AuthorTopic: Method: Shape & Cluster approach  (Read 4750 times)

Offline Groggeneral

  • 0001
  • *
  • Posts: 7
  • Karma: +0/-0
    • View Profile

Method: Shape & Cluster approach

on: September 08, 2014, 07:35:13 pm
Hi everyone,

 i'm new to Pixelart and I have a few questions about the Shape & Cluster approach (as explained by cyangmou at deviantart

I was looking at a progress roll gif of one work by Thu (See here (For the progress roll see here and wondered whether that is a good example for the shape & cluster approach. In the garlic example by Cyangmou, it looks as if he just did the whole shape perfectly right from the start. As I understand it at the moment, the Shape & Cluster approach tries to develop shapes a bit more organically by staying away from the line-art approach. The organical development comes out quite remarkable in the work of Thu, so I wondered whether that is actually meant by Shaping & Clustering ?

I'd greatly appreciate more examples or tips on how people approach their pixel art from the scratch other than going by line art.
Thanks so much! :)

Offline Cyangmou

  • 0011
  • **
  • Posts: 929
  • Karma: +3/-0
    • cyangmou
    • http://pixeljoint.com/p/32234.htm
    • cyangmou
    • View Profile
    • Pixwerk Homepage

Re: Method: Shape & Cluster approach

Reply #1 on: September 08, 2014, 08:15:09 pm
First of all consider that there is no wrong or right approach to go over pixelart.

The main problem with beginner pixel art and lineart is, that if lines are drawn in, those lines usually don't vanish.
The process of thu's knight is a lot more messier, but he adjusts his proportions as he adds shading.

Some quickly grabbed examples from Deviantart:
http://blazt01.deviantart.com/art/Some-Monsters-474028619
http://shademist030.deviantart.com/art/Gift-Levi-Pixel-Doll-Neko-Systeme-480594917
http://jazminj323.deviantart.com/art/eos-sticker-version-ART-TRADE-480971622

As you see all of them have lines but there is no underlying form. With underlying form I mean that there is no feel of depth or 3-dimensionality.
Even if we are painting on a 2-dimensional surface it should be a goal to convey 3D forms, unless you don't explicitely want to break it.
 the difficulty is to get away from "thinking with outlines" and to advance to "thinking with 3d forms"

the painterly approach with refining usually leads faster to a 3-dimensional impression, because value-shapes are more powerful in describing forms than lines.
And it forces you to see the other side of the coin, if you exclusively have worked with lines.

Just look at this example, there is just a line but no shading at all.
http://minakie.deviantart.com/art/Minakie-Skull-2-0-480876150

the other extreme: here are no lines:


A cluster is just a blob of color. If you are going for descriptive illustration, I would describe it as  planar shape of an object.
Like if you simplify things down in simpler forms every form could be a different colored cluster.
But more importantly you have to know the structure of a form that you can recognize how clusters have to be put in in order to achieve the form you want to draw.

« Last Edit: September 08, 2014, 08:24:02 pm by Cyangmou »
"Because the beauty of the human body is that it hasn't a single muscle which doesn't serve its purpose; that there's not a line wasted; that every detail of it fits one idea, the idea of a man and the life of a man."

Dev-Art
Twitter

Offline Groggeneral

  • 0001
  • *
  • Posts: 7
  • Karma: +0/-0
    • View Profile

Re: Method: Shape & Cluster approach

Reply #2 on: September 08, 2014, 09:00:04 pm

Some quickly grabbed examples from Deviantart:
http://blazt01.deviantart.com/art/Some-Monsters-474028619
http://shademist030.deviantart.com/art/Gift-Levi-Pixel-Doll-Neko-Systeme-480594917
http://jazminj323.deviantart.com/art/eos-sticker-version-ART-TRADE-480971622

As you see all of them have lines but there is no underlying form. With underlying form I mean that there is no feel of depth or 3-dimensionality.
Even if we are painting on a 2-dimensional surface it should be a goal to convey 3D forms, unless you don't explicitely want to break it.
 the difficulty is to get away from "thinking with outlines" and to advance to "thinking with 3d forms"

The dimensional difference between outlined sprites and ones without outlines really is amazing, I didn't realize that until now. Thanks for the good examples!
So clear outlines always foster a sense of "sticker optic" / "lack of depth"? I thought it was mainly a matter of good shadowing to reach a high sense of depth/dimensionality but an ordinary (real) 3d-form (like a garlic) doesn't have an outline that distinguishes between the body of the form and its border. It's all just a set of bodies forming a whole object, no borders inbetween there.

