Aunt Herbert的論壇貼

This is just a question out of curiosity. I feel like certain references appear so often, that I almost start to know them by heart. Not just the model, but the same exact pose, prop, lighting,... On the other hand, it can't be that the overall pool of references is just so small, that I basically know all of them by now, as every now and then an image pops up, that I have never seen before.

I have to add, that I had the same phenomenon, when I was drawing on quickposes.com . Are those just the downsides of working with an imperfect random number generator, or are some images deliberately referenced multiple times in the pool?

Yes Polyvios, but generally the problem in drawing faces is to NOT get them to look like caricatures. Which requires a whole lot of detail, with quite carefully measured proportions. Famous cartoonists and caricaturists often use specific shortcuts, which can indeed be drawn quite quickly. But drawing Elmer Fudd, Eric Cartman or Charlie Brown in under 30 seconds will neither teach me a lot about portrait drawing, nor even about developing my own caricature shortcuts. Even Manga faces are already a bit too complex to gain very much from one minute drawings, although they might be at the border if you really train on repeating them quickly.

I tried quicksketching portraits myself and found no value, but I also found no one online, who promotes quicksketching as a good entry into portrait drawing. The usual beginner tips for portraits to more than 50% start with Loomis, and you can't do Loomis under a minute. And those, that do not center on Loomis tend to be more complex, not more simplified.

I personally haven't found a solution either, pretty much out of the same reasons. I just use "all of the same length" mode with 3600 seconds as timer.

I mean, there are sometimes street artists in touristy places, who sell quick portraits, but even they take 5 mins.

Probably no, but probably yes.

What I mean is, if I look at my own 30 second sketches, they vary wildly, basically determined by what my goal for the session is, and what my mood in the moment is. If I want to go for line quality and consistent drawing rhythm, there might be barely two or three lines on the paper at all. If I want to focus on perspectivic drawing, there will be a few first primitives, circles or initial sketches of cuboids visible, but rarely enough to make sense to publish, because most audiences could barely even guess the purpose of my lines. If I want to focus on pose, I might start with the line of action stick figure, as it is proposed in the tutorial. If I want to focus on proportion, there might be a lot of details indicated, but not as a goal in itself, but to use as landmarks.

Or, for F's sake, if I am stressed and tired, drawing in the break between two shifts, I might be happy to just watch the pen moving over the paper for a while, and when I am in a bad mood I'll just desecrate the paper until the timer runs out.

So, the one "better" technique? Probably no. But the clearer you learn to distinguish between all the different aspects and subskills of drawing, the more purpose you can assign to your shortys, and the clearer the purpose, the clearer your lines.

Looking at your shorties, I can see, what you are working at, and there is nothing crying out to me: "Oh! my! God! No!". You are focusing on displaying a pose, well, I can understand each of those poses at a glance. The measurement and proportions look natural and convincing, as much as it can be determined on a single glance at that low level of detail.

You say, you wish for clea-R-er lines, you could probably achieve a bit of clea-N-er lines, for example for your heads, by just grinding a bit of oval drawing to improve your manual dexterity there. Or by drawing fewer lines at a slower pace. But I am not certain if I should recommend that, as that does not seem to fit to the purpose of those sketches. You could do that, but it would probably compromise on the goal of simply communicating the pose. Also, these poses are clearly drawn to prepare an underlying construction, while line quality ultimatively is an aspect of rendering. If you aren't dedicated to draw from first lines with a big fat ink brush, those first lines will probably be erased or overdrawn anyways before you end up on the final polish.

The way those timed classes work for me: I try to imagine a specific goal for the session, I use the shorties to test out and reinforce the concept for the session, and then the 5 minute and 10 minute drawings as proof of concept.

