PDA

View Full Version : How to practice eye measurements


McWolfe
10-25-2010, 03:34 PM
I'm currently (as so many others) struggling with becoming a better artist. Somehow it seems that the longer I practice, the more I get back to the basics. So, now I've arrived at the stage where I feel that I can't improve anymore if I don't improve my sight, my feeling for shapes, proportions and sizes (you know, the basic stuff).

So, I was wondering if anyone of you here have some good tips on how to practice and develop my ability to draw something accurately from sight. Nothing fancy, but rather the oh-so-important basics.

For instance, would you recommend med to use the grid-method (you know, overlaying a grid on an image, and using that as a support), and then gradually use grids with bigger squares? Or something else entirely?

Lunatique
10-26-2010, 05:41 AM
This is covered pretty extensively in the first week of my workshop, but I'll try and give a condensed and simpler version here, without all the visual examples.

Basically, you must learn to compare sizes, distances, angles, curves, association with geometric shapes...etc. You look for shapes in the scene that have similar distances or angles and you use them as anchors. You compare sizes so you can tell that one shape is roughly 3 times the width of another shape, and you make sure you maintain that relationship. You look at angles or lines and curvatures and make mental notes like, the contour of that object is a slight curve and angled at about 45 degrees. You look at patches of values or general shapes of objects or details within objects and note when they look similar to geometric shapes. Use the more prominent shapes as guides--for example, a particular object's height is 2/3 the overall length of the image, and you use that as a simple mental landmark to measure other objects in the scene.

The grid method is really just overlaying easy and consistent landmarks on top of an image so it's much easier to do measurements and compare. But that's all it is--easy landmarks. You can establish your own landmarks for simply observing the composition of your scene--the only difference is that they won't be a consistent grid.

McWolfe
10-26-2010, 06:01 AM
Thanks for the reply!

On a side note: I'm saving up for a future workshop with you, what I've heard/seen so far looks very good. It'll take hard work, but it's worth it

Now, back to the subject: I find this webpage: Observational Drawing (http://www.conceptart.org/forums/showthread.php?t=199211) on conceptart.org that has some hints and tips. Is this something along the lines (no pun intended) that you are talking about?

Also, since I am not particulary good at judging angles, I started to do a mental overlay of a watch, basically trying to visualise things like "that line is in the 2 o'clock direction". That makes it, at least for me, a bit easier to judge.

I'll keep on practicing (with this as a specific goal), and do a couple of posts in my sketchbook thread.

Another question: I've come to understand that before you can draw convincingly from observation, drawing from your mind/construction is a lot harder. It would be interesting to hear your thoughts on that relationship. Are we talking about things like after doing a couple of hundred (thousands) hands from observation, we are building up a mental reference/understanding about how the hand works, or are there other factors at work as well?

Lunatique
10-26-2010, 06:14 AM
The conceptart link shows similar mental process. Different artists have slightly different approaches, but the underlying principles of visual comparison and measurements are very similar. For example, your own unique approach of using the clock is a great example--it woks and it's easy to relate to. someone else might have another approach to measure angle, but as long as it's reliable and practical, it's fine.

Working out of your head isn't as simple as simply imprinting your brain by hardcore memorization of how things look in real life or in photos. The ideal way to work out of your head is to combine observing and analyzing how things look in real life with studies of its underlying structure--in other words, combining visual observation with logic and analysis. This is why you shouldn't just draw nude models and think that's all it take--you must study anatomy from good books on the subject, learning how the muscles are shaped and wrapped around the skeleton, how they flex, or what are the limitations of the skeleton itself in terms of joint rotation, and so on. Only when you understand the underlying structures would your visual observations make sense, and when they make sense, you will be able to reconstruct them logically out of your head.

