Value in Light and Shadow

[MChesleyJohnson]

Here's a quote from Bongart that I've heard from other instructors:

Nothing in the light is as dark as the shadow. Nothing in the shadow is as light as the light. In other words, you can have all the detail you want in the lights and all the detail you want in the shadows, but the lights should stay light, the shadows dark. The two should never mix.

True or not?

I've observed that this isn't true. I've seen the white, shadowed side of a house brighter than nearby, dark objects in sun.

I can see where this dictum can be useful in illustration - it certainly removes any confusion in the viewer's mind.

[WFMartin]

I've heard this dictum several times, and when one begins to dissect just what is meant by it, it begins to be less and less understandable.

A proper selection of a middle value is truly all that is needed. Then, all this "lights-within-dark-areas", and "darks-within-light-areas" concern rather takes care of itself, automatically.

Buy yourself one of these grayscales, or something similar


A neutral density of about .65 was a standard for middle value that I worked with for over 40 years, reproducing all sorts of images, and it seems to work equally as well, when applied to the creating of fine art.

If one can effectively reproduce all the light value steps between the left and the middle, white triangles on this grayscale, and, in turn, reproduce all the value steps between the middle triangle, and the right triangle, one can pretty well reproduce any subject in front of him, with not much concern about all this "lights in shadows" and "darks in light area" stuff. In real life, and nature, that takes care of itself quite well, and all we, as artists, need to do (for the effective creation of a realistic scene) is to reproduce it.

When painting, it is important to accurately place your paint, mixed to these 3 values (those marked by the white triangles) in the areas of the subject that actually require it. If you begin by doing this quite accurately, the remainder of the values will all fall into place, because you will only need to compare every further tone you mix up to these three.

Your choice, from that point on, is merely to determine whether a given value that you are going to be painting is lighter or darker than these values that you've already placed.

I realize that many artists use some sort of a 10-step, self-prepared step tablet, but there are by far many more values in a natural subject than 9 or 10. A .65 density step makes a very good choice of a middle value, and these 9 or 10-step scales that artists use do not seem to indicate any sort of density values on them. Without such a measurement, I'd have a tough time determining which one of these steps actually should be considered a middle value. (A value that numerically lands halfway between the two limiits on those 10 step scales is not, necessarily a middle value-it is merely a value that has "landed" halfway between the two extremes, and this is not usually representative of a true, functional, middle value.) Notice how, on the above grayscale, the "middle value" (.65 density step) is truly NOT in the logistical "middle of the density steps, at all, but is much closer to the highlight area. This is as it should be, for proper reproduction of an original scene or reference photo.

This placement of the middle value can of course be modified, but then so will the subject you're trying to reproduce.

For accurate reproduction of any subject, it is almost always more appropriate to favor the accuracy of reproducing the values from highlight to middle value, and to sacrifice, if needed, the middle values to shadow values, than to do it the other way around. It has to do with what the human eye is most capable of seeing and detecting, and where the human eye needs the most help in detecting value differences.

Here's a link to my tone/value management workshop on Wet Canvas, that I did quite some time ago. Many approaches are explained in this workshop, and the proper use of a middle value is one of them.

http://www.wetcanvas.com/forums/showthread.php?t=362343

Bill

[DAK723]

I've always interpreted this as "on the same object". In other words, nothing on the shadow side of an object can be as light as something on the light side. Often you will see reflected light on an object or a persons face or skin depicted as light or lighter in value than values on the lighted side, which is to be avoided by the above dictum.

Don

[WFMartin]

Originally Posted by DAK723

I've always interpreted this as "on the same object". In other words, nothing on the shadow side of an object can be as light as something on the light side. Often you will see reflected light on an object or a persons face or skin depicted as light or lighter in value than values on the lighted side, which is to be avoided by the above dictum.

Don

Well, THAT surely makes the most sense, I think.

