The Great Falsehood

August 12, 2010

All images are included here strictly for educational purposes- all rights belong to the proper copyright owners. I don’t own squat.

Brian Stelfreeze once told me “There’s no such thing as one point perspective”, I argued, and I was wrong. What he meant was when we’re taught one point perspective as a shortcut, and usually we’re only shown half the theory. It’s used as an introduction to the complex nature of perspective, but most of us never get the whole truth and accept the stripped down version as an absolute over time. It isn’t. The simple fact is, there’s no such thing as true one-point perspective, it’s all multiple point perspective, but the limitations of a vantage point make the other Vanishing Points (VPs) less obvious. It’s only how lazy we are that dictates how few points we use. We limit it to two or three for simplicity, but the fact is, in real life we’re surrounded with billions of VPs, and in some cases multiple Horizon Lines*!

When we learn one-point perspective as kids, it’s usually drawing a train tunnel, hall way, or some such nonsense. We start with the tracks as usual, vanishing off to the Horizon Line (HL) and this is where we got side-swiped as kids. When we place the wooden ties under the rails, as kids, we’re taught to just draw them straight across and we’re done. Truth is, the moment you move to the right or left of those tracks, it slowly falls apart. At center point of the image, there’s the illusion that things fall straight across in the middle. What is really happening is a subtle slope to two VPs on both sides; it’s gradual but it’s there. If we were only drawing the tracks we might get away with this, but when we move past this ultra simple image and start to create an environment around them, the image will start to fail. Any structure we place around the train ties, to the left or right, above or below the center point immediately falls victim to this falsehood. Andrew Loomis did a good job explaining it in the illustration below.

One-point, you can stretch it, but eventually like a rubber band- it'll break!

See the little box around the viewer’s head- in essence that’s the viewing area where you can cheat the concept of one-point perspective before things begin to fall apart, it’s not to say it works, it just means at a glance it won’t scream “failure”.

Below is a brick bridge built over some train tracks, and at first glance it looks straight across at all points. Our one-point perspective is safe! ~ Wrong.

the good old train track trick

When we rule lines straight across and space them out evenly up and down the image, first glance, things are still looking pretty good.

1-point theory- all we need are straight lines across the middle!

No, wait a second, look the top of the bridge? Or the interior cabin of the train we’re on? Or the ties on the lower right hand corner of the image! We might have to pull out our type ruler for the perspective on this one, because that’s what drawing things properly requires! We’re not in this to make things easy on ourselves, we do it because we enjoy creating an image that we want our viewers to believe, and just like crummy anatomy on a figure, lazy perspective can do just as much damage.

Whew- it's way out there- glad we know that type ruler trick!

When we set up our outside VPs, we can rule out where the edges truly fall. It’s then that we can see things aren’t quite what they’ve seemed.

Blimey- there be slopes at the edges of of me image!

Not everything works going straight across as promised by one-point perspective, especially further away from the center point our edges get!

Well just look at that!

As you move closer to that center point you can see how what was once our one-point perspective begins to dominate. This is the transition point to the left side VP and it’s influence away from the right side VP. What’s really happening is 3 point perspective (at the very least) with the center point dominating, and the two side points pulling out to the far edges, but ever so slightly.

Wrap your head around this-dat!

So can you cheat, and do simple one point perspective as we were taught in grade school? Sure, just don’t draw anything except cartoon train tunnels and tracks. You’ll often see the failings of it in amateur drawings. Where hallways have doors that appear to be painted on, and things are obviously out of proportions. Part of the problem is that people believe just by following the perspective lines, it means the image must be correct. Usually it isn’t and things end up looking flat and sterile, but worse they ignore the proportions of the image. For example, do these doors have enough space to open? Do people have enough room to move down the hall? What about when they reach the end? Can they turn the corner and go down the other end of the hall?

Note the flat nature of the doors and width of hall

If you want to see some amazing perspective work, check out Cannabis Works by Tatsuyuki Tanaka, he’s a Japanese artist who has designed backgrounds for films such as Akira.

Available thru companies like animebooks.com

It’s an amazing use of perspective, and puts mine to shame. He does some wonderful images involving stairwells with about 50 VPs and multiple HLs going on. It’s simply amazing. Here are a couple of examples of his “one-point” perspective, see if you can tell how many VPs he’s actually using.

Practically the same shot as above- much more developed

Now look at this “fish-eye” effect he’s done- try that with only one-point.

I wanna break his hands- he's too good!

*As mentioned in the beginning, multiple horizon lines are complex, but in truth it’s any change in the viewer’s line of vision up or down. Since HLs are based on the “POV” of the viewer; changing that POV changes the HL. This happens when one is looking down a stairwell or up into the sky. It’s a difficult concept to explain and the best I can do is give you this illustration by Andrew Loomis.

