So your question actually made for a bit of a conundrum for me, because it's not a presumption I've ever made before. The rules for perfect circles in 3D space always adhered to the simple criteria of a minor axis aligned to the vanishing point perpendicular to the circle's surface, and the contact points aligning towards the other two VPs. That said, logically speaking if a 90 degree ellipse (a perfect circle) is facing the viewer dead-on, and a 0 degree ellipse (a flat line) is facing perpendicular to the viewer's orientation, then logically that should mean that maybe the sum of 90 degrees for that angle is maintained?

Now it's almost 6am (I had a late start on this critiques, and then ended up getting distracted by this conundrum several hours ago, since I decided to take an early peek at your submission), but I think I have an answer. And that answer is no. Or at least, I don't think so.

I analyzed the issue in three different ways, starting with looking at what I had drawn in my Measuring to Scale video. As shown here, the left ellipse has a degree of 38, and the right ellipse has a degree of 52. Add them together and what do you get? 90! So far, things look promising for your theory, but one data point doth not make a rule.

Then I took a couple cubes in Blender and drew ellipses on them, abiding by the two criteria to represent circles on those faces. Here we get 22/52 for the lower one, which wasn't rotated too dramatically to one side or the other, and 7.4/72.5 for one that was rotated more dramatically. Two points against.

Given the fact that here we're working more with the precise perspective of a 3D modeling program, it looks to be pretty much a given that the degree of the two ellipses does not add up to 90, but I had one last idea that would at least in theory, disprove the notion.

We know that a 90 degree ellipse is a circle, and that it would also have a surface that was facing the viewer head-on. If it were to sit on the face of a box, that box would be drawn in 1 point perspective, due to its alignment with the viewer. Based on your theory, this means that the ellipse drawn on the side plane of the box would be 0 degrees - basically meaning we wouldn't see any face for it at all. But that unfortunately would break the rules of perspective - since that face would be converging towards a vanishing point. We'd only get that 0 degree ellipse if there was absolutely no foreshortening whatsoever.

As shown here, given my little test, I ended up with a side ellipse with a degree of 14.

Now while that tells us that your 'rule' doesn't work in every situation (or even most situations), it doesn't mean it's a complete loss. Due to the need for sleep, I can't pursue this any further, but I can leave you with this:

There is one circumstance in which your rule does work in the 1 point perspective test - if there is no foreshortening at all. And that makes sense, because you were working off the 90 degree relationship in 3D space. But we're working in 2D space, where the way we see things is subject to the focal length of the viewer's "camera" lens. So is there a missing variable that allows us to find the "sum" angle of two perpendicular ellipses' degrees? Ellipse1 + Ellipse2 = 90 - X, perhaps? Or more likely = 90 * X...

Anyway, that's far too much math for one night. Let's get to your critique. To put it simply, your work here is fantastic. You've ended this course on a ridiculously high note. There are a couple things I could point out - for example, this tractor's front wheel being lopsided, the fact that the quad bike's handlebars not being taken into consideration for the main enclosing box, or the somewhat sporadic use of form shading despite lesson 2 mentioning that it should not be included in this course.

In truth, none of those things really matter. What you've demonstrated here are all the qualities that stand as the backbone of this course:

• Incredible patience and tenacity. While they weren't the most time consuming drawings I've seen from students, they're definitely up there in the top 5, and regardless, you've shown the willingness to touch it out through a drawing, committing only the best of your ability and to demonstrate the peak of what you are currently capable. Doing one or two 4+ hour drawings would be impressive, but you clocked in a bunch. Consider me humbled.

• Your capacity for spatial reasoning is spectacular. You've constructed these vehicles to be believably solid and heavy beasts, and you've done so with apparent ease. And to think that when you submitted your Lesson 2 work, you didn't even attempt the intersections themselves until you were forced to with a revision. Now you're marking out those relationships between your forms as though you were merely tracing upon their shared surfaces.

• Hesitating in the face of the intersections back in Lesson 2 was understandable, but it did show a lack of confidence. Confidence is important, both when it comes to how we draw, as well as how we approach every task. Here you've shown immense confidence - the capacity to jump into a difficult construction and figure your way out of it, knowing either that you can, or that if you fail, there'd be nothing lost (except maybe some time).

I could go on, but I think you get the point. You've captured here precisely what I aim to provide, and while I know for a fact that the vast majority of it was earned through your own hard work and toil, I can be proud to have played some small role.

So, with that, I will go ahead and mark this lesson - and with it, this course - complete. Congratulations.