Often the problems of my concept won't become apparent within the shorties themselves. For example: I try to find big curves, looks fine at 30 seconds, 1 minute, then when the 5 minute drawing is up and I start to flesh out the details, it becomes apparent, that I took massive liberties with proportions. I got another attempt at 5 minutes, same stuff happens, but I try to make it look at good as possible none the less. When the 10 minute figure is in, and the timer is running low, it's crunch time. If all went well, and I feel like I still have to add to the figure, I'll stop or restart the timer to get going. If the result instead looks funky in not a good way, I'll then try to pin down, WHAT aspect I don't like. And then try to come up with a concept for the next batch.

In the example the proportions went wild, so my next batch of shorties will focus more on measuring and determining landmarks. Or, possibly, I decide, that in spite of the strange proportions I really like the 10 minute results, and even wished that it would look even cooler with even less details and a more abstract rendering, so my next batch will focus on spending less time scribbling and more time on observing the reference and looking for even bigger lines....

The thing is, I don't believe there are "perfect" shorties, and can't be by design. The purpose of shorties is to emphasize one aspect of your final rendered work, they aren't meant to be published, and they don't necessarily look pretty. I could upload pages full of shorties, and basically everybody would just search for polite words to tell me, that is just a bunch of mad scribbles, depending on what aspect I am working on.

Lines are clear, when they have a clear purpose, but the purpose of shorties is only revealed in the final drawing. Which is only a waypoint to define a clearer purpose for your next set of shorties.

Other artists can only tell you about what they experienced on their path, and that may or may not be informative for you, but it can't take the task of defining the purpose of your lines away from you.

Wow, that sounded so preposterously Zen, I'll better stop writing and go back to drawing, before I embarass myself even more.

One specific detail... I think ribs are a bit treacherous. They have such a beautiful parallel structure, so everyone is tempted to overdo them, and then the model looks like it is close to dying from starvation. Especially they shouldn't break the curvature outline of the ribcage, but even the general value should be double checked to not overdue them. I try to form a habit to only very lightly indicate them in the early stages, and only start to work on them, once the overall values of the figure are strongly defined.

OK, you are working with charcoal on a rough underground, starting from light searching lines and coalescing towards deeper values. I have seen it done before, it's just a very different style than my usual technique, which focuses more on a geometric construction from clean and decisive lines as basis (at least in idealized theory).

I generally can't say much about fluidity, because I don't get the concept. To me it is such a strange word, that everyone loves to throw around, but it seems no one really agrees what it is supposed to mean, and no two people use it the same way. Something about drawing speed and rhythm, but then it is also supposed to somehow be a line quality, or lead to a line quality, but it also has something to do with exaggeration of the underlying construction, except no actual construction is generally used, and the concept of the piece works from observing value first, and lines last, and it isn't about good old fashioned stuff like just mere line weight or dynamic, and controlling your lines is suddenly a bad no-no, ...? Just don't ask me, you can't good answers from me.

To me what you do mostly looks like the early stages of what other people, who primarily work value centric with charcoal, do. So, I guess you are doing it right? Probably the test would be to do longer form drawings, until you either end up with a finished work, or run into discernible problems, that you then can start to analyze.

I am not certain, whether this style of drawing even uses timed drawing and short sketches for training. I found this Tony Swab guy on youtube, and I think he generally works in a similar way to what you are going for, but don't ask me to explain it. And I remember him ranting in one of his videos against repetitive exercise, and that sketching doesn't work, because you work differently when you get used to always finish your works (IIRC?).

I'll link one of his videos, randomly picked; if he clicks with you, you might check out his other stuff, and the youtube algorithm almighty might lead you to more artists embracing that approach to art.

I just discovered a bit of an annoying bug. Including links to my sketchbook in a comment is highly useful, as they immediately display. Now, after entering the link into the text box, the site display a warning note, that the image will not be displayed immediatly, but only after posting. Which is all fine and dandy, except. the. same. warning. pops. up. after. every. key. stroke. after. pasting. the. link. which. is. kinda. annoying.....

I think the potential harm in just removing that warning and let users figure it out for themselves is lower than the current functionality.

Honestly, #1 quality: cheap! The cheaper your paper, the less you have motivation to worry about turning each sketch into a museum piece, and the more relaxed you can just scribble away and waste huge stacks of it. I mostly use printer paper made for business correspondence from a stationery store or even better a dollar shop. It's mass produced for all sort of administration tasks and therefor almost unbeatable in price.

