Adventure Gamers - Forums

... What do you really start with, and what’s where on the timeline?

With an intentionally naive approach, I combined looks from the outside and inside at the production of a few games, and came up with some odd observations.

The obvious choice seemed to be:

1. Write a story, design main characters, make it a novella if that’s what it takes.
2. Convert it to a playable script on an engine of choice - simplistic, basic text and placeholder images only, but properly representing the game logic and story flow from start to end. Test it on a screening audience, check reactions, make adjustments until it’s “perfect”.
3. Write proper dialogue (playable script could’ve had just “hello” “give me quest” “ok” dialogue placeholders), thus fleshing out characters’ interactions.
3. Give it graphics, music, movies, record dialogue, translate to other languages. Improve and polish and refine until it’s goddamn perfect.

Maybe I’m missing something obvious or are astonishingly inexperienced, but… is there something wrong with an approach like that?

See, I’ve seen quite a few games - commercial or indie - take a completely different route. With a whole set of backgrounds and characters and sometimes even recorded dialogue, they have as little as 1/3rd of the story playable, and nobody finds it wrong.

Isn’t this taking a huge, unnecessary risk?

The engine of choice often influences the way the game is perceived. A sequence that was planned to be played across several rooms (for which graphics have by now been finalized and paid for) might now prove to drag forever. A puzzle ends up turning backwards, because the player can pick up the key item earlier than the puzzle is shown - it didn’t look like it in the screenplay, but the player’s freedom of movement made it so. The initial idea had a character come up with something on their own, but after being converted to a game puzzle it turns out either pointless or ridiculous.

Now, if the game is made playable (though ugly and placeholder-based) enough to do a closed test on live audience, they would pick out all kinds of uneven story flow, difficulty spikes or cakewalks, out-of-character actions or unintuitive puzzle designs.

Is there some fatal flaw in that line of thinking, stemming from my lack of 30 years of professional gamedev experience?

Ailinon - 04 July 2013 12:05 PM

The engine of choice often influences the way the game is perceived. A sequence that was planned to be played across several rooms (for which graphics have by now been finalized and paid for) might now prove to drag forever. A puzzle ends up turning backwards, because the player can pick up the key item earlier than the puzzle is shown - it didn’t look like it in the screenplay, but the player’s freedom of movement made it so. The initial idea had a character come up with something on their own, but after being converted to a game puzzle it turns out either pointless or ridiculous.

I’m not sure I understand what you are trying to say but I have definitely noticed the problem of being able to solve a puzzle before you find out why you would want to. This is very common in the one-click method where every item has only one action (look, or use). When there are so few actions, you often know what you have to do from the layout before finding it out the way the game wants you to.

For example in Jack Keane 2 you can prepare and shoot a cannon, knocking over a statue to provide a path to a cave. But this only makes sense if you blow up the path to the cave that is already there, which I happened to do after I knocked over the statue. And then I had characters telling me I shouldn’t have blown up the path in the first place! 

And that’s why there are now so many cutscenes, because the player’s freedom ruins the puzzles. It makes you appreciate how much work went into some of the more non-linear AGs.

As for the order of making an AG, I don’t know. I do think the puzzles are just as important and part of the story, and that the game isn’t a story plus added puzzles. If you look at the Grim Fandango puzzle book it actually treats the story as a series of puzzles, which I think is a good method for 3rd person games.

My main point was that I see many games take (or at least appear to take) an “art and even dialogue first, playable prototype later” - which can (and often does) result in the game just trying to make the best use of its already made assets, instead of only now deciding what assets are going to be needed in the first place. Of course, a story and puzzles are definitely designed beforehand - but a FULLY PLAYABLE prototype is seemingly unheard of.

I’m trying to think how you can have a playable prototype before the art, and it’s difficult. I am accustomed to seeing those art designs you see when a game is announced, so maybe you’re right that they are done first. Or maybe we are always shown those first because they are more attractive and it’s harder to show a prototype to the public than a cohesive story and artwork.

To answer your question, yes it is a risk. Which I guess explains why many games fail.

I’ve worked on loads of artwork that probably represents a hundred or so hours of work, for games that will never be released. It’s not all wasted time (because you’re learning and getting better at your craft) but I’ve learned my lesson. No more drawing detailed backgrounds before the design is done, and the game is running.

This is what I recommend:

- Plan, plan, plan
- Once you’re convinced that your plan isn’t too ambitious…
- Create a rough prototype. Get the game working from start to finish with dummy dialogue, rough artwork & sound
- Play it, get others to play it, iterate and fine-tune the progression/puzzles/story
- Implement final artwork & sound

Working like this can be difficult if you’re a perfectionist/artist type, but you’ve just got to get over it. Ultimately it limits the risk of redoing massive chunks of game and saves time.

orient - 05 July 2013 01:53 AM

I’ve worked on loads of artwork that probably represents a hundred or so hours of work, for games that will never be released. It’s not all wasted time (because you’re learning and getting better at your craft) but I’ve learned my lesson. No more drawing detailed backgrounds before the design is done, and the game is running.

This is what I recommend:

- Plan, plan, plan
- Once you’re convinced that your plan isn’t too ambitious…
- Create a rough prototype. Get the game working from start to finish with dummy dialogue, rough artwork & sound
- Play it, get others to play it, iterate and fine-tune the progression/puzzles/story
- Implement final artwork & sound

Working like this can be difficult if you’re a perfectionist/artist type, but you’ve just got to get over it. Ultimately it limits the risk of redoing massive chunks of game and saves time.

That’s a great plan.  I couldn’t agree more.  The only thing I’d add is just the idea that artwork can be consistently worked on and added as you go.  But maybe that was implied.

Yeah, definitely. The same thing with dialogue—sometimes you’ll write something that’s perfect at the very beginning and you’ll never have to change it, but it’s always good to be prepared to change things.