Bill

[Wrichards]

Great explaination guys, this is somthing im always wrestling with, but sooner or later it will click

[Einion]

Originally Posted by DAK723

I've always interpreted this as "on the same object".

Agreed. The shadowed side of an egg might be lighter than the lit side of a piece of coal, as an example of a possible exception if it were not assumed to be a single object.

It's not going to be 100% correct in every possible circumstance* but there are certainly much worse rules of thumb bandied about in painting.

*One could imagine reflected light from polished metal falling onto the shadowed side of something that provides (in a small zone) light of the same intensity as is falling on the lit side.

Einion

[nit-wit]

Francis Crick (of DNA fame) wrote an interesting book about the brain. Included was a chapter on seeing. To cut a long chapter short, the brain, apparently, compensates greatly for the eye's deficiencies. That is you're not always seeing what you think your seeing. Make sense? The best example that I can remember refers to the blind spot. This is where the optic nerve leaves the retina, and thus there is no retina at this point. No retina, no vision. Why don't we see this blind spot? Because our brains 'trick' us into overlooking it. In fact it's possible to make the tip of your finger 'disappear', by aligning it with your blind spot.

A lot of art training is concerned with helping us to see what is there and not what we think is there.

And when we stop to think about visual representation, it becomes apparent that it's plainly full of flaws/contrivances. Which to cut my post short, is why Cezanne developed from an unremarkable dauber to a lynch pin of modern art.

Andrew

[MChesleyJohnson]

The point is that we don't want to break up the integrity of shadows; a shadowed mass still must read as a shadowed mass. This goal is something to keep in mind as we paint, but the "rule" I note shouldn't be taken as an absolute. There are definitely times when it doesn't hold.

Thanks for the .65 density scale, Bill. It's a concept I'm not familiar with, being a painter and not a photographer. And thanks for the link to your online value workshop!

[WFMartin]

Originally Posted by MChesleyJohnson

The point is that we don't want to break up the integrity of shadows; a shadowed mass still must read as a shadowed mass. This goal is something to keep in mind as we paint, but the "rule" I note shouldn't be taken as an absolute. There are definitely times when it doesn't hold.

Thanks for the .65 density scale, Bill. It's a concept I'm not familiar with, being a painter and not a photographer. And thanks for the link to your online value workshop!

I believe that the real crux of the matter is simply that if we could effectively interpret those values that are important to our reproduced art, by looking at the subject, we should do that, rather than relying upon these extraneous, one-step-removed tools. All these grayscales, color guides, Munsell color chips, and the like are only for the isolating of these various values and colors that we can't easily discern from the "copy", itself.

Quite often a simple "spotter" such as I described in my little tutorial is sufficient for isolating those hard-to-see value and color areas. But, again, if you are able to easily interpret the values in your original (scene or photo) without being misled by a degree of inherent optical illusions, it is always best to paint "what you see" from the subject, itself, if possible. Doing so puts each area into more accurate context with the rest of the subject.

To use some intermediate "tool" often serves no more purpose than to create another opportunity for possible error, in interpreting the value, and mixing the paint to match it. But, such a tool can be helpful in SOME cases, so I believe it is good to be at least familiar with their uses, as long as we don't find ourselves becoming such a slave to such tools that we lose all concept of the art, itself.

Bill

[MChesleyJohnson]

Indeed, Bill. I work mostly from life, and teach workshops on painting from life. At one recent workshop, I discussed value with students. While visiting students at their easels, I walked around the property with a value finder and made a curious observation about value finders: you can't trust them.

For example, here's a paved driveway as seen through a value finder. The red circle indicates the value of the pavement.



Now, here's exactly the same scene. You'll note that the value of the pavement seems to have changed significantly.



Why? Well, neither the light on the pavement nor the pavement's intrinsic value has changed. What has changed, however, is the incidence of light on the value finder itself. The first photo shows the value finder in full sunlight; the second shows it tilted into near-shadow. I found I could get the pavement to read as any value I wanted just by tilting the tool a little more or a little less. In fact, it was almost impossible to avoid tilting the tool and getting erroneous readings.

Tilting the tool changes the angle of light hitting the cardboard, allowing more or less light to illuminate it.

Two properties are at work. As the amount of light hitting the cardboard increases, the value scale becomes brighter, thus making the pavement seem darker. Also, your pupil constricts as it adjusts to the stronger light. (The camera adjusts, too.) This allows less light from the pavement to enter your eye, and again, the pavement seems darker.

The best value finder is your naked eye. You just have to get better at comparing values.

[Daniel_OB]

All theory I saw always talk about only one single light source but never mentioning the same. I think it is a point of all confusions.
We have one light source (not 100%) at night in the room with only one bulb.
Outside, at day light, we never have only one one light source. The core of the problems is densitometry which painters do not undestand very well.
Imagine Sunny day and a tree. Object exposed to the Sun is lighted with light from the Sun and from the sky too (weak). Object in shade of the tree is not lighted from light of the Sun but only from the sky. If we wish to put this on the canvas, we get a problem with not enough contrast. The one problem here is that our eye accomodate to different light sources very fast so we actually do not notice that difference in contrast so easy.
It is important in using "gray scale card" to hold in main that it is valid for one single light source, NOT two or more. It tell us about reflectance of the objects, going from 3%-black velvet, to 95%-white paper, and in the middle being that 18% reflectance.
Also object in shade of the tree is lighted only from "sky light" and so being "blueish".
Keep in mind that buildings also are light source, but being week compared to direct Sunlight, but important when thr object is in the shade of the building.

ALWAYS TAKE CARE ABOUT ABOUT ALL LIGHT SOURCES IN THE SCENE. Dark object can be much ligher on strong light source than white object in low light source (in the shade e.g.) DO NOT MIX THAT THINGS. Eyes only are not very useful here, or also is valid, eyes have to be so well trained to use correctly in such condition.

It is really benefit to have and to know how to use photgraphic lightmeter.

Because paiting is flat, two-dimensional, it is always exposed to only one light source, while object shown on it was exposed to multiple light source.When we convert that outside multiple light source to one single light source we have 3 to 1000% (e.g.) reflectance. All that objects as painted shows reflectance 3-95% (say Titanium white). This is a point of confusion to all painters I saw.

3-1000% reflectance means the brightest object is 8 times ligher than darkest.
3-95% reflectance means the brightest object is 5 times ligher than darkest.

Hope it will help.

[V-A]

Lately i have been conserned about the value structure in my paintings and it has helped that i have focused on this matter. Before that i was conserned about edges and colors.. Even after we break it down to simple values (without over modelling, hopefully) we have to be sure that the part we are painting will match the whole. Its WAY too easy to go too bright in the shadows. This "rule" usually propably work, but then again we have to trust your eyes and be true to what we see. Squinting down is a way to see where it all belongs in the bigger scale.
So the hardest thing is, in my opinion, is to keep the focus and to be mentally awake about whats the next color/value that i am going to put down.

[Rosic]

Some great discussion going on here... I'm learning a ton... 5 stars!

[Einion]

Originally Posted by MChesleyJohnson

You'll note that the value of the pavement seems to have changed significantly.

Very valid point you're making about the importance of the angle of the incidental light on a value scale but I think I should point out that the pavement in the two photos is two different values, doesn't just seem to be (because of the camera's automatic exposure I presume).

This should be obvious to most viewers, but just in case it wasn't I didn't want anyone who sees the top one as darker and after reading the explanation to then think it was only because of simultaneous contrast due to the lighter card in the centre of the image.

BTW Michael, does the card come with instructions on how to hold it (i.e at the same angle as the plane of the canvas)?

Einion

[AnnieA]

I agree with Bernie - this is a 5-star thread. I've recommended it. There's a lot to study here. Thanks Michael and all the other contributors.