If you go for more elaborate paper, I would think less in terms of good or bad, and more like an alchemist: if I mix this kind of paper with that kind of drawing medium, what will be the effect? Probably the reason to make a sketchbook from rough paper is to disuade the artist from focusing on minute details too much, and emphasize the flimsical nature of initial sketches. Now, you decided to focus on rendering, which is on the opposite end of the scale in terms of the work process. (sketching, then construction, then line work, then rendering) You can do detail on rough paper, it just takes more time and effort to get the lines and values look sharp, and the key is to pay attention to the grittyness and draw repeatedly over the same spot to catch all the white spots left between the bumps. (you will need quite a sharp point on your pencil to even get between the bumps; having a spare pencil prepared, so you don't have to break your workflow for resharpening might be a good idea) On the upside, this gives you an additional contrast that you can use for expression, with crisp and sharp values where you want to focus on, and a more gritty, spontaneous look on the periphery of your work.

I just finished a session with your post in mind, and observed how I do rendering. So, a) idiosyncrasy warning, I am not an accomplished master myself, this isn't immutable law, just the way I do it momentarily, and b) I usually use ink brush, which is almost digital in terms of darkness value, and does not even allow for arbitrarily thin lines, so I need to plan for hatching.

What I generally do, is that I draw the outlines for most shadow areas before I actually start entering the value. If done right, visual habits will already kick in and "fill in" the values to turn the figure 3-D, before I even start hatching, and occassionally, if I want to indicate a very light darkness value, it is even enough to just let a very light outline do the work, without any hatching at all.

Basically every detail on a figure, that isn't an actual surface color (like on a piece of clothing, a tattoo, body hair or strong variations in skin tone) IS a shadow. If you just bury them all in smooth gradients, they will be gone. If that is the result of you purposefully simplifying your construction and linework, that can look extremely stylish, if that happens as an accident of your final rendering, it will just look boring and flat.

Here is an example of me preparing a figure for shading. It isn't a textbook example for a number of unrelated reasons (line quality and proportions are questionable at best, for example), but I just happened to end the drawing process right before starting to shade.

It's a bit hard to explain, but especially when you have a darker shadow inside a brighter shadow, you tend to focus a lot on smoothing out the transition, which kills the illusion of volumn and makes the area look like one flat papercut. Try to use the edges of shadows, or even of different values of darkness within one shadow area, as construction lines, that emphasize either the volumn or interesting details.

Also, you use a quite rough paper, which leaves some of your shadows very grainy. You still achieve a deeper darkness by applying more pressure, but the grainyness remains. I am afraid you either have to switch to smoother paper, or get used to drawing repeatedly over the same area, until the grainyness is at least largely diminished, as especially in your darker shadow, it also seems to kill the volumn. Maybe drawing, rubbing, drawing over it could help?

Btw, I am starting with the problems solely to honor my national reputation of painful directness and social ineptitude (kraut here). The problems aren't consistent, they don't ruin your entire work, and you make good progress. These are just the things which I think would have the most impact if successfully changed.

Grats. I still remember working through the robobean, and starting to realize, that my scribbles finally start to resemble actual human beings.

Partially you are already there, partially you have to start a whole new chapter.

The part, where you are already there: you already extremely simplified your bodies, and you are already indicating parts of the body as voluminous objects in space. For example those circles around legs you already use. Draw an outline around that thing and call it a tube, voila, simple 3-D object.

The part with the reason, why you would consider calling your beautiful leg a silly tube, and the part with the new chapter: geometric objects are assumed to be easy to rotate around any axis. I am saying "assumed", because in reality it makes a lot of sense to actually practice that quite a bit. If you need ideas how to practice rotating geometric objects, I suggest a bit of a dive into drawabox.com, it's really good. The trick is for you to develop a set of useful building blocks, that a) you can substitute the body parts of your gesture drawings with, and b) you have practiced how to rotate.

