girl in a field

[moonskittles]

Hi guys! I'm making a new piece. I really want to work on developing settings for my characters, but I'm not too experienced in this matter. I'm also not used to creating a composition that takes up the full workspace. So, here's my beginning work on this piece. I would much appreciate your tips.
I know her leg is a little wonky; I will be making a reference for that and fixing it when I have the chance.

I am mostly looking for advice in composition, contrast, values, etc. But if you see anything else amiss, or have anything else you'd like to share, I'm all ears. Are the pine trees in the background too dark? Do I need darker darks in the foreground? The lighting is supposed to be slightly overcast, so I don't want to shoot out the highlights.

Also included my refs.

[krysjez]

Top of the picture near her horns feels claustrophobic. You've got pretty strong light on the girl coming from the left, but the background has like no shadows and it's a little incongruous, as though she's been pasted on. And I don't think the horizon is that high in your reference photo, so that has to be moved down accordingly.

[arenhaus]

Notice how you tend to make symbols instead of shapes, even when you have reference. You draw repetitive triangle shapes which don't look like anything other than "here is a christmas tree". Look at the real trees in the photo you attached - they have varied branch angles, clumping, asymmetries, no two trees are alike. You draw repetitive, evenly distributed dot-and-dash for flowers. Look at the real flowers in the same photo - they grow in clumps and rows, they are different height, and so on. You don't construct the figure in perspective, either; you are just guessing, even when there is reference.

You should start looking at the reference.

[moonskittles]

Notice how you tend to make symbols instead of shapes, even when you have reference. You draw repetitive triangle shapes which don't look like anything other than "here is a christmas tree". Look at the real trees in the photo you attached - they have varied branch angles, clumping, asymmetries, no two trees are alike. You draw repetitive, evenly distributed dot-and-dash for flowers. Look at the real flowers in the same photo - they grow in clumps and rows, they are different height, and so on. You don't construct the figure in perspective, either; you are just guessing, even when there is reference.

You should start looking at the reference.

Hi there.

I see what you mean; I am familiar with this concept. I guess I was shying away from actually doing the background ref because it was complicated, and instead for now I was using these symbols more as placeholders. Anyway, I can fix that. But I'm not sure what you meant about the figure. There I was actually trying to do as the ref showed, though I changed the hair and hand (intentionally). Could you elaborate a bit more on this point, or possibly show me something?

[arenhaus]

But I'm not sure what you meant about the figure. There I was actually trying to do as the ref showed, though I changed the hair and hand (intentionally). Could you elaborate a bit more on this point, or possibly show me something?

I just see all the big differences between the ref and your version which tell me you are not reading the form well. You didn't track the volume in many places. You didn't get the symmetry right. You have even changed the curve of the spine, which would make quite a different pose, yet kept the overall contour.

Where you tried to change or invent things, like the hand and the shin, you haven't tracked the volume either.

If this is the typical result of your work with reference, then you probably aren't ready to just sit and paint the form at a glance. You have to work tighter for a while, spend more time planning, and make sure you pay attention to the volume, perspective and structure.

[moonskittles]

You have to work tighter for a while, spend more time planning, and make sure you pay attention to the volume, perspective and structure.

Thanks for your help. That is exactly what I'm working on right now. This piece is practice, and I am still planning. Now you've pointed out the things I need to work on and I'll fix it.

About the values on the form, I was under the impression that since she is in the foreground, her contrast should be higher. I'm assuming that's not what you mean, but rather that I don't have the values in the right places, right? And, am I wrong to up the contrast in this case?

As an aside, I'm curious -- I've lurked in other threads where you give advice, and I usually find that you are technically correct, but your tone is a little demeaning. I don't mean the difference between solid crit and an asspat, or anything like that. The difference is entirely in how you treat the person. Is there a reason for that? Is anything accomplished by that? Honestly curious.

[Lulie]

'Lo!

To expand a tiny bit on Arenhaus' comment about the spine: In the reference, there is this beautiful C curve (or S curve if you count the head) of the spine, which wasn't replicated in your figure. I really recommend Force: Dynamic Life Drawing -- it'll change the way you see figures and think about drawing.

Contrast should generally be higher in the foreground, due to atmospheric perspective (and composition reasons), yes. If anything I'd push the darks to be darker, like what you did with her left horn -- it's a bit 'mid value'-y overall.

Arenhaus: When you said 'volume', did you mean something like mass and form, or did you mean values? (I.e. are you saying the shape is wrong and the 3Dness isn't understood, or are you saying it's not shaded correctly to convey the shape?)

[Lulie]

PS: I don't think Arenhaus' tone is demeaning. It's just his comments are usually about overall things the artists should work on by themselves, and pointing out some problem areas, rather than detailed comments saying which lines to change to what. He also says it in the most straightforward way, with no 'friendly filler'. So it can seem curt, but I think that's just a way of being efficient rather than unfriendly. Keep in mind he comments on a lot of threads -- I'd guess spending the extra time would mean he would have less time to give comments.

There are people who are demeaning in their critiques (e.g. they will be insulting), but I don't think Arenhaus is. If anything, I think his tone is respectful: it's assuming the artist appreciates criticism and is here to be serious about art, and doesn't need to be treated delicately.

[zephyri]

Hey Moonskittles - Thought I'd drop in and expand on what Arenhouse has been advising you on. His tone isn't meant to be offensive I'm sure... he's actually giving you great feedback very directly without the need to back it up with niceties. He's saving you time in the long run!

I've done a quick 'Paintover' - though I did it from scratch quickly - to try and show you how you might take the advice you've been given into visual form. The main composition problems you have at the moment are your figure is cramped in the picture frame you've given her - she's touching the frame at bottom and top almost, which makes the more quiet scene you've chosen feel tense. The treeline cuts right through her head and it's almost straight across the picture plane - same with the shadow at the bottom combine that with the fact your figure as a whole shape is very vertical and it's making the composition quite rigid and boring for such an organic subject matter.

But your main issue is value, since your figure is being lost in the identical value you've chosen for the grass. Looking at the reference, the green is actually pretty dark, and it would make a great contrast for your figure, and you can use those 'clumps' of flowers to break up the dark mass and help lead the eye around the image to the girl. Ask yourself what is your image meant to be about.. what do you want the viewer to look at most, and then use the rest of the elements of the image (grass, treeline, clouds etc) to help pull the focus back to that. And don't be afraid to use darker tones - ultimately it'll give your image way more realism, and you'll generally get a more natural feel if you then go in and paint in the lighter area around the darks.

The paintover I've done is fairly rudimentary and you could be much cleverer with the way the head could be framed by the sky and trees, but hopefully it'll give you an idea of how it could look.

[moonskittles]

Thank you all for your help!

@Arenhouse: sorry if I was oversensitive. I do appreciate your pointers.
@Lulie: Thanks, I'll have to add that book to my list.
@Zephyri: Thank you so much for taking the time to do a paintover. That helped enormously. You'll see now I've slightly blatantly used a lot of your composition now.

Also took my extra ref for the leg. I couldn't get the pose exactly right, as I have a 3-second timer, which is hardly enough time to dash over and get into position, but I got the leg at least. I see now what I was suspecting, that the second foot would naturally peek out under the thigh. Is this now distracting? Should I leave it hidden beneath her leg?
Am I doing this right, now?

[krysjez]

I think you can leave the foot, and also that working these things out with drawings instead of having to repaint every time until you settle on something you like will help save you time.
A bit more light on the girl could help bring her out a bit more. Right now I'm most drawn to her head because of the contrast, and her body's sort of lost in the general mid-tone of the background.

edit: Also, look at the ref you took, then look at how you've got her leg positioned now. See crude ASCII art below: on the left is how your ref looks, on the right is how you've drawn the leg, which means the ref was basically taken for nothing. I also like the more natural curve of the back and sense of weight you'd get if you followed the ref pose a bit more closely, but your call on that.

``___````____
`/```````\
/`````````\

[Giacomo]

Also took my extra ref for the leg. I couldn't get the pose exactly right, as I have a 3-second timer, which is hardly enough time to dash over and get into position, but I got the leg at least....Am I doing this right, now?

Not really. You're rushing too much here. The problems are 1) that your reference material is weak and 2) you're being pretty sloppy about copying it. You need to get a well-shot photo that clearly defines the lighting, gesture and the forms-- which means shoot it somewhere where there aren't shadows interfering with the lighting, with a model who's in the exact pose you want (and hopefully, with something better than a low-end camera phone). If the figure in the final product is supposed to be nude, you should shoot your reference with a figure as close to nude as possible. Basically, you want the reference photo to do as much work as you can before you start drawing. Once that's done, you should make a line drawing paying close attention to the contours and edges (and a graphite pencil will work a lot better for this than a Wacom stylus) Once you have a tight line comp, you can scan it in and begin painting.

Realistic painting is a slow and systematic process requiring a lot of stamina and focus. Being impatient and just gobbing down pixels is not going to produce a good end result.

[moonskittles]

@Giacomo thanks for the advice. Unfortunately, my scanner is not working right now, so I can't scan any paper sketches. Reading over your crit, I realized that part of the awkward pose was because I was copying half the pose from the ref in the first post, and half from the crappy pic I took. I hope you didn't think I was so horrible at following a reference as that! Also, as far as the background, I am not intending to copy the ref exactly; I will rearrange the flowers and trees to my bidding. I don't have those kind of landscapes where I live, so that's about the best ref I'm going to get. The values I have on the pic right now are more how I would like to arrange them; they will of course be finalized later. This is far from my final!
From your advice, however, I dug out my digital camera and decided to take a better ref shot for the pose. This has pretty much the lighting I'm looking for. I also decided this pose felt more natural and interesting than what the other girl has going on.

I'm leaning toward making her smaller and more in the midground, and adding flowers close up for the foreground. Does this put too much dead space on the left side of the picture?

[Giacomo]

That new reference photo is about one quadrillion times better than what you had before! Nice job.

The reason I suggested you work in pencil is that you're not looking at contours and edges as precisely as you need to be, and it's really hard to do that with a stylus, especially for a beginner. (For several reasons, the tool just isn't very good at putting down a precise, controlled line and that's what you need to be doing at this point. The drawing of the figure is looking really clumsy and I suspect it's because you're using a tablet.) If it were me, I'd get as far as I could in pencil (do separate drawings for the figure and the setting) and find a scanner somewhere (there's got to be one at the local copy shop.) Also, you obviously don't need to copy your background reference exactly....you do, however, need to be observing the forms a bit more carefully than you're doing right now--even if you're going to be putting stuff in different locations.

As far as the composition, it's up to you. I personally like to crop in close to the figure, but if you're interested enough in those plants to make them visually engaging, then the composition you have now will work fine.

[krysjez]

Good job on the new ref! And I'm envious of your Florida weather - 34 inches of snow here...

[arenhaus]

Arenhaus: When you said 'volume', did you mean something like mass and form, or did you mean values? (I.e. are you saying the shape is wrong and the 3Dness isn't understood, or are you saying it's not shaded correctly to convey the shape?)

I meant the former. The form in 3D was misunderstood or not tracked. It can't end up in correct shading anyway, but sometimes you do see a good grasp of the form and structure but a poor reading of value. In this case, however, the OP focused more on his idea of shading than on the form or the value in the reference. It's a common thing.

Generally when I say "the volume" or "the form" I mean the form in 3D, not shading. When I say "value" I mean value or shading. When I say "shape" I usually mean 2-D shape or a smaller component of 3-D form.

[arenhaus]

About the values on the form, I was under the impression that since she is in the foreground, her contrast should be higher. I'm assuming that's not what you mean, but rather that I don't have the values in the right places, right? And, am I wrong to up the contrast in this case?

Somewhat wrong. The problems were mostly with the underlying form, not the contrast or something. You weren't reading the anatomy well, and were inventing things without a lot of experience in inventing.

Any value has to make sense in the context of the environment of the whole picture. There is no rule saying that the foreground must have more contrast than the background - for some cases, the opposite might work. But the value, if you are going towards realistic treatment, should represent the lighting consistently.

As an aside, I'm curious -- I've lurked in other threads where you give advice, and I usually find that you are technically correct, but your tone is a little demeaning. I don't mean the difference between solid crit and an asspat, or anything like that. The difference is entirely in how you treat the person. Is there a reason for that? Is anything accomplished by that? Honestly curious.

How is it demeaning? I am usually brief and direct, but I don't call people names or discourage them (unless telling them they should use a different method is a discouragement). I only point out problems and ways to fix them. If I like something despite the problems in it, I say so.

I could add more fluff, but that would mean I'd have to spend longer on each reply, and so help fewer people. I have limited time in which to do this, and this is pure volunteering on my part.

[JavierP]

Hi moonskittles,
Both arenhaus and Giacomo have given you some great advice, and your determination to get good reference is very admirable. However, I would like to add something else for you you to think about as you work on your image, (arenhaus mentioned it briefly in his last comment) which is the arrangement of the shapes and values in your image as a WHOLE. It is very important to use the right values and shapes within the individual elements of a picture to create a convincing representation. It is also equally important to consider the overall pattern that those individual elements make, when viewed together as an image. For example using many disparate values in small fragmented shapes throughout an image creates a sense of energy or chaos, while using large shapes of closely related values can create a feeling of serenity or stability. I made a rough painting of your image using only 5 distinct values. This is simply one way, out of many, to arrange the elements in this image. The thumbnail is the stage at which this type of work is conceptualized and implemented. This saves time, but it also illustrates the idea that if your image is 'read-able' at a small size, then the larger final image will have much more visual impact. Hope this helps, and good luck.Attachment 1714413

[arenhaus]

It is very important to use the right values and shapes within the individual elements of a picture to create a convincing representation. It is also equally important to consider the overall pattern that those individual elements make, when viewed together as an image. For example using many disparate values in small fragmented shapes throughout an image creates a sense of energy or chaos, while using large shapes of closely related values can create a feeling of serenity or stability. I made a rough painting of your image using only 5 distinct values. This is simply one way, out of many, to arrange the elements in this image. The thumbnail is the stage at which this type of work is conceptualized and implemented. This saves time, but it also illustrates the idea that if your image is 'read-able' at a small size, then the larger final image will have much more visual impact.

Excellent advice, there. I fully subscribe to it.

OP, notice also how Javier frames the head of the figure in the gap between the clumps of trees, and how he uses the dark horns to both rhyme with the trees and frame the lighter head. There are also tangents and arcs there, e.g. on the left you can see a half-ellipse formed by the left clump of trees, the left side of the figure and the top edge of the flower clump in the bottom-left corner - and there is an almost straight line going through that corner, the figure's foot, knee, shoulder, head, and the right clump of trees' left edge and then close to the top right corner. These arcs anchor the figure as the center of composition; by emphasizing them, reducing the busy contrast in the background, clumping value blocks together and increasing the contrast between the figure and the background Javier makes the figure the focus.

On the other hand, there may be potential problems in this treatment because of the sky being darker than the figure, and because of the figure's falling shadow not fitting into the value range. Color tests would be needed.

(BTW, Javier, what does the Mayan inscription in your userpic read?)

[Pavel Sokov]

JavierP

Is it just me or is running the horizon line through her head look a little weird? If we lowered your horizon line to her shoulders, or made an opening around the character's head like OP has done, wouldn't that get rid of the weird tangent?

[JavierP]

On the other hand, there may be potential problems in this treatment because of the sky being darker than the figure, and because of the figure's falling shadow not fitting into the value range. Color tests would be needed.

(BTW, Javier, what does the Mayan inscription in your userpic read?)

Arenhaus makes a very good point, and it shows that these type of sketches and thumbnails are only starting points. There are many things that could be improved and refined. In fact, starting with 5 values that are more closely related may produce a more pleasing result, or making the figure larger, or the trees smaller, or the flowers more plentiful ... etc. There are almost an infinite number of variations to try. The important thing is to make the effort and try a few to find one that works best for you and your image.

EDIT: I forgot to answer your question. The glyph in my user pic reads, "ah-tzib". In Yucatec Maya, it means he/she/it wrote/drew. Writing and drawing were the same activity for the ancient Mayas.

Is it just me or is running the horizon line through her head look a little weird? If we lowered your horizon line to her shoulders, or made an opening around the character's head like OP has done, wouldn't that get rid of the weird tangent?

I don't know Pavel, it may just be you. Why don't you make an overpaint to explain your solution and help the OP out?

[Pavel Sokov]

Yea, here it is.

[vineris]

Yea, here it is.

No, because now it forms a tangent with the shoulders. In general, cutting the body at the joints or having lines running through the body at the joints (and the neck is a joint) is shitty composition. The horizon must either go through the middle of the head, or through the upper arms.

[LaCan]

No, because now it forms a tangent with the shoulders. In general, cutting the body at the joints or having lines running through the body at the joints (and the neck is a joint) is shitty composition. The horizon must either go through the middle of the head, or through the upper arms.

Yeah, it's kind of "natural" for horizon to go through the middle of the head since that's where the eyes are. If your subject's head is roughly on level with your head (which is likely), the horizon will - by definition - go through their eyes.

[Pavel Sokov]

Oh I didn't know that, sorry! I learned something though

[Avvatar]

This piece could benefit from going back to the thumbnail stage, the composition right now is very static and awkward. The values are everywhere too, there is no depth.

[vineris]

Oh I didn't know that, sorry! I learned something though

Yeah, I learned 10 times more stuff by giving critique and occasionally being yelled at than by getting comments on my own work. It turns out that you can learn from advice given to everybody -- even when you're the one giving it.

[moonskittles]

Okay, I've reworked this, and I've come up with this new composition. This is actually more what I originally had in mind for this piece, and honestly I think it works way better. I've tried to incorporate what you've told me, simplify the values, only use 5, and not have the horizon going through her joints at all.

In my plan, she is sitting on a hill overlooking the field and trees below. I think that translates. I will probably simplify the field into mostly grass, because I think the composition gets messy with all that detail everywhere. Instead, I will have one clump of the flowers to her left.

I'm personally much happier with this layout. What say you?

And oh jesus this is a rough thumb. I know her horns are lopsided, the flowers and her anatomy are rough guesses, etc. I'll fix that before I move forward, ofc.

[Pavel Sokov]

You know, I came back here with a fresh mind, and am looking again at the head/horizon line....

I don't get it, I still find it weird and prefer the head from my overpaint that is exposed. What is wrong with me?

[moonskittles]

I don't know. In my OP I had the horizon going through her head and was told to fix it, now I'm told it's the other way around. :X