The answer to your question is, due to the question itself, kind of mixed. Cylinders are forms with circular ends - therefore if we are to enclose them within a box, the ends of those boxes would be proportionally square. This is actually something I'll get into a bit further down in my critique, as it is tied to the purpose of the cylinders-in-boxes exercise. But basically you're asking if the minor axis of your ellipses only need to align with that vanishing point if the ends of the box are more square, and the answer is that the ends of those boxes are always going to be square if they're meant to contain cylinders. But again, there's a bit more to this that I'll get to shortly.

First, looking at your cylinders around arbitrary minor axes, unfortunately your work for this first section does not follow the instructions, and will need to be redone. As shown here in the assignment section of the challenge, the cylinders you draw for this part of the challenge need to have lots of variety in their rate of foreshortening, between shallow foreshortening where there is only very gradual convergence of your side edges, and more dramatic foreshortening where the convergence is more rapid. You appear to have forced all of your vanishing points to infinity, which actually goes well beyond shallow foreshortening and isn't correct in the context of perspective projection.

We don't actually control the nature of our vanishing points - not directly, anyway. We control the desired orientation for a given form, and thus the orientation of the different edges that it's composed of, but it is this orientation which determines where that vanishing point will fall. Thus, we only get infinite vanishing points for a limited, specific set of orientations - specifically those which run perpendicularly to the angle at which the viewer is looking out into the world. In other words, the vanishing point only goes to infinity if the edges it governs run across our field of vision, not slanting towards or away from the viewer through the depth of the scene. If we intend for any other kind of orientation, then it would not be correct to force that vanishing point to infinity. This is the same reason we can't have a box with only infinite vanishing points.

In this challenge, just like the 250 box challenge, we are rotating our forms randomly in space, and therefore the chances of such a perfect alignment with the viewer's angle of sight are unlikely enough to be ignored altogether, instead making a point of ensuring there is at least some convergence to those side edges, whether shallow or dramatic.

Now, there are two additional points I wanted to call out for you to keep in mind when tackling this part of the challenge again:

• Be sure to draw each and every mark using the ghosting method, as is required for every mark we freehand throughout this course. Your ellipses tend to get a little shaky and uneven due to the hesitation that comes from not applying the approach in its entirety, and there is some wavering in your side edges as well.

• When marking the "true" minor axis line, make sure it cuts all the way through a given ellipse. You seem to be drawing shorter red lines, which give us a far less clear idea of whether or not that minor axis line is correct at a glance.

Moving onto your cylinders in boxes, your work here is considerably better. You've definitely adhered more closely to the instructions in this section than the last (although I can see the line extensions relating to the minor axes of your ellipses were skipped). This exercise is really all about helping develop students' understanding of how to construct boxes which feature two opposite faces which are proportionally square, regardless of how the form is oriented in space - this goes back to your question, as to. We do this not by memorizing every possible configuration, but rather by continuing to develop your subconscious understanding of space through repetition, and through analysis (by way of the line extensions).

Where the box challenge's line extensions helped to develop a stronger sense of how to achieve more consistent convergences in our lines, here we add three more lines for each ellipse: the minor axis, and the two contact point lines. In checking how far off these are from converging towards the box's own vanishing points, we can see how far off we were from having the ellipse represent a circle in 3D space, and in turn how far off we were from having the plane that encloses it from representing a square.

While skipping on the minor axis lines in your extensions is definitely an issue, and one we will address with a few revisions as well, it is far less significant of a concern than the other ones I've called out. To explain why it still matters however, is that all of the line extensions operate together. If we neglect to extend the minor axes and check their alignment with the boxes' vanishing points, we leave a little gap in which those proportional errors can accumulate under our noses, without our being aware of them.

Based on your question, it seems like you noticed that the ellipses' vanishing points were not aligning correctly, and instead of considering how your approach could be adjusted to bring them back into line, you decided to drop that line extension. Going forward, it's extremely important that you not alter the instructions on your own. While you may not know how it's all meant to come together yet, you can trust that if you follow the instructions to the letter without modification, if those instructions end up being somehow incorrect, that will not be held against you.

You'll find your revisions listed below.