Finished Ahmed Aldoori's 100 heads challenge. I'd appreciate some tips
3yr
Yanislav Ivanov
Hello, I'd appreciate some tips on drawing less defined faces.
The problem:
Less defined faces are harder to capture accurately. They look cartoonish at best.
Older people have more forms to "anchor onto" and they flow from one to another easily.
Younger people have... not-so-tightly knit together forms(?). Example: You'd have the nose sink into the void of a smooth face and I can't relate to it for other features.
Notes:
- I can spend 30 mins on 1 sketch of an older head, this can't apply to younger heads since there's just not that much to add.
- When it comes to likeness, I found it's better for me to just estimate and triangulate/ drop verticals instead of going for Loomis' approach.
- Old people have many bumps, meaning if I don't draw a bump accurately, it's barely noticeable.
- Young people have only facial features and smooth skin, meaning if I miss the eyelashes' shape by a tiny margin it's instantly noticeable.
- If I measure everything perfectly, capture accurate shapes and proportions, the best I get is a cartoonish version of the reference. (or an ugly version)
- If I don't accurately capture the same things on older people though, it's not that big of a deal and still looks similar.
- shading smooth skin is harder than stark plane transitions resulting in high contrast.
- If I don't put any shading at all it looks even worse.
Great job on the 100 heads challenge, i had a hard time finishing it and it took me more than 20 days.
I also struggled with the younger faces in the challenge, the more wrinkles and expression the better the likeness. With younger faces I tend to shade less, but i always apply a layer of graphite (by rubbing with cotton over the first lay-in) and then I can use a eraser pencil (https://www.google.com/search?q=koh+i+noor-artist-eraser-pencil) to get subtle features on the face, but pure white I only use for highlights .
One of the issues I had was because I tried putting 10 faces on a piece of paper that the margin for error gets smaller as well (although i do love working small)
I used some construction lines but I didn't construct a full Loomis head I think for any of them.
Thank you, I think your lines and values look insane!
I also tried to put 10 heads on a page but when I got to the older heads I decided to just blow up the size on them since I want to add everything lol. I think time spent = better results but on young heads you just can't as much time as on an old head.
giant progress between the first once and the last once. splendid
I would recommend looking at Andrew Loomis’s “Drawing the Head and Hands” for the simpler faces. The women’s faces, especially, have simple features but still look like they could be portraits (they’re not like cartoons). You should be able to find the book online. Great faces! You clearly got a lot from this challenge!
Hey men, really nice work! Especially you beard-with faces.
If it is helpful. I remark that you need to use al least some structure, some guide lines. Even if they are a first attempt in a clear-light line to draw over or change it and do it again. It is actually a step in when drawing from observation.
Second. I personally agree with the difficulty of young faces. For me it has been useful to not go for examples seen from exactly the front or the side. Search for something with features close or over the others; like a 3/4 and so on.
Another trick from my experience is that, if the face is not helping you, then put the light on your side. I mean: make the shadows enough present that you can use them as guide lines, plus they give more interest on the subject.
Hope to be helpful!
Thank you for the tips! I usually use perspective to align features. Also yes, that's exactly what I found out with higher contrast lighting - It's way easier to capture the form, but most model photos have diffused light on top of their smooth faces lol.
I did this challenge too (it was 100 heads in 10 days). I noticed some of the same things as you. Even though older people were harder to draw because they had more details, they were a lot more forgiving if you made mistakes, whereas with a young face if you got a line slightly off it was obvious. Here are my drawings. I spent more time on some than others.
These are so fun! You'll get so good at this rate. If I could offer a word of advice: don't be afraid to draw through the form. And maybe draw bigger if you can (using cheap copy paper so you don't spend a fortune on sketchbooks). I mainly say that for the more complex ones- drawings that are very detailed need space to breath. If you cram too much detail into a small drawing it won't read from far away
Nice! If I may give a tip: I found out that highly defined faces are hard to do in small sizes so you could try to make the next in similar size to your pen. For the younger ones I just tried to sketch them like Ahmed Aldoori's approach to get something semi-good.
