Congratulations on getting through all 250! This challenge is definitely a doozy, and it's commendable that you pushed through.

Throughout the set, I can see that you've been quite fastidious in checking the true alignment of your ellipses (in comparison to the original minor axis you were aiming for). I can also see that while your ellipses are at times a little more hesitant, you have been improving on this over the set. Just remember, above all else - drawing from the shoulder, and doing so with a confident pace will yield the most evenly shaped ellipse. Leaning into the confidence of that stroke will inevitably sacrifice some of your control and accuracy, but that will be regained with conscious use of the ghosting method, specifically investing your time in the planning and preparation phases, and of course practice.

Now there is a big problem that stands out in your first section of this challenge - you appear to be drawing your cylinders - those around arbitrary minor axes - with absolutely no convergence of those side edges. You're keeping them completely parallel to one another. This is incorrect for two reasons:

• Firstly, the rate of convergence is one of two signs we use to convey how much foreshortening is being applied to the form. Those two signs are the shift in scale from one end to the other (caused by the convergence of the side edges), and the shift in degree from one end to the other. You've got varying amounts of the second sign present, but none of the first - these should always be applied in equal measure in order to provide a consistent signal to the viewer of how much foreshortening is being applied.

• The other reason this is incorrect relates to what the foreshortening itself represents. Because we're drawing in two dimensions, on a flat piece of paper, we can convey to the viewer the length of a given form (like a cylinder) that runs in a dimension parallel to the picture plane - so from left to right, top to bottom, or rotated in any fashion along that 2D slice of space. As soon as the form starts tilting towards or away from the viewer however, a portion of its length will no longer be visible in this dimension parallel to the picture plane. It can only be conveyed through foreshortening - so by increasing the foreshortening, we can tell the viewer that in calculating the length of the form, they should be multiplying it by some factor. The only circumstance when there would be no foreshortening is if the form is running parallel to the picture plane, which would put the vanishing point at infinity, as discussed way back in Lesson 1. Any other orientation will demand some foreshortening, even if it's only very slight. Because this challenge has us drawing forms in completely random orientations, we can pretty much assume that none of these will run so perfectly parallel to the picture plane so as to eliminate all foreshortening.

It's worth mentioning as well that in the assignment section, I specfically did request that students include lots of different rates of foreshortening. This was mainly to make it easier to pick up on the first mistake (where students wildly differ the shift in scale vs. the shift in degree).

Continuing onto your cylinders in boxes, here you have done a better job of incorporating at least some foreshortening. I can also see that while your markmaking earlier on was pretty hesitant and uncertain (when drawing the ellipses), you got much more confident as you pushed through the set. I'm also pleased to see that you were applying all of the necessary line extensions - this is an important component of this exercise, where the focus is to develop a student's instincts for constructing the boxes themselves, such that they feature two opposite faces that are proportionally square.

We work towards that by using the cylinder - or rather, its ellipses - as part of an error checking method, building upon the box challenge's line extensions. By checking the ellipses' own line extensions, and how far off they are from converging towards the box's vanishing points, we can check how far off we are from having the ellipses represent circles in 3D space, and in turn, how far off the planes that enclose them are from representing squares in 3D space.

In this regard, I have just one recommendation to continue making this exercise more effective. When extending your minor axis lines, extend them all the way as far back as your box's line extensions. In extending them only partially, it may be somewhat more difficult to factor them into your analysis at a glance. While your minor axis alignments have definitely improved over the set, I think they are the ones that are lagging behind the most.

Anyway, your second section is coming along well, but the first section as a critical issue. As such, I'm going to assign some revisions to demonstrate your understanding of what I called out.