While the lighting is an improvement, I still think it looks a bit flat when each clump of leaves looks like they have a different light source than the right side and trunk. The black outlines that are around each clump make it look more like an island of leaves. Since it's a top-left lightsource, each clump should also be effected in the same way that the right part of the tree has been (lighter on the left, darker on the right).



Making the clumps around the edges of the trees (that sort of, dark outline of leaves) lighter as it gets hit by the light more would also help.

As for the trunk, it would pop a lot more if the cast shadows from the leaves would darken the top.

Take a look at this, 3/4 down the page where it talks about trees -might help you a lot http://gas13.ru/v3/tutorials/sywtbapa_de-mystifying_greats_1.php

I think the pose itself comes off as very odd, and somewhat symmetrical (especially with the arms in the way that they are). Maybe consider experimenting with something more dynamic?


(not a great example -it still looks pretty stiff, but hopefully it gives you something to at least consider) You could exaggerate the bend in the spine and stretch the torso out even more for a better effect.

@Mav Thanks!

Finished. Lighting on the face is real boring, too bad Also could probably use more contrast

I think you're on the right track with the face. The most glaring issue that stands out the most is the pupils/iris, they should be bigger. Also if you flip the image I think you'll see how lopsided the head looks because of the hair.

There's also another issue here I think and that's how far the shading under the cheek bones is being pushed. With the light already striking the cheekbones above it already suggests there is some sort of elevation going on there -having the shadow under them makes it look like there's a cave or something there.

 I also think you should be more frugal with that highlight, if you use it in more select places, and in smaller quantities (like one or two pixel clusters) you'll be able to make certain aspects of the face stand out a lot more -like where the highlight on the nose is, is perfect but having it on the chin and the cheekbones sort gives it a puffy look.

I don't know if you've thought about it this way, but since it's pixelart and you're working at this res you have to think about suggesting features with light, rather than copying the face 1x1. I think for the sake of learning you should cut down on some of your darker colours too so you can concentrate on other things! You don't need too if you're comfortable right now, but yea, really on the right track with this!

Hi zizka,  I would really recommend that you go bigger with your current study. I feel like when you're learning fundamentals like anatomy, especially when using pixelart as your medium, you ought to go bigger. Right now I feel like when you're working at such a small scale, it becomes more about managing which pixel will symbolize a certain body part, and removes putting form into practice.

I think a good example of this is Ashcrimson, he has a massive thread on pj about his anatomy/sprite studies, but I feel like for the time he has been working on it not much has changed as I would have hoped (no hate to him ofc).

Even just upscaling by a little bit would make it less about managing each pixel- it makes it more about learning anatomy instead of learning how to pixel tiny sprites. If you do enjoy working at the size you currently are then keep on going!

quick edit to demonstrate.