Lesson 4: Applying Construction to Insects and Arachnids

1:46 AM, Saturday August 26th 2023

Imgur: The magic of the Internet

Direct Link: https://i.imgur.com/Y5JKeTw.jpg

Discover the magic of the internet at Imgur, a community powered enterta...

This lesson was very hard and looking back I can see many mistakes. However, I believe I have learned a lot through these exercises. Please do not pull your punches with my work, I am only here to improve. Thank you for your time and effort!

0 users agree
7:52 AM, Sunday August 27th 2023

Hello MurcDirty, I'll be the teaching assistant handling your lesson 4 critique.

Starting with your organic forms it is clear that you're working towards the characteristics of simple sausages that are introduced here, though there are still some inconsistencies such as ends of different sizes, pointy or lopsided ends, and the occasional bulging or pinching midsection, which you'll want to try and iron out through practising this exercise as part of your warm up routine.

It is good to see that you're experimenting with varying the degree of your contour curves. Keep in mind that the degree of your contour lines should be shifting wider as we slide along the sausage form, moving farther away from the viewer. This is also influenced by the way in which the sausages themselves turn in space, but farther = wider is a good rule of thumb to follow. If you're unsure as to why that is, review the Lesson 1 ellipses video. You can also see a good example of how to vary your contour curves in this diagram showing the different ways in which our contour lines can change the way in which the sausage is perceived.

Moving on to your insect constructions, most of your demo drawings are very well done, though I did notice the wasp was unfinished. When drawing along with a demo it is important to follow all of the steps, so you can fully understand the techniques Uncomfortable is using, and this will help you to use the techniques in your own original constructions later. Maybe you were under the impression that the antennae and all the additional forms on the legs fall under "texture and detail" but they do constitute constructed forms and so should be included, even in the "construction only" pages.

Looking through your original constructions, I can see that you're using the techniques taught in the demos, and there's a clear sense that you're thinking about how all the pieces of your constructions exist in 3D space and connect together in with specific relationships. It is good to see that you've drawn through most of your forms throughout these pages, as this helps to reinforce your understanding of 3D space and build solid constructions.

Speaking of building solid constructions, I do have some points that should help you get more out of these constructional exercises in the future.

The first of these relates to differentiating between the actions we can take when interacting with a construction, which fall into two groups:

1 Actions in 2D space, where we're just putting lines down on a page, without necessarily considering the specific nature of the relationships between the forms they're meant to represent and the forms that already exist in the scene.

2 Actions in 3D space, where we're actually thinking about how each form we draw exists in 3D space, and how it relates to the existing 3D structures already present. We draw them in a manner that actually respects the 3D nature of what's already there, and even reinforces it.

Because we're drawing on a flat piece of paper, we have a lot of freedom to make whatever marks we choose, but many of those marks would contradict the illusion you're trying to create and remind the viewer that they're just looking at a series of lines on a flat piece of paper. In order to avoid this and stick only to the marks that reinforce the illusion we're creating, we can force ourselves to adhere to certain rules as we build up our constructions. Rules that respect the solidity of our construction.

For example - once you've put a form down on the page, do not attempt to alter its silhouette. Its silhouette is just a shape on the page which represents the form we're drawing, but its connection to that form is entirely based on its current shape. If you change that shape, you won't alter the form it represents - you'll just break the connection, leaving yourself with a flat shape. We can see this most easily in this example of what happens when we cut back into the silhouette of a form.

This is happening on quite a few of your pages. For example, I've marked on this page in red where you cut back inside the silhouette of forms you had already drawn. Sometimes I think you cut back inside your forms accidentally, when your ellipses would come out a little loose (which is totally normal), and then you'd pick one of the inner edges to serve as the silhouette of the ball form you were constructing. This unfortunately leaves some stray marks outside of its silhouette, which does create some visual issues. Generally it is best to treat the outermost perimeter of the ellipse as the edge of the silhouette, so everything else remains contained within it. This diagram shows which lines to use on a loose ellipse.

On this tree hopper I marked in blue where the opposite is happening. Rather than cutting back inside forms, you'd extended off existing forms using partial, flat shapes, not quite providing enough information for us to understand how they actually connect to the existing structure in 3D space.

Instead, when we want to build on our construction or alter something we add new 3D forms to the existing structure. Forms with their own complete silhouettes - and by establishing how those forms either connect or relate to what's already present in our 3D scene. We can do this either by defining the intersection between them with contour lines (like in lesson 2's form intersections exercise), or by wrapping the silhouette of the new form around the existing structure as shown here.

This is all part of understanding that everything we draw is 3D, and therefore needs to be treated as such in order for both you and the viewer to believe in that lie.

You can see this in practice in this beetle horn demo, as well as in this ant head demo. You can also see some good examples of this in the lobster and shrimp demos on the informal demos page. As Uncomfortable has been pushing this concept more recently, it hasn't been fully integrated into the lesson material yet (it will be when the overhaul reaches Lesson 4). Until then, those submitting for official critiques basically get a preview of what is to come.

