The dynamic aspect is interesting, but I'm not really seeing how this would appeal to a moderately competent pixeller (with the features that are currently demoed).
There are a few things that would be required for me to be interested:

* Outline should be able to use a round kernel rather than a square one, since a square one won't get pixel-clean results for most cases
* Some kind of noise reduction/abstraction filter (in accordance with the general artistic principle 'start general, incrementally make important things more detailed', I don't want to have to remove extra details -- it's much better IMO to add details). If I was implementing this, I would probably use gmic's Anisotropic Smoothing filter and possibly the Segmentation filter too. They aren't fast enough for real-time use though....

@Ambivorous: have you checked out SNK stuff (King of Fighters, etc).. or Segmentation filter (GIMP+GMIC is the easiest way I know to experiment with this)? Your simpler painting actually looks a lot like the result of segmentation filter on a somewhat-simplified photo.

Segmentation filter is far from perfect but it can often abstract clusters very well, so it can give many hints on how to paint a subject.

Good call on the shoes, the reddish color unifies it more. You might be able to push the depth more and get a bit more hint of local color by using greyish shadow tone.

@Samaramon: I love those, the clean architectural look and the well-unified pattern work (my Pinterest is filled with these drafting-related type of things, heh). Some of the later ones where you quickly cycle light grey/ dark grey/black on a face get a bit busy looking though.


---

Still working on the problem of 'keep it general until you *have* to go specific!' :

Thank you for the response. If you don't mind me asking for clarification though, are you talking about a single "sprite?"

Do you mean a single frame, or one animation, by 'sprite'?

(i suggest avoiding the term sprite, as it is not specific enough. RO also has animation that ranges from good to pretty damn cheap. Ideally, you would cite one or more particular animations or static frames as a concrete example of the scope of your question)

Experimenting with controlled opacity. Feeling like it gives more control of colour and doesn't look so muddy.

Yet another empty post from skittlefuck.
Interestingly, if I copy-paste the url, it shows up ok..
Not sure what is going on there, referer checking?

Interesting use of shapes. the polygonization and atmosphere reminds me a lot of Out Of This World.

I know I need to learn to "see". I have figured that out on my own. From copying pixel art pixel-by-pixel I feel I have improved. I guess I'll copy pixel art hair too. I need to get an answer from someone about this.

Ai- I don't understand how on that loomis on hair 59-60 is learning about the hair. It's Analysis of facial markings and drawing faces of all ages. Did you give me the right page?

It also covers hair. Look how he starts by marking out chunks of hair. dividing them into broad planes and shading those planes. Then he divides up those chunks and refines the shading. You can see he uses this for the face too (meaning that the exact same methods are applicable no matter what you are shading)

Oh! I may have had an epiphany. I feel that the hair strands that are sticking out to your camera (pixel art or not) are highlighted so that's why the second character from the right is highlighted there. The hair that is top right is also sticking out towards you so it's highlighted. If the hair is depressed, there is less light than usual. Please tell me if it's true.

Sort of.. Shading is all about "how much light can reach your eye". Light hits a surface and bounces off. If it hits your eye at all, then the surface doesn't appear totally black. The more rays hit your eye, the more the color of the surface is revealed. Specular Highlights occur when there are an overwhelming excess of rays hitting your eye.

The whole thing is about angles. Highlights mostly appear on parts of a surface that are facing fairly directly towards the viewer (this is the circumstance in which it's easy for the most light to get to your eye). On faces that are not pointing directly at the camera, highlights are possible but it requires proportionally more light to produce a highlight (since a percentage of these light rays will bounce off in a direction that *doesn't* hit the eye)

Naturally, less light rays manage to hit recessed areas to begin with.

So 'what is sticking out' is one part. 'Where is the light source' is another (for example, you wouldn't expect highlights on the front of the object if the lightsource was behind the object)

The third guy from the right doesn't seem any different in this regard -- I guess he is lit from the front+above, creating the ring of highlights you see.
But this is one of the reasons people recommend study from life. An artist's portrayal of light is always imperfect and inconsistent to some extent, nature is not.