Ambient Lighting and Colour Temperature

[Alex Chow]

Well, amidst the drama this past week, I bring more questions. Hopefully this is the right time to do so, however. This time, it pertains to the secondary light sources.

a) If ambience is the global lighting in a region, does that mean its strength is ultimately determined by the strength of the main lighting and any secondary light sources bouncing off a billion surfaces in an area? Does the amount of light/mid/dark values occupying a picture space indicate the level of ambience?

b) How do people determine the hue of an ambient light? This is also tied into the question of where ambient lighting comes from. If a picture is predominantly a certain hue, can we automatically assume the ambience is that certain hue? I ask of this because the definition of ambience implies it's just light bouncing off a billion surfaces until it hits every plane. If it bounces off a billion things, what will the resulting hue be anyways?

c) Not really ambience, but if the sky shoots down light at many many angles, can we assume any plane that remotely faces upwards will have a sky specular highlight (though some of the upward planes will reflect more than others)? This is, of course, not including other objects causing "disruption" or stronger light sources shrouding those areas.

I hope these are worthwhile questions to ask and hopefully the questions will help people, more than just me.

Thank you in advance.

[dpaint]

a) Ambient light is based on the strength of the primary light source in an area. Anything that has light falling on it becomes a source of light itself taking some of the energy depending on its surface qualities.

b)Ambient lights hue is a product of local color of objects in the light and the strength of the primary light. Ambient light must be dependant on the airquality and strength and color of the primary lightsource combined with local color of other objects in the light. You can prove this by placing two objects in full light one in front of the other; if the first object has a local color, or the primary light has a strong local color, it will alter the color of the shadow of the second object. In space there is no ambient light because there is no atmosphere.
c) When outdoors the primary lightsource is the sun and the secondary source of light is the sky itself. When there is no direct primary light falling outside, (think of a cloudy day or forest canopy) the specular highlights on an object come from the sky if nothing occludes it from the object.

[Carl Dobsky]

nice to see a good question with a really good response especially what with all the hoo-ha.

I've been thinking about this lately too.

[Craig D]

Isn't the sky too diffuse to have a specular highlight?
Although my confusion might have to do with not specifically understanding
the word specular.

edit: OK I got it confused. Whether the object has a specular highlight or not is
pretty much dependant on its reflective qualities instead.

[PieterV]

I think specular surfaces wil show specularity but not in a hotspot like fashion as the light is hitting it from all angles and therefore bounces into our view at all angles. If you have a strong light source such as the sun you will see a clear specular highlight. Otherwise it would change the temperature and value mostly.
Ofcourse, it also depends on the amount of specularity and the smoothness of the surface.

[Bruce Pluto]

AC


a) Yes

b) Yes

c) Yes, also the further the subject moves away from the highlight the truer the color till it reaches shadow. In the shadow the ambient light will reflect and the color reflection is the color you use.


I hope that clears things up, Bruce

[dpaint]

The color and strength of the same ambient light will appear different under different conditions with a primary light, this is why direct observation is so important for how you approach painting it; most artists overpaint the secondary and tertiary light sources. This is a function of your eyes ability to adjust to conditions. If you move from a bright office or daylight setting to a dark basement there is a period of adjustment for your eyes but the rooms lighting hasn't really changed.
When painting, the longer you stare at a dark area in the shadows or a bright area in the light the more you will see in those areas. The tendency is too overpaint the value and color separations as an isolated passage, but doing so undermines the sense of light for the whole painting. This is why it is important when painting to keep the big divisions of light and shadow in your painting relative to each other and the details are added to those areas without replacing the initial comparisons.

[Elwell]

The color and strength of the same ambient light will appear different under different conditions with a primary light, this is why direct observation is so important for how you approach painting it; most artists overpaint the secondary and tertiary light sources. This is a function of your eyes ability to adjust to conditions. If you move from a bright office or daylight setting to a dark basement there is a period of adjustment for your eyes but the rooms lighting hasn't really changed.
When painting, the longer you stare at a dark area in the shadows or a bright area in the light the more you will see in those areas. The tendency is too overpaint the value and color separations as an isolated passage, but doing so undermines the sense of light for the whole painting. This is why it is important when painting to keep the big divisions of light and shadow in your painting relative to each other and the details are added to those areas without replacing the initial comparisons.

