Pixel Art Wiki

[tcaud]

I've been thinking that a pixel art wiki would be a useful resource. A wiki would serve two purposes:

• educate visitors about pixel art technique
• tag copyrighted pixel art for easy reference

Regarding the technique issue, a look around Moby Games reveals that there were and are many independent techniques for designing pixel art, each with their merits and drawbacks. Studies of these techniques could educate a new generation of pixel artists and result in improvements and refinements of existing technique.

The purpose of the art library is to observe technique in action. It would also serve as a source of inspiration and discussion. While there are archives of pixel arts already available, they tend to categorize by game (Spriter's Resource) or by entire sheets at the discretion of the contributor (OGA). There is need for a resource that permits viewers to observe the history of an artistic idea in all its variation and detail. A wiki (or content management site) could make this feasible.

[Ai]

.. Use the search function. It's been discussed (and tried!) before.

[tcaud]

Why not? Why not use the search function and just sort through reams and reams of posts interspersed over dozens of pages, taking 16x the amount of time to get the basic idea as you would with a single well written debrief?

[Mathias]

There's a new version of Pixelation in the works. It could have a built-in wiki. Canon, to the believers.

[tcaud]

There's a new version of Pixelation in the works. It could have a built-in wiki. Canon, to the believers.

Don't you think a continuation of Tsugumo's original endeavor, in wiki form, would be good for the site?

[ptoing]

We have discussed this probably 2 - 4 times now. The conclusion that we who run this place have come to, and I am sure many members agree to as well, is that just having some tutorial somewhere would take away from the human interfacing and back and forth, which is a big part of learning.

Also there really is not enough technique in pixelart to warrant a full on wiki, unless you wanted to have super specific entries on stuff like grass and dirt tiles, which imo would be pointless as well.

[Pix3M]

Also there really is not enough technique in pixelart to warrant a full on wiki, unless you wanted to have super specific entries on stuff like grass and dirt tiles, which imo would be pointless as well.

Is it more pointless than I realize?

I've seen enough badly executed knockoffs of SD3's grass tile to know that people LOVE to use super-specific tutorials like that. Not to mention, while it would be kind of obvious to the more serious artists that we ought to use references from real life and of other artists's, maybe such super specific tutorials can be useful. Only difference from other tutorials floating out there on the internet is that we introduce special considerations that come with the medium, and we cover the range of possibilities (fully textured vs simplified approaches).

Maybe tutorials aren't as great as their users think they are, but people still love them.

[ptoing]

Learn to draw, observe nature, abstract shapes and textures from nature into pixel things like grass tiles.
This will get you further and you will learn more than just reading some grass tile tutorial.

If someone would want to go through the forum and search for super helpful threads, like the ramblethread and such, it might be viable to move those all into a childforum under pixelart or something. That I would not have anything against. But a Wiki is overkill.

[tcaud]

I didn't intend a wiki for tutorials, but for analyses. Notes that one particular artistic goal was accomplished in several different ways, with an accounting of each way and its draw. Something that a newbie wouldn't find interesting, but that a serious artist could learn well from. And I don't think such a wiki would reduce discussion at all... if anything, it would increase knowledge of specific techniques and create a reference point for the same. Now of course if you're just gonna rehash the same old thing over and over and focus on communicating that knowledge one person at a time, then I suppose a forum-only solution is indeed superior (or a chat room). However, this won't make the best artists any better.

[tcaud]

Alright well, I'm going to make such a system/service/whatever on my own. Because although a forum is a great way to exchange ideas, it's a piss poor reference system. If there are any who want to participate, mail me at tcaudilllg@gmail.com. I have my own site for the purpose, so that's not an issue.

[tcaud]

WikiPixels is in business. Link: http://wikipixels.gamestargcs.net/

Thanks to Dustin Bales for the logo.

[Ai]

I've added some software to the list and fixed up the EGA description somewhat. Not sure what to do with the 'EGA is sort-of 6bit' statement -- it's basically wrong, but I'm not sure how to replace it (the actual restriction usually used in 'EGA art' is the "fixed" 16-color EGA palette. AFAIK the ability to select any of the 64 'master palette' colors is only available in higher-resolution/tweaked modes (for example the stock 640x350 res available on 256k EGA cards))

Anyhow, this link (Ptoing's 'Restriction guide' thread) should be a good resource to help populate the History section. Gives a pretty good overview of CGA, EGA, C64, Amiga 500, CPC, Spectrum, GBC in Ptoing's original post, and respondents also covered NES, SMS, GameGear  (and CPC+, KC85, but they are overly complex, not historically important, or both.)

Is the aim of History pages partly to show some exemplary pixelart of each particular system? Obviously they should provide some more relevant information than say, what can be gotten by looking up the relevant system in Wikipedia.

[Mathias]

Good job, tcaud! I hope its content continues to expand.

[tcaud]

We're off to a good start. We need more contributors!

Here is some content I think would make the wiki more useful:
- pixel art lessons
- palette collections

I'm no professional so my ability to contribute lessons is limited. As for the palettes, I know of Arne's and Dawnbringer's... any others?

[Mr. Fahrenheit]

Manupix's and Neot's are some good ones. There are also a lot of game palettes, not sure if you are including those in this list.

[tcaud]

The problem with game palettes is that if you use them knowing what they are, you can be accused of derivative work by the copyright holder of the game. But it depends on whether or not the colors are forced, I think (NES, Master System), and the breadth of the master palette. Regardless I'd make sure I had permission before using any palette besides my own.