Another way you've accidentally undermined the solidity of some of these constructions is by tracing back over large sections of the silhouette to reinforce it with extra line weight. We can see an example with this tree hopper. Going back over your lines causes your initially smooth and confident marks to get wobblier, causing small alterations to your forms, making them appear less solid. This effect is even more pronounced in the places where you've added thick line weight with choppy little chicken scratch marks.

I find that the most effective use of line weight, given the bounds and limitations of this course, is to reserve it for clarifying overlaps as explained here, and restricting it to localised areas where these overlaps occur. What this keeps us from doing is adding line weight to more random places, or worse, attempting to correct or hide mistakes with additional line weight. Keep your line weight subtle, it should be a whisper, not a shout. Usually a single confident, ghosted, super imposed stroke will be enough to create the desired effect.

I noticed on some of your pages you'd added quite a few extra contour curves to the surface of a single form, such as on the head and body of this construction. Contour lines themselves fall into two categories. You've got those that sit along the surface of a single form (this is how they were first introduced in the organic forms with contour lines exercise, because it is the easiest way to do so), and you've got those that define the relationship and intersection between multiple forms - like those from the form intersections exercise. By their very nature, the form intersection type only really allows you to draw one such contour line per intersection, but the first type allows you to draw as many as you want. The question comes down to this: "how many do you really need?"

Unfortunately, that first type of contour line suffers from diminishing returns. The first one you add will probably help a great deal in making that given form feel three dimensional. The second however will help much less - but this still may be enough to be useful. The third, the fourth... their effectiveness and contribution will continue to drop off sharply, and you're very quickly going to end up in a situation where adding another will not help. I find it pretty rare that more than two is really necessary. Anything else just becomes excessive.

Be sure to consider this when you go through the planning phase of the contour lines you wish to add. Ask yourself what they're meant to contribute. Furthermore, ask yourself if you can actually use the second (form intersection) type instead - these are by their very nature vastly more effective, because of how they actually define the relationship between forms. This relationship causes each form to reinforce the other, solidifying the illusion that they exist in three dimensions. They'll often make the first type somewhat obsolete in many cases.

The last thing I wanted to talk about is leg construction. It looks like you've made an effort to use the sausage method for constructing legs, though sometimes the forms you've drawn aren't sticking to simple sausages and occasionally you've dropped in a partial shape rather than a complete form. It's not uncommon for students to be aware of the sausage method as introduced here, but to decide that the legs they're looking at don't actually seem to look like a chain of sausages, so they use some other strategy.

The key to keep in mind here is that the sausage method is not about capturing the legs precisely as they are - it is about laying in a base structure or armature that captures both the solidity and the gestural flow of a limb in equal measure, where the majority of other techniques lean too far to one side, either looking solid and stiff or gestural but flat. Once in place, we can then build on top of this base structure with more additional forms as shown in these examples here, here, and in this ant leg demo and also here on this dog leg demo as this method should be used throughout lesson 5 too.

So - I've outlined some things work on, but your underlying understanding of 3D space appears strong, and these are all things that can continue to be addressed into the next lesson. I'll go ahead and mark this one as complete, just be sure to actively tackle these points as you handle your animals. It's not uncommon for students to acknowledge these points here, but forget about them once they move on, resulting in me having to repeat it in the next critique (which we certainly want to avoid).

Next Steps:

Lesson 5

This critique marks this lesson as complete.
6:10 PM, Sunday August 27th 2023

Thank you very much for the extremely helpful information. I'll try to remember to refer to this specific critique when I feel unsure about some of these subjects. Again thank you so much for the detailed response, it seriously helps a lot.

6:19 PM, Sunday August 27th 2023

No problem! I'm happy to hear that this was helpful. Best of luck in the next lesson.

The recommendation below is an advertisement. Most of the links here are part of Amazon's affiliate program (unless otherwise stated), which helps support this website. It's also more than that - it's a hand-picked recommendation of something I've used myself. If you're interested, here is a full list.


This is another one of those things that aren't sold through Amazon, so I don't get a commission on it - but it's just too good to leave out. PureRef is a fantastic piece of software that is both Windows and Mac compatible. It's used for collecting reference and compiling them into a moodboard. You can move them around freely, have them automatically arranged, zoom in/out and even scale/flip/rotate images as you please. If needed, you can also add little text notes.

When starting on a project, I'll often open it up and start dragging reference images off the internet onto the board. When I'm done, I'll save out a '.pur' file, which embeds all the images. They can get pretty big, but are way more convenient than hauling around folders full of separate images.

Did I mention you can get it for free? The developer allows you to pay whatever amount you want for it. They recommend $5, but they'll allow you to take it for nothing. Really though, with software this versatile and polished, you really should throw them a few bucks if you pick it up. It's more than worth it.

This website uses cookies. You can read more about what we do with them, read our privacy policy.