Project Failure

[Beck]

Hi guys,

Some of you may be aware of Nysuatro's "Project Failure". He made another blog post about it last night from a level design perspective.

http://nysuatro.blog...te-failure.html

I'm going to be working with him on this project and it'd be nice to hear what you guys think about it and what you guys might like to see come from this.

In the blog post Robb talks about how a blank canvas can be very daunting. For example, pressing the "New Level" button in a world editor (or similar) will probably show on screen four windows (Perspective, Top, Left, Right), each of which will have an empty grid inside. This is fine for people who already have a level planned out; they would probably start blocking things out and that empty grid starts to fill up pretty quickly. But what if you aren't sure what you want to create or just want a bit of a challenge?

A "New Challenge" button will effectively fill the empty grid for you, placing geometry at random or modifying geometry you place at random. This random geometry could then be a starting point for creating a level. Sure, the layout of the geometry will probably be bad and not make sense within a game world. But the challenge comes from taking that geometry and trying to make it better. You might have blocks which you could turn into buildings, vehicles or hills. You might have dips in the terrain which could be made into rivers, streams, streets or even canyons.

People would be able to gain experience by experimenting with this geometry to see what works best. It could be a great learning tool for people who want to learn how to design a level. They could play with the geometry to figure out what works and what doesn't. They may make a tower too high so players cannot see who is shooting them. They may make a corner too tight in a racing game out of the geometry they've been given. They could get this feedback and then play about with the geometry some more to try and figure it out.

From experimenting and playing what they create in the "New Challenge" mode people will be able to learn about play testing and iterating. Skills which can be used when they come to create their own level. Challenges could also be shared between people (like you guys here!) to see who could create the best level out of a given set of geometry.

It'd be nice to hear what you guys think of this and if you'd be interested in such a button within your world editor of choice.

Edited December 4, 2012 by Beck

[2d-chris]

Don't get me wrong there's nothing wrong with failure, or maybe I'm not getting the purpose of this. But the last thing I try to do is fail on purpose just to learn from it. Isn't the key to fail fast, and then do it right before you start putting your company into a financial burden? Sure with personal projects it doesn't matter and you can learn from it, but when money is on the line you have to succeed fast.

[Nysuatro]

I am afraid I was possible not clear enough with the article.

This is part of a project of mine where I am looking for a different way of making games. What I am researching here is how to use failure to speed up the workflow, allow for experimentation, have more interaction with the user and create more motivation of using the tool.

With agile development you plan out your whole process and you already know most of the end result before working on it.

By creating failure as in a form of a challenge, I am putting more focus on the process then on the end result. I want to find out if what the relation is of fun development and the end product. So you end up spending less time planning the end result, but more working on the challenge of making the best out of what you have to start with. The mapping competition where you guys applied some rules as a restriction is a good example. It forced the designer to be more creative and gave some amazing results.

I hope this makes it a bit more clear

One of the reasons why I post this publicly is because I want more conversation about this. This whole project is constantly developing and needs to identify itself more. And I want to do this by hearing other peoples reactions and questions.

Edited December 4, 2012 by Nysuatro

[Steppenwolf]

I don't really get it tbh but then i don't find empty canvases daunting and i'm usualy quite good in implementing the ideas that i have in my head and make them work.

So you end up spending less time planning the end result, but more working on the challenge of making the best out of what you have to start with.

I don't know man but that sounds like the opposite of what i would recommend to anyone. It's a recipe to throw work down the drain. People need to learn to spent more time in a proper planning phase not the other way around.

For instance i'm currently working on an armored car for my work. Before i even open max for the first time i spent a whole day collecting refs, reading about the subject of my work, analyzing where best i split the mesh in pieces etc.

Your idea for a work flow sounds like your program creates a completely messed up random mesh and then i spent the next couple weeks trying to making it look somewhat like an armoured car that was designed by Homer Simpson.

If anything i think your procedural software should be aimed at creating success not failure. Think of a level design equivalent to dDo.

[Chimeray]

It does sound interesting as a training tool perhaps, but not really in production or anything. So I'd have to agree, workflow wise it doesn't sound too appealing.

If you compare it to Alchemy... People create cool stuff with it but often they don't go into that program with the mindset of: Today I will draw a deer but I don't know exactly how it'll look. No, they go into that program and suddenly end up with a castle. A castle that looks more interesting than they perhaps would've designed but nonetheless nothing that was needed for the project... (Which was a deer). If you get my analogy

So yeah, planning is important indeed. If you have a plan then an empty canvas isn't all that daunting. I'm not afraid of failure, it's part of the iteration process. It is very important however to get something in there as soon as you can and twist and shape it to work, but not in a random way.

My 2 cents but I'm interested in seeing where it'll go nonetheless!