GAME JOBS
Latest Jobs
spacer View All     Post a Job     RSS spacer
 
June 18, 2013
 
Red Storm Entertainment, a Ubisoft Studio
Online / Networking Engineer (Back End)
 
Airtight Games
Senior Graphics Programmer
 
Red Storm Entertainment, a Ubisoft Studio
Online / Networking Engineer (Front End)
 
Gameloft - New Orleans
Senior Online Developer
 
Trendy Entertainment
QA Lead
 
Wharton County Junior College
Instructor of Computer Science-Gaming
spacer
Blogs

  On Game Structure
by Adam Saltsman on 02/24/10 12:09:00 pm   Expert Blogs   Featured Blogs
8 comments Share on Twitter Share on Facebook RSS
 
 
The following blog was, unless otherwise noted, independently written by a member of Gamasutra's game development community. The thoughts and opinions expressed here are not necessarily those of Gamasutra or its parent company.

Want to write your own blog post on Gamasutra? It's easy! Click here to get started. Your post could be featured on Gamasutra's home page, right alongside our award-winning articles and news stories.
 

STRUCTURE

Finally, after months/years I found time to play through Braid (in a single sitting), and the next day I played the first 6 hours or so of inFAMOUS, the latter a gift to myself for finishing a couple big pieces of art.  Anyways, this morning it got me thinking about the basic structure of these games, how they remind me of some classics and how they differ too.  I won't be comparing Braid to or against inFAMOUS, they're just both on my mind and both mirror some of my old favorites.

 

BRAID

Basic Structure: Braid is set up as a series of nested "hub worlds", or levels that serve primarily as gateways to other levels.  Within each actual game level there are multiple objectives as well.  This is a great way to organize a game, the player stays with the character as they journey around and you don't have to program annoying menu systems.  It also provides some non-linear or player-directed advancement; you don't have to play through each level in order.  Some of Braid's hub worlds also incorporate optional (and for me, incomprehensible) story elements in a nice way.

Classic Comparison: Aptly, the classic hub world example is Mario 64, where you run through an ever expanding castle, granting you access to rooms with paintings in them, which in turn grant you access to multi-objective playgrounds.  On the surface Braid seems very similar!

My Take: Braid doesn't really live up to Mario 64's organization.  It's hard to tell why the developer made certain choices with such an obvious and constantly referenced example at hand.  Braid's hub worlds are not smoothly connected like Mario's; in Mario 64, you open a door in the castle (analogous to Tim's house), and pass into another room in the castle.  In Braid, you enter an abstract cloud room full of doors that lead to other levels.

By itself that is in no way a damning choice, but it's indicative of the rest of the layout.  Like Mario 64, Braid offers up levels with multiple objectives (puzzle pieces instead of stars), but most levels can be completed by simply walking from one side to the other, skipping all of the objectives.  The levels themselves are essentially linear hubworlds, providing you access to a few puzzles at a time.  This extra layer of hub-world-ism seems like a good idea at first; if you get yourself into a place that you simply can't handle, you can just leave again.

However, in order to actually complete the game and read the epilogue, you must complete every single puzzle.  The advantages of using a hub world layout are that the player chooses which challenge to dedicate time and thought to solving.  Braid, which uses a hub world that leads to hub worlds which lead to even more hubs, requires you to solve every single puzzle in the entire game before you can play the final challenge.  Oddly, the final challenge requires none of the tricks required to actually solve all the game puzzles, and thusly plays out as an anticlimax.

This is symptomatic of the game's design, though; where Mario 64 gives you a wide variety of abilities you can use in interesting ways on every level, Braid's worlds are segmented by specific mechanics, and most of the puzzles only have a single solution.  It strikes me as a very odd decision to embrace a non-linear hub structure when every other aspect of the game requires specific, precise, and full linear completion in order to advance.  It seems like an afterthought, rather than a conscious design decision.

I would love to play through a version of Braid that let you toggle the time controls and use the ring on EVERY level.  If these things are too overwhelming for new players, unlock them after some quota of puzzle completion.  Why allow me to replay old levels after learning new skills and mechanics if I can't apply them to those first spaces?  Finally, there is no reason why the player shouldn't be able to play through the final levels and read the epilogue with say 50% of the puzzle pieces collected.

 

INFAMOUS

Basic Structure: inFAMOUS is a big open-world game, in the style of Grand Theft Auto or Assassin's Creed.  You are provided with a constantly changing array of waypoints, which either advance the story or help you reclaim territory or get new powers.  inFAMOUS rather elegantly handles the whole problem of invisible walls and expanding the player's world by creating a citywide power outage at the start of the game.  Since the player requires electricity for ammunition and health, these blackout areas are very dangerous to explore.  Restoring power via excursions into the sewers seems to be the main progress system (so far).

Classic Comparison: This game reminds me more of the original Legend of Zelda than anything else I've played in a long time, except maybe (and obviously) Link's Awakening.  In the original Legend of Zelda, you start out with a huge world to explore, punctuated by the occasional dungeon, which gives you an item to help you continue your explorations into more dangerous areas.

My Take: inFAMOUS outdoes Zelda in the micro, but fails in the macro.  By "micro" I mean the moment-to-moment challenges in inFAMOUS have a lot of player-determined elements. Rarely are you forced to engage an enemy in an awkward or uncomfortable way.  Choice is omnipresent and meaningful through most of the combat and world navigation.  GTA4 and AC2 both kind of exist in this same space, but it's a really wonderful space to occupy.

Where inFAMOUS does not succeed nearly as well is in the "macro".  Just like Zelda you have a big open space, but you must dive down into dungeons (i.e. sewers) in order to increase your abilities and gain access to new areas.  That sounds like a positive comparison, but here the contrast becomes evident.  In Zelda dungeons could largely be completed in any order, and each dungeon had new puzzles, new enemies, and was itself a nonlinear space for the player to explore.  Compare to inFAMOUS's sewers, which must be completed in a specific order, are all the same color, do not involve problem solving, and have extremely simple, linear layouts.

Sadly, the linearity and similarity of each sewer stands out so much because the overworld is so expansive and so open.  Perhaps even more than Zelda, inFAMOUS needs to allow the player to restore power to the city in whatever order they want, simply because that is the grammar of the game.  Reclaiming territory, engaging enemies, navigating the city, and assaulting strongholds can all be accomplished with a really exciting degree of freedom.

Repetitive, monotonous sewer dungeons mirror the game's only other major flaw, which is the occasional and oddly lengthy battle sequence.  Two or three times now the game has asked me to partake in a fight that simply goes on for far too long.  In a game this open-ended, it feels odd to occasionally be corralled into such a tedious space.

 

I AM ERROR

I'm not feeling bold enough today to claim that either of these games are categorically worse than these classic games.  Legend of Zelda has buckets of flaws itself!  But these specific organizational issues caught my interest because the games are so different in origin and execution; big teams and little teams, big budgets and little budgets.  But in both cases the basic structural choices work against the designs by their own internal rules despite having successful and high-profile predecessors with phenomenal organization.

"If it ain't broke, don't fix it" is a phrase that has always bothered me.  I once drew a comic which featured two cavemen; one had carved a square wheel, another a circular wheel.  The caveman with the square wheel was deriding the circular craftsman for literally "reinventing the wheel."  I believe there is always room for improvement, but it's hard to do so if you don't have a solid understanding of what wheel shapes have been tried, and why they don't work well.

 
 
Comments

Wylie Garvin
profile image
Interesting comparisons. I would quibble a bit about the dungeons in the original Zelda; several of them require an item (e.g. the stepladder) that you get from an earlier dungeon, in order to progress. Its certainly true that you can start the dungeons in whatever order you want, but you might not get very far in them.

Shay Pierce
profile image
Your criticism of Braid's hub structure kinda seems to miss the point. The puzzles in Braid are super mind-bending; sometimes they can actually seem impossible, and the player can become exasperated. The solution then is simple: the player just moves on to some other puzzle. Even if they then get stuck at the next one and choose to go back to the earlier one they were stuck at, it's quite possible that they'll be able to solve the first one now that they've played with the mechanic more; at the very least they've taken a break and can tackle the first puzzle fresh this time. I think it's a great option to have in the game, especially one with puzzles that players can easily burn out on.



On the other hand, Portal was based entirely around mind-bending puzzles just like Braid was; but it was completely linear; and of course like Braid, it was a fantastic game. I don't think that Braid would have been better for NOT having its nested hub worlds; but then, I don't think that Portal would have necessarily been better for somehow having the option of walking away from a puzzle and coming back to it later, either. Hard to say why, but it just seems to fit in Braid but wouldn't fit in Portal.



The suggestion of letting the player take the different time mechanics into different worlds baffles me a bit... sure you COULD, but why WOULD you?

Adam Saltsman
profile image
Hey Shay, those are totally valid points I think. Braid's difficult puzzles definitely require some kind of "skip this one" type of functionality, which hub worlds totally provide. The thing that surprised me is that it doesn't appear to take advantage of all those other wonderful things that a hub world provides. If the levels were designed such that taking different mechanics into older levels (which is how many of the levels are actually designed - the multiple Pits, Hunts, etc) did yield interesting new results, I think it would be a stronger game.

Dean Longmore
profile image
I quiet enjoyed that each new world gave me a new set of mechanics to play with. It kept the game fresh and challenging, without becoming overly complex.

To allow the player to use all the available abilities in each level would be a nightmare to both design and play, half the beauty of Braid is to look at a puzzle and think 'how can I achieve this with the current power'.



EDIT: In reference to Braid, obviously.

Adam Bishop
profile image
I too found the levels/puzzles in Braid frustratingly linear at times. It often felt like "guess what the developer was thinking the day he programmed this". But I disagree that it would have been better to allow the player to go through the ending sequences with only some of the puzzles completed. The game is largely about Tim sorting through his mental anguish and dealing with problems from his past. These issues are portrayed metaphorically through the puzzles. It makes perfect sense for Tim to not be able to deal with his final, over-arching problem until he deals with the smaller ones first.

Daniel Felice
profile image
I'd say there is no reason why the player SHOULD be able to see the ending of Braid without collecting all