Aunt Herbert的論壇貼

OK, those are 15 individual sketches, which is a lot to critique. I'll try a general overview of my impressions instead. I like the curves, and I find your lines mostly very readable and deliberate, but sometimes you end up with a bundle of lines, where a single line would have been more convincing.

Anatomy and proportions work mostly well, with minor mistakes sprinkled in between now and then. You emphasize the big forms, and avoid getting lost in details, which makes your bodies appear quite compact and gives them a pleasing stylized appearance.

I am aware, that the owners of this site do not so much shoot their own photographies, so this isn't as much a suggestion to them, but to photo artists maybe looking for inspiration on what images this site could profit from.

And that is really the portrait library. There seem to be a lot less portraits than poses available. And furthermore, all the portraits have a lot in common. There is the one fact, that almost all portraits stem from the same 7 or 8 models, which is unfortunate, but maybe can't be changed. But on top of that, about 95% of all the portraits are shot in a very similar manner. Shoulders are almost exclusively relaxed and parallel to the camera. The lighting is pretty much universally soft room lighting from above and to one of the sides, background is always neutral.

When you are already in the business of shooting images of poses, could you just occassionally zoom in on the heads and shoot a set of "portraits" in the same go? Fullbody acting and posing really has a stark effect on the facial expression.

Or just be more inventive with the lighting situations? Stark direct lighting, natural lighting, atmospheric lighting, backlit portraits, there are so many possibilities to experiment with and to explore the planes of the human face.

In hope of having given someone a few ideas

AuntHerbert

You show a decent understanding of abstracting simple lines and forms, and draw clean and readable poses. I suggest you try to include anatomical observations next.

I think the next step forward for you is to understand how the ribcage actually looks and functions in the chest, and try to "see" it in a reference before drawing. Short description: a flattened egg-like form, with the "pointy" top ending between the shoulder joints, just underneath the neck, and the bottom cut off along the line of the lower ribbs. That cut-off extends on the back down to about a hand-width above the hip, about two handwidth on the front, where the belly is. Understanding the ribcage is also the first step towards observing shoulder and breast muscles, as they are mostly what obscures the ribcage. Be aware, overestimating the distance between hips and ribcage is a very common mistake, which tends to lead to unproportionally long torsos.

In your drawings the chest area has a tendency to be a bit slimmer than natural and looks like a simple tube. Finding the ribcage would add more substance to the construction.

Btw, female breasts are best imagined as formed like an inverted heart shape, hanging over the ribcage with the tip attached underneath the throat. A good shortcut to look for is the line underneath the breaths. Finding that line and the center line of the heartshape is often more indicative than starting with sketching out one of the spheres and then finding the other one. In constructing the foundation consider even ignoring them, as they don't actually influence the pose a lot, but merely feature in drawing the outline.

Also, check out proko.com for better explanations and examples.

Not a 100% silver bullet, but some steps that helped me to somehow overcome the problem (Also only somehow, it still keeps creeping up on me)

The first step is meassuring. That doesn't necessarily involve holding a pencil in front of you and making a smart face, what it does involve is finding a habit to check what you already have drawn during drawing. Usually we are good at checking horizontal lines and vertical lines. So when you plan your next line, compare the point where you want it to end, to the lines you have already drawn. Should it end a bit deeper or a bit higher or at the same height as that line you already drew over there?

When I mess up a drawing, the cause is usually not that I meassured wrong, but that I didn't meassure at all. I was so concentrated on that details of how the shadows on that right knee looked, that I not once looked left to check how long I already drew the other leg, and such stuff. Or I didn't look at the torso above for more than a minute and now the leg is broader than the entire chest.

In my experience the "meassuring" part falls in two areas: Really building the habit to do it regularly while drawing, and also sort of a training effect to just develop more capacity to keep more relationships on the page in mind while drawing. It's a bit like, once you pay attention to it, it's really easy, but learning to keep paying attention to it while concentrating hard on all the other stuff, like forms, shapes, line quality, controlling your pen, analysing the reference, etc.. is really hard.

A second step is training specific methods. For heads and faces, the Loomis construction is by far the most well known, for bodies, there is for example the O'Reilly method. You basically just grind to know the proportions of such specific objects as the human head or body by heart. It's good to have at least a basic grasp of this methods, and when you find, that you keep struggling with the same parts of your sketches over and over again you can always go back to the foundation to find out more about what mistakes actually keep bugging you.

Then, generally the advice to learn to draw big forms first, before you concentrate on details. Sounds extremely easy, but it absolutely isn't, as it runs a bit counter to our intuition of looking at art. When we look at art naively, we tend to be impressed by all the pretty, pretty details first, and that is what we want to copy. Then we end up with a page of multiple mediocre details, that are somewhat randomly scattered over the page, and just don't form a cohesive image.

