8
\$\begingroup\$

I've done some digital paintings and concept sketches in the past, and I was thinking about employing those skills in my game making efforts.

From an artists standpoint, what does he have to do differently/additionally to make his artwork "game ready" or "game friendly"?

Mostly I'm interested in using 2D in Unity, perhaps with an isometric view (so called 2.5D)

\$\endgroup\$
  • \$\begingroup\$ @iQue you must seriosly be underestimating me. Im not asking how to put a spritesheet into unity. Please read the question carefully. I can make, say, a sprite sheet by pasting each frame I drew into a file, but that may not be the most effective way to do that. I might not know about anchor points and such. You dont know what you dont know, so I am asking what I need to know. If you need further clarification, I will elaborate. \$\endgroup\$ – K.L. Jan 30 '14 at 14:12
  • \$\begingroup\$ You should discuss these topics with the programmers of your project, because the answers to your question depend on lots of project-specific technical details. \$\endgroup\$ – Philipp Jan 30 '14 at 14:15
  • \$\begingroup\$ @Philipp please see my further clarifications \$\endgroup\$ – K.L. Jan 30 '14 at 14:19
  • 1
    \$\begingroup\$ @K.L. Oki then I misunderstood ur question. Better after edit, but afraid I cant help you. \$\endgroup\$ – Green_qaue Jan 30 '14 at 14:23
  • \$\begingroup\$ I've removed the portions of the question that are more opinion based/polling. There's no correct answer to best practices, everyone has their own. I'm afraid that's more of a discussion based question. Even now, this question is likely too broad/open ended for the site. Can you be more specific about what differences you're interested in? \$\endgroup\$ – MichaelHouse Jan 30 '14 at 15:37
4
\$\begingroup\$

Game sprites move over all kinds of different background together with other sprites. For that reason it is important to work cleanly regarding transparent and non-transparent parts of an image.

When you create digital paintings or concept scetches, it is usually not that noticeable when you draw a bit over the outlines, don't fill the outlines completely, have slight transparency on layers which shouldn't, or have some almost unnoticeable smears outside of the object.

But when the sprite is moving over a background, such artefacts become clearly visible.

For that reason it is important that each pixel has an opacity of exactly 0,00% or 100,00%, unless you want an intentional(!) transparency effect.

With game sprites you also have the problem that you often have little control over which context they appear in. Sprites need to stand out on any background, no matter if it's dark or bright or if it's noisy or plain. Using thick, dark outlines often helps to make a sprite more background-agnostic, but the resulting comic-look might not fit any aesthetic direction.

When your graphic engine doesn't make lots of use of scaling, you should try to create all 2d assets in the same resolution you want them to have in the final game. For performance reasons most 2d engines don't use image scaling algorithms which are as advanced as those available to graphic editors, so any rescaling will blur your artwork and should thus be avoided, especially when some images are scaled more heavily than others.

It's hardly possible to give you any advise about your workflow, because it's highly dependent on the project and its technical details. But I made the experience that it pays off to set it up in a way that testing your graphics in the actual game while you make them is as painless as possible. Most non-obvious problems don't appear before you see the graphic in the game.

In the ideal case, the game engine should be able to load graphics directly from image files in a format your graphic editor can export with one click. In one project I added a hotkey to reload all assets from hard drive while the game was running. It proved to be a real time-safer.

\$\endgroup\$
  • \$\begingroup\$ Good points on transparency and sprite context! Thanks for your answer. \$\endgroup\$ – K.L. Jan 30 '14 at 14:53
1
\$\begingroup\$

I would say that when it comes to creation, there are advantages and disadvantages to each method of art creation. So let us cover these first.

Let's start with standard art. One benefit that standard art has, is that you can usually get away with making black and white pictures that still look good, using only a pencil. I believe that controlling the shade of objects is a whole lot easier than having to pick colors.

Even when you do create color images, it is still usually easier to find colors. I think this is because you generally have a fixed pallet of colors, so you don't have to mess with RGB trying to find the right color (more on that later). If you need to control saturation of a color, all you need to do is control how light you are when applying the color.

Another reason that standard art is easier, is that you have the most intuitive image editor ever! Wherever you put then pencil, is where it draws. It is much easier to control a pencil than a mouse.