Quoted for emphasis. This is SOOOOOOOOO important.

[Alex Chow]

I think I often paint secondary and ambient lighting too weak. :/ Does that statement imply that the secondary/tetiary lighting is often painted too strongly or weakly? Both?

dpaint, do you mean that it's better to paint the light/dark relationships before the eye adjusts or after?

This thread has been great so far! Thank you!

[dpaint]

For realism you want the initial idea of relative strength between light and shadow, this, more than anything describes the quality of light to the viewer. all other information must conform to these broad masses of color and value or you weaken the affect. Most people exagerate the fill lights, not de-emphasize them.
Anything can be manipulated for style and emphasis but everything must be considered when you deviate from the norm, otherwise you fracture the painting.

[JeffX99]

Can I get an AMEN?! Halleluja Bruthah! That's why painting/studying from life is so important...but I'm just preachin to the choir so...great thread!

[Alex Chow]

This thread has been great Thanks everyone!

But one more quick question

From multiple sources, I'm told white is inherently a cool colour. How true is this? In Alla Prima by Richard Schmid, it implies ANY hue gets cooler (and lighter obviously) when mixed with white. For warm colours, I have no problems with it. I find this hard to believe when it applies to cool colours; how is a blue mixed with white cooler than a saturated blue?

I'm beginning to play with very subtle temperature shifts and something like this may hurt me in the end.

Thanks.

[Téa_Passer]

This thread has been great Thanks everyone!

But one more quick question

From multiple sources, I'm told white is inherently a cool colour. How true is this? In Alla Prima by Richard Schmid, it implies ANY hue gets cooler (and lighter obviously) when mixed with white. For warm colours, I have no problems with it. I find this hard to believe when it applies to cool colours; how is a blue mixed with white cooler than a saturated blue?

I'm beginning to play with very subtle temperature shifts and something like this may hurt me in the end.

Thanks.

yes awesome thread indeed!
to your question, blue mixed with white gets more violet. I know this from digital art experience. Not sure about traditional but as far as I can tell, there's subtle change in temperature when you mix blue with white towards cooler.

[Elwell]

Alex, are you working digitally or traditionally? The cooling effect of white is far more of an issue when dealing with actual paint (it has to do with the physics of pigments), and, on top of that, varies among different whites. In other words, titanium, zinc, and lead whites all shift tints slightly differently.

[Alex Chow]

Alex, are you working digitally or traditionally? The cooling effect of white is far more of an issue when dealing with actual paint (it has to do with the physics of pigments), and, on top of that, varies among different whites. In other words, titanium, zinc, and lead whites all shift tints slightly differently.

I work both digitally and traditionally. I use Zinc White when I paint with gouache.

I'm not so much concerned about the mixing consequences. My main concern is that if the addition of white truly does cool all hues down, it may explain why I'm screwing up temperature situations, especially when I'm establishing a cool light/warm shadow setting. I found it hard to believe when Schmid stated how cool hues with white look more cool than the saturated versions of those hues. My painting instructor also stated this before.

I must be blind because I don't see how this can be true :/ . I was sure that white simply "neutralizes" in that it will cause warms to feel cooler and cools to feel warmer.

EDIT: I think "feel" is a better word to describe this situation.

[dcorc]

I'd suggest that rather than talking about "warm" colours and "cool" colours, it makes more sense to identify the exact nature of the shifts taking place (or a difference, in comparing two colours), and be specific about them. Do you mean that the colour changes (differs in) hue? ..or chroma? ...or value? - or where the effect is a combination of these, then what exactly is the contribution of each of the three to the overall change.


Dave

[Alex Chow]

Well, here's a diagram to clear things up.

The hues get progressively mixed with white, so value and satuation shifts. If Schmid is right, all the 4s should look cooler than the 1s. It works for the "warm" hues but I just don't see it for the "cool" ones. Are others seeing that or do I have it backwards?

Of course, I'm not taking into account how some colours in traditional painting changes in hue as white is added, but I think it's close enough.