[Pix3M]

I would recommend to show only palettes that does something particularly well to show what is cream of the crop. Otherwise, knowledge of actual color theory will better serve artists than mindless ripping off palettes of other works.

[tcaud]

My personal preference is a straight up 9-bit palette, and I don't find the default VGA palette that horrible, or even the websafe palette. As long as a palette has a decent range of skin tones, and enough shades for effective dither, it's workable. I think people like using small palettes because they don't want to have to think about what color to use when they draw something. But for work I'd consider professional, it's about always necessary to just experiment with color until you get just the right shade. I was going for a Super Metroid "Brimstar"-style effect a few months ago which had spores on the ceiling... 9-bits didn't give me the color variation I needed to pull it off.

Regardless, I think palettes are something people like to talk about. They are a part of the fun of the medium. Thing is, people take them too seriously. If FF Legacy redid its sprites and tiles and levels and names and still used the same colors as FFIV, would it taken down by Square again? Possibly, because anything made with that palette will ultimately end up looking like Final Fantasy IV, just as any game which makes use of the SNES FF's french horn will sound like Final Fantasy. I dunno, maybe I'm overblowing it.

In my opinion, anyone who would want to use a commercial game's palette, is most likely trying to recreate the original game, and that's not worth encouraging anyhow.

Pix3M I agree with you. I guess for a palette to do something well, it would have be particularly well suited to drawing particular things?

[ptoing]

Nice effort, good luck with it. Maybe I will add something to it sometime.
I suggest changing "EGA Technique" to something else. There is no such thing as an EGA Technique. Try not to invent terms if you can avoid it.

[tcaud]

I think if you develop a specific clustering technique to accommodate for the limitations of the CGA palette in EGA mode, that's an EGA technique. Just as if you develop a set of guidelines for rendering with MS Paint, that's MS Paint technique.

[PixelPiledriver]

I think people like using small palettes because they don't want to have to think about what color to use when they draw something.

While that's a valid approach to using colors, just as using a random palette is valid, it only barely touches on the use of limited palettes.
There are many popular ways to select and use colors, but it is also easily invented.

I guess for a palette to do something well, it would have be particularly well suited to drawing particular things?

It depends on the purpose of the palette and the goal of the artist at the time.
Some purposes:

Global
Colors used to represent all objects in the image/world.
Can be broken down into sub palettes, using a small group of colors per object.

Object
Colors specific to a single object making it unique from other objects.

Group
Colors used on several objects.
This creates a relation between them in the mind and eye.

Theme
Colors used to represent a specific event/object/character/idea.
Some examples:
Event -> Christmas: Red + Green
Object -> Egg: White + Yellow/Orange
Character -> Santa Claus: Red + White + Black
Idea -> Magic: Rainbow!

Mood
Colors that evoke a certain emotional response.
These overlap into other moods and are easily repurposed, but some examples:
Anger: Red
Happy: Yellow
Sad: Blue

Natural
Colors specifically chosen to represent an object as it is seen.
Drawing from life.

Iconic
Colors chosen to represent an object as it is thought of/ known as.
Green grass, blue sky, brown dog.
Drawing from the mind.

I'm missing some purposes and each has many more properties, challenges etc.
It's worth investing some time into later.

anything made with that palette will ultimately end up looking like Final Fantasy IV

Only if the groups of colors are used in the same combinations and balances.
The meaning and usage of a palette can be tweaked.
It's a short step to take happy colors and draw something horrific.

They are a part of the fun of the medium. Thing is, people take them too seriously.

It's ok to take fun things seriously. 
Closely analyzing a subject allows for ability to manipulate and be creative.
As a mostly graphics programmer I do a lot of color math.
While it is not always palette related I utilize some knowledge of color theory every day.

[Pix3M]

What do we really mean when we say that people take palettes too seriously? I think colors are a huge subject when it comes to art.

Plus not all of us are aimless and largely experimental with color. It just takes a bit of knowledge of color theory to just know what you're doing and just constantly make purposeful, minute adjustments to get the effect you want.

[ptoing]

I think if you develop a specific clustering technique to accommodate for the limitations of the CGA palette in EGA mode, that's an EGA technique. Just as if you develop a set of guidelines for rendering with MS Paint, that's MS Paint technique.

I think this is pretty silly. There are no special techniques that people use to render with different palettes. The only thing you do is how you use the colours, which is something that falls under colour theory.

Reading "MS Paint technique" made me laugh.
At best you could boil it down to certain workflow steps which are ideosyncratic to a certain program, like MS Paint, or Deluxe Paint. All you are doing in the end is adding pixels, subtracting pixels, or changing the colour of pixels (which really is ALL you are doing if you reduce it to the lowest level), different programs give you different ways to do so, but there are no tools I am aware of that have things which are fundamentally different at a workflow level that you could call it a technique, and MS Paint certainly not.

You could call certain ways of working a technique, like the technique of working from an outline, or from polygonal fills or from a scanned pencil drawing. Just like there are different techniques of applying paint only a canvas. But calling something MS Paint Technique or EGA Technique makes no real sense. It is similar to an artist only using red, yellow and blue paint to mix everything he paints and calling it the Red-Yellow-Blue or Primary Colour Technique.

[tcaud]

Reading "MS Paint technique" made me laugh.

I'd appreciate if you didn't troll the thread.

Lately the wiki is being used as a resource link station. I think there is a lot of potential there and I really do think that if we work hard then there will come a time when a person will be able to set aside a day and just browse the wiki in one sitting (or over a week) and come out a pixel art expert.

