Photographer
digital Artform
Posts: 49326
Los Angeles, California, US
NewBoldPhoto wrote:
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? You multiply.
You can have, for example a 50 percent filter that takes out 1 f-stop (multiplies by half)
Photographer
NewBoldPhoto
Posts: 5216
PORT MURRAY, New Jersey, US
digital Artform wrote:
You multiply.
You can have, for example a 50 percent filter that takes out 1 f-stop (multiplies by half) Multiply? Really?
Let's see I have one dozen eggs and I filter out 50%... 12X.5=6 so I got rid of six ow let's see I multiply the result (according to you) by the total 12 to find out how many remain OK six times two is twelve plus six time ten is sixty so thats seventy-two...Uhhh Huston we have a problem!
Let me suggest that you try subtracting the change from the original quantity when you are talking about filters... It works much better.
and have a good night ... it has been fun
Photographer
digital Artform
Posts: 49326
Los Angeles, California, US
NewBoldPhoto wrote:
Multiply? Really?
and have a good night ... it has been fun Colors arise in different ways on different reproduction media. From the earliest times it has been known from painting that from just three different pigments, namely yellow, blue-green and purple-red, all intermediate hues can be produced by mixing the primary colors mentioned. Primary colors are understood to be those colors which cannot be mixed from other colors but from which all other colors can be mixed. In chromatics nowadays this type of color mixing is referred to as subtractive color mixing. The term subtractive color mixing is derived from the fact that a pigment layer absorbs certain spectral components of incident white light and reflects others, as a result of which the color impression arises for the viewer.
Other types of color mixing were initially not known.
It was not until a long time later that Isaac Newton recognized that the spectral colors of the light, the so-called color stimuli, can also be mixed. With this type of color mixing the jargon uses the term additive mixing in contrast to subtractive color mixing explained above in the case of pigments. Additive color mixing is governed by relatively simple rules, known as Grassmann's laws, which also apply to self-luminous screens, such as, for example monitors based on cathode ray tubes.
A special case of subtractive color mixing is the combination or superposition of optical filters. The transmission-of the filter combination is equal to the product of the respective transmissions of the individual filters, which is why the jargon also uses the term multiplicative color mixing in this case. This last-mentioned type of color mixing is also critical for color reproduction in the projection of color films which have three different color layers lying one above the other.
http://www.wipo.int/pctdb/en/wo.jsp?WO= … SPLAY=DESC
Photographer
digital Artform
Posts: 49326
Los Angeles, California, US
So my point in this thread is that experts in the physics of color understand that 'subtractive' model is really a multiplicative one. And 'multiplicative' is often the term they use.
Is our own jargon too ingrained too change for the more accurate?
Photographer
digital Artform
Posts: 49326
Los Angeles, California, US
While on a monitor colors are created by an additive color mixing (color creation by active light sources), on the film colors are created by multiplicative color mixing (color creation by filtering gelatine layers in a white light source). http://www.amaranthine.tv/files/alice_cms_bro_e.pdf
Photographer
digital Artform
Posts: 49326
Los Angeles, California, US
So the bottom line is that for normal people, you only need three primary colors. In additive color, they're red, green, and blue (RGB). In multiplicative color (misleadingly called "subtractive" color), they're cyan, yellow, and magenta (CMY), usually combined with a pure black (CMYK) http://www.cs.uaf.edu/2006/fall/cs381/l … color.html
Photographer
digital Artform
Posts: 49326
Los Angeles, California, US
The visible spectrum contains all colors, which the human eye can process, i.e. the color portions of the flash spectrum, to which the receptors in the eye react. All these colors together result in after additives the color synthesis white. Since an object, so that it can be noticed as colored, absorbs all color portions except the self-color, one speaks of a subtraction. If two colors are subtraktiv mixed, both decrease the spectrum. From system-theoretical view the subtraktive color synthesis is a serial connection of filters, which absorb certain in each case parts of the spectrum. The spectrum of the incident light is multiplied by a filter characteristic. If several filters A and B are switched one behind the other, this corresponds to a filter C, whose characteristic is equal the product of the individual filter characteristics A*B. Therefore one speaks in the theory also of a multiplicative color synthesis. http://www.economy-point.org/s/subtrakt … hesis.html
Photographer
digital Artform
Posts: 49326
Los Angeles, California, US
There are two types of color mixing, additive and subtractive (or multiplicative). For example, the picture on a TV screen (RGB color space) is formed by dots of the basis colors red, green and blue. When all of them are present, white color is perceived, and when none of them is present, black is perceived. This is called additive color mixing. Unlike this case, a piece of paper doesnât add energy to the illuminating light and the printed inks can be seen as filters absorbing some of the incoming light before it is reflected (CMYK color space). The more colors printed on the same spot, the more filters will subtract energy from the illumination. This is called subtractive (multiplicative) color mixing. p 34 http://webstaff.itn.liu.se/~sasgo/Avhan … _Sasan.pdf
Photographer
Lumondo Photography
Posts: 779
Vancouver, British Columbia, Canada
Subtractive is a more general term than multiplicative.
