts green photons) and we see red light.

Additive mixing: start with no light (e.g. black) and add some to make a new color.  If you add some red light and then add some green light you create the perception of yellow.

"Multiplicative mixing"  No, not really.  Now we're talking real, actual, physical photons vs. numbers in a computer.

Please help me here!  Can you point me to where I went wrong?

You used the word 'subtraction'

HarveyT wrote:
Call it iguana...
Does the picture look good when your done?

If you want a better understanding of some details, then why not try to call it something more accurate?

If you don't care, then enjoy your camera. Off you go.

Le Beck Photography wrote:
From the physics department at Georgia State University:

First there is no such thing as multiplicative color. Light is additive color. Dyes, pigments, and colored filters such as tricolor filters used to make color separation negatives, are subtractive color. I learned this in the third grade during art class at The Cabrillo School in Hawthorne California back in 1961.

Additive color mixing is the kind of mixing you get if you overlap spotlights in a dark room, as illustrated at left. The commonly used additive primary colors are red, green and blue, and if you overlap all three in effectively equal mixture, you get white light as shown at the center. Additive color mixing is conceptually simpler than the subtractive color mixing you get with paints and pigments since you are just adding light energy in different ranges of the visible spectrum.




Subtractive color mixing is the kind of mixing you get if you illuminate colored filters with white light from behind, as illustrated at left. The commonly used subtractive primary colors are cyan, magenta and yellow, and if you overlap all three in effectively equal mixture, all the light is subtracted giving black. Subtractive color mixing is more complex than the additive color mixing you get with colored spotlights.

You have a lot correct there, except for the reference to subtraction.

Monad Studios wrote:

To be fair: this whole thread started with a nit-picking complaint about terminology.  You can't really expect it to rise to art.

I'll be happy if it just approaches some form of science.

BodyartBabes wrote:
What filter 2 is subtracting is what passes THROUGH filter 1, *not* filter 1 itself.  So, the reason a 50% red followed by a 50% red = 25% red, is that the first filter removes 50% of the red from the original source, and the second filter removes 50% of the red from the 50% reduced red that made it through the first filter.

And that process is called? Bueller? Anyone? Bueller?

Multiplication.

Just the mere admission that a '50% filter' exists is an admission that the process is multiplicative. 'Percentage?' See? There is a 50-50 probability that a photon will pass.

digital Artform wrote:
(multiplied downward)

Can you show me that in a formula?

NewBoldPhoto wrote:
Can you show me that in a formula?

How many more formulas do you need?

NewBoldPhoto wrote:

Can you show me that in a formula?

Can you show me subtractive color with filters?

Call me when you get below black. I want to see the negative colors.

digital Artform wrote:
Can you show me subtractive color with filters?

Call me when you get below black. I want to see the negative colors.

You want to see a stack of filters that will absorb more light than your light source can produce????   OK. show me your light source.

NewBoldPhoto wrote:

You want to see a stack of filters that will absorb more light than your light source can produce????   OK. show me your light source.

Who cares about the light source?

You have the magic SUBTRACTING filters! Go for it!

Start with a dim light, you'll go negative faster, right?

digital Artform wrote:

Who cares about the light source?

You have the magic SUBTRACTING filters! Go for it!

Start with a dim light, you'll go negative faster, right?

You don't quite see where the apples meet the orangutans... Hmmmmm...
The problem seems to be that you are fixated on how one calculates the magnitude of the change in color. This is problematic because the process is name for the direction of the shift not the magnitude.

NewBoldPhoto wrote:
The problem seems to be that you are fixated on how one calculates the magnitude of the change in color. This is problematic because the process is name for the direction of the shift not the magnitude.

Of course I'm fixated on it. It is the entire subject of the thread.

And multiplication is also the name for the direction of the shift, plus it has the added benefit of being a more accurate name, so why not use it?

digital Artform wrote:

Of course I'm fixated on it. It is the entire subject of the thread.

And multiplication is also the name for the direction of the shift, plus it has the added benefit of being a more accurate name, so why not use it?

Because no one else on the planet  would have a clue as to what I was talking about and everyone would think that I was either an idiot or a lunatic.

NewBoldPhoto wrote:

Because no one else on the planet  would have a clue as to what I was talking about and everyone would think that I was either an idiot or a lunatic.

Well, keep it under your hat then. It may come in handy. Who knows?

digital Artform wrote:
(not sure if lumens is the right unit I need for this explanation, but anyway...)

If you had a filter that subtracted 50 lumens, lets call it a '50 lumen filter,' and you shined a 500 lumen light through it, it would transmit 450 lumens. If you shined a 50 lumen light through it it would transmit  nothing and the result would be black.

That (above) is the 'subtractive filter' universe.

In the real world filters don't subtract amounts of light. They transmit percentages of light. There are no such things as '50 lumen filters' as far as I know.

Yes and in the real world do you add or subtract the result from the total light to get your new value for the remaining light?