[Steven Stahlberg]

I recently tried converting Andrew Loomis' cool method of constructing heads in a 2d drawing, into 3d - you know, the sphere with the sides cut off and a bulge at the back, and the equator and meridian marked out, with a line hanging down the front...

It really does help, makes it a lot easier when you don't have perfect profile and front shots of a reference person, or when you want to create an average but handsome face.

But when I was finished, it seemed that the skull would be too small in relation to the face, if I followed the template all the way. I know there may be some freaks out there who look like that, but I was going for the handsome/average, the kind of idealized male caucasian face (20 - 40 years old).

So I enlarged the skull a bit.
My thinking is, Loomis never intended this method to be used so strictly, with perfect geometrical surfaces, but meant it as a rough 2d visual aid. What do you guys think? Has anyone else tried this before, and noticed the same effect?

[aquamire]

I've followed Loomis's teachings on head construction, as well as the previous lectures by Fred Flickstones pertaining the same method. For some time I kept to the basic form and tried to apply features. It's impossible to get a realistic head out of just sticking to the underlying form he presents. In Creative Figure Drawing for All it's Worth, it goes on to describe the plains of the head briefly. I believe he intended it to be a rough understructure so the beginner can grasp the form of the head in varying angles and perspectives. After you define the head with the basic construction you go on to define the planes, and mold the face to your character. The form he presents is mostly a rough guide as to where to apply the features and planes in a proportional manner. Thats how I take it anyways.

[Tendril]

Ive never tried this but I know that whenever I try and model a head that I have drawn. ie with front and side veiw. The skull is always too small....
I always put this down to perspective- If you look at a photograph of someones head front on, they can look quite different depending on the feild of veiw. You can see lots of skull on either side of the eyes when far away, and not much when close...

Maybe loomis tries to teach people to make the skull smaller because it looks right in a drawing with perspective... if you get what i mean....

[Steven Stahlberg]

Yes, Tendril, I've noticed the same effect, it's very dependent on the angle of view.

Aquamire, that sounds interesting, is there a link I can follow?

To me, though, the planes are not as crucial as the basic measurements of the face, when I work in 3d - width, height, depth, how high to place the eyes, how far back the ears etc. (I know this varies a lot but in 3d it seems when I make a mistake it's always towards the ugly and freakish, never towards the nicer side of the Gaussian bellcurve.) Actually I should just buy a skull...

but

how would I know the previous owner of that skull was not the most freakishly ugly man that ever lived? You see my problem? (And also, is there no shrinkage or warpage in skulls?)
I can't very well ring the doorbell of 10 or 20 male supermodels or actors and ask to measure their heads... Measuring my own skull certainly won't help me at all.

[immi]

Steven, check out http://www.anticz.com/learn.htm (at the bottom) for a few of Ron Lemen's aka fred flickstone's tutorials on the subject. He showed up the other day on the forum after a long absence...he'd be a good guy to get an opinion from.

[quaternius]

Yep Steven, you're not alone.

The back of the skull and skull size gets shorted using Loomis' sphere method too strictly. He develops this a bit more in his heads and hands book, but it's the one thing that always bothered me, (except for establishing the hairline location, which establishes the nose and chin location...)

Let me dig up a couple of diagrams and edit this post to show you what I mean. Don't get me wrong, I'm a big fan of Loomis and it's what I grew up with and how I learned to draw. I also applaud Ron, he does a great job of expanding and describing the method. However, IMHO I think everyone is entitled to develop their own system of proportion - you can't follow the Loomis "ideal" blindly. In fairness to Loomis, he does say in his book to "add a little to the back of the head".

Now first, the obvious problem for people who start using the Loomis head method is they say to me "How do I know where to put the hairline to establish the other points?" Although you still have to use an educated guess, I developed a curve at 45 deg. angle to help.


Second, here's my sketch of the Loomis head from his separate book on heads and hands. This shows his idealized measurement system for the male. Note that his "sphere" is not drawn - but if drawn would meet at the edges of the "box" shown for the front view as well as the top of the skull. You'll notice I've also added some 45 deg. angles in the side view to clarify positioning and aid in establishing other reference points.


Third, some sketches developed directly from skulls to show how "reality" differs from Loomis. You will notice immediately how much more skull is in back of the ears in reality, vs. the Loomis construction. Loomis gets away with it most of the time because in a 2D drawing the curve of the skull disappears in a 3/4 front view - so no one will notice, (IMHO). In a 3D model it won't work.