I can see how it would be more difficult to develop a sense of depth working with a lineart-approach. So a lack of clear border lines and good shadowing could be considered the core of creating dimensionality? Is there something else?

Offline Cyangmou

  • 0011
  • **
  • Posts: 929
  • Karma: +3/-0
    • cyangmou
    • http://pixeljoint.com/p/32234.htm
    • cyangmou
    • View Profile
    • Pixwerk Homepage

Re: Method: Shape & Cluster approach

Reply #3 on: September 08, 2014, 09:23:18 pm
So clear outlines always foster a sense of "sticker optic" / "lack of depth"? I

nope, but you need to know a lot about form, to get an outline which conveys form.


thought it was mainly a matter of good shadowing to reach a high sense of depth/dimensionality but an ordinary (real) 3d-form (like a garlic) doesn't have an outline that distinguishes between the body of the form and its border. It's all just a set of bodies forming a whole object, no borders inbetween there.

the shilouette (outline) and the shading are equally important.
If you actually check the garlic with the photo, it's slightly off in a lot of spots. But it's close enough to look believable.
You just need to get close enough.
Reality is the best source for outlineless objects.

I can see how it would be more difficult to develop a sense of depth working with a lineart-approach. So a lack of clear border lines and good shadowing could be considered the core of creating dimensionality? Is there something else?

It's not more difficult to reach good forms with lines, it's about as hard as correct shading is. It's just the 2 smaller building blocks which lead to the solution of a bigger problem.
The thing is that you *have* to understand how form is working in reality.
And then you can abstract the form in lines or clusters - means to an end to actually describe form.

the basic technique to master of how to get a solid form drawing is perspective.


Although it's possible to use clusters in a graphical pattern-wise approach as well.


take for example cubism styled art

form approach: (the goal is to get depth with clusters)

pattern approach (the shapes look flat, but there is also no actual attempt to achieve any understandable 3d-form)
« Last Edit: September 08, 2014, 09:55:45 pm by Cyangmou »
"Because the beauty of the human body is that it hasn't a single muscle which doesn't serve its purpose; that there's not a line wasted; that every detail of it fits one idea, the idea of a man and the life of a man."

Dev-Art
Twitter

Offline Ai

  • 0100
  • ***
  • Posts: 1057
  • Karma: +2/-0
  • finti
    • http://pixeljoint.com/pixels/profile.asp?id=1996
    • finticemo
    • View Profile

Re: Method: Shape & Cluster approach

Reply #4 on: September 09, 2014, 11:06:30 am
...

You should probably rehost those images rather than hotlinking them.
Spoiler #2 in this post, and "here are no lines" image in your earlier post, both fail to display.

For others' use, here are the links involved:

"Here are no lines":
http://cudl.deviantart.com/art/C-Alfred-480816756

"pattern based approach":

http://www.ebsqart.com/Art/Cubist/Media-Style/732646/650/650/Cubist-123-2424-W-Original-Cubist-Art-Just-A-Dream.jpg
If you insist on being pessimistic about your own abilities, consider also being pessimistic about the accuracy of that pessimistic judgement.

Offline N0ixZ3

  • 0001
  • *
  • Posts: 7
  • Karma: +0/-0
    • View Profile

Re: Method: Shape & Cluster approach

Reply #5 on: September 13, 2014, 11:08:03 am
This method takes a lot of traditional light and form theory and lots of practice with it. I recommend you start learning about drawing/painting forms of all kinds and understanding light and how it interacts with forms and stuff of that nature. It'll be a head buster for a long time but at least you'll learn this shape and cluster approach.