Lesson 1: Lines, Ellipses and Boxes

This is the last exercise of lesson 1, and it's a doozy. Like the rotated boxes, what I want most is for you to complete the recommended number of pages to the best of your current ability. No more, no less. We're not looking for you to impress anyone, but rather to face the challenges head on so they can start making you think and consider a new kind of spatial problem.

You'll probably be pretty used to this by now, but it's the same process we follow through the plotted perspective and rough perspective exercises. Split up your page into 3 separate frames, using a ruler.

This exercise is going to involve drawing a lot of boxes that flow along a gently swooping path through the depth of our scene. In order to help us keep track of our desired path, we'll start out by drawing it.

Be sure to use your shoulder for this, and draw confidently. Accuracy isn't particularly important here - just let your arm flow naturally to create a swooping stroke that goes back and forth.

Using the ghosting method, draw a Y as shown here. The lengths of the various arms is up to you, but do make sure that the angles between them are of at least 90 degrees - when the angles get smaller than that, things tend to get a little... weird.

This Y gives us 3 edges of our box, radiating out from the corner (the center of the Y) which is closest to the viewer. This is very much 3 point perspective - meaning that we're dealing with 3 concrete vanishing points, and we're going to have convergence for all three sets of parallel edges. That suits this exercise just fine, as we'll be drawing boxes that are rotated randomly, rather than aligned in any particular fashion to the direction the viewer is looking.

This is the first step of the "Y Method", something we'll be using extensively throughout this course.

The Y method hinges on drawing a Y, but it doesn't necessarily dictate what order in which you should be drawing your other edges. What we can do however is combine it with the tools given to us by the ghosting method - specifically the points we use to mark out where a line should start and end - to consider a variety of sources of information before deciding on and committing to a particular line to represent an edge.

As shown here, I've got a blue arrow, ending in a little dot. The blue arrow isn't something you'll draw on the page, but rather it represents the path we think our edge will need to follow. Since we're at the early stages and only have one edge actually drawn (from the original Y), technically anything we choose here will be correct as long as it converges in the same direction as the existing edge.

As we add more however, we'll have to focus on having those edges converge consistently. To do that, you can ghost in the direction you think the edge should go, and place a little point somewhere along that path (ideally a little beyond where you think that edge should actually end).

The great thing about these points is that they're so small. Mine has been drawn a little bigger than I normally would so it can be visible in the example, but these points can be quite tiny. This also means that unlike a whole drawn line, they don't really carry a lot of weight and are very easy to ignore.

So, using these points, we can consider certain trajectories, mark them out on the page, and then consider other edges to see if our idea is going to work out.

A corner is where multiple edges meet - but if we only consider an edge coming from one direction to decide where a corner should be, we've only got part of the information. So, let's consider a different edge.

Here, just like before, I've got an arrow in red representing the trajectory I think this edge will need to follow. Rather than drawing it, I've ghosted through its path, and like before, placed a dot at its end.

We've got two points, but a single corner must exist in a single location. There are a few ways we can decide where this corner should be, given the information we've got. We can either choose to pick one of the two we've already drawn, or we can create a third point based on the other two - a sort of "average".

It doesn't specifically have to be perfectly in between the two - at the end of the day, you're making that call based on the information you have.

I ended up adding a third point, somewhere in between the two others.

We can do the same thing for our other three visible corners. As noted in the video at the top of the page, when you're first doing this exercise as part of your Lesson 1 homework, you won't need to draw "through" the boxes as though you have x-ray vision, so each corner's going to be the result of 2 edges.

When doing this exercise in your warmups, and when doing the 250 box challenge, you'll be drawing through your boxes, meaning that there will be a third edge coming in from the opposite side of the box, adding yet more information to help you make an informed decision.

Similarly, as we consider the other corners of this box, we should also be thinking about more than just the two edges at play - because each edge also has to consistently converge with the others in its set. This means you'll be jumping around a lot, considering not only the 2 edges that meet at a given corner, but also the other edges from elsewhere in the box.

This may be a lot for you to worry about all at once, especially this early on, so take it slow. It's okay to limit yourself to focusing on just the two edges for now - as you get used to it, you'll find the number of things you can be considering simultaneously will increase, especially with all the practice we'll be getting through the 250 box challenge.

There's really no reason to rush into things or push yourself beyond your limits here. We're simply introducing the concept, so do what you can. Students who have done their current best to follow the instructions for this exercise (and who despite that still end up with all kinds of issues) do not get assigned revisions.

With your corners decided, all that's left to do is connect the right dots, and your box is done!

All that's left is to fill in the rest of the frame. You might have noticed here that the example is entirely different from the last one - I just switched to the example from the video.

Keep in mind that as you draw boxes farther away along your given path, they'll be getting smaller. For the purposes of this exercise, avoid particularly dramatic foreshortening - keep the convergences of your box pretty gradual, so we can avoid ending up with scale that feels inconsistent across the whole set.