Foreshortening or drawing stuff in angles... well there is a quite famous and good page, named drawabox.com. It focuses on perspectivic drawing first and foremost, and does it in a very methodical way. If you do the lessons, and the homeworks, you will have a good idea on how geometric shapes behave in perspective. One of the homeworks, probably the most famous one, and the one the site is named after is: Draw 100 boxes and check each of them for correct perspective! Definitely check the site out. And try it out. Doing it yourself is less gruesome than it sounds, and it feels great, when you are done.

Here is a freebie tip from me, if you ever venture into urban sketching or architectural drawing: Drawing horizontal lines horizontally greatly enhances the final result of your image! I keep repeating it to myself when drawing outside, as my righthanded body has a biomechanical tendency to let horizontal lines tumble more and more to the right when drawing on the right side of a page. I assume lefties have the same problem on the left side of the page.

I am not sure, whether 10 minutes is a good time frame for you right now. You are still struggling with proportions quite a bit, and neither you nor your piece gain a lot by spending time on attempting to hatch out a mediocre foundation. The shapes you found for belly, hips and thighs actually look nice, but the upper body, head and arms are oversized. Also, you don't find your lines immediately, and you don't use long lines, instead going more for a trial and error approach with quite short strokes.

I would recommend practicing one or two minute drawings for now, focusing on simple shapes and clear lines, and consciously reducing the amount of finer details you use for meassuring proportions.

Hatching properly is quite a science in itself, and it has to start from a good constructional basis, as finding a good orientation for the hatching pattern, that doesn't distract from the pose can't be immediately observed from the reference, and has to be partially deduced from abstract methods. I know that showing volume without halftones is sometimes hard, but for now you should try to keep it to a minimum until you become very confident with basic anatomy and find the excess focus to also take a first look into how lighting works, and what tones indicate which types of shade on a body.

Can you elaborate on how the S curve is being "antithiesed"?

The theory behind using (ideally few) simple curves (CSI) is called line economy. Simple curves can be used by the draftsperson in a very controlled way, and reducing the overall complexity of the form leads to very easily readable shapes that look pleasant. There are some minor caveats, like only using single curves ("C"s) on the whole figure, placing them symmetrical and have their ends always match, can lead to snowmansyndrom, where the body appears to be assembled out of a collection of elongated spheres, like a balloon animal.

I am not aware of specific problems with the S-curve, and use it a lot. The human spine, which is often the basis for the Line of action, naturally follows an S-curve when seen from the side in an upright posture, and S-curves appear all the time on various limbs.

I am by far not a master draftsman, but I think I can handle line economy decently well, you can check my results on my sketchbook here or on artstation.com/stefanbast. If you have concrete questions about line economy, I will try to answer them to the best of my knowledge.

• Aunt Herbert edited this post on July 6, 2021 10:25am.

Hi, I am a bit worried,... if I flip through the first 5 pages of "student work", literally half of the stuff posted is from me. And, frankly, I AM holding back quite a bit. I mean, it's the 21st century and we all live in an attention economy. I don't want to suck up all the oxygen in the room, so I wonder, how much exposing my own stuff is OK without appearing to attempt to conquer the site?

I find it quite convenient to be able to post the results of my daily practice here, but, OTOH... Well, a lot of other students seem to do their entire lesson on a single page, which I find questionable from a didactic side, just because only practicing tiny scribbles isn't a very good way to learn good line art. But if I finish even a half-hour lesson, I am holding 18 pages of results. I am usually already only picking the ones I like best, which kind of feels like bragging. Then I have a discussion page in the forum on how to apply the Loomis method to actual photo reference. I finally got a response from someone who seems to have working experience with the method, and I am all giddy and feel the urge to show and discuss all my failed and successful attempts of the last two days, but that would be another 8 page bomb, and I am not even done practicing today.

So, I would love to have some pointers from the owners of this site and other artists, how much posting max is still on the polite and civil side of things, before I start to annoy the living heck out of everyone else.

Or would it be better to host all but a few pieces elsewhere and only post links in the forum? If so, which pieces would be most suitable to be hosted here? The best ones? The ones I struggle the most with?

Thx

Aunt Herbert.

Here is an example of why I find assessing the barebones Loomis construct hard:

Untitled by Aunt Herbert

The head obviously has 2 right ears. The outer one is where I expected it to be, merely from following the construct method, only after I indicated the right eye and the outline of the shadows along her right cheek did I realize, that the distance from her cheek to her ear was way too big, and indeed the cut-off of the side of the head on that side should have been way closer to the centre, making the right side of her head too wide.

Thanks a lot CCC, you truly answered the most important questions, that led me to create this thread.

You are probably right, that I should temporarily switch back to pencil for a deeper dive. Unfortunately, at the moment it seems like all my graphite pencils have been swallowed up by the creative chaos around my workplace. It's probably time to put tidying up my workplace on the schedule. :(

You mentioned how on my 1 minute drawings the Loomis construction is barely visible. I noticed that too, but chalked it off as somehow a natural development: When I draw actual brows, the browline will mostly disappear, when I draw nose and chin, the markers for their placement will be usurped by the lines, the only lines that stay visible in my experience are the excess parts, that don't indicate facial features, like the lower front quarter of the original sphere, and the rim of the side planes.