This reminds me: how many pixel art tutorials have been archived to the Way-back Machine? It would be a shame to lose the all that knowledge and effort to circumstance, don't you think?

[ptoing]

Why do you assume that I am trolling? I am just giving feedback and my opinion here.

[Helm]

Perhaps telling people what they say makes you laugh is a touch rude. But on the point, you will find many old hands in Pixelation to be very resistant to making up new terms, because we grew up in an environment (old Pixelation) when this sort of talk happened way too often and led to very weird inter-community results, so if we don't need a new term for something, we won't endorse it.

[tcaud]

"Technique" seems like the only way to explain the subject to the technically disinclined. Sure a very technically (even mathematically) inclined individual can appreciate the term "color theory", but what we're really talking about is blends. I think the word "blend" requires too much experience for newbies to appreciate, because "blend" is both a verb and a noun. Such words must be treated with caution when writing for laypeople. And, the article touches on topics other than blends besides.

Whatever happened to the old board anyway? Is all that art gone forever or did things just fall apart somehow?

[Helm]

Sadly 'technique' sounds cool. THAT is bad for laymen.

[ptoing]

"Technique" seems like the only way to explain the subject to the technically disinclined. Sure a very technically (even mathematically) inclined individual can appreciate the term "color theory", but what we're really talking about is blends. I think the word "blend" requires too much experience for newbies to appreciate, because "blend" is both a verb and a noun. Such words must be treated with caution when writing for laypeople. And, the article touches on topics other than blends besides.

Do you realise that this is a highly patronising sentiment? Blending is not a very complex concept at all. I understood basic colour blending when I was in kindergarten and so do most kids if you explain it to them.

Really, it is a much better idea to go into concepts like colour theory and do write ups on that instead of making up stuff like EGA technique. What Helm wrote about the "it sounds cool" is very true as well. Stuff that sounds cool might attract laypeople to it, even if it does not have any great value to the subject.

And to the specifics of EGA technique, how would you differenciate EGA technique from CPC technique from +4 technique from MSX techique from C64 technique? The only thing you can say is that you have different colours which go together in different ways and you might have some system specific limitations.
But that does imo not warrant calling the different ways you use different sets of colours together a technique. Arne 16 colour palette technique, Dawnbringer palette technique... Try to focus on broader things like clusters, general colour usage, antialiasing, dithering and stuff like that.

Having a whole bunch of tutorials on how to make grass tiles potentially will serve more as a crutch for lazy people than anything. It is better to make it clear to people that they should try and use reference, ideally from real life, and try to abstract from there, to whatever suits their needs. There are enough trees with very round crowns and blue reflected light in their shadow already.

Whatever happened to the old board anyway? Is all that art gone forever or did things just fall apart somehow?

The original board got hacked back in 2004 I think it was, that stuff is lost. The posts anyway.

[PixelPiledriver]

Do you realise that this is a highly patronising sentiment?

Exactly.
Relax tcaud.
There is some benefit to unveiling information 1 layer at a time as things become complex or specific.
But there is little benefit to assuming people might not get it.
If the information is clear and helpful then people will take the time they need to understand.
Or they will ask for another perspective.

[tcaud]

Blending is jargon. Jargon should be avoided when explaining things to newbies/laymen/what-have-you.

Yeah "technique" sounds cool. Cool matters. It matters to make people use their noggins in a variety of different ways. The experiences stay with them. It's worthwhile to see people experimenting.

The EGA uses a very specific, non-programmable palette. There is a specific sets of rules you need to follow to get the most out of it, as with any palette. Maybe create a set of articles just for blends, discussion of specific color blends, interaction between colors of different palettes.

Tutorials could be an excuse for laziness, but I think the sense of dissatisfaction with rehashes and "the usual" is its own motivation for creativity.

ptoing you seem to be suggesting that effective instruction has both color and technique aspects.

[ptoing]

I do not buy the "specific set of rules" you say is needed to be followed when doing EGA. I never thought about it that way, ever. If you have a decent knowledge off colour theory then you will figure out which colours work together in ANY given palette.

One exception I can think off would be something like PAL blending, which occurs if 2 colours have the same luma and are displayed on a proper CRT monitor with a PAL signal. Those 2 colours will make a totally new colour if alternated horizontally, because of how the hardware works and the PAL signal is encoded. This is something that works on the C64 and maybe on other Commodore 8 bits as well. This could be called a technique, but it is not C64 technique.

And about blending being jargon, I think everyone understands what blending means. If you tell someone about colours blending in natural media such as oils, they will get it, if you tell someone how printing is done with dots grids of a bunch of colours and that the colours to a degree blend visually, they will get it. If you stop assuming people will be confused by such terms, and treat people like intelligent human beings, you will get further in what you are trying to do. Art is not quantum physics or anything like that. The concepts of colour blending and such on an application level are not very hard to understand.

I think that effective instruction has a lot to do with looking outside of pixelart, which has a very tiny insular set of specific things which are kinda unique to it, such as manual antialiasing and dither (dither already is borderline as you could dither in non pixelart too). Colour theory, perspective, anatomy, composition. This is the kinda stuff people should learn. I believe that any artist who has solid classical training could be made into a superb pixelartist in a matter of months if given the right instructions or given good self-motivation.

[tcaud]

That's all pretty good for replicating stuff you've actually seen before, but maybe not so great for creating exotic art like we'd like to see more of in games.