The exact geometric forms actually vary from artist to artist. Some people only use cuboids, because they are easiest to rotate, with just right angles and straight lines to think about. Rotating a tube is already a bit more challenging, as you have to get a good grip on drawing ellipses in all sizes and with rotating axises, and getting an idea how to manipulate your ellipse to indicate that you are looking at a circular disk from a specific angle, but being able to use objects with round sides to indicate a human body let's you get so much closer to the original organic form, that the extra hazzle is worth it.

In the end, how simple or complicated you make those building blocks is up to you, the decisive quality is, whether you can rotate them around all axis. If you need inspiration, you can buy an artist's mascot of a human body, they are already made from quite simplified forms.

So, now you can substitute each part of the human body on your gesture drawings with an object, that you can easily rotate around any axis. Great. You can either rotate the whole body to switch your point of view, or you can just rotate a single limb around a joint to change the pose. Complete freedom at last. All that stands between you and that freedom is a **** load of torn hairs, sweat and tears when trying this nice theory out in practice.

My first experience with rotating a gesture: It took me about 5 minutes to draw a simplified gesture, with building blocks of my choice. 7 hours and half a drawing block filled with wild scribbles later, I managed to recreate approximately the same gesture, but viewed from a point of view, that had vertically shifted by 90 degrees. I congratulated myself on the success and swore to include rotating gestures as a fixture into my daily training, but so far resident trauma from that experience somehow always makes me skip to a more comfortable drawing practice, when gesture rotating is up. But it's totally a great idea, and I would immediately do it, if I wouldn't feel so tired right now. I'll do it tomorrow. Or at least the day after tomorrow, I promise....

I mean, your idea of images flowing from your head definitely has a lot going for it: You have fun drawing those images, and it translates well, and makes it fun to look at your results.

With the tools: You are already using digital tools, which are basically just other people's ideas, just interpreted by a machine. And you are obviously having fun using them, like they were NOT intended. It's the same with most of the art theory stuff you learn in courses. Doing the course isn't someone telling you how to draw, it is just introducing you to some additional ideas, downloading some code for a new and exciting toy. And then it is up to you to either stick with it, be all boring and prissy and try to channel the person, who introduced you to it, or to run amock with it and figure out a bunch of interesting ways to break it.

I think I have a hunch to what will be your priorities.

There is a bit of a gap between 30 second sketches and rendering... 30 seconds at most gives you a faint idea of a pose, a concept to develop. The next step would be to start building a foundation on that concept. I would propose to first start to develop a more complex, anatomical form for your torso, as it defines the body in space, and determines where the major joints need to be.

My most recommended course on the topic of how to develop a foundation of a figure would be https://www.proko.com/course/figure-drawing-fundamentals/overview, basic course has all you need. I can't guarantee that it is the best course evaaaaa, because I only took that one, and so I have no comparison, but I can guarantee that it taught me a lot of what I use today.

You got such a strange style. 90% of each drawing is really, really good, but then 10% is somehow off, and because the other parts are so good it sticks out like a sore thumb and pulls down the overall impression.

I think you would really do yourself a favor getting used to some underlying construction for the human figure, And I don't mean "anatomy", like in getting to know each bone and muscle in the human body by latin name, postcode and tax number, but more "landmarks", like in which proportions you absolutely have to measure correctly to make the end result look good, and which ones you can (and occassionally even should) play around with to your delight without anybody noticing.

Like your Freddy Mercury, so much of it looks really good, but that just aint how a human shoulder works.

Also on Freddy, I love the way you scaled down the digital blending enough to leave some structure to the shadows, and then accentuated it with a few drawn lines over it, that is a really cool style to render an image. It has a bit of an old school early digitalisation MTV vibe, but it looks very deliberate, and perfectly fits the timeline of Queen at their heydays, which makes it even better. But then the placement of the features of the face show your struggle with the underlying foundation, which makes it look like you slightly changed your point of view on different areas of the face.

Great job. To an equally productive May!