PDA

View Full Version : Depth Question With Color


jaymarvin
06-23-2002, 11:25 AM
In order to show depth with color is the object nearest the eye the brighter, or hotter color, and the object furthest from the eye the duller, or cooler color? Or is it the other way around?

Patrick1
06-23-2002, 03:02 PM
Jay, it sounds like you're asking about aerial perspective (it's what causes mountains to look purple or blue as they recede in the distance, for example).

Yes...usually, an object will be cooler (bluer) off in the distance, and get warmer (or more precisely, closer to its true color) as it comes closer.

The farthest objects show little contrast in value, while as they get closer, the darks get darker and the lights get lighter.

This is a very general rule, applying to the same object, under the same lighting conditions.
If, for example, the farther object is in full sun, and the same object closer by is in shadow, the farther object might well be warmer.

An easier way to think about this is to just think that the farther away an object is, the closer its color becomes to the sky color behind it. That's a vast oversimplification, but it's a good starting point.


Wow...a WLS host. I live in Ottawa, Canada and I sometimes listen to WLS at night when reception allows.

jaymarvin
06-23-2002, 05:16 PM
OK, thanks. So if you had two panes, or squares, and they were side by side or one overlapping the other, the one on top would be lighter, right?

P.S. I worked at night on WLS for years. First 10am to 2am, that was about six years ago, then 6pm to 9pm. They just moved me to 9am to 11am, and back again at 6pm to 7pm. You should be able to hear us when winter comes again. If you can real audio we stream at our web site!

Patrick1
06-24-2002, 11:39 AM
It depends on how light or dark the panes are.

If the objects are dark, they get lighter as they recede into the distance; look at the brown portion of the squares below. See how they get lighter off in the distance? And more blue too.

If the objects are light (lighter than the sky behind them) they will get darker as they recede into the distance; look at the white portion of the squares. They get bluer and slightly darker (only because blue is darker than white) off in the distance.

http://www.wetcanvas.com/Community/images/24-Jun-2002/panes.jpg

But this atmospheric perspective usually needs lots of distcnce to become noticable; sometimes miles of distance are needed (unless it's hazy or foggy).

Note: nature is a little more complex than this. You'll often get this 'blueing' effect even under a grey sky. Also, you'll often get a bit of purple to the objects that are a little closer than the bluest ones.

By your question I assume you're not talking about light and shadow, because all this assumes all the objects are under equal illumination.


I'll have a listen your website. But nothing beats
good old radio...(no I'm not a grumpy old man) :)

Scott Methvin
06-24-2002, 11:59 AM
Generally speaking, cool colors recede and warm come forward.

Darks are darker up close, lights get whiter as you move closer.

Texture should also be more pronounced as you move closer.

(ex., Texture in a sky will spoil the illusion)

A famous artist once said, "The first job of the painter is to break the plane of two dimentions." He invented the sfumato technique.

However, like most "rules" in oil painting, feel free to break and make your own.

jaymarvin
06-24-2002, 03:48 PM
Thanks to all! I learn more here every day!

Jay

"a grumpy old man"!

Patrick1
06-24-2002, 06:28 PM
Here's what I meant when I said "nature is a bit more complicated than that". The first picture I showed in my previous post is more of a fog/haze effect; the color of the object merges in a linear way with the sky color behind it as it recedes into the distance.

Here's the slight twist that nature throws in. With mountains, for example, there will be some purple introduced (as I mentioned before) somewhere between the blue and the foreground, and the distant blues can be surprisingly dark and intense before fading to the sky color. This is where the idea of distant objects looking cooler really has a lot of truth and impact to it.

So as you recede into the distance, hue change (towards blue) seems to occur sooner than value change; at least this is what I've noticed:

http://www.wetcanvas.com/Community/images/24-Jun-2002/panes3.jpg

If you comapre the two pictures, I think most people will agree that the first looks like fog or haze, while this one looks more like the color of receding mountains on a clear, sunny day.

jaymarvin
06-24-2002, 06:48 PM
What I'm trying to do is put a two dimensional effect on one dimensional canvas aka the Hanns Hoffman theory. So if I have two panels next to each other, or off set how do you get that depth? Is the panel closest darker in order to get that effect?