Ow, ow, ow, ow!

Note how the environment at the bottom of the hill is in perspective, but he’s actually dropped a VP to a lower level to make those items accurate. In essence he’s made a second HL at that level to accomplish that, and it’s in line with what our heads would do and the angle our eye line would take. Since a HL is based on our eye line, by looking down to see the town at the bottom of the hill, we instantly establish a second HL. Whew- makes my brain hurt!

When I was still in college, and a few years after, I lived in Atlanta, the home of Gaijin Studios. For those who don’t know, that’s the studio Brian Stelfreeze calls home, and shares with the likes of Cully Hamner, Karl Story, and Laura Martin, and in the past with the likes of Tony Harris, Jason Pearson, Joe Phillips, and Adam Hughes. Brian’s an incredible talent, and an amazing teacher. I believe he does lectures for SCAD both in Savannah, GA, and now at their acquired Atlanta branch, which used to be my old school, the former Atlanta College of Art. Sadly, when I was in school they didn’t bring in comic artists to speak, so I sought out Gaijin Studios on my own. I pestered the guys (bringing boxes of candy didn’t hurt my chances), but I’ll admit, at the time I wanted to bug Adam, as he was my idol. It was Brian who taught me the tricks, gave me the bulk of his time, and passed on to me the most important lesson of all… the wheel! Naw, but darn near as important- perspective.

When you look at pages by other artists, anatomy is the first thing you notice- because we all know what it’s supposed to look like. We’ve seen it our whole lives, so it stands out- but immediately behind that (literally and figuratively) is perspective! You’ll find guys who are bad at both, but often you’ll have someone who is good at anatomy, but lousy at perspective. Funny thing is, you rarely (if ever) find a guy who’s good at perspective but lousy at anatomy; it just seems to go against the learning curve. I’ll never do justice here to what Brian taught me, and I could never cover all the things that he passed on to me (it’s like he did some wacky Vulcan mind meld and downloaded way more than I can convey), but I’ll try to pass along a few tricks that have always helped me. They’re not the only things you need to know, but they will help save you time, put you on the right path to finding a solution, or at least help you spot the problems early on. I know that Brian didn’t originate a lot of the information he passed on to me, but like any great artist, he took it and made it his own.

The first is something I believe both he and Adam Hughes got from Andrew Loomis, and because of that, I’ll use the images from the out of print book that they got it from… Andrew Loomis‘ “Figure Drawing for All it’s Worth“. If you don’t have his books, as most of us don’t, you can probably track down PDFs from various art sites, or spend crazy amounts of money (like I did over the years) buying second hand copies of the originals. In Figure Drawing for All it’s Worth, Loomis discusses “hanging the figure”, something Brian passed onto me when he saw my first pages for DC and promptly pointed out all my failings. :::sigh:::: What it means is, whatever Horizon Line (HL) you establish for a shot, all of your figures will be subject to the same spacial relationship to that line. So if we have a HL at the waist level of Mr. Main Guy, then all the figures the same height as Mr.Main Guy will also hit the HL at their waists. Without doing accurate math, that means if the HL and Mr. Guy’s waist are 41″ from the ground, then the HL (in theory) should hit all figures at approx. 41″. If they’re shorter than Mr. Guy it will hit above their waists, and if they’re taller it will hit below. If a person of equal height to Mr. Guy is sitting (considering most chair seats are set at the same height) then odds are their shoulders & head will be at about waist level of our standing Mr. Guy. Similar to before, everyone seated who is just as tall as Mr. Guy, will also have their head/shoulders hitting our HL as well. Now all of this is subject to what a person is doing of course. If they’re bending over to pick something up, jumping, dancing, blah, blah, blah- then of course that alters their relationship to the HL, but this is a simple gauge to use quickly to see if there are any glaring mistakes in your image. Here is the image from Loomis’ book that illustrates this more clearly than I can with words.

Andrew Loomis- making it simple

The next trick needs a quick recap of a basic perspective theory, which is that any image (excluding cubist ideals) is seen from ONE view point. That image is being shown from one set of eyes, and one position in space, so the HL and all Vanishing Points (VP) contained are subject to that view point regardless of where they are. So if they’re standing on a ladder we see things from slightly above (bird eye), and if they’re laying on the ground, the view is from slightly below (worm eye). A simple note to remember is that where the view point is in relationship to an object will dictate the distance between VPs so if we’re standing just to the left of an object then the left side VP is much closer to our center and the right side VP is further away from us just as that side of the object is. The closer we get to the left side of the same object, the closer the left side VP becomes and the further away the right side of the object, the right side VP moves further away. Here’s a couple of diagrams to show that as well.

We start out almost in the middle.

Then move a little to the left.

Then just a little more to the left.

Notice anything on odd about the last image? Well that’s because we ran out of table- there was nowhere to put our far right VP. When doing perspective, we learn to draw out our HL and then add our VPs, so we can draw boxes, buildings, cars, whatever. When you start getting into trickier shots, the VPs tend to get further and further out on the sides of the drawing, if pushed to the extreme- you run out of table. Now there are tricks to doing this in the computer, but I’m a proponent of learning how to do things by hand. That isn’t because I hate computers, but I’ve found that a lot of guys who don’t learn it by hand first, and rely on the computer to do these types of things for them, never learn why stuff happens the way it does to begin with. Learning how to do something by hand, trains your brain to recognize perspective properly, and over time your drawings will adjust to being more accurate without the aid of devices. Not to toot my own horn, but thru the years I’ve noticed that my perspective layouts are practically spot on from the thumbnails to begin with and all the ruler tricks only help to tighten up the lines.

For speed you’ll want a specific type of ruler, you can find it in most art shops, or if you live someplace without a good art store, you can order it on line from any major supplier. It’s called a “Type Gauge”, used in the old days of type setting to see how much space a column of type would take up at various point sizes. Of course there aren’t a lot of people who use these much anymore, so you better get one quick before they stop selling them! Here’s what they look like.

Now establish your horizon line, but when you know your VP is going to be very gradual and end up off the edge of the table, you whip out the ruler. If both VPs fall off the table you’ll have to establish this for both sides, but for now lets only deal with the idea that the VP on the right side of the drawing has fallen off the side of the table. On both sides of the image draw a perpendicular line to the HL, one going thru the VP on the left the other can just be on the right side of the paper.

The left will be the larger side and the right will be the smaller just as the VP would do if it were on the paper. Take your type gauge, you’ll see the point sizes are clearly marked according to height, most range on one side from 7- 10 and on the other from 11-15, skipping 14 (don’t ask me why, I don’t know- you’d think they’d skip 13 out of superstition if they had to drop one). Decide how extreme you want the VP to be, if you want it to be very slight pick numbers closer together like 8 and 10. If you want to make the VP more extreme obviously pick number further apart, such as 7 and 15. Usually you’ll end up with numbers only a couple of steps apart, the more extreme you get, the odds increase that the VP would have ended up on the table to begin with.

Now lay the ruler to where you can see your perpendicular line in the space of the type gauge, if you’re on the left it should be inside the gauge for the larger size and reversed when doing the right side (smaller). Tick off marks going above and below the line, I tend to only tick at the even numbers and then make double marks at every fifth point for quick reference when I’m done.

Do the same process on the right side with the smaller type gauge. You should end up with something like this.

Now all you have to do is start connecting the ticks. The numbers don’t mean anything other than to reference where you are, but if you have a line that falls at tick #4 on one side, the find #4 on the other and you have your perspective line! It’s that easy, and all that struggling just falls away.

That’s it- quick easy and ready for perspective ruling! If it doesn’t seem extreme enough or too much, simply erase one set of ticks and make the spacing more or less accordingly. The first couple of times may feel a bit odd, but you’ll get the hang of it in no time. After a while, you’ll start to notice how you really get a feel for perspective and that you observations when sketching outside or doing interiors will speed up.

Eureka!

Next time- the great falsehood of 1 point perspective!

Sketch, rinse, repeat.

May 23, 2010

Almost every time I see someone getting a portfolio review it ends with the same comment, “Sketch from life”. Invariably you can tell which artists listen to this and which ones don’t based on one simple fact, what their work looks like the next time you see it. I know artists, and even some friends, who never heed this advice. Over the course of years, I never see any improvement in their work, yet they always talk about how they’re on the verge of getting a big break at Marvel or DC. It never happens.

When I moved to Washington, not knowing a soul, I spent a lot of time at the local coffee house sketching, and sketching, and sketching. Within six months I felt I had improved more than I had in the previous five years. Why? Sketching. It’s not something I have a ton of time for anymore, but I try to get out there and do it whenever possible.

Why do it though? What does it actually help with, other than filling time and justifying the adage of “practice makes perfect”? What’s so damn helpful about it that drawing super heroes from imagination all day doesn’t do?

Sketching from life isn’t some snooty requirement of art schools- it serves a purpose. It trains the eye and brain in all the basics of drafting at once. Every sketch done from real life is a crash course in drawing. It trains one to see proportions, perspective, anatomy, shading, textures, dynamics, composition, etc, all at the same time. There’s not a wasted moment with sketching from life, not that I’ve found. Drawing from imagination, before you’ve learned how to draw to begin with, only reinforces your mistakes and poor decision making.