Now that I have discussed standard art, I will cover the advantages to digital art and how you can best use your skills for game creation.

One benefit that you get with digital art, is that it is very forgiving. Erasing doesn't leave behind traces of what was once there. This means that you don't have to worry about making a line perfect on the first try, which can make you spend less time on practice draws, and make you more productive.

Another benefit to digital art, is that you have the benefit of layers. This allows you to easily duplicate a part of your image before editing it in case you mess it up. This can also be used to keep parts of an image separate.

But one of the biggest advantages is the fact that your image editor usually provides you with tools that do the work for you. For example, instead of spending a bunch of time looking for reference photos, you can simply click a button, and you have a lens flare, for example.

The final and biggest difference, is that in regular art, digital or not, you are drawing an entire image, containing shading, a subject, and in some cases, scenery. Basically a final image. Where as in game art, you are only drawing individual components. Such as a player (or in some cases, you are only drawing individual body parts), a grass tile, or a weapon.

Adapting

Based on what I have discussed, I will now discuss how you can adapt. And remember, use the advantages of digital art to your advatage.

First off is color. To make the job of color picking easier, you can create a pallet of colors to use in your image. If you need variation, you can simply adjust the hue and saturation. For this, I recommend switching to HSV.

I also discussed the issue of a mouse being harder to control than a mouse. To alleviate this issue, you may consider using a graphics tablet. A very popular company in the realm of tablets is Wacom. The only problem with Wacom is that they are a bit expensive, which is why I haven't bought my tablet yet. The advantage with graphics tablets, is that they usually have pressure detection. This makes the experience a lot closer to using a pencil. This can be useful for shading.

Philipp's answer (which is very helpful) discussed scale. "You should try to create all 2D assets in the same resolution you want them to have in the final game". And though you want your assets to be the same resolution when exported, doesn't mean you can't create you assets in a different resolution. One trick I learned (I don't remember where I read this) is to start with a large image that can be scaled down to the proper size (I recommend doubling the size a few times. This is also why I like to use image sizes that are powers of two, as 8x8, 16x16, 32x32, 64x64, etc.). Draw the sprite at the large size being mindful that you are going to scale the image down, and then when you are done, scale it down to the proper size. Once you do this, you will usually need to repair edges, and fix small details. Make any additional edits, and save the image at the corrected size. Remember, you may want to keep the large version of the image too.

The final piece of advice I have for switching to game art is making sure you use your editor to your advantage. If you need to make dirt, all your really need to do, is fill a brown area, and then apply noise to the image. And BAM! You have dirt. This is just one example of using your editor to your advantage.

The truth is, that it is hard to understand what you meant by game-ready, but I have tried my best to help answer the question.

\$\endgroup\$
0
\$\begingroup\$
  1. With general art, you can use any size, any style and any colors. In games it's important to plan so the sizes, colors and styles of sprites make sense and are compatible when they are together on any of the games backgrounds.
  2. Often times game-play mechanics and needs dictates small sprites. It is often good to keep the art clean (instead of trying to scale down a big detailed sprite) so it will look not look too distracting.
  3. Animation may require planning. Too much detail means you need to redraw that detail for every frame of animation. Keep it clean and simple, at least until 99% of the work is done.
\$\endgroup\$
-1
\$\begingroup\$

Philipp's answer is great but he missed one key part in the design of sprites and 2d game art and that is repetition. Sprites, tiles, everything that the user sees should have the ability to be reused somewhere else (even if they are not). For example, the bushes in Mario NES are the same sprite as the clouds, just with a different color code applied to them. Some of the enemies in the NES Zelda are the same sprites as other just flipped 180. Also, the trains and boats in the Pokemon games are the same sprites but with the ability to be elongated by repeating or omitting the use of the internal sprites. Thus a speed boat and a luxury liner are the same sprite base. I hope this makes sense.

\$\endgroup\$
  • 2
    \$\begingroup\$ I think this is only relevant to 8-bit and perhaps 16-bit (from the early 90's) machines that rely on cartridges that have very limited storage. Nobody needs to worry about reusing most assets nowadays over 20 years later because the storage capacity available today is about a million times greater. \$\endgroup\$ – AturSams Jan 30 '14 at 19:12

Not the answer you're looking for? Browse other questions tagged or ask your own question.