Design from Process or Set Pieces?

[RatheMcGrath]

In the discussion of Yahtzee's review of Uncharted 3, a topic of discussion was raised that I thought was worth looking into a little more closely. Apparently, in their discussion of the game process, the designers of Uncharted 3 said that they first decided what the set pieces would be, and then created the game around those. Yahtzee took exception to this, and several of his readers did as well, saying that this was bad writing 101.

My response to that was that MOST writing figures out the big moments first, and the connecting material second, if any pre-writing was involved at all. It then occured to me that one could see there as being two basic approaches to Game Design, depending on the emphasis one takes. Do you design the set pieces first, or the general process? I think I even have two timeless examples: Mario and Zelda.

In pretty much every traditional Mario game, it is the process that is important. For the most part, Mario has his complete skillset from the beginning of every game. Sure, he may not have the stars (or whatever) to get into every level from the outset, but the meat of what he does, the platforming, is there from the get go. The game is a process of raising the difficulty curve, adding challenges without necessarily adding new dynamics. Power-ups tend to be optional ways to decrease the difficulty... you could theoretically get through most Mario games without even a single power-up. That said, it tends to lack "Oh HELL YEAH!" moments... sure, it may be difficult to, at full speed, hit a jump perfectly, weave between two floating Koopa Paratroopas and then hit the flag... but it isn't hugely more difficult than the thing that came before it, or the thing that came before that. Boss fights might change things up a little, but they tend to be little more than highly animated obstacles.

The Legend of Zelda, on the other hand, has worked differently ever since A Link to the Past. In Zelda, power-ups are necessary to advance, and the dungeons, (always the set pieces of the series) are exquisitely designed to use the new tools. A power-up doesn't make life easier... it makes life POSSIBLE, but in so doing it unlocks a whole new range of techniques that have to be learned an mastered. Link at the end of the game barely resembles the hero who starts. As a trade-off, though, the stuff between the dungeons can sometimes lack panache. There is usually a story present, but that isn't what pulls you along. You want to get to the new areas, find the new toys, unlock the new powers, and get to the "Hell Yeah" moments when you use your new equipment in awesome ways. Combat in the open world tends to be underwhelming, simply because the enemies do not adapt the way Link does... a problematic goblin becomes a hilarious (and short lived) bump in the road.

My point is that neither approach is BAD. They each have weaknesses, and biases that have to be overcome, but they CAN be overcome.

This isn't a defense of Uncharted 3... haven't played it yet to see if it even needs defending. But my point is that just because the set pieces were designed first doesn't make the game bad inherently. It is very possible to take a bunch of "This would be AWESOME" moments and create something breathtaking as a whole... just as it is also possible to take a game about world experience and process and give it some "AWESOME!" moments (See: Skyrim.)

 

[tharglet]

Well, the big issue with coming up with set pieces first, especially if you make them varied, then it's going to make it hard to shoehorn in a plot that actually makes sense.
When writing, it's usually best to come up with the vague thread of what you're going to do along the way, roughly how things are going to pan out, and then flesh out the skeleton, piece by piece (but not necessarily in story order).

The things you seem to be talking about don't seem to relate to what Yahtzee was on about. Yes, in Zelda there's large progress markers, but I'm pretty sure the game writers planned out in advance how often you get these markers and roughly how they would occur. From what I watched in the Uncharted vid, it sounds like the game writers were pretty much "umm, yeah.. so we want stuff to blow up and err.. yeah... big fight that goes like this... ok now let's write a plot", which has led to some rather silly bits in between the set pieces that don't necessarily make much sense.

 

[Kezboard]

I read an interview with Rhianna Pratchett once, where she explained how video games get their plot: Three to six months before the game ships, they hire a writer who is given a mess of disjointed levels and expect him to pull a plot out of their nether regions that joins them together with minimal changes.

So yeah, I see how Uncharted happened.

 

[burningdragoon]

So your point basically being 'it depends on the game' obviously is fine. There is a good way to design from set pieces, but having a sprawling adventure story is not the kind of thing you want to start from set pieces.

Mario games work because it's a fantasy world where a mode of transportation is sewer pipes. Super Mario 64 is good example of how to design by set pieces. You jump into different paintings that have different worlds connected to them. There's no need for each world to be connected. Any hub-based game that isn't very story-driven like that works.

Games like Zelda that have one world to explore mostly at the will of the player can also be designed from set pieces, but the fact that all the pieces are literally connected to each other will at least force the designer to think about how they all fit together. Not just 'okay, so first Link goes into the Water temple and then in the next scene he goes into the Fire temple, and then he rides on the back of a dragon while space aliens attack.'

 

[JMeganSnow]

My response to that was that MOST writing figures out the big moments first, and the connecting material second, if any pre-writing was involved at all.

Um, how many writers have you asked about this? Because all of the writers I know use different processes. Some will detail out the climactic showdown first, then write toward it. Some only figure out what the climax should be after they've been working on the story for months.

The *problem* with starting out with your big spectacle set-pieces first is that you make your story rigid. What happens when, later on, you find out that it makes NO SENSE WHATSOEVER to have a big running gun battle through a sinking submarine? If you're a good writer, you cut your beloved scene, maybe throw it in your desk somewhere so you can ponder it's awesomeness later, and move on with the story. If you don't, you'll wind up with glaring cracks in your story where it was obvious you tried to crowbar this particular scene in. It will torpedo your writing.

This is why building the set-pieces first for a video game is the Kiss of Death as far as story quality goes. If a writer sits down and types up a few scenes that later cannot be properly included in their novel, they've lost a few hours, maybe a day or two of work. If a video game designer does this, they may have a big pile of art assets, voiceovers, level designing, scripting etc. associated with that set piece. Are they going to throw it out? No. If they're a good designer, they may decide to yank this section and make it into DLC. IF they're a GREAT designer, they won't get themselves into this stupid mess in the first place.

So, in summary: coming up with set-pieces first is fine if you're still in the brainstorming stage and prepared to throw them out later. Doing this as your design process is NOT a good idea.

 

[ResonanceGames]

Either one can work, they're just different approaches. I suspect that they were oversimplifying when they said that they designed the set pieces and then worked the gameplay into them. AAA game design requires a lot of things to happen all at once.

The Uncharted games generally have excellent level design for their genre, so whatever they're doing is clearly working for them.

I personally prefer games that build on a strong gameplay core, then hone the levels into something that suits that gameplay perfectly. I think Thief 2 is the ultimate example of this, where they took mechanics of the first game and dropped them into vastly more intricate and well designed levels. Just pure perfection. But that's not to say other approaches don't work for other types of games.

 

[RatheMcGrath]

Um, how many writers have you asked about this? Because all of the writers I know use different processes. Some will detail out the climactic showdown first, then write toward it. Some only figure out what the climax should be after they've been working on the story for months.

Like I said, if Pre-writing is involved, which tends to be a big chunk of writers in the commercial world. Wandering writing until you find your point does not include pre-writing, and so not applicable to the point. And yes, that can look different depending on who is doing it, but if you started with an outline, you started by finding your "set pieces."

 

[RatheMcGrath]

Games like Zelda that have one world to explore mostly at the will of the player can also be designed from set pieces, but the fact that all the pieces are literally connected to each other will at least force the designer to think about how they all fit together. Not just 'okay, so first Link goes into the Water temple and then in the next scene he goes into the Fire temple, and then he rides on the back of a dragon while space aliens attack.'

I think maybe my point got missed a bit, but yes, they do eventually have to think about how the dungeons fit together, I am just saying they tend to do so later in the design process, and that doing so does not automatically make a game "worse."

The things you seem to be talking about don't seem to relate to what Yahtzee was on about. Yes, in Zelda there's large progress markers, but I'm pretty sure the game writers planned out in advance how often you get these markers and roughly how they would occur. From what I watched in the Uncharted vid, it sounds like the game writers were pretty much "umm, yeah.. so we want stuff to blow up and err.. yeah... big fight that goes like this... ok now let's write a plot", which has led to some rather silly bits in between the set pieces that don't necessarily make much sense.

You're the second to reply this way, with a quote that makes the designers sound like idiots. (The other was in the comments of the ZP vid.) I thought that was interesting... like you have to characterize the designers to sound like idiots, or it doesn't immediately follow that they ARE idiots.

Yahtzee only commented on the issue in a throw away fashion, it wasn't a big point, as I remember it, but it came in huge in the comments and that was what interested me. Why NOT design some really cool moments, and then figure out how to make a game around them? Like I said in the comment thread, JRR Tolkien designed his world first, and it was only decades afterwards that he concocted plot and characters to lead people through his world.

I will not contest that it makes the writing plot HARDER, but it does not necessarily follow that the plot would then have to be bad. Writing in meter makes poetry harder, as well. And like in my Zelda example, story can suffer for this... particularly if you don't really bother to actually connect the pieces, as apparently Uncharted did.