[dcorc]

Of what possible use is the term "cooler", here? What meaningful information does it actually contribute?

[Elwell]

Yes, the problem is that "cooler" can refer to both a chroma shift and a hue shift. Also, warm and cool are strictly relative terms, not absolute ones.
Either way, I suspect that what you are worrying about isn't actually the source of your problem.

[Alex Chow]

Of what possible use is the term "cooler", here? What meaningful information does it actually contribute?

Well, I've been working on temperature and its role in lighting environments. The use of the term "cooler" is just so that I can understand the temperature shifts occuring when a hue gets progressively "whiter", ultimately allowing me to see what colours to mix and to avoid when I want a certain temperature shift. I'm asking on the possible temperature shifts when white is used in traditional media or, in the digital realm, decreasing saturation while increasing saturation for a certain hue.

If this is a question deemed unimportant by the community, I have no problems letting this thread die. Sometimes my mind just wanders on...

EDIT: Elwell's above post has a point. I have no need to drag this onwards because it's a much bigger problem than I may have anticipated. Thanks everyone and until next time! I'm going to have to do more studying on this before I make myself look worse.

[briggsy@ashtons]

Alex

I don't think your question is unimportant, it's just that you should switch to a more helpful frame of reference.

Try to take on board dcorc's advice to think instead in terms of just hue, value and chroma. Once you do you'll see that trying to analyse colour problems in terms of "warm/cool" is both redundant and confusing.

In the context of adding white paint, a "cool" shift might refer to the colour becoming more neutral, or it might refer to the subtle hue shift that you see with some pigments of a few degrees, up to at most about 30 degrees, around the colour wheel. For example, adding white paint to ultramarine tends to make the hue less purplish, which I would call making it cooler, but which some people here would call making it warmer (another confusion!).

In the context of a light/shadow setting, on the other hand, a "warm/cool" shift refers to quite a different thing - a hue shift of 180 degrees. If you just think in terms of warm-cool you are asking to confuse these things. Much better to think more specifically in terms of say a yellow vs blue or a red vs cyan light/shadow setting.

[JeffX99]

Hmmm...I have a little different perspective on it Alex. The best advice I've always heard is to keep it simple. So to that end I do think "cooler/warmer" are very valid terms - yes, they are also relative terms but color is relative anyway. Color is a very subjective element between people - at least "cooler/warmer" are easy to understand terms and actually more definitive than talking about a particular hue.

For example - you could tell a student "That purplish, royal blue note has to go more toward a sky-aqua" or somehting like that - they may have different notions of what those "hues" are. Whereas if you just say "You need to cool and push that blue note back and raise the value a bit" it is a more definite, and easier to follow statement. I hope that makes sense.

In your example the 4s are all definitely cooler to my eye. So I guess my advice is to not get too wrapped up in the technicalities - focus instead on observation and push your interpretation to get the results you're after.

[Elwell]

For example - you could tell a student "That purplish, royal blue note has to go more toward a sky-aqua" or somehting like that - they may have different notions of what those "hues" are. Whereas if you just say "You need to cool and push that blue note back and raise the value a bit" it is a more definite, and easier to follow statement. I hope that makes sense.

I agree with you about simplicity, and often fall back on warm/cool terminology when I'm talking to someone about a specific case. But your example actually points out the problems with it, since which blues are "warm" and which blues are "cool" depends on where you're placing the warm/cool axis, and there's no universal consensus on that. And even if you did agree on which blue was the "coolest," then shifting it towards purple, towards green, or towards neutrality would all be "warming" it. So, in your example, I would say, "that blue needs to be lighter, greener, and more neutral." Or, if we were dealing with paint, I might refer to specific pigments. J. Crew-style goofy color names aren't what anybody here means when they say "hue."
(see http://www.handprint.com/HP/WCL/color12.html and http://www.huevaluechroma.com/077.php)

[JeffX99]

You're right - Elwell - because the next question is generally, "How do I gray that color?..." And you talk about mixing in a little complementary color, green, orange, etc. That's the problem with talking about this stuff instead of showing it- and why I was saying keeping it simple helps.

[Chris Bennett]

Interesting inputs from everyone here.
I'd like to pick up on something dpaint touched on regarding the painting as a totality and something Elwell said about there being no absolutes, only relatives in painting:

Bringing a painting together, so that it feels 'right' (and we all recognise that elusive moment when it comes) can be thought of as the point at which you understand the DNA of what is making everything tick within all the painting's elements. At this moment the painting seems to take charge and you willingly fall into its arms whereupon they carry you to the finishing line.
In terms of colour and its function as light organisation in the painting, i.e. luminance, questions of how colours 'turn' or modulate (a purple gradually feeling like its got green in it for example) will always be dictated by the engine governing the colour scheme as a complete entity. How colours behave is dictated by the scheme of the painting as a whole, not by any 'rules' outside it. This is what I mean by the painting's 'DNA'. Problems and difficulties arise when there are two or more DNA systems going on at the same time and the painter has either failed to realise it or, more commonly, unable to decide which one to put their creative money on.

[dpaint]

This is devolving into semantics but there are some things worth observing in all this for future reference.
The problem with the Munsell system is, it doesn't help you when you are working, it does help to build a vocabulary for teaching and study though. I swear by it. Its what Frank Reilly used and I have taught it in my workshops for years, but telling someone your chroma needs to be adjusted by ten percent and your hue five percent, is still an opinion, better to say "cool that color" because the decision then becomes cumulative and relative as opposed to painting it exactly as it is measured by an instrument or anothers observation.
I like to say the weight of all your decisions determine the outcome of your painting. Every stoke every color has to be judged in total and the problem with Munsell is people get caught up in the numbers and that doesn't get you art. This happens to allot of the DuMond Theory students who are also using a Munsell/ derivation for fine art landscapes, and spend most of their time mixing color strings, but their landscapes don't look any better than people who don't use it. The legacy of that system then fails, if it produces no better result in the same time frame.
On the other hand the Reilly method has produced more successful illustrators than any other US method in the last hundred years. Guys like Jeff Watts and Glen Orbik are continuing the method with the same results. So what is different? I think Reilly was more flexible, and so his students often morphed the system Munsell started and produced high caliber painters across many disciplines.
So to wrap up my rambling; Munsell is great but it produces better results when the theory and terminology stays flexible because in reality it is just information and that information has to be used in practice. And in practice
there is an emotional component to art that must be recognised. What is the right mixture of science and emotion is left for us to decide individually with the market place the final arbiter.

[briggsy@ashtons]

Thanks very much for that opinion on Munsell dpaint. There are of course strongly held opinions in either direction from it, so I hope this thread isn't about to "devolve" any further.

Putting aside the idea of "percentages", do you really believe it's better to just say "cool that colour" than to say specifically either "make that cooler in hue" or "more neutral" or both?

[dpaint]

No I don't . As I said, I believe in, and teach Munsell theory, it is the best and most accurate theory out there. But in my experience some very fine artists who have passed through my classes respond to a lesser version than the by the numbers approach, I don't know why, but they do and they have become fine painters. So as I have gotten older, I have mellowed, I start with Munsell and the people that want it, I give it to them, but the people that get that blank hundred yard stare... I find another approach to reach them. Something not so absolute. Afterall it is about getting people to see and be the best artist they can be. It is not about making them think like me. You disagree?

[JeffX99]

I'm not sure if you were asking dpaint specifically but thought I'd share my experience (I think this is all still positive, good discussion - the same as we'd have if we were all sitting around). Anyway, Jim Gurney is the guy who finally got me to go out and start painting - with the advice to keep it simple - that really worked for me and is what I try to communicate to my students.

So for me the answer would be yes - to using both. But I'm talking about fairly limited situations that occur when helping beginning/intermediate students painting en plein air.

I think it is completely appropriate and useful to be as specific as possible - given the level of the student or even peer. One should actually push them a bit and challenge them to grow by presenting them with theory and approach somewhat beyond their current level. On the other hand I think it is best to not over-complicate things, especially to the point where the beginning student becomes overwhelmed.

Just my two cents - I'm sure it works differently for other situations, mediums and digital vs. traditional as Elwell pointed out.

[cdejong]

This thread is awesome. I just wanted to come here and say thank you guys, these kind of threads are what make CA a great place.