I may have mentioned this in the past (or perhaps not, in which case it's a liiiittle late for me to be saying it now), but I generally don't pay too much attention to students' self-critique, simply because it tends to be more distracting than anything else, influencing where I pay attention based on what the student believes to be important, rather than what I actually consider to be relevant.

I did however, against my own judgment, skim through the part about your x-wing. Honestly I'm not entirely sure what you're referring to - from what I can see, each of the 4 wings tucks roughly into one of the corners, and it all seems to more or less check out. I do see a few small gaps here and there, little inaccuracies that certainly can accumulate to create the kind of perspective anomalies you were talking about (especially when we're dealing with a vanishing point that is really close by, which causes more rapid foreshortening/distortion) but as far as this course goes, perfect perspective was never really a big concern. Drawabox is intended to equip students to work within the generalities of perspective, without getting bogged down by them. To that point, your drawing came out fine.

Anyway, before I get further distracted, let's get into your main critique. The first thing that jumped out at me had to do with the many wheel constructions you included - they by design are starting off with a pretty severe flaw: since all the horizontal lines remain parallel, there are a lot of cases where you end up drawing those wheels without any foreshortening (or at least attempt to). Sometimes you end up with foreshortening in there, likely because your gut knows better, but this kind of setup should definitely be avoided. We can see its impact more notably in the wider wheel constructions, where they visibly fall into the patterns of axonometric projection rather than perspective projection.

For what it's worth, I still don't see you attempting to capture the chunkier tire treads using cast shadows (you're still outlining them first and foremost). I do see you using more implicit drawing techniques in other cases, although there you're still drawing individual lines rather than outlining and filling purposefully designed cast shadow shapes. Breaking away from this considerable reliance on line requires making an active decision to do so. That's all on you.

So the main assignment for this lesson is broken down into two parts. The "form intersection vehicles" and the main vehicle drawings. It seems that you more or less did the same thing for both - which isn't really a big deal, but I figured I would call it out. The initial drawings are supposed to just basically be the form intersections from lesson 2, but arranged as though one were trying to block out a vehicle. No worries of specific proportion and whatnot, just placing forms in the world and welding them together.

As a whole, there are some concerns, but overall you've done a pretty good job. The main concern is that while some of these came together really well in the time spent on each, all things considered you didn't actually end up spending all that much time compared to a lot of the other students who reach this stage.

Now when I notice this, usually it suggests that a given student has rushed through, skipped important steps, relied on guesswork, and generally not done the work as well as they could have. That isn't really the case with you - not for the most part, anyway. Most of your drawings are actually quite well done.

Now this 45 minute car is definitely falling pretty short, simply because you skipped a ton of steps and did a lot more "winging it", basically building the overall bounding box and then jumping into drawing a car inside of it. It was however your first vehicle drawing. Your second one showed considerably more care (and took twice as long) and as a whole the structure came out really nicely. The wheels on the side definitely could be better (and the linework there is pretty scratchy) but an ellipse guide would definitely have made that easier.

Now I generally don't like showing other students' work because it gives the impression of a comparison of skill - and while I'm going to do so here, that is not my intent. You've already shown that your grasp of 3D space and overall level of skill is quite solid. I want you to take a look at veedraws' submission just to serve as an example of how far each drawing can be taken, both in terms of construction and subdivision, and in investment of time.

For example, this bugatti took an investment of about 1hr 45min of analysis/study (for the proportions), and 4h 15min for the drawing itself. The ellipse guide definitely helped her nail the wheels (and the use of ballpoint was effective in keeping the linework from getting clumsy or too heavy), but as a whole, one of the most important things to pay attention to is the fact that the vast majority of the time was spent on putting down construction lines, pinning down specific subdivisions, and finding very specific positioning for every component.

I am genuinely surprised that you've been able to capture constructions like the x-wing, the galleon, and the star destroyer in as little time as you did - but I do believe you can achieve far greater results than you realize, by slowing down your overall process, and really taking your time.

So, before I mark this lesson as complete, I'm going to ask you for just one more drawing, with a few restrictions:

• I'd like you to spend at least 4 hours on it. You can break it up across as many days as you wish.

• I'd like you to use ballpoint pen for the whole drawing, just to help avoid your lines piling up as thickly as they certainly can when digging as deep into construction.

• Don't be afraid to put lines down to help flesh out the overall structure.

• If at all possible, please try to get your hands on a "master" ellipse template. That's the cheaper option which has more limited size options with decent degree coverage, but can still be extremely useful.

• It has to be a car. Nothing ficticious - use a real car, and try to get high resolution reference images to work from.