Whew, for a second I was confused, since you drew your cylinders on the back of what appears to be your box challenge. I thought for a moment that maybe you'd totally misunderstood what the cylinders-in-boxes section was about.

Starting with your cylinders around arbitrary minor axes, these are largely looking very well done. You're doing a great job of checking your ellipses' true alignments, and you're drawing each ellipse with confidence to maintain a smooth, even shape.

One thing I did catch - more in the earlier cylinders than the later ones - was that on occasion you'll draw cylinders whose side edges remain completely parallel (on the page, in 2D space), resulting in no convergence and no actual shift in scale between the two ellipses. This is actually incorrect.

Foreshortening manifests in our cylinders (and really in all forms) in two ways. It is reflected in the shift in scale from one end to the other, where the closer ellipse is larger and the farther ellipse is smaller, caused by the convergence of those side edges. It also manifests in the shift in degree, where the closer end is narrower and the farther end is wider. By looking at either of these signs, the viewer can see that because their cylinder is tilted slightly towards or away from the viewer (rather than running perfectly parallel to the picture plane), some of its length exists in the "unseen" dimension of depth. That rate of foreshortening tells us just how much more we should expect to multiply the length we can see on the page to actually understand how long it is in 3D.

So ending up with no convergence to those side edges, and therefore no shift in scale from one end to the other, presents two problems:

• Firstly, it makes the two shifts inconsistent. The shift in scale tells us there's no "unseen" component, but the shift in degree tells us there is. These should happen mostly in tandem (roughly, you don't have to be perfect), with a greater shift in degree corresponding to a greater shift in scale. I say mostly because there are other factors that'll make the degree shift (which I discuss in the lesson 1 ellipses video).

• Secondly, the only situation where we end up with no convergence for those side edges is where the cylinder runs parallel to the picture plane - meaning, it's not slanting towards or away from the viewer at all. Because we're rotating our cylinders totally randomly here, we can basically assume that this perfect alignment will never occur, so we should, at least in the scope of this exercise, always incorporate some convergence, and some shift in scale.

For the most part, across your considerable number of cylinders, you did this correctly. There is the odd case where the two shifts are definitely out of sync - like 115 where the degree becomes vastly wider, but the scale remains fairly similar, but it's definitely an outlier.

Continuing onto your cylinders in boxes, the issue you so boldly fessed up to, as if it'd actually keep me from sending you to cylinder hell should I choose to do so, is actually not as much of a problem as you may think. There is actually another minor issue related to it, but because of the actual focus of this exercise, drawing the minor axis while constructing the cylinder isn't that important. It does help, though.

So this exercise really comes down to the boxes - specifically, training students to construct the kind of boxes that would contain a cylinder. Boxes with a pair of faces that are proportionally square (so they could contain actual circles in 3D space).

We develop the student's instinct for this kind of understanding of how what we draw impacts the proportions of what exists in 3D space in the same way we have the student get used to drawing lines that converge more consistently in the box challenge - through the line extensions. The ellipses themselves add their own line extensions - that of the minor axis and the two contact point lines, and by checking how far off those lines are from converging with the box's own vanishing points, we're able to test how far off we were from the ellipses representing circles in 3D space, and in turn, how far off we were from the planes that enclose the ellipses representing squares in 3D space.

Now, here's the important part - when it comes to extending the minor axis line during this analysis/extension phase, it's important that we extend the true minor axis for each ellipse, not the one defined by the box. It seems that you were mostly drawing from the center of one plane, through the center of the other, which gives us one line rather. We want two lines, one for each ellipse, splitting it into two equal, symmetrical halves down its narrowest dimension.

Since you weren't testing for that, your work could definitely have been more efficient as you worked through this challenge (there's the possibility that your ellipses weren't lining up correctly), but it's just really something to keep in mind when practicing this exercise in the future.

From what I can see, I do feel that your sense of those proportions has developed nicely, and I think you should be well prepared for what you'll encounter in the next lesson - where being able to start your cylindrical structures as boxes (with the right proportions) will serve an important role.

So! I'll go ahead and mark this challenge as complete. Keep up the good work.