Einion
06-24-2002, 11:25 PM
Jay, it sounds like you're basically trying to achieve a trompe l'oeil effect, in which case atmospheric perspective rules probably don't apply. If I understand the effect you want to achieve correctly, a lot depends on the colour(s) you intend to use, the size you're painting and other factors - for example are you going to use any edge shading or just use flat colour? If you are using shading or other lighting effects you could pretty much use any colours and still have it work visually.

If you have something like Paint Shop Pro you should be able to knock up some images in no time that will show you what works best for your specific goal. If you don't have any painting programs just do some small sketches in your chosen medium, hold them at arm's length and judge for yourself. Oh BTW, I think you'll find it's a three-dimensional effect on a two-dimensional canvas :D


Patrick and Scott posted some good stuff here I just want to empahsise a little. The important parts of aerial perspective are that value tends to lighten, contrast gets compressed and colour moves towards blue the further away something is. Distant object tend to look bluer, even under a grey sky, because longer wavelengths (i.e. closer to red) are scattered more easily than shorter wavelengths which is exactly why light under water becomes blue so quickly. In painting terms we can use this, and other tricks, to push and pull pictorial elements - more texture and/or distinct brushstrokes, higher chroma and contrast in the foreground elements, lighter value and cooler hue with increased distance.

However I think we should move away from the idea that cool colours recede and warm colours advance. Like so many visual effects, value, contrast and placement are all more significant than colour "temperature". An example, sunset-tinted clouds over the ocean don't look closer than the water do they? This was discussed here recently in <a href=http://www.wetcanvas.com/forums/showthread.php?s=&threadid=31563>this</a> thread.

Einion

bruin70
06-27-2002, 07:15 AM
it all depends. what if you were painting a landscape, and the day was filled with clouds that shadowed the landscape except for a small patch of sunlight breaking through the clouds in the distance. :) just thought i'd throw that one in.

here's one way to answer your question. take a red ball and view it on a very GREY foggy morn at different distances. or view that red ball beneath the surface of the BLUE sea at different distances.

......what happens to the coloring of the ball?

color loses its intensity the further back it goes.

now, you can use the formula that domer suggests. it is easy to visualize and straight forward. it is a formula that works best outdoors because atmosphere plays a strong roll in what occurs in far distance when the "air" gets "thicker".

however, all you really have to do is stick with the concept of color losing intensity, and you'll be ok. indoors, for instance, the atmosphere is negiligable. and if you used domer's formula here, you would be creating the unwanted effect of thick atmosphere. so chroma intensity and contrast works best here,,,,,and a good overall rule in general.....{M}

bruin70
06-27-2002, 05:31 PM
Originally posted by Scott Methvin
He invented the sfumato technique.



i order sfumato ice cream all the time when i go italian. love it.:)

bruin70
06-27-2002, 05:45 PM
Originally posted by jaymarvin
What I'm trying to do is put a two dimensional effect on one dimensional canvas aka the Hanns Hoffman theory. So if I have two panels next to each other, or off set how do you get that depth? Is the panel closest darker in order to get that effect?

you mean you're trying to get a three-D effect on a two-D canvas, yes? do you have two different canvas panels next to each other, or are you PAINTING an image of two different panels on one canvas?

if the first, try domer's suggestion,,,,,only substitute the color of the wall for the blue he uses to fade the squares. if it is the second, fade one "panel" with whatever color you decide to use for the background of your canvas. ie, if your canvas is red, then make one panel fade by using the red as domer used his blue......{M}

jaymarvin
06-27-2002, 05:51 PM
Yes! I'm trying figure out how to paint two panels next to each other on one canvas. Does that makes sense? I'm doing this from work so I'm in a hurry!

jay

bruin70
06-27-2002, 06:31 PM
then the domer approach makes sense for you. however,,,how much of the background color you use, blue in domer's case, depends on how much you want to drop the panel into distance. domer shows the full range of extreme extreme extreme distance. you don't have to go that far. just take your color to domer's mid range sample, because it sounds like you only want to drop a panel back a bit. in fact, the second or third square sample would do. domer's extreme end is for portraying mountains 40 miles away.

if you go to too much extreme, you will lose the effectiveness of your illusion. the illusion will be best served in a tighter range of chroma and contrast because what you want is for viewers to COMPARE the closer panel to the farther one. the comparison of the two side by side will be enough for you. you'll be giving the viewer parameters of which they can make logic sense. and besides,,,,,if one panel were far in the distance, it would have to be considerably smaller in size as well.....{M}