When I mentioned the lack of individual informations on a Loomis head, that was mostly a reaction to the proposal to quick-sketch them. After 30 secs there just isn't enough visible on a paper to actually critique myself, just the beginnings of a very generic geometric pattern that gives little clues as to whether its proportions or alignments are where they need to be or slightly off.

I still think spotting such inaccuracies without filling in the features first is hard, but your post just inspired me to give it another try.

No, plastic is a good thing, I wanted to say it gives a good idea of their volume.

How to improve,.... well, my house rule, if less than 30% of my practice drawings fail, it's time to raise the stakes. You seem extremely good at what you do, so you should find something more difficult. You got good foundations, now go build a house on them.

What you chose is an extremely personal artistic decision. You could go for stylization, try to go the extreme of exagerating movement, or test out how much you can simplify your shapes without losing information. Or start exploring lighting and shading, or add more anatomical details, or both, to go for an extremely naturalistic finish.

Alternatively you could train to draw a lot quicker or from memory, so you can take passer-bys on the street as reference. Or go drawing from your mind altogether without reference, so you can start storyboarding. Maybe instead of single figures you could go for complete compositions. Or start looking out for different artists, that can really drop your jaw and try how closely you can copy what they do. There is a free pdf of Arthur Guptil's "Rendering with pen and ink" on the net, just browse it, and you'll find a boatload of great artists to be inspired from. Talking about pen and ink, experimenting with different pens, pencils, brushes could be worth a try...

You seem to meet your goals. I love your lines a lot, your torsos look very plastic, the movement looks natural and dynamic. If those figures would be printed in a drawing tutorial as examples, they wouldn't look out of place.

Thanks, I think I understand now what you mean with "redlining", and see the advantages. I'll have to spend some time scratching my head to decide how to best integrate it into my daily practice routine, though.

I somewhat disagree with you on your mistakes. I definitely find your figures very readable, I understood on first glance what poses they are in, and the fact, that the proportions of the body masses and legs seem natural helps that a lot. Incompleteness is kind of a problem of shifting goalposts anyway. If you go with the Atelier style, these guys spend weeks and months on finishing a single draft, while Minamoto Musashi was famous for drawing a bird, a tree or a flower with a single stroke of a brush.

A better knowledge of anatomy will probably improve how you draw joints, but your already well established sense of proportions takes you 90% of the way to the goal.

What I think is more of a problem is, that you don't trust your instincts. You use a whole lot of lines, searching for the one line you like. That's somewhat natural, as your brain isn't completely used to adjusting how your pen moves on the paper to what your eyes perceive, but it's also a bad habit, and very hard to shake. I would recommend trying to live with your first stroke and not correcting it after the fact, even if it is off. You have another attempt in just 30 sec time, when the next reference shows up. You will end up at first with 80% images that you really hate, 10% that you can live with, and 10% that really surprise yourself. If you keep doing it daily, that percentage will shift very quickly. And if the percentage of images you hate drops below 30% or so, it's time to up the game and define more challenging goals for yourself.

Remember, there are at least 5000 bad images inside every artist, and the only way to get them out is by putting them down on paper. :D

The problem with light hatching, especially if you work with ink, is, that you can't really dodge the decision how to orient your hatching lines. There needs to be a clear graphical separation between additional features, like wrinkles, scars or face hair, and the lines of the hatching pattern.

In my experiments the best results appeared, when I actually found the constructional underpinning and oriented the hatching lines along them. The hard part is, these are not anything that can really be observed from the reference, they only become painfully visible when I ignore them with my hatching.

I think I understand what you mean, and it works well if the reference is extremely dramatically lit, just a stark separation of dark shapes on a light background, but if the reference is lighted out somewhat more naturally the image loses a whole lot of information if I just ignore the middle tones.

The effect, that the edge of the hatched out shapes is pretty much defined by where my hatching lines end doesn't make it any easier. Sometimes I draw in those edges with an extra line, but that is often a heavyhanded solution that impacts the overall style more than I intend. If I want to define those edges in more details OTOH I am forced to put the hatching lines closer together, which I would need to correspond with making the lines even finer, to avoid darkening the tone. Drawing the finest possible lines with a brush without losing contact to the paper and ending the line prematurely is a bit of a pain in the butt, too.

So my theory is, if I get that "plane thingy" nailed down to a far further extent than where I am currently at, it would free my focus to concentrate more on my brush work.

I would say the time isn't too short, you use too many lines, as you try to go from details to overall structure. Simply increasing the amount of time might just lead you further down the dead end.

Eventually you will have to learn to see and use overall structure first, before you worry about details. There are some very basic tips displayed by this site throughout the classes, about line of action and finding the masses of the body. A more complete and useful explanation of the concept, as it refers to the human figure I found on proko.com. The free courses are totally sufficient, and I would highly recommend them to anyone starting out with tackling the human figure.