Despite slight variations in design, we all know the approximate height that a seat of a chair should be in relationship to the human body, how high a counter top in a kitchen is, the width that a chair/couch needs to be to accommodate an adult human, or how wide/tall a door needs to be to allow a person through it. The practice of drawing those proportions makes us more aware of the relationships between objects, so when we are drawing from imagination, we can gauge these fabricated spacial relationships more accurately. Artists who fail to understand these proportions will convey it in their images, often that is what feels off about a drawing. One common mistake I see is in overhead (bird eye angle) shots of people standing next to a car that they could never fit in. Sometimes, it’s a dining table way too short for people to fit their knees under should they have a meal at it. These are obvious mistakes to spot, mainly because we see these spacial relationships between humans and objects every day.

Clothing is the same, many artists can not draw a suit jacket to save their lives. It’s one of the most common costumes seen in comic books, yet so many artists can’t do it. Clark Kent = suit, office workers on street when heroes are fighting = suits, mobsters = suits, secret service agents = suits, hell, any government agency guys = suits. Folds in fabric are like snowflakes, never the same twice, but subject to certain consistencies by their nature. Cloth has certain tendencies, but no hard fast rules, yet again, it’s obvious to the eye when it’s drawn incorrectly. Drawing from life is the best education an artist can get for rendering fabric. Yes, understanding the basics of drapery is beneficial, such as how folds tend to radiate out from a pinch point, and how they stretch from whatever structure they snag upon; from an elbow to a knee. But sketching from life trains the mind to understand how clothing moves on the form, so when you draw cloth from imagination, you have a sense of how it should look. Pant cuffs around the ankle are some of the most difficult as they change shape drastically from standing position to sitting down. An artist has to be aware of these tendencies for in those subtle, but highly important, details one can make or break a drawing.

Drawing from life will help an artist more than all the books, lectures, suggestions, or insights anyone can give. My wife read an article that said it takes 10,000 hours of practice (on average) for a person to master any skill. I know I’m not even half way there yet, and we all need the practice. This doesn’t mean an artist should labor all of those 10,000 hours on one drawing trying to get it perfect. Jason Pearson told me about an artist who used to show him artwork at conventions, Jason would critique it, and months later that same artist would show Jason the same pages with the corrections. Jason said the artist did this several times, always showing him the same pages for critique, as if he got those pages perfect, then he would get work. Jason told the guy, “You need to do new pages every time, not just continually correct these. You’ll never learn how to handle new challenges if you only address these over and over. Plus, when working on a book your schedule only allows about a day to each page. What publisher is going to give you six months to get one page right?” The basic idea is to sketch, rinse, repeat. As an artist you learn more from the challenges in a blank sheet of paper, than you ever will in perfecting one image. By the end of a month, what would you rather have? 100 sketches showing a clear, forward progression of skill? Or one drawing you think is perfect now, but will find fault in within the week? I’ve never known a great artist who thought they had done a perfect drawing. I have known plenty of artists who think they’re ready to go pro, and all the guys working just don’t know it yet.

Now, shuddap and go sketch something.

My wife passed along an interesting article about the use of “optical aides” by Renaissance painters and various artists through history in their works. I’ve heard, and read, arguments that Vermeer used a “camera obscura” to trace out his images, and that the pinpoints of white he used as highlights, are actually visual proof of it because that’s not how naturally reflected light appears to the eye.

The article she passed on from the New Scientist website discusses how painters who used similar methods to trace out their images, but they left behind clues of their method in the mistakes made when shifting the equipment to bring various points of the subject into focus. It’s interesting that the article also provides a diagram of the mistakes in perspective to justify the claims. A nice visual aide for those who may not be familiar with the theories involved.

I’ve recently been considering doing a few posts on tricks of the studio, one of which is what Brian Stelfreeze taught me about doing perspective to make it easier. It’s something all artists fight with, from basic levels to insane works like those by M.C. Escher.  Until then, take a look at this article from New Scientist, very interesting.

Lastly, I’d like to pass on my condolences to the family of Dick Giordano, who I mentioned recently in my posts as someone who gave me an informative review of my work. Dick was a legend in comics, having worked as an artist, writer, and editor for decades. He was influential in the careers of countless professionals, and was always willing to give advice and support. I only met him briefly a few times, but he was always kind, but his words had the weight of gold about them. He will be missed by the comic industry as a whole.

Here’s an article on Newsarama about his passing for those who are interested.