I think the cartoonish look is because you're outlining the features, which is not how they are in reality. And yes, in my experience, the Loomis approach (as with other constructive methods) does not lend itself to a perfect likeness. The atelier block-in method is better for that. The Loomis method is better for inventing and turning heads in space.
That's exactly what I found out about outlining the features!
Before I started this, I decided to study how Ahmed Aldoori does his sketches since mine were outlining most features (approaching it as a construction drawing, non-organic shapes). I saw that he would outline lightly the features but then adding values makes the whole difference. After value, the next thing I saw was line weight and line-art getting included in sketches to add to the believability. I think that's because if you lack space you have to just indicate stuff with value (again), which is the line art's purpose.
Example: tear ducts, eyelids and laugh lines - if you have too little space, just capture them with value instead of line (which also would make them look older).
Is it possible that you're either just perceiving this "cartoonish" look on your younger faces due to the fact that some stylistic choices which we associate with cartoons (larger eyes, simpler silhouettes, etc.) are actually a natural distinguishing aspect of youth as opposed to age? Or, further, that you actually make such stylistic deviations unconsciously due to the fact that those subjects actually do possess those variations, and you're just unconsciously exaggerating those bits just a little bit more than is accounted for by the age difference?
I don't want to discount what you've specified, that this occurs even when you measure perfectly and nail the shapes and proportions, just thought I'd throw out those considerations as possibilities. I would try to use a process of elimination to work through all of the possible reasons this is occurring; suppose you eliminate shape and proportion issues, what else could it be? Perhaps you could use finer/subtler mark-making, as it makes sense that bold and sketchy technique would lend itself to the weathered visage of an aged person, while a younger face would be better served with more careful, gentle linework. If you do some studies and find that making that change doesn't positively affect your results like you're looking for, move on to the next item for elimination, and so forth.
Iterating on the same subject while only making limited changes to your process each time might help you isolate what it is that's causing your dissatisfaction. I hope you're able to puzzle it out!
This is exactly what I'm kinda getting at too!
While studying Ahmed Aldoori's style of sketching I noticed that line art and values play a huge role. Before that I mostly outlined everything in a technical approach, like I'm outlining a robot's plane changes and not an organic blob with smooth transitions.
The "cartoonish" look, I think, mostly comes from the need to use mostly line art to cover small details: unless you blow up the size and have a head on a whole page, you just can't add those tiny details like the tear ducts, eyelids and skin folds, resulting in a need to convey many things in 1 stroke. But if you can add value, then there's a possibility to make it more realistic. Unfortunately most young people have just a smooth blend on top of diffused lighting, resulting in a comic/cartoon-styled cell shading and lineweight.
But when it comes to old people, as you said - a messy, strong values approach works because it's more similar to painting with tones rather than capturing form with line only. Using tone instead of lineweight as the driving force.
Wait, I think I got it.... a hypothesis:
Inherently, using line as indication of form makes things look cartoonish (flat). On top of that, the scenario of a young person with diffused light kills the chances for adding impactful tones to convey form.
Inherently, using values as indication of form makes thing look realistic (3d). On top of that, the scenario of an old person with high contrast gives more chances for adding impactful tones to convey form.
So, old people are easier to make realistic on smaller sized canvases since there are conditions to use values as indication of form. Young people do not give such conditions, resulting in a comic/cartoonish style of a sketch.
Unless you blow up the size to add tiny details to make younger heads realistic or add more contrast, you would end up with a comic-esk face even if you've captured correct proportion and likeness.
Maybe I'm just spilling crap, but this the most I can make out of it lol.
It looks like you made some good progress between 20 and 100
There is one major thing that stands out to me as far as what you could work on.
It look like you are not adding structure before putting in details which could be fixed by using constructive drawing methods. When you render before the structure is in place correctly the images may still feel flat or proportions are off even if you get the details right. I think if you work on defining how the forms sit in 3D space and the proportions are correct before rendering you will see some good improvements.
Thank you! I think it's mostly because young people tend to have smoother faces, resulting in the need of very subtle shading (which I'm bad at for now) and using lineart to convey small details. Before starting this I used to define all of the planes, resulting in a robot/odd look, which works on older people since they have many bumps and adding value later makes it look realistic.
