More reasons to add art early in development

[cmoleman]

Something I would add to the Game 3 Roadmap lecture:

There is another reason to start adding (some) art at a much earlier stage in development: and not just placeholders, but rough blocking first passes that don’t need to look finished but DO need to resemble the intent and feel of what is needed.

Here’s why: Art is not just wallpaper that looks nice and makes a game easier to sell; it is an integral part of the game experience. If you’re designing and building your whole game first and then stick art on top of it at the end, what you get is a disjointed mess that feels exactly like you just stuck some random art on it. If you want your game to be a cohesive complete experience, art (especially where it concerns movement and in-game feedback) should be part of your prototyping and design process.

Absolutely cannot agree with “art isn’t that important till later on”.

Art communicates information, so if you’re neglecting that part of your game, you’re basically missing out on a whole extra channel of direct communication to your players’ brains. Now that doesn’t mean you should be making final art at an early stage - agreed there - but you should start putting elements in there that communicate what they need to communicate and feel the way they need to feel. Because when you’re playtesting the feel and experience of a game, you’re also testing those parts. And if you’re just testing boxes running around grey boxes, you are not testing the complete experience.

And by the way, the same is true of sound also. Like that story where player feedback was that a weapon felt weak and should be re-balanced. Then they just changed the sound to make it boomier without touching the numbers and suddenly it was everyone’s favourite weapon… All the parts matter.

[MegaTrond]

As a non-artist myself, my take away from the video and your comment is that one should focus on… for lack of a better term… rudimentary art and sound, to start. Not just plain old boxes and such (fine for concepting and getting movement down and such maybe).

By this I mean that maybe you have the general look of what you want, but you might not have included any shadows or highlights and such, maybe the colors aren’t even quite right, but close enough?
Then have that be an iterative process as the game “matures” so to speak, and likewise as you say, with sound.

[cmoleman]

Yeah, that’s it. Don’t waste time on polished art in the early stages (unless it’s required for promotional stuff or a demo or something) but DO put something simple in that has all the important bits so you can test how that affects the feel of the game and how players react to and understand it.

For instance, for animation you might have rough blocking of movement that doesn’t look great but gets the main action poses and overall flow of your moves in so you can see if this fits with your movement mechanics and then adjust as needed (and maybe even do stuff like communicate NPC AI state through body language, like a relaxed or distracted idle pose vs an alert pose… useful in a stealth game). When doing 2D animation what I’ve sometimes done is use my rough thumbnail drawings as placeholder art. This way I get the feel of the movement in without having to bother with final art and can iterate quickly.

Or for environment art and level design you might have rough shapes and placeholder props for what the space is more or less supposed to be like, without spending a bunch of time modeling anything detailed. This way you can communicate the vibe and test how the layout affects navigation without too much work.

For sound maybe you can get some of the main action sounds (like weapons and reactions) in early but don’t worry too much about ambient sounds initially. And so on.

It all depends on what the game needs and also what skills you have, but as much as possible start iterating early on all the parts without committing too much time to any one detail.

[eric_v]

I find this comment/thread really valuable. I’m currently working on my first game, so I’m nowhere near the game 3 roadmap, but I was struggling to stay engaged while trying to move quickly and not get bogged down in details.

Every level I tried to create seemed to land with a “thud,” and I eventually realized it was because there was so little sensory feedback. I went back and spent some time adding more detail to the level art, adding music and basic SFX, and creating a basic character animation. Suddenly, everything started to sing, the world felt alive, and I could see what the game was going to become.

This kind of detail is especially important for my game, because the mechanic is dead simple and most of the player experience is about the feeling of exploring/navigating new spaces. But I think it’s a good point in general that sensory information is not secondary to building out the core mechanics.

When I was in film school, we used to submit rough cuts of our short films for feedback without doing even basic color correction or sound editing, or even adding temp music. The problem with that is you end up misleading yourself about what you actually have, because those elements will completely change the way a film is perceived. Eventually an editing teacher yelled at us about it and made clear she wouldn’t consider a rough cut without some basic color/audio work to be a valid cut at all for the purposes of the class. That’s been on my mind a lot as I stumble through making this first game.