Munsell Cubes - A Value Painting Exercise, Part Two
This post follows on from last Saturday's, which was the beginning of some simple exercises with cubes painted in the Munsell neutral scale. Simple in that all I've done so far is paint three cubes, but I have a feeling that it was one of the most useful exercises I've done yet with regard to tone.
Having come up with a way of working out the relationships between the tones of the white cube, the most obvious next step was to put all three cubes together in one painting. That will extend the range further, and dramatically increase the number of tones to control. I planned to use the same system as I did for the white cube, but to be honest I was too excited and steamed straight in with what you might call guesses for the tonal relationships. I've never been good with numbers.
At least they were educated guesses though, I already knew at this point the steps needed between the values of each plane of the three cubes.
Apart from dropping the tones on the light and half tone planes of the grey cube, it should be a simple matter of adding black to the values from the painting of the white cube.
The values of the white and grey cubes hold pretty true to Loomis' form principle I think, the relationships between the planes of those two cubes are the same: one step down from the light plane to the half tone (the top plane), and three steps from the half tone to the shadow plane. The relationship between the shadow plane and the top plane of the black cube is compressed though, exacerbated by the fact that ivory black paint is higher than a value 1 in reality. On the other cubes, there's three steps between the top plane and the shadow plane. On the black one, it's only 1.5. That can't be right. Loomis would rap my knuckles for departing from the form principle there, I've compressed the dark end of the scale.
Despite that, I think these cubes came out ok, although I'm less convinced by the white cube. Perhaps because I'd used my lightest light on the light plane, so couldn't add the little white highlight on the top edge of that plane, which the other two cubes have. That might seem a bit obsessive, but these exercises have to be related back to the real world of painting pictures at some point. That's another small thing I'd like to work out in paint before going too much further.
What's interesting though is that on the final painting, I lightened the light and half tone planes on the black cube, it seemed to need it. Perhaps I was feeling what I'd missed in the numbers. The relationships are still compressed on that cube when compared to the other two though. I can't help wondering what this painting would look like with the tones worked out again, allowing for that, so there'll probably be another couple of versions of this set up.
The point about the missing highlight on the white cube is pretty well shown by this painting I think. It all seems to work pretty well until you get to the highlights on the saucer. Someone emailed me once to say they liked the picture, mentioning that when they originally saw the painting in a small thumbnail view, they thought it was a photo on which the highlights had been painted. That's pretty funny. I take it to mean that most of the values in the painting are pretty convincing, but the highlights aren't light enough in relation to the other elements to work. A lot of my sketches from last year are like that, with big problems in the tonal balance. That's the kind of thing I want to resolve, if I can.
Despite the departure from strict adherence to the form principle, on the black cube, I think the values on the cube painting have come out pretty well. Each cube seems to me to be reading right against it's neighbours in terms of values. To me, the yardstick for whether or not this way of compressing the tonal range helps is whether it will make for a better painting, whether it helps to create a more convincing impression of the light. I'm pretty encouraged by the first go at this, and am immediately itching to try some colour cubes.
I'm not alone...
My painter friend Marsha is painting Munsell cubes too, this is one of hers. Marsha's done the three cubes exercise though, and the 5YR cubes - cubes of different value and chroma in the hue of 5YR in Munsell (orange, to us mortals). She started on a blue cube because she just couldn't wait.
I know just how she feels. This exercise has fired me up, and I'm immediately thinking about other exercises I can do with the cubes and spheres. After doing a few colour ones and getting at least some grasp of chroma (the second of Munsell's three elements of colour, hue being the third) I can perhaps stick some bits of fruit in and see what happens.
This here is one nicely painted cube. The tones look completely convincing to me, and are pretty close to the value relationships I got for my cubes, I think. To my eye, this little blue cube has a palpable reality and solidity to it, and that's what these exercises are all about.
Amanda, another painter friend, is also doing these tone exercises, but she's gone clear. She's got cones too, this is her Munsell set.
Adding cones strikes me as a pretty good idea. It also harks back to Cezanne's idea that all nature could be reduced down to the basic forms of the cube, sphere and cone. I'm not a big fan of Cezanne personally, but drawing of geometric shapes has been featured in some well respected courses. The Boston Museum School (I think) had an exercise in drawing these three shapes.
Amanda also gets extra appreciation from me for being civilised and using the word 'tone' instead of 'value'. Just to confuse matters, I use both words interchangeably these days. See how the Internet can confuse you.
I've also been corresponding with my painter friend Robert about this tone problem. We think alike, in a lot of ways, and have been mulling over exactly the same problems of tone. In particular, Robert has listed a lot of possible approaches to this 'range problem', like compressing the tonal range as I've done here, or accepting the upper limit and averaging lights, or doing the same at the bottom end of the range with the darks.
These approaches are mentioned by Harold Speed. Me and Robert are down with Mr. Speed, as all clear thinking people are. Robert pointed me to a couple of pages of Speed's "The Practice and Science of Drawing" which are so relevant here I'm going to quote them. He's talking about tonal relationships in his chapter on 'Unity of Mass'. In particular, he's talking here about the fact that paint can't match the tonal range of nature, and looking at various approaches to the problem. To be fair though, these passages deal with catching the more brilliant effects of light in nature with paint. Harold does say that the range of paint is sufficient for catching normal interior light. My white cube tells me otherwise, but perhaps I'm being pedantic.
"In all quieter aspects of lighting this range from black to white paint is sufficient. But where strong, brilliantly lit effects are wanted, something has to be sacrificed
In order to increase the relationship between some of the tones others must be sacrificed. There are two ways of doing this. The first, which was the earliest method adopted, is to begin near the light end of the scale, and taking something very near pure white as your highest light, to get the relationships between this and the next nearest tone, and to proceed thus, tone by tone, from the lightest to the darkest."
Harold points out that if you do this you'll hit your darkest dark, black, pretty early. What you're effectively doing is shifting the tones you see down the scale and running out of space at the bottom. He cites Rembrandt and his dark paintings as an example of this. He then talks about the approach of starting with the lightest light and darkest dark, and progressing from there. I've used that approach, on Harold's advice, and I think it's pretty sound. Then he describes the opposite approach to the first one:
"The third way, and this is the more modern, is to begin from the dark end of the scale, getting the true relationship felt between the greatest dark and the next darkest tone to it, and so on, proceeding towards the light. By this method, you will arrive at your highest light in paint before the highest light in nature has been reached. All variety of tone at the light end of the scale will have to be modified in this case, instead of at the dark end in the other case"
Harold says that this will result in a more brilliant picture, and is effective for representing sunlight. It strikes me that it would be very interesting to try the first and third approaches, using the Munsell tags, on the white cube. The second approach he mentions, placing the lightest light and darkest dark first, is a practical way of compressing the tones, since you'll be working from both ends to the middle. More like what I've done here with the aid of the Munsell value tags. I can feel two more white cube paintings coming on.
I'm pretty sure that the approach of finding the full range of Munsell steps in the subject, locating values and then compressing the range is effective. I've also got a strong feeling it's possible to do this by eye, maybe through experience, sheer talent, or both, but without the safety net of the Munsell tags. For me though, for now, this is a way to practice and to develop this faculty of judging tones. I'm also hoping that a close study of tone will help me to understand better how light works in nature, and how better to translate it to a picture. Having a good command of that, and putting it in the service of the creation of something with a bit more feeling than a painting of grey cubes, is the eventual goal.
Which ever way you look at it, there's no doubt that you can't hit the entire range of tones in nature within the range of paint. I didn't even have a wide enough range to hit all the values of a white cube on a grey cloth. So some kind of transformation has to be done to paint a representational picture of nature. That thought struck me some time ago, but it's nice to have my little Munsell tags bear it out.
These Munsell Value Exercises are posted in sic parts. The links below will take you to more posts with more exercises: