"PAINTOVER PLEASE" - painted crits on demand - with Steven Stahlberg


Hi, Ido, thanks.
My paintovers don’t work very well on stylized characters like this. There needs to be a certain level of realistic human anatomy that I can use as a guide. Stylized and cartoony images are very difficult to ‘correct’, because I don’t know what the artist is going for, it could be any of a thousand different styles and looks.
What are you going for in this case? And what exactly do you think is the problem with the image? To me it looks pretty good. :slight_smile:


Here’s number 504 and 505


Thanks a lot Steven,
I don’t really know if there is something wrong with this sculpt because I’m not sure I can look at it objectively. I’m sorry, I thought that this may be a problematic image for your paintovers, but I couldn’t miss the chance of getting any kind of feedback from a great artist like you. I will try again with my next character that is going to be a realistic female, if that’s o.k with you.


Thank you steven! I really like the modifications on the legs and waist area. The head does seem too small too me, i’m not sure if its cause i was used to the old bigger head version. I already can see more clearly how i can play around with the proportions.

I was wondering, if it were ok, if i could have another paintover please? This one is actually a finished piece, that got rejected from the 3d showcase gallery. I think something about the lighting was off.


Ido, sure, post as many as you want, just give me time to respond to each one. :slight_smile:
About this frog creature, as I said, it looks real good, anatomy very realistic and nicely exaggerated in the proper ways. The head/face looks a little unfinished, and if you showed some references of what you’re going for maybe I could help you further.


The devil guy
I think the background is way too busy, even in my version it’s still too busy with 2 strong light sources visible in the image, I recommend choosing one to focus on, making the other one less obvious.


Hello there Mr. Stahlberg. It’s been a while. First off, I’d just like to credit your paintovers as being one of the, if not the, most influential aspect of my development as a digital painter. I always learn something that I try (try being the keyword) to keep in mind in my future work, and it’s a resource that I’m afraid to constantly tap for fear that I may become too reliant on your skill instead of developing my own. So I hope you don’t mind if I approach this as inquisitively as a possibly can since my goal is to take what I learn from this paint over and take it to that “next level.”

Anyway, I would love your input on this since I feel like I’m at an impasse with it.

OK, this takes a bit of explanation. This monster is a wereparrot. The idea comes from my conceptualization of facebook. It’s a place where people you know parrot ideas at each other via memes and transform in to monsters before your very eyes. Especially during the election, I’m getting to the point where I want to murder my family. (OK, not really, but still, they are obnoxious) And, as anyone who owns a parrot can attest, after a while your instinctive response to a parrot is “oh god, just shut up.” So conceptually it’s supposed to be this mixture of horrifying and obnoxious.

In terms of what’s going on the pic, the idea is that this is a during a full moon in a jungle, and this creature has just been interrupted during a meal by the flash of a camera. I’m trying to go for this mixture of http://images4.wikia.nocookie.net/__cb20111207010633/powerlisting/images/0/0c/Werewolf2.jpg and http://farm8.staticflickr.com/7063/6836470522_2bc9e54ba0_z.jpg but I’m having troubles.

Here are the problems I’m aware of but not entirely sure about how to approach fixing them. The leaves in the foreground, and many of the compositional elements such as the trees are too even. The light sources are fighting each other too heavily. And the background is too barren for a jungle. The figure is too “cut out” and seems like a wax figure.

All of these problems are interconnected. The leaves and the Trees don’t seem like that much of a problem, but I’m afraid of things becoming so cluttered that you can’t see the moon which provides that rim light. Since the idea of a moon is important to the idea of a werewolf and by association to the idea of a wereparrot, that rim light and its source are important. The jungle itself in the background is barren because the image itself is so dark that I’m afraid of loosing the parrot itself. That’s why the light sources are about equal. My problems stem from the piece being dark and me wanting to make the creature and the idea readable

I’m just stuck in this loop of going back and forth on these ideas. I could be wrong about my assumptions about what is and is not important and I look forward to your input. Thank you. (sorry if I wrote too much)

Oh, and anatomically I’m going for kind of a stretched out and awkward figure, since something about that kind of a figure seems bird-ish to me. I’m not sure if that’s a problem or not.


Ok, here’s the paintover finally.
Let’s start with the sky and the moon: people always seem to get this wrong, and I’m kind of wondering why, because it’s very simple, as you can see my version is simpler than yours. It’s also very logical and there’s not much room for variation: a full moon in a sky will have a slightly yellowish tinge, which together with the blue sky often gives a faint greenish tinge in the halo. The sky can NEVER be darker than the ground in this scenario, the full moon is a very strong light source. Clouds are never straight lines across the moon, as many people like to paint, probably because it’s easier. :slight_smile: The center part of the cloud is darker than the sky, obviously, because it’s facing us and away from the light, blocking it. The fringes of the clouds are always almost as bright as the moon itself.

The trees, too straight and regular. Highlight should be thinner.

edit: I know I’m breaking my own rules about placement of contrast with those leaves in the foreground, so I might actually delete those. I just used them to show more clearly the placement of the light source.
I also note that I screwed up the parrot monster anatomy a bit in my quest for a lighting-change, I think the original was better.

Parrot monster, cool originality, it was a bit asymmetrical and the beak was both a little twisted, and a little impractical, it wouldn’t be possible to close it (use reference).

The front light on the monster, it would be better if it looks like it’s coming from the camera or nearby.


An interposed question:

Sometimes you can’t avoid more busy backgrounds, for instance when they are important for the context or the story of the image. Landscapes, jungles, cityscapes (at night!) and the like can make a very problematic background.
I usually manage with DOF and/or desaturation, but after a while this seems a little lame and I wonder what you would suggest for these cases to highlight the images focus part?


True, sometimes in real life you get very busy backgrounds, in photos. What I usually try to do is to exaggerate the “atmospheric perspective”, the notion that contrast drops the further away an object is. That is, the distance between the lightest and the darkest pixels grows less and less. I call it ‘flattening’ because as distance increases certain parts of the background start to look flatter and flatter, until at the extreme the background starts to look like a series of silhouettes cut out from thin walls.

This technique will enable you to convey highly complex information, like a tree for instance, in great detail, but with the contrast turned WAY down it will look harmonious and not demand the attention of the eye constantly. Which, if the focus of the painting is supposed to be somewhere else, can be extremely bad.
Also, you may be surprised how much information a silhouette can convey.
For a city scape at night, one trick is to add mist. Or rain!

This technique works for smaller scenes too, without a deep background. Just be very very careful with where you put the deepest darks and strongest lights, if there are too many spots where these are too close to each other - that’s the problem. Not the amount of detail, but the placement of the contrasts.


I see. Cool.
Thanks for the tip!


Hello Mr. Stahlberg !

You make a incredible job here, I went through the 107 pages and I think I have learned more here than anywhere else.
And I hope it will be a bit visible in my image.

This is my most recent work, done for school.

The subject was “a rope bridge is suspended above a ravine where a river flows. Far away, beyond the treetops, a Mayan city”

I wanted to show a fantasy version of a Mayan city. With the light beam from the temples to the sky as a signal to space, during a sunrise.

Thanks for taking the time to help me !


A very nice painting, but a couple things I want to note: the waterfall looks small here, though everything else around has a larger scale. Also it’s brighter at the top and darker at the bottom, this is usually the ‘shorthand’ method to describe a shape, as if the light is coming from above (because in human experience it usually does). But in many cases this is wrong, so you have to look out for using this shorthand method. In the case of a large waterfall it’s the other way around, it gets brighter further down due to more water turning to foam.

The bridge, you missed an opportunity to play with shadow, check my version to see what I mean. Any time you can do something like that, have realistic shadow and light playing on a surface, it usually turns out better.

The big rock on the right also had a perspective problem, the water surface didn’t look flat due to the shape of where the rock met the water.

Other than that it’s mainly the common problem of not consolidating large shapes enough, but splitting them up with a lot of slightly higher-than-necessary contrasty details. See the vegetation, the pyramids, and the rocks on left and right.


Hey Stahlberg, i only want said you are awesome, i dont know how you can acheive this in so less time!
A little question, do you use painter or photoshop?



It’s Photoshop. Thanks


That awesome !

It always seems so obvious when we look at your changes… But in practice it’s a bit more complicated.

I take good note of all your advices. Thank you very much !


Hi Steven! I know you already helped me with this model previously, but i am really stuck at the moment and cannot find the problem with the proportions, and was hoping you could help out again.

Your previous paintover helped out alot with the structure but i still went for the slimmer design. But now i cannot get the model to not look like it is an action figure… and i think it’s due to the proportions. I tried and do a paintover myself to see if i could work it out… and i think it’s better but it still looks like a toy:

I am trying to match something like this

i really like the proportions on the above image but i cant replicate it on my own model… everything keeps looking stretched out.
Hope you dont mind asking help on this one again, but i am really stuck.

Ok so i tried as hard as i could to match the proportions of the image i posted above and i came up with this:

I think it looks better but i still cant find exactly what looks off about it. I have changed the characters proportions about 10 times already, and i cant find it. Hope you can help me with this.


Awesome work man, can you please tell me what to improve here:


M1KES, there are a few parts to the recipe. One part is you still have to make the head smaller, a little bit. Then you have to make the feet much smaller. Then you have a problem with the transition between feet and legs, and between hands and arms. They look like it’s a toy now, due to the sharp deep cuts there. Thicken that out and smooth it over, more like Master Chief.
His armor in the front view bulges on the lower arm and lower leg, but look at Master Chief’s lower arms, they more closely capture the physique of a naked muscular man, Try to go more for that. There’s no part of the human body that’s bulging symmetrically and smoothly like a bottle, instead the bulges are almost always offset strongly, more like braided hair, if you know what I mean. ( I know this is not a human body, but we have to work within the confines of people’s gut reactions and if it’s this close to a human they’re going to compare it to a human.)

Also to make him look bigger make those Robbie-the robot ring shaped bulges going from the top of the thighs into the crotch, smaller. Thinner, and more numerous.


Greetings Mr Stahlberg, :bowdown:

im currently working on this head modeling. i wanted to make my work to be come more believable, but yet still contain my cartoon style.

the main reference of this work is baron samedi, the hoodoo guy with skull painted on his face (although this one are still bare, without anything beside the normal skin paints)

please help me with bits of this and that, as this is my first character i modeled ever in 3D i want to make it look good.

thank you so much, and sorry for the broken english.