Subtractive simply indicates that spectrcal components are being attenuated. Light is being taken away.
Multiplicative is a specific terms that describes the quantitative nature of this attenuation.
While subtractive may not be the most precise term, it is, IMHO, a good practical definition.
You must remember that most people have no clue about how quantities scale. The notion of additive, logarithmic, exponential (now here's a word that's misused), etc, is foreign to them.
Photographer
Anthony Stubbs
Posts: 5399
digital Artform wrote:
I guess. Well how scientific would that be!
I think a new term is called for .....d'Admultiplicative ...where rgbcmy are used to dictate a particular colour within space.
Free chocolates for all those who can say d'Admultiplicative three times rapidly. 
Photographer
digital Artform
Posts: 49326
Los Angeles, California, US
Anthony Stubbs wrote:
Well how scientific would that be!
I think a new term is called for .....d'Admultiplicative ...where rgbcmy are used to dictate a particular colour within space.
Free chocolates for all those who can say d'Admultiplicative three times rapidly. How about if you want to switch you just use the term of the numerous unrelated scientists I linked to above?
That would be good enough for me.
That is, if you want to deepen your understanding of your craft.
It's up to you.
Photographer
Anthony Stubbs
Posts: 5399
Now, now, dA..I think you may have missed the humour. Rgbcmy is a serious topic....both additive, subtractive and multiplicath... multiplicath ...multiplicathive ...in terms of photographic lighting. Just because you may not understand something.... 
Photographer
digital Artform
Posts: 49326
Los Angeles, California, US
Anthony Stubbs wrote:
Now, now, dA..I think you may have missed the humour.
Rgbcmy is a serious topic....both additive, subtractive and multiplicath... multiplicath ...multiplicathive ...in terms of photographic lighting.
Just because you may not understand something....
Photographer
digital Artform
Posts: 49326
Los Angeles, California, US
![https://www.abovethemark.com/howtos/colormix/bluebook.jpg]() Blue and Yellow Don't Make Green by Michael Wilcox is an interesting book on pigment mixing for painters. One point he makes is that if blue paint of a pure blue wavelength spike mixed with yellow paint of a pure yellow wavelength spike the result would be not green but, in fact, black. When two colors multiply what survives is what they have in common. So only a somewhat greenish blue mixed with a somewhat greenish yellow will make a good green.
Body Painter
Monad Studios
Posts: 10131
Santa Rosa, California, US
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? 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. LOL! Thank you for that, NBP!
Body Painter
Monad Studios
Posts: 10131
Santa Rosa, California, US
digital Artform wrote:
And that removal occurs by percentages. Not fixed amounts. We call such removal 'multiplication' (1) No, the removal doesn't occur by percentages. We calculate the removal by percentages. That removal was taking place in nature eons before percentages were invented.
(2) Percentages can't be fixed amounts? Explain!
Body Painter
Monad Studios
Posts: 10131
Santa Rosa, California, US
digital Artform wrote:
Conceptualizing a spectrum as a wheel is wrapping it around. Conceptualizing don't make it so. You still haven't explained where the non-spectral colors come from.
Body Painter
Monad Studios
Posts: 10131
Santa Rosa, California, US
digital Artform wrote:
So my point in this thread is that experts in the physics of color understand that 'subtractive' model is really a multiplicative one. And 'multiplicative' is often the term they use. Links, please.
Photographer
digital Artform
Posts: 49326
Los Angeles, California, US
Monad Studios wrote:
Links, please. Wow, you are lazy.
Look up at all the links on this page.
Body Painter
Monad Studios
Posts: 10131
Santa Rosa, California, US
digital Artform wrote:
Wow, you are lazy.
Look up at all the links on this page. Those aren't experts in the physics of color; they're experts (or at least practitioners) in the digital representation of color.
The digital representation of color as commonly practiced is a simplification. You do seem to have a thorough understanding of that simplified mathematical representation.
What's really going on with light is richer and more complex, and that's the part you're not seeing.
Photographer
digital Artform
Posts: 49326
Los Angeles, California, US
Monad Studios wrote:
Those aren't experts in the physics of color; they're experts (or at least practitioners) in the digital representation of color.
The digital representation of color as commonly practiced is a simplification. You do seem to have a thorough understanding of that simplified mathematical representation.