McWolfe
10-26-2010, 06:31 AM
Working out of your head isn't as simple as simply imprinting your brain by hardcore memorization of how things look in real life or in photos. The ideal way to work out of your head is to combine observing and analyzing who things look in real life with studies of its underlying structure--in other words, combining visual observation with logic and analysis. This is why you shouldn't just draw nude models and think that's all it take--you must study anatomy from good books on the subject, learning how the muscles are shaped and wrapped around the skeleton, how they flex, or what are the limitations of the skeleton itself in terms of joint rotation, and so on. Only when you understand the underlying structures would your visual observations make sense, and when they make sense, you will be able to reconstruct them logically out of your head.

Yupp, I figured it was something like that. (and that was more or less what I meant above, although maybe expressed a bit to condensed ;-) ).

It seems like an iterative process, where you (like in my case, since I'm a fairly abstract thinker/ programmer by nature) understand the structure, but can't really put it on paper. Then you develop your observation skills, and learn what things look like. Then, basically, you ask yourself "WHY does it look like this?", applying what you know about structure. From my own meager experience with this kind of practicing, that in turn increases my understanding of the structure, because I'm seeing and analyzing the structure in action, so to speak.

halen
10-26-2010, 10:27 AM
Just started reading Betty Edwards "(the new) Drawing with the right side of the brain" (http://www.drawright.com/). It seems to have some exercises related to this. I think it was also recommended somewhere in the sticky threads.

I'm also working on how to see and then how to draw/paint. Currently having only "flashes of seeing" things in shapes/lines/masses/values, not as what they present. Feels weird, but I think it helps. Had these also before every now and then, but newer knew what it is about and couln'd take an advantage of it or consciously work with it, measuring, thinking things on a flat surface.

edit: and I think that there is iteration also in seing details - first sizes, shapes and large blobs, then smaller things inside em, untill you have something like you see.

McWolfe
10-26-2010, 11:03 AM
Just started reading Betty Edwards "(the new) Drawing with the right side of the brain" (http://www.drawright.com/). It seems to have some exercises related to this. I think it was also recommended somewhere in the sticky threads.

I'm also working on how to see and then how to draw/paint. Currently having only "flashes of seeing" things in shapes/lines/masses/values, not as what they present. Feels weird, but I think it helps. Had these also before every now and then, but newer knew what it is about and couln'd take an advantage of it or consciously work with it, measuring, thinking things on a flat surface.

edit: and I think that there is iteration also in seing details - first sizes, shapes and large blobs, then smaller things inside em, untill you have something like you see.

I think so to. Sometimes I've found it easier to squint a little, so that what I'm looking at becomes more blurry. That way, the details disappear, and I have an easier time to see the bigger shapes.

Seems like I have to get the new edition. I read the first one years ago, but never really did anything with it.

McWOlfe

halen
10-26-2010, 01:28 PM
Seems like I have to get the new edition. I read the first one years ago, but never really did anything with it.

Don't know it there is mutch difference to the old one. :shrug: Got one from local library. Impression so far is that it might help me to see things at this point. There are also other books that have slightly different approach - some more exact (towards measuring) and some more loose (towards impression of the line / direction of the force or what ever it is called in English). All seem to have some benefits. I just enjoy learning and try to add things learned to the next work.

dbclemons
11-22-2010, 10:37 PM
I would add that it's good to simplify the basic forms at the beginning. Block out the forms roughly and look for landmarks from one side to the next as well as triangulating the points as you go to improve your measurements. Otherwise there's a tendency to stretch things out too much. I also like to start in the center of my page/surface rather than at the top, and mark off the extremes so that everything will fit. When you get your extreme points placed it's easier to fill it all in.

The old "thumb" measuring device is a tried & true method. This refers to holding out your arm straight in front of you and measuring distances using either your thumb, pencil, or brush handle. A ruler or "L" square can also be used and will show you actual measured units as well as help you line things up. The distance between you and the model should remain constant.

CGTalk Moderation
11-22-2010, 10:37 PM
This thread has been automatically closed as it remained inactive for 12 months. If you wish to continue the discussion, please create a new thread in the appropriate forum.