Proportion Experiments: Head

[manfredkooistra]

Unfortunately I had to remove the images for copyright reasons.

[IMBass]

The one on the right looks normal and better.

[Maxine Schacker]

I like them both equally. They look like almost identical twins.

[IanE]

Right.

[ZenzybaR]

Right looks better but left looks more "normal" IMO.

[Suira]

The one of the right seems more normal and armonic to me.

[drd]

The one on the right.

[rinjii]

i'll say the right one- the left just seems too unaturally round.

[georgi911]

The left is the normal one, and the right is more beautiful.

[Leonor]

Right (from my PoV) is more beautiful, left is the normal one.

[dose]

The use of photos and Photoshop are mucking up the validity of your test. In the second one I'm reacting to loss of contrast, compromised values, and slight mistakes in the placement of the features mixed in with the proportions you've changed. If you're trying to be scientific, you haven't held the other variables constant.

And anyway, you already have the answer to the problem- if it doesn't look right when you draw it a certain way, don't draw it that way. If you can already recognize that it's better or worse one way or another, you don't need to get out the ruler and figure the numeric proportion unless you're doing some kind of scientific study. It's all going to go out the window when you're drawing someone who's not facing directly forward...

[manfredkooistra]

Hey dose, I am sure you read that I recommend to squint your eyes and blur the images to even out the "faults" you mentioned. Contrast and values should not be a problem that way.

If you find the features have been placed wrongly, then that is exactly what I want to show: that the schema or system developed by Andrew Loomis is faulty. Thank you for verifying my opinion on that.

But basically I'm not trying to be scientific. I am learning to draw human beings, and among many other things I use some books. All of these books (and all of the teachers I have seen DVDs of or visited courses from) teach some kind of proportional schema. This thread is not meant to discuss the validity of such a schema, but if everybody teaches it, there seems to be some use for it. Personally, I am almost convinced that pure life drawing, without any abstraction, is the best way to learn to draw figures, but it is certainly the longest path and the most difficult. For those of us who want to learn to draw figures from their mind quickly (for comics, for example), a schema can help. But of course we want to learn one that provides us with the proportions that we feel "real" or "average" or "beautiful" people have (and that is flexible enough to allow for the creation of un-real, non-average, ugly people as well). Since the books of Loomis are praised to be the best more often than any other book, I picked them up and worked with them -- and found many faults and inconsistencies in them. Some of his ideas I find helpful, others I need to adapt to my own taste. Strangely enough, while the faults of Hogarth are always mentioned when his books are, almost never does anyone criticize Loomis, although his mistakes are obvious and urgent for anyone actually working with his books. Mabe no-one ever really works with them?

Now, constructing a head and face in a drawing, it is very hard to see and understand what might be wrong with it. The drawing is usually not exact enough in detail to allow for that kind of judgement. After all, the features in real people are very variable, too, so why should a drawing look wrong, when the eyes are too far apart or too low? The same "aberration" does not look wrong in a real face.

So the only method to test a schema for proportions is with a real face. Only if you change a real face to fit the schema, can you see why it is wrong. That is what I tried, and I share this experiment with you. That is all there is to this.

If you find that thinking about proportions in a face facing forward is stupid, because the proportions will change when the head turns, then proportions are not for you. Other artists find them helpful. They are able to think three-dimensionally and turn the schema along with the head in their minds (or construct it using the laws of perspective).

Finally, I do this only because I enjoy reflecting and analyzing what other people prefer to take as is. I like to question tradition, if I find it does not do what it promises, and Loomis does not do for me what people claim he did for them. He achieves other things, and for these I do not want to simply throw away his books. I want to find out, name and correct the faults and thus be able to use the rest.

But you may, if you want to, explain, how you draw, what proportion and abstraction means to you, how you use it or why you don't use it etc.

[armando]

The proof is in the picture. I've got the drawing heads book open and the manipulated photo matches it except for the eyebrows being a little high. If you conducted the test with an unattractive man on the right, and a more polished Loomised guy on the left I think the Loomised guy would generally win, but I could be wrong. A Loomised head would definitely beat a Hogarthed head. I would like to see Loomis paint the original head, now that would be the real comparison. In Creative Illustration he actually shows just that, and you can see how much more artistic and beautiful the painting is in comparison to the photo. He harmonizes all parts of the picture, and contrasts where he needs to, in the service of an idea.
Loomis was a great artist, there's no denying that. The difference between him and Hogarth is that Hogarth uses his concept of the head throughout his book, Loomis uses his concept only as a quide to draw real people's heads either from life or photos. In the beginning pages of "Drawing the Head and Hands" he shows a concept of squeezing the skull in order to get other head types, and gives a method of building different types of heads out of different forms. He encourages artists to get models, and reference, to make the head as believable as possible. At one point he even says that if it comes down to it - trace. There's no getting around the fact that Loomis drew more appealing heads than Hogarth, his women are way more beautiful than hogarth's trannies. Loomis says everything Hogarth says and more, and his pictures are better examples to copy. Hogarth is very idosyncratic.

"After all, the features in real people are very variable, too, so why should a drawing look wrong, when the eyes are too far apart or too low? The same "aberration" does not look wrong in a real face."

We are artists, not recorders of facts. So many people have said basically - art is poetry. Everything we put in a picture serves a purpose, to communicate an idea. If something doesn't comunicate that idea it doesn't belong in the picture.

[Suira]

Hey dose, I am sure you read that I recommend to squint your eyes and blur the images to even out the "faults" you mentioned. Contrast and values should not be a problem that way.

If you find the features have been placed wrongly, then that is exactly what I want to show: that the schema or system developed by Andrew Loomis is faulty. Thank you for verifying my opinion on that.

But basically I'm not trying to be scientific. I am learning to draw human beings, and among many other things I use some books. All of these books (and all of the teachers I have seen DVDs of or visited courses from) teach some kind of proportional schema. This thread is not meant to discuss the validity of such a schema, but if everybody teaches it, there seems to be some use for it. Personally, I am almost convinced that pure life drawing, without any abstraction, is the best way to learn to draw figures, but it is certainly the longest path and the most difficult. For those of us who want to learn to draw figures from their mind quickly (for comics, for example), a schema can help. But of course we want to learn one that provides us with the proportions that we feel "real" or "average" or "beautiful" people have (and that is flexible enough to allow for the creation of un-real, non-average, ugly people as well). Since the books of Loomis are praised to be the best more often than any other book, I picked them up and worked with them -- and found many faults and inconsistencies in them. Some of his ideas I find helpful, others I need to adapt to my own taste. Strangely enough, while the faults of Hogarth are always mentioned when his books are, almost never does anyone criticize Loomis, although his mistakes are obvious and urgent for anyone actually working with his books. Mabe no-one ever really works with them?

Now, constructing a head and face in a drawing, it is very hard to see and understand what might be wrong with it. The drawing is usually not exact enough in detail to allow for that kind of judgement. After all, the features in real people are very variable, too, so why should a drawing look wrong, when the eyes are too far apart or too low? The same "aberration" does not look wrong in a real face.

So the only method to test a schema for proportions is with a real face. Only if you change a real face to fit the schema, can you see why it is wrong. That is what I tried, and I share this experiment with you. That is all there is to this.

If you find that thinking about proportions in a face facing forward is stupid, because the proportions will change when the head turns, then proportions are not for you. Other artists find them helpful. They are able to think three-dimensionally and turn the schema along with the head in their minds (or construct it using the laws of perspective).

Finally, I do this only because I enjoy reflecting and analyzing what other people prefer to take as is. I like to question tradition, if I find it does not do what it promises, and Loomis does not do for me what people claim he did for them. He achieves other things, and for these I do not want to simply throw away his books. I want to find out, name and correct the faults and thus be able to use the rest.

But you may, if you want to, explain, how you draw, what proportion and abstraction means to you, how you use it or why you don't use it etc.


Hey there. Well I think Armando is kinda right in the sense that Loomis should be interpreted more as a guideline than anything else. And, frankly, in most of the cases I've seen of people drawing good proportions and all, they don't have so much of a system, it just came out with practice and work.

The typical books recomendede here might help to some people (me included) but I think that in my personal experience I got more out of drawing from references , studies and all that, than using a method , wich works almost always worse, as your natural way of drawing will probably provide the option of a less forced drawing. On the other hand if I had to follow an instructor's method I would rather follow Kevin Chen's demo drawings ( wich taight me slightly more than Loomis / Hogart ) that are based on a more comprehensive structure and try to avoid typical methods.

Much love :mod:

[tandy1000]

Not to sound like an ass but any break down of the figure is meant to be taken on the basis of generalization. Generalized proportions help you understand the human form but it is not supposed to capture the personality.

Construction or any other break down of the figure helps the student along the path of drawing well, but it does not teach the student how to capture the likeness. That is up to the student himself to be able to discern between the difference and tweak accordingly.

You might think that Loomis teaches an unrealistic schema, but it's merely a demonstration of proportion and basic head construction.

But, it is good that you are looking at it from a point of skepticism, asking questions of how and why things work will help you grow.