What's really going on with light is richer and more complex, and that's the part you're not seeing. Oh, is that a fact? And your solution is to FURTHER simplify this richness by calling it SUBTRACTIVE?
Nice!
Photographer
digital Artform
Posts: 49326
Los Angeles, California, US
digital Artform wrote:
"Mixing colored dyes is commonly called subtractive color mixing, but properly it should be called multiplicative color mixing instead, for the following reason: Suppose you know a certain amount of ink attenuates a certain wavelength by a factor of X. If you have twice as much ink, the attenuation will be X2, not 2X."
http://www.av8n.com/imaging/dye-spectra.htm Above is a quote from the very first post in this thread. It explains how if ink darkens by a factor of X then twice the ink will darken by X times X, not X PLUS X.
This is not mathematical theory divorced from reality.
This is experimentally tested and confirmed real world behavior.
Body Painter
Monad Studios
Posts: 10131
Santa Rosa, California, US
digital Artform wrote:
Oh, is that a fact? And YOUR solution is to FURTHER SIMPLIFY this richness by calling it SUBTRACTIVE?
Nice! I think it's safe to call it a fact, although "richer" is really a fairly subjective word.
Color is modeled in digital systems as a three-dimensional quantity (whether RGB, HLS, or CYM). But colored light in the real world is almost always a mixture of many different wavelengths of light; it takes a lot more than three variables to describe the reflective or transmissive spectrum of a colored material (paint, filter, etc.). I'd say that makes it richer, but that's just me.
The trouble with treating the 3-dimensional model as reality -- not as a simplified model -- is that one can't explain where the non-spectral colors come from; and one can't explain why there are three variables instead of, say, 2 or 5; and there's much else one can't explain, besides.
So "subtractive" isn't a simplification here. "Multiplicative" describes the phenomenon within the simplified mathematical model. "Subtractive" describes the phenomenon within a richer, more complex physical model.
Photographer
Studio 144
Posts: 394
Mayfield, Kentucky, US
I use additive and subtractive color because I think they are much simpler terms for most people to understand.
Additive color references colors which are projected or transmitted. (Slides, displays, etc.) When you "add" the 3 additive primary colors in equal amounts the viewer perceives white.
Subtractive colors references colors which are viewed reflectively. (Prints) When you "subtract" all 3 primary color pigments from the substrate (white paper) you see white.
No it is not completely technically correct but simpler for most people to understand.
What I really think messes with most people's mind, is that "color" really does not really exist it is just a "personal perception" of the wavelengths of light.
Body Painter
Monad Studios
Posts: 10131
Santa Rosa, California, US
digital Artform wrote:
Above is a quote from the very first post in this thread. It explains how if ink darkens by a factor of X then twice the ink will darken by X times X, not X PLUS X.
This is not mathematical theory divorced from reality.
This is experimentally tested and confirmed real world behavior. No, not divorced from reality at all. Of course the mathematical model corresponds to the reality. That's what the model was built for.
Photographer
digital Artform
Posts: 49326
Los Angeles, California, US
Monad Studios wrote:
No, not divorced from reality at all. Of course the mathematical model corresponds to the reality. That's what the model was built for. So what is your objection, then?
Photographer
digital Artform
Posts: 49326
Los Angeles, California, US
Studio 144 wrote:
What I really think messes with most people's mind, is that "color" really does not really exist it is just a "personal perception" of the wavelengths of light. I wonder if you and I see red differently? There is no way to really know.
Body Painter
Monad Studios
Posts: 10131
Santa Rosa, California, US
Monad Studios wrote:
No, not divorced from reality at all. Of course the mathematical model corresponds to the reality. That's what the model was built for. digital Artform wrote:
So what is your objection, then? My objection is to treating the model used in digital imagery as if it told the whole story about color and light.
Sure, within the model, "multiplicative" makes sense.
But if one is looking at the phenomena in more detail (wavelengths, absorption and transmission and reflection, the sensitivity spectra of color receptors, etc.) one can see that "subtractive" makes more sense in that realm.
Denying that "subtractive" makes sense here is tantamount to denying that there's anything more going on than what the simplified math model captures.
Body Painter
Monad Studios
Posts: 10131
Santa Rosa, California, US
digital Artform wrote:
Boy, those eggheads. Don't they know light is much more subtle than their silly models? Presumably they don't care. Their models are completely sufficient for the task of image rendering.
Photographer
digital Artform
Posts: 49326
Los Angeles, California, US
Monad Studios wrote:
But if one is looking at the phenomena in more detail (wavelengths, absorption and transmission and reflection, the sensitivity spectra of color receptors, etc.) one can see that "subtractive" makes more sense in that realm. I just looked at it in that much detail. I gave you the ink example, which makes color printing possible, and I showed you what happens when you used the unvarnished equations of Physics to create what is called an unbiased renderer.