These sketches are my overlay revisions of drawings found in John Raynes figure drawing book. (Good basic reference book by the way) Notice how the top portion of the skull is actually egg-shaped; not sphere-shaped, especially in plan view. Granted, the skulls developed here are female - so the dimensions will be a litle different for the male head. However, the main point of "egg" shape and large amount of mass behind the head centerline is still valid.

Loomis shaves off the sides of his sphere to approximate this shape - but leaves the back of the skull "shorted" as well as the question I've always been asked - "How much do you shave off?". Again it's a guesstimate which grows more accurate the more you draw.

Hmmm... don't know if this helps or not - but it sure is an interesting discussion.

By the way Steven, if you're still looking for some good reference photos I happened across a pro weight-lifter dude that is a dead-ringer for Superman. I could maybe post a couple photos and also check to see if he has his own web site if that helps.

[ July 18, 2001: Message edited by: quaternius ]

[ July 18, 2001: Message edited by: quaternius ]

[Steven Stahlberg]

Thanks Immi and Quaternius!

Edit: Great images Quaternius! Yes, I'd like to see those other images of that weightlifter guy too, thanks!

[ July 18, 2001: Message edited by: Steven Stahlberg ]

[S4Sb]

Steven, I don't know if I understand what you mean... (translation problems maybe) Can you perhaps show some shots of the head you modelled? Before you adjusted it to an average male head?

[Anthony]

2d never translates perfectly into 3d, cause of the way we draw. When you draw something, even a straight forward figure, you utilize the perspective of our eyes to make it look "natural." So we distort it. Loomis' head method utilizes this. In 3d we're using orthographic views, which means things like the back of the head which look small in perspective look wider in the view. Pretty straight forward, eh? :] Also, we tend to make figures much taller in 2d than looks natural in 3d.

[Steven Stahlberg]

Ok, here's the head I'm modelling. You can see that even though I kept the skull larger than Loomis suggested, it's a bit on the small side. Now that I hear other people have had similar experiences as me with Loomis method I've found the confindence to tweak my Loomis template larger on the skull, all that remains is to see exactly how much to enlarge it. (Why am I so concerned with this? The basic shape of the face is the hardest thing in 3d, unless you're 3d scanning, and a simple template is better to use for a new head starting from scratch, than an old head, AS LONG AS the template is reliably average... if you see what I mean)

[Plouffe]

Ok ive been starring at that head for the past 30 mins.. MAN that is SCARY lol.. looks almost real.... I was waiting for it to talk to me or something

[Freddio]

So you worked on final Fantasy?



cool job m8

[S4Sb]

U don't use max, right. Sub-D?
I got what you mean

[Steven Stahlberg]

Thanks Plouffe!
No I didn't work on FF.
It's Maya, yes and Subdivided surfaces. I'll be speaking at Siggraph about this face. Not sure I'm going to like that...

[Shirotsugh]

Wow! That head is unreal, in a real-sorta way.

[KOryH]

SEPERATED at BIRTH or Steven and I have met in another life!!! When I saw the thing that Steven was working on it gave me the creeps you can see why...

WEIRD HUH?

[]Aratex[]

Good God, he's beautiful!!

Err.. Not in a homosexual way.. more in a "I suck at 3D" way..

Damn.. Maybe I should just give it up already... I'll never be good enough...

[KiNgStiNg]

really awesome looking model dude

[quaternius]

Woa! That's a great head you're workin' on Steven.
On the photos of that guy I thought looked like Superman, I can't find the darn magazine, but his name is Christian Boeving and he's actually an actor, not a pro weightlifter. And I take back the Superman look-a-like thing after seeing more photos. But he's still a good reference I think.
Turns out he's actually co-host of a TV fitness show and the "commander" in the TV show Battle Dome. (guess I missed that series)
His own website seems to be down - http://www.christianboeving.com

but here's a couple others -
http://www.darkwoods.com/bodybuilder/male/bb/gallery/fullsize/boevin03.jpg
http://www.naturalmuscle.net/nm0600/christian.htm
http://www.menofmuscle.bigstep.com/generic.html?pid=0

Geez KoryH - the resembance IS creepy!

[ July 26, 2001: Message edited by: quaternius ]

[silber]

awsome head steven !
can't wait to see the very final work

quaternius: thanks for your input
and even bigger thanks for sharing Loomis at your site
keep it up