Results 1 to 13 of 24
Thread: Construction Woes with boxes
November 28th, 2010 #1
Construction Woes with boxes
I've been trying to practice construction with box forms using photo refs, but I'm still stuck at Square one after like 4 weeks of hard trying.
I've decided to post some photos with lines superimposed on the photos to show you guys how I usually construct with box forms. I'm currently reading Vilppu and Hampton regarding this subject.
For the front body, I use the ends of the clavicle and ends of the 10th rib for the torso box, and the ant-sup illiac spines for the top of the pelvic box, and the pubic arch for the bottom of the pelvic box:
For the back, I use the ends of where I think the acromion processes are, and use that as the top of the box. I use the end of the rib cage (12th rib) for the bottom of the torso box. For the pelvic box, I use the posterior superior illiac spine for the top of the box, and the area just slightly above the bottom of the gluteus maximus as the bottom of the pelvic box (that's where the bottom of the pelvis is):
I'm super sure I got the following wrong, especially the torso box:
In poses like these, I totally don't even know where to start. It's cases like these that truly makes me feel that anatomy is useless and I wanna slam my Peck's anatomy book across the room. The 10th rib and illiac spines are ****ing hidden for goodness sake! I can't even see the posterior superior illiac spines! And there are poses that are even more challenging that this. How the hell do you construct shit like these??!!!! Damn it!
I really need some advice here. In the past, I dunno about things like landmarks, but if you show me a nude model now, at least I can tell where are those bony landmarks, but even then, landmarks are not helping that much, and it's confusing. Maybe if someone is kind enough, do draw over my photos here so that I can see where I did wrong.
Hide this ad by registering as a memberNovember 28th, 2010 #2
The boxes for the pelvis are too low but I'll show you why you are also getting the rib cage wrong. You need to identify the end of the ribcage too.
The Following User Says Thank You to Arshes Nei For This Useful Post:
November 28th, 2010 #3
For a start, the ribcage doesn't taper down that way. Take a look at this page (and the next) from Bridgeman's Constructive Anatomy to see how he handles it. Personally, I've stopped using boxes per se and moved on to a rough oval with a section cut out in the front for the end of the ribcage, more like Loomis' basic mannikins, except I do the same with the hips instead of doing a simplified pelvis. Centre lines help keep track of the tilt of the different masses, which helps me a lot when I'm working with posing.
The Nezumi Works Sketchbook - Now in progress
My online portfolio
"Skill is the result of trying again and again, applying our ability and proving our knowledge as we gain it. Let us get used to throwing away the unsuccessful effort and doing the job over. Let us consider obstacles as something to be expected in any endeavor; then they won't seem quite so insurmountable or so defeating." - Andrew Loomis
November 28th, 2010 #4
Boxes are more of a conceptual aid than a drawing aid, you don't need that degree of precision. Their use is to get you thinking in terms of perspective and space when dealing with the rounded, lumpy, squishy forms of the body. Also, in general I think it's more useful to think of bounding boxes that contain the whole form rather than ones that are inside of it.
**Finished Work Thread **Process Thread **Edges Tutorial
Crash Course for Artists, Illustrators, and Cartoonists, NYC, the 2013 Edition!
"Work is more fun than fun."
"Art is supposed to punch you in the brain, and it's supposed to stay punched."
November 28th, 2010 #5
Yes, in general you want to think "which way is this form tilting", and then draw a corresponding box that is tilting generally the same way and has the same general size. Emphasis on "general", so that you have room to adjust later.
I wouldn't focus so much on lining your box up with specific landmarks, though it's good to use the landmarks as cues for which way things are actually tilting if you're confused.
In a sense, you are missing the forest for the trees in terms of anatomy. On the one where "The 10th rib and illiac spines are ****ing hidden"- you have the pelvis tipping away from you and the rib cage tilted slightly back towards you. It's plain as day- start there, not with the Anterior Superior Iliac Spine. If you can't draw a box that accurately displays the tilt of the whole pelvis or ribcage, then your problem is with perspective- not anatomy.
November 28th, 2010 #6
Yeah what Elwell said. The anterior superior iliac spines are very good landmarks to keep in mind, but when constructing a figure out of your imagination, you want to think more about the movement, direction, or thrust of the mass of the hip, ribcage and skull. When doing practices like this on photos, don't look for exactly where the clavicles are, but try to see the movement the ribcage makes when it goes this way or that. Also, try to consciously think of what the side, front and back, top and bottom planes of these boxes are. For example, in your third picture here, both the torso, head and pelvis are turned almost completely with their side planes straight at the camera, so you wouldn't really see the back planes of their boxes, as you have drawn. In fact I think they're closer to showing their front planes! Also, the thrust of your ribcage box is off, you need to rotate your box a bit clockwise, so that it aligns with the thrust of the ribcage. The point of this kind of construction is so that you can construct the large masses in perspective, and then add anatomy on them. Like so:
This paintover was done quickly and with a mouse, so I'm not guaranteeing that it is extremely accurate, but I hope it gets the point across..
EDIT: and as my amazingly skilled crosshatcing on the leg shows, the thigh in this particular photo is better thought of as a cylinder, not a box. I recommend reading Robert Beverly Hale's Drawing Lessons From the Great Masters if you don't already have. If you have, read it again. It's a great book.