Your best bet is to say 'Subtractive is easier to explain to the layman,' and leave it at that.
Any other line of reasoning is a non-starter for you.
Body Painter
Monad Studios
Posts: 10131
Santa Rosa, California, US
digital Artform wrote:
Your best bet is to say 'Subtractive is easier to explain to the layman,' and leave it at that. Thanks, but I'll stick with 'Multiplicative is easier to explain to those who look at everything from the point of view of digital image processing.'
EDIT: Here's the bottom line: you say that the models used in digital imaging tell the whole story about color mixing, and that therefore only the digital-imaging terminology ("multiplicative") is legitimate.
I say that those models don't capture what's really going on, and I gave several examples of things they don't explain, the main ones being non-spectral colors (purple, magenta) and the fact that it takes three numbers, not two or four or seven, to model color in a computer.
I say there are other ways of looking at color phenomena, and that when viewed in those ways "subtractive" makes perfect sense. But someone who can't see things in those ways of course won't get that sense.
I think this is as far as the discussion is going to go.
Photographer
Why Dangle
Posts: 2791
Manchester, England, United Kingdom
My understanding is.............. (admittedly from a zillion yers ago when I worked for a photographer with his own colour lab).
............ that it simply means subtracting colour.
The printing was always done in values of CMY and when you had a colour cast to a print eg. magenta you would subtract magenta to correct it.
The other way way is to add it's opposite colour but because the filters were CMY you always subtracted the colour of the cast not add it's opposite.
Photographer
digital Artform
Posts: 49326
Los Angeles, California, US
Monad Studios wrote:
I say that those models don't capture what's really going on, and I gave several examples of things they don't explain, the main ones being non-spectral colors (purple, magenta) and the fact that it takes three numbers, not two or four or seven, to model color in a computer. Here is something I've been bugging the 'Painter' people about for years:
Suggestion:
MULTISPECTRAL COLOR MIXING
Internally converts 3D color to a higher dimensional color space with 9 or even 12 spectral samples. Down-converts back to 3 samples for display purposes. In this world, colors that appear pure blue and pure yellow mix to green because the software gives pure-looking blue and yellow some hidden green qualities in the higher dimensions, and the green from each pigment survives the mix. ** possibly exists in the Artists Oils. Possibly does not exist in Painter.
(incidentally hi quality renderers compute in more than 3 color dimensions. They sample wavelengths at many more intervals)
Photographer
digital Artform
Posts: 49326
Los Angeles, California, US
Monad Studios wrote:
non-spectral colors (purple, magenta) There are two kinds of yellow, yes?
There is red + green which a prism can break apart into red and green again.
And there is actual yellow at around 575 nanometers that a prism cannot decompose.
Same with the 'non-spectral' colors. They are represented as combinations of other colors. So what?
Photographer
digital Artform
Posts: 49326
Los Angeles, California, US
Why Dangle wrote:
My understanding is.............. (admittedly from a zillion yers ago when I worked for a photographer with his own colour lab).
............ that it simply means subtracting colour.
The printing was always done in values of CMY and when you had a colour cast to a print eg. magenta you would subtract magenta to correct it.
The other way way is to add it's opposite colour but because the filters were CMY you always subtracted the colour of the cast not add it's opposite. I address that exact example in post #2 of this thread:
digital Artform wrote:
![https://www.digitalartform.com/assets/greenFilter.jpg]()
Using a pale green filter over a magenta scene doesn't subtract magenta, it multiplies by green.
![https://www.digitalartform.com/assets/ccorrect.jpg]()
...which makes sense when you think of the colors in zero to one space, and not 8-bit 0 to 255 space.
Body Painter
Monad Studios
Posts: 10131
Santa Rosa, California, US
digital Artform wrote:
MULTISPECTRAL COLOR MIXING
Internally converts 3D color to a higher dimensional color space with 9 or even 12 spectral samples. Down-converts back to 3 samples for display purposes. In this world, colors that appear pure blue and pure yellow mix to green because the software gives pure-looking blue and yellow some hidden green qualities in the higher dimensions, and the green from each pigment survives the mix. ** possibly exists in the Artists Oils. Possibly does not exist in Painter. That's an interesting possibility! But wouldn't the up-convert would only capture the info that was in the original 3D representation; and wouldn't the down-convert lose whatever extra information had been incorporated into the the higher-dimensional representation?
What if the the higher-dimensional representation were maintained all the way through to display? I've been experimenting with something similar using six or eight different colors of LED.
|