[Helm]

That's all pretty good for replicating stuff you've actually seen before, but maybe not so great for creating exotic art like we'd like to see more of in games.

Are you making a wiki to assist in learning or to influence the direction of future pixel art?

[Mathias]

A wiki should be as unbiased and neutral as possible, just presenting the facts.

[ptoing]

That's all pretty good for replicating stuff you've actually seen before, but maybe not so great for creating exotic art like we'd like to see more of in games.

I don't even know what your thought process here is. Do you think any artist that made something new got that taught somehow? Like Dali, or Escher, or van Gogh, or any other great artist who broke some new ground. They all had more or less amounts of traditional art training and then brought themselves into their art. That is what you bring to it. How you see the world, how you recombine things you have experienced and seen into something new or something relevant to you you feel is worth expressing. You are a collection of your experiences, and the more you know and have seen the better you will be able to do certain things. For example a creature concept artist who actually spends lots of time looking at real animals, drawing them, studying their anatomy, will be able to make much more convincing new creatures which do not exist.

[tcaud]

Well yeah but the point is to create a conversation. To create a conversation of a certain level, you have to elevate everyone to that level of relative sophistication. That's what the wiki is for: to allow people to attain a baseline quickly. The more information on the wiki, the higher the baseline, the better they can contribute to the discussion. I'm actually much more concerned about seeing people create novel designs than discussing particulars of form in specific circumstances, as people do here. The introduction of novelty, even at low quality, has a way of working itself out when the socio-environmental conditions are right.

[Helm]

People will create novel designs because that's what people do. I'd be more interested in solid education.

[Mathias]

http://wikipixels.gamestargcs.net/index.php?title=Palette_Collections

Several "obsolete" typos.



*EDIT - nevermind; appears intentional . . .

[tcaud]

People have been taught it's bad to be creative. That needs to change. Creativity can be good, in moderation. Questioning is what's important.

[Mathias]

People have been taught it's bad to be creative.   . . .

Whoa, what now? All I ever see is creativity being promoted in creative circles. Where/how is it being suppressed?

[Helm]

I don't understand what you mean.

[ptoing]

You can argue that thinking for yourself and being creative gets discouraged in certain school scenarios, but certainly not in any online art forums I have been to.

[Helm]

In any case, a pixel art wiki should be presenting history. Not inventing terms. Not nudging people towards any one direction.

[ptoing]

Wholeheartedly agree with that.

[tcaud]

I look at it this way: we see all this creativity on this forum... but when do we ever see that in a game? Practically never.

[Helm]

First of all there is a lot of pixelation technique from pixelation users in a lot of games, not all of them high profile. But if you look around you'll find them. Check out the stuff Cyangmou's doing, for example. He's one of the best westerner pixel artists right now and he uses a lot of the technique we discuss on this forum.

Second, the reasons not every asset in a videogame can be worked on the level people work their stuff in here is due to pipeline and deadline issues most of all. Of course people are going to spend more time in their personal projects, and even go a bit 'out there'.

[tcaud]

That's what needs to change, then. If the pipeline can't produce compelling product, it needs to be questioned. This leads me to believe that a free-lance industry would be a stronger industry, if there is indeed a negative correlation between producer control and creativity.

I see three solutions:
- accelerated creativity
- longer-term financing
- broader employment base

[ptoing]

It is all about budget and estimated return in the end. If a developer pays a freelancer to make pixelart they will have a certain budget limit of how much they can shell out. The pixel artist then can say, OK, I can do this, and this will be about the standard of work I can do, if you want me to go super fancy it will come out more expensive. It is as simple as that. Pixelart is a lot of work, and if you had an AAA sized team work on a pixelart game with something like 30 of the best pixelartists, a brilliant art director and great programmers and all that, and of course a good concept, you could make something totally amazing as far as art goes. But it is very unlikely to happen because it would be a very high risk investment that no publisher at this point in time will be willing to make.

Then you got something like kickstarter, but even that wont allow for something of this scope. A lot of the goals on there are unrealistically low to begin with.

[Helm]

All that's fascinating but pixel art technique exists right now and it can be taught right now and a wiki should mirror that history, not hypothesize on 'what should change so more great pixel art arrives in future products'.

[ptoing]

What you should do for the wiki is gather information on games which have very solid pixelart (within their timeframe in history, compared to contemporary games at the same time) and the development of pixel art due to changes in hardware and such. And then also articles about specific pixel artists. Getting info about Western pixel artists should be a lot easier than getting info about Asian ones, though.

[tcaud]

That sounds like a very good idea... specific attention to portfolios has been missing. Leaving it to contributors to create their own page has the downside of no one but them being able to edit their page. So you're right in that biography should be relied upon, rather than autobiography.

So on the matter of unoriginality, I take it that NDAs hamper artists and basically put a choice in their hands of "draw what we tell you to draw, or go it completely alone". It's a matter of power imbalance between the project leads and their relationships with the suits, and the people who are essentially. When teams were small there was balance between leader and staff, which resulted in the wild creativity of the early years. Now with larger teams there is risk of outright mutiny, so the project lead is given more control and the power to make people do as they are told. In other words, egos with high charisma and low talent are ruining the industry.

What if games were marketed explicitly for art? I've noticed a few Kickstarters around that theme but they don't appear to have been very successful, maybe because they didn't involve many people.

[Helm]

Are you involved with game development? Then or now?

[tcaud]

Are you involved with game development? Then or now?

Professionally? Yes and no. (and it's irrelevant to the matter at hand... I can offer opinion if I will, free discussion).

I see problems and I'm looking for solutions.

I guess raising the profile of artists is one way to go about it. Nameless artists have little creative control.


On another tangent, what do you think about this edit?

http://wikipixels.gamestargcs.net/index.php?title=Palettes_Applied&oldid=165

[ptoing]

To me that looks like something that should not be on a pixel art wiki and it sounds like it is taking the piss.

[Ymedron]

From my point of view, I think color should not be suggested upon anyone newly learning pixel art - or any art for that matter. The princibles of contrast, harmony and color scheme should be taught as a theory but offering specific palettes doesn't seem very helpful for newbies, especially since it puts specific meaning to certain combinations of color ('painted metal has x y and z aspects in its colors') which don't really even work. I think it's possible to make an object appear metallic with just about any three colors if you're skilled enough.

Also in general providing palettes for specific aspects of an illustration fails to educate on how light affects the local color of an item or the effects of color on a scene. Falls into a slightly more elaborate version of "sky is blue so I'm coloring it with my blue crayon"

Did my message make any sense?

[ptoing]

Yeah, that makes a lot of sense and I agree. It is one thing to provide an archive for old computer and console specs and palettes and another to write about specific use of those colours. It is pretty pointless.

This image I made is as far as I would go as far as specifics are concerned. And the greys in the C64 palette are a bit special case here.


But I would not go: This ramp is good for skin, this is good for metal, this is good for whatever. A lot of people have made nicely working stuff with unexpected colour combinations, which as Helm pointed out is just something that humans do, being creative. Telling people about super specific ways to use palettes will not help them be creative necessarily.

[tcaud]

As I said before, foundations matter. You must understand the standard before you can begin to rebel against it.

[hapiel]

As I said before, foundations matter. You must understand the standard before you can begin to rebel against it.

That is something I mostly agree with.

But are you suggesting that what is written on that wiki is the standard? If so, why does it lack any references?
To me it seems to try and create rules rather than point them out. Also the information is very weird...

For example the Palettes Applied page..

Painted metals are defined in 3 shades.. since when? What is wrong with 2, 4 or any other number?
I would agree that highly reflective surfaces such as metallic ones show big contrast between the darker (especially if it reflects a dark or shadowed object) and the lighter spots. However the palette you show does not always have big contrast, for example the light teal and medium teal are super close.
Why would you suggest a palette for a specific purpose instead of giving it as an example? (preferably an example showing rendered metal and not just the palette)

No I have not seen these colors at my local art store. Most of these colors are for me very typical screen colors as they can not be recreated without them being a light source. Way too high saturation to look like something I might find on a pencil!

Sorry for the hostility against your wiki project, but it really is not taking a good approach at collecting and displaying information.
This is the 'what would tcaud do' guide, not a pixel art encyclopaedia...

[tcaud]

My goodness... have you ever looked at a car? It'll gleam white at the reflection of the light and be two shades pretty much everywhere else. It's direct observation.

It's turning out to be a pretty good resource for my benefit... if you don't agree, don't contribute.

[ptoing]

Three colours, yeah? There are some very high gloss metals and stuff like metalic paint, but even those need a few more shades especially if you have curved surfaces.

And I think hapiel is correct with his last sentence, it is a wiki written from your subjective perspective and lacks objectivity.

[ErekT]

Hehe. Well to be fair you could represent that with 3 colors and get the idea across. Or 4, or 20, or three thousand etc. Hapiel and Ptoing make good points. Best not be dogmatic about this stuff. Many many different types of metal surfaces, light conditions to illuminate them, and art styles to represent them with. Instead, maybe word it something like "one (often-used?) method to represent metal surfaces is to divide them into three main shades of high contrast etc"?

[ptoing]

Better would be to have a couple of example pieces by different artists which show good metal surfaces, and then perhaps a couple of shaded spheres with different amount of colours and lighting situations. Just saying: "Here, these are good metal colours, use 3 shades, bam.", is not helpful.

[tcaud]

Three colours, yeah? There are some very high gloss metals and stuff like metalic paint, but even those need a few more shades especially if you have curved surfaces.

And I think hapiel is correct with his last sentence, it is a wiki written from your subjective perspective and lacks objectivity.

That's not spray painted metal.

This is.



And I spy three shades.

I know you want to think I'm some kind of idiot, or overall not as intelligent as you, but it's really not a good idea. For one thing, you know nothing about me. For all you know, I'm one of the most if not the most intelligent individual alive. Whether I am or not, it's really not a good idea to underestimate me, as it tends to result in major humiliation on the part of the underestimator, which results in turn in lost reputation and attitude hardening which tends to trigger descent into delusional disconnect, among other psychological maladies.

By the by, you might want to stick around for a particularly good show I intend to put on at Bugzilla tonight. You'll need to do some catching up to understand what's happening, but it'll be worth the price of admission I assure you.

[PixelPiledriver]

You asked for critique about the wiki.
People are giving suggestions that would improve it.
Based on what you stated your intentions of the wiki are, the current content doesn't fit that well.
You're bringing a lot of emotion and opinion based information into many of the articles.
If the purpose of the wiki has evolved to be a place where you write about stuff, and not an educational source, that's fine.
But it would be better to recognize that and ask for critique based on that perspective.

[tcaud]

I think a lot of amateur pixel artists (read: people who like drawing NES sprites) would find the painted metal perspective rather useful. If I didn't think it a useful epiphany I wouldn't have written about it. You need to understand that a lot of people who like pixel art will never have the mastery of form required to say, shade clothing accurately. They don't have the ability because they lack fundamental interest in the actual process of considering how one fold determines another fold. These simple rules are enough for them to shade small sprites, which is all they should quite frankly ever aspire to (and all they will ever stick with besides). If everyone had talent in the visual arts, there wouldn't be anywhere near as many demos and half-baked engines with bad art. But if we can just raise the caliber of that bad art a little bit, I think it'll be worth it.

[PixelPiledriver]

Also please be mindful of Rule #1.

[Jeremy]

I think a lot of amateur pixel artists (read: people who like drawing NES sprites) would find the painted metal perspective rather useful. If I didn't think it a useful epiphany I wouldn't have written about it. You need to understand that a lot of people who like pixel art will never have the mastery of form required to say, shade clothing accurately. They don't have the ability because they lack fundamental interest in the actual process of considering how one fold determines another fold. These simple rules are enough for them to shade small sprites, which is all they should quite frankly ever aspire to (and all they will ever stick with besides). If everyone had talent in the visual arts, there wouldn't be anywhere near as many demos and half-baked engines with bad art. But if we can just raise the caliber of that bad art a little bit, I think it'll be worth it.

That wiki says "The ArseGA palette is an excellent choice for metals, employing an appropriate distance between shades". That doesn't make any sense! Are you saying you should only pixel metal surfaces with that palette? Or that you should grab a ramp from that palette when pixelling metal? Either way they're hypersaturated; not easy at the best of times let alone when just starting out with PA. The paint by numbers type of tutorial you're trying to find and collate are nowhere near as useful as general art/colour theory, maybe with some pixel-specific issues to deal with.

Oh, and flat walls are all very well and good, but that's not what the world is made from (even disregarding obvious colour reflections)

[ErekT]

These simple rules are enough for them to shade small sprites, which is all they should quite frankly ever aspire to (and all they will ever stick with besides). If everyone had talent in the visual arts, there wouldn't be anywhere near as many demos and half-baked engines with bad art. But if we can just raise the caliber of that bad art a little bit, I think it'll be worth it.

Know thy limits isn't a mindset to be encouraged imo. Anyone can learn to become a pretty good artist I think, at least technically, just as anyone can learn to read or pick up a new language. If they're not willing to learn because of lack of a fundamental interest or whatever tho, well tough shit. You can't expect to be good at anything if you can't be arsed to put the effort into learning the ropes. Shortcuts will at best give people a limited and inflexible skillset and maybe also a false sense of security along the lines of "yeh I can do something that looks a bit like that with my bag o tricks", that stunts their development.

[Helm]

As I said before, foundations matter. You must understand the standard before you can begin to rebel against it.

What is becoming increasingly apparent is that whereas this is true, you're not setting foundations based on historic research but based on your own opinions. And we can shoot down these opinions all day and maybe that's useful to you but it's not a good foundation for an objective, informed wiki.

[ptoing]

I never said that you were stupid or anything, I just think that your wiki is too subjective and as such not really helpful. Also that "spray paint" metal you posted is a bit of pretty much flat metal with raindrops. Is that a representative example? I don't think so.



I think this is a better example, and it could probably be done in 3 shades, but it would look better using more.

[Helm]

You need to understand that a lot of people who like pixel art will never have the mastery of form required to say, shade clothing accurately.

It is not your place (or mine or anyone's) to dictate what the limits of knowledge of a third party are and for someone who claims to be interested in helping people this is a really rotten mindset to have.

[Cyangmou]

I read through the wiki.
It doesn't seem really helpful at all for anybody.
It's too complex for beginners with giving them a point to start and at the same point too generalized for experienced artists.

People who wants to get better at any thing usually find their own solutions with the common and hard way called "practise" and through "experience"
People who wants to get better at art have various places where they can receive individual critique and opinions which usually help them a lot more than a generalized text.
Good art don't comes from great resources, it comes from hard work and there are no shortcuts at all. Even if you tell somebody some rules he won't neither maste rthem immediately nor uses them because he lacks the experiences of all his failed attempts.

Fundamentals of art are no secret, they are shown over and over in the web, in art books, in videos, schools etc. None of those resources helped to improve the general art skill for all people around.
Pixel art is just another technique of art, like oil color or charcoal are, the basics are the same as for any other art technique, there is already enoguh around.

You don't help anyone if you show him a supposed "shortcut" or better "it's done like that".
You rather help people if you encourage them to make their mistakes and learn from them that they can come up with their own ways of solving problems and have fun with finding ways how to solve problems.

I'd recommend that you should rather start a piece of art and post it here, that everybody can see it, learn from it and maybe give critique and help to improve it - that's the purpose of this place here.

[Pix3M]

I mentioned earlier that knowledge of actual color theory is more helpful than blindly following another artist's palette.

If you were to have a painted metal object outdoors on a sunny day, the color of the highlights should be white as I've observed that highlights will generally be the exact color of the light source - highlights are essentially reflections. Shadows should be shifted toward the cyan range due to the sky becoming the dominant light source in shadowed areas. Even then, this is putting aside reflected light which can complicate lighting a bit further. When you start to understand lighting, colors really depend on an subject's surroundings.

If we were to approach colors for a specific substance material like, spray-painted metals, I think an approach that would be a lot more helpful is to explain the physics underlying how different substances interacts with light. That sort of knowledge is part of the fundamentals of learning how shading and lighting is done. That sort of knowledge can be individually applied to individual works; the artist themselves can make their own decisions for what they specifically require. If we know how to properly convey different substances, and we probably won't need an encyclopedia of what colors to choose for spray-painted metals once we understand what colors are supposed to be used for the situation. Even then, doesn't hurt to use a ref for safety measure. 

[tcaud]

You need to understand that a lot of people who like pixel art will never have the mastery of form required to say, shade clothing accurately.

It is not your place (or mine or anyone's) to dictate what the limits of knowledge of a third party are and for someone who claims to be interested in helping people this is a really rotten mindset to have.

(this is not Tsugumo's Pixelation...)

My critique stands: you don't think like dot artists, but like artists who happen to draw at a high zoom. You speak little to the NES spriters, the GBA spriters, the RPGMaker peeps. Frankly you all seem stuck in a time warp fixated on the Commodore 64. And I maintain that the failure of this forum compared to what Pixelation was is a testament to the greater pixel art world's opinion of your opinion of the art. You all have talent, but you lack for -EVERYTHING- else.

[Helm]

If you feel that way about Pixelation, well, nobody's keeping you. Go forth and be the change you want to see in the world. Make a community around your wiki, support your approach with, say, a decade of community building and see what comes of it. I wish you success.

[ptoing]

You are making such misinformed assumptions, it's not even funny. I have worked on GBA, DS titles. Other people here have as well. I also have made pixelart to a lot of different specifications for the fun of it, such as (yes) C64, Amiga, Atari ST, ZX Spectrum, MSX, and others. Quite a few people here who do pixelart for a living, myself included, do not do gamey pixelart in their freetime. I do enough of that when I work.

But any of those people you mentioned, GBA Spriters, RPG-Maker folks or whoever else, could come here, post their stuff, and will get crits with what they are trying to achieve kept in mind. It is not like we are trying to force some artistic vision on anyone. Have you read Helm's thread on clusters? That is at the ground fundamentals of what makes "dot" art as you call it, taking the smallest element into account and how a single pixel change can make a big difference, esp in small resolutions.

So really, I have no clue what you are even trying to say, are you suggesting what we do and "dot" art is somehow different?

[Carnivac]

oi...

You speak little to the NES spriters.

Like me (to some extent)?

Frankly you all seem stuck in a time warp fixated on the Commodore 64.

Not me...(I use the other 8-bit computer with the wide-pixels)

[tcaud]

The peculiar nature of clusters is the main reason I enjoy making pixel art. However, by their very nature they are impossible to study, only refine.

But we see things fundamentally differently. So I won't waste our time anymore.

[Ymedron]

if you don't agree, don't contribute.

Nobody ever improved with the sentiment "Don't like don't look."
Furthermore, many real problems in the world are perpetuated by this same thought.

The idea of a wiki is to employ the knowledge of the masses, and to rely on the chance that someone more knowledgeable will come to correct the mistakes.
However, of course a private wiki can be someone's personal recording of what they think about this or that. You should clearly indicate it in that case.

I'm also extremely insulted that you think pixelation is stuck in the c64 era or whatever, since most of the art I've even seen on this forum has been way more modern (snes, gba or ds to use consoles I know anything about). You're discounting a lot of talent with petty arguments because you can't accept that your work isn't fit for the purpose you've stated before. It works for you (though without examples I can't even say that much) but to educate others it's inadequate.

I suppose you won't even be looking here anymore, oh well.

The best way to get me to rant is by doing the deviantart-dance regarding critique.

[Carnivac]

I'm also extremely insulted that you think pixelation is stuck in the c64 era or whatever,

Heh.  This caught my eye straight away in that post.  The word 'stuck'.  I often (most recently this morning in fact) get accused of being stuck in the past, stuck in the 8-bit era, stuck doing NES or CPC style graphics... I'm not stuck doing any era or style, nobody here is.  It's a choice and not one anyone has to be stuck with.  I bet I'll still get those annoying emails that say I oughta 'learn' how do draw HD graphics and such cos low res/color is apparently crappy and out of date.

[Ymedron]

Ah yes, sorry Carnivac. I guess I think of "stuck" in the same way as being stuck in traditional art vs digital... Blindly hating the newer inventions because they make it too easy or whatever.
Anyway, an informed choice to do a certain style is an admirable and fine thing to do. c:

[ptoing]

And speaking specifically about pixelart, working with restrictions actually can make you learn a lot. In other art too, but especially in pixelart, I think.

[Ai]

Restrictions make it easier in a pretty real sense IME, just by reducing the amount of variables you need to deal with. Enabling much greater conceptual clarity by removing a lot of peripheral concerns.

@Ymedron, Carnivac:
There's also 'modern retro art' in the good sense as distinct from 'contemporary retro art'; for example Fres Fighter which is artistically well beyond 98% of the CPC games that were available when CPCs were popular, enabled by the use of tools (paint programs, emulators) that are just plain better. The standards of CPC, C64, NES art are still improving.

@tcaud: I agree with Ymedron and others that I don't want to see this subjective-content-written-in-a-style-suggesting-objectivity representing pixel art or pixel artists on the web, and a private wiki is more appropriate. Or more specifically, a personal wiki -- I can recommend Zim from personal experience. Personal wikis avoid the maintainability, cost, and security issues incurred by server-hosted wikis, while also generally having a nicer UI, more robustness, and more flexibility.

[tcaud]

I use the same site for several things, including my game maker (I couldn't find stable hosting for it anywhere else).

Honestly I think the site is coming into its own as a starting point for learning. Now honestly PixelJoint would probably be a better home for this kind of project. I notice that the materials there are very good, and I'm linking to several of them from the wiki. But I hope people understand that neatly organized information has value. Search is good for getting the latest information about something you have basic familiarity with, but the world's internet forums are no match for Wikipedia when it comes to gaining intellectual foothold.

If I see something like this happen at PixelJoint, I will probably take the wiki down. I am also open to handing the wiki's backup over to PixelJoint proper, if they want to host it. At the very least, I think WikiPixels is a good name and that its logo should be preserved, because it was very gracious of its author to make it.

I am adamant that extensive, well organized reference material is key to the usefulness of a wiki such as this. Putting all the palettes you can find on one page, or even giving them separate articles, is not a bad thing. But one thing the wiki should not be subject to is immature trolls. I've seen them in action on other wikis for other topics, and they literally strangle their evolution. It's controversial to ban them but I think it's for the best.

In some ways the wiki is treated like my personal resource but it's simply not my nature to stockpile private information. If I'm going to exert my mental activity in novel directions, then I want it to be of use to others. I understand their needs may differ and I'm willing to negotiate to the extent that the resource in question continues to serve my own intellectual need. Although I can be gracious I am no altruist.

I do want people to participate, so long as they respect my contributions. In general I think contributed information represents well-intentioned mental effort which should be respected.

[Carnivac]

@Ymedron, Carnivac:
There's also 'modern retro art' in the good sense as distinct from 'contemporary retro art'; for example Fres Fighter which is artistically well beyond 98% of the CPC games that were available when CPCs were popular, enabled by the use of tools (paint programs, emulators) that are just plain better. The standards of CPC, C64, NES art are still improving.

If you say so...

[ptoing]

To what is that "If you say so..." referring? To Fres Fighter (which looks like arse imo both gameplay and art wise), or to that the standard is still improving (which is actually correct, or at least what now is doable because of way more powerful crossdev.)

[Ai]

^ some of the art style is a bit silly (anime faces, anime-ish bodies) but it's unquestionably better animated and presented then -- well, the only CPC game I can think of that it doesn't clearly beat in its area (mode 1) is Stryker and the Crypts of Trogan.  The graphics are more detailed and more readable, the animation two to three times as smooth, and the performance is also excellent (important in a system that doesn't have any hardware sprites). I guess what I'm saying is there are more stylish CPC games (Fairlight is a nice example), but very few CPC games that are better -presented-.

If there are actually any other better-presented mode1 games I would definitely like to know them.

[Carnivac]

I downloaded the game a few years ago (to run on emulator as it doesn't run on my real CPC 464 likely needing more memory for the intro and animations and such) and didn't enjoy it much.  Found the gameplay very slow and clunky and didn't care for the graphics at all, finding the style extremely bland and it's use of colour pretty poor.  Too much of the flat Spectrum-style of shading and not enough of the kind that made better use of the 4 color mode for a bit of extra shape and volume.   Scaling up the character portraits (which I didn't think were well drawn anyways so scaling them just made them look even worse) doesn't make much sense either (what's the point of using that resolution if you're going to just make bits of it super blocky?).    Character designs weren't very appealing either.  Come to think of it the little animated bits of the intro were my fave bits of the game.   

Regarding beat 'em ups on 8-bit hardware it kinda amused me that a lot of CPC owners were waiting for Street Fighter II which never actually made it onto the CPC (there was all that waiting in hope for it by Amstrad Action back in the day, even being on the cover of one issue) and that despite loving my CPC and loving SFII I didn't want the game on that computer.  I bought a SNES to play that game which was more suited to it and I loved playing all sorts of other games on my CPC that weren't available on a SNES (I actually prefer western and european graphic styles over japanese for the most part... which is a bit strange cos one of my all time fave 8-bit games is SwitchBlade which was one of the early western games to have a clearly anime-influenced hero but still quite a dark atmosphere rather than all cutesy and over the top dynamic despite the SD-like player sprite) and even the Amiga version of SFII failed to play anywhere near as good as the SNES version (and despite some decent attempts there never really was an Amiga beat 'em up that matched SFII's gameplay and graphics). 

Street Fighter Alpha on Gameboy Color however played very well despite having extremely downscaled graphics (and thankfully didn't resort to the cutesy SD style of the Neo Geo Pocket fighting games).  It had the feel of Street Fighter even if the graphics looked quite scrappy and primitive though easily recognisable as to what was what and who was who.

I really gotta cut down on my use of brackets...

Also this has gone very off the original topic... Sorry.

[Helm]

http://youtu.be/Xgm2biUQO2Y this game looks pretty awesome, really!

[Ai]

@ Helm: That's because it -is-. Both in graphics and gameplay (and sound; less so music). I think they made a bit of a mistake on the CPC+ version of Stryker (overuse of raster palette effects in a way that occasionally feels dislocated from the rest of the game's aesthetics), so personally I prefer the plain CPC version.
Not quite as awesome as Switchblade (done by the same guys), but very close.

@Carnivac: that's a pretty fair review, although I felt that the flat-color style was somewhat needed to make the characters read consistently against the BG. I'm pretty sure the scaling was just 'because we can'
SFII was somewhat constricted even in the SNES version, a CPC version would just be laughably limited. (there was a CPC version of SFI, but noone wants to play SFI SFA3 was the best done of the SNES ports IMO, it seems pretty solid.

I haven't played SFA for GBC, though I've seen some shots and it seems fairly well done.