November 28th, 2010 #7
I really appreciate the fact you are so curious and willing to ask for information. Think of it this way, the box is to help you imagine the big divisions of light and shadow, thats all. Once you establish those in your mind you can then decide how you want to represent the other information so that it is easily identifiable to the viewer. Why? because sometimes the reference is ambiguous and you as the artist must correct that. This is what the Reilly method does so well. It teaches you to abstract the shadow shapes from the halftones and lights. All good systems do this but I happen to like Reilly. Here is an Asaro who was a Reilly student, in this painting he has left the planes exagerated and visible. Go back and look at your photos and try and imagine where you would abstracxt the curved surfaces into planes. Decide what the plane change is, a value shift, a hue shift, or a combination of the two. Squint at the Asaro and you will see the hard edges of the geometric planes disapear.
Last edited by dpaint; November 28th, 2010 at 02:53 PM.
November 28th, 2010 #8
The Following User Says Thank You to Kamber Parrk For This Useful Post:
November 29th, 2010 #9
Since someone else got the other one, I'll try to explain this one.. As others have said it's good to find the landmarks, and definitely read Hale's books they give you a greater understanding.
The building blocks whether you use a sphere, oval, cube are good for the imagination or memory drawing until you learn how to be able to construct without it.
The reason cubes were used was that you'd learn how to view the figure in the most basic form in 3 dimensions, remember they are guides since the human form is more organic, but breaking it down to basic shapes is what helps you see the bigger picture than worrying about each little form.
The spine is the best landmark as it divides the body in half vertically. It's a good landmark to place the shape you're looking for. you'll know where 'half a box" will sit by that line.
It's also a good idea why learning basic shapes in perspective help so much, it's not just about pretty buildings but understanding overlaps and how those shapes work. I'd practice trying to draw cubes and and other objects in different perspectives and then apply that knowledge to the figure. So everyone else said it in shorter words but you get the idea.
November 29th, 2010 #10
This thread is trying to fit a round box through a square hole.
The human body doesn't look like that.
This visualization is only useful to understand how the body twists and where the sides of the parts are facing.
If the shoulders are rounded off, a box no longer fits.
The hips are in no way box like.
This is just another mannequin that you can use to help you visualize.
Ultimately, YOU have to visualize it, not Bridgeman's mannequin, or Loomis's mannequin or any of the others - they are only fingers pointing the way; they can't take you there and they aren't the solution.
YOU have to find a way that YOU can visualize it.
November 29th, 2010 #11
Wow, thanks a lot, everyone!
There's so much info here to digest, and I'll re-read and re-read this page again and again.
Currently, my way of constructing the figure is to find the landmarks and draw out the boxes based on the landmarks since it seems to be easier (based on my own interpretation of what Vilppu and Hampton did, and my interpretation may be wrong). For Vilppu and guys, they seem to start the top of their pelvic box at where the ant-sup illiac spines are, and not at the peak of the illiac crest.
That's why I'm having such a hard time cos' sometimes, due to certain poses or the condition of the lighting, the landmarks can't even be seen.
The paint-overs and advice by Arshes Nei, Serpian and dose are very informative, though,and I intend to do some copies of those to get a "feel" of it.
Just one more question to Arshes: When you were constructing the rib cage and pelvis using boxes like in the paintover you did, were you constructing them based on landmarks or based on "movement, direction, or thrust of the mass of the hip, ribcage and skull" (like what Serpian says)? What Serpian says seems to be a much easier way to visualize the construction process instead of painfully looking for landmarks and trying to line them all up in boxes, like what I've been doing all these time.
*PS*: I'm getting Hale's book from the library end of this week!
Last edited by Xeon_OND; November 29th, 2010 at 09:09 PM.
November 29th, 2010 #12
Went ahead and took a stab at this myself. Here's my thoughts: Don't let anyone ever tell you that using boxes for construction is a crutch, or something that you'll grow out of when you REALLY learn human anatomy. The way I see it, the box is an abstract tool AND a reality. The two major masses of our bodies, the torso and the pelvis, are very boxy indeed. They have top planes, bottom planes, and side planes. But they are modified boxes, which turn out to be very complicated shapes. Drawing a box in perspective is easy. Drawing a complicated curved-here-straight-there box-like form, not so much. But it all starts with the box.
The problem I see with most attempts at boxing in the torso is that they fail to take into account the fact that the back is a very curved surface, compared to the torso's relatively flat plane. Take a look at the small diagrams I drew up, especially the straight top-down view. That's why you can't see the far corners of the torso from a 3/4 angle -- because they've vanished as the form turns very quickly near where the far scapula begins.
Here's my interpretation of the pose if you care to take a look, I hope it is helpful!
November 29th, 2010 #13
Here's the thing -- you'll need to learn the shape of the body masses, and learn to draw them in perspective. Once you understand the shape, you only need a few landmarks. Here's an example: By seeing the angle of two landmarks (say, the line across the back from the innermost tip of one scapula to the other), you get a line with a particular tilt. From there, you automatically know the tilt of all lines parallel to that line on the form (say, the bottommost edges of the front and back planes of the torso). From there you automatically know the tilt of all lines at right or nearly-right angles (so long as you can draw a cube in space, you can construct right angles in perspective!) and with those lines you can draw the planes.
Knowing the geometric relationships of the landmarks means that you know where they are (which means you can construct the planes) EVEN when you can't see them cuz some lady's big butt is blocking your view.
(Edited a bit for clarity. Probably still makes no sense anyway, it's very tough to type out a visual explanation!)
Last edited by Lamp; November 29th, 2010 at 11:01 PM.
The Following User Says Thank You to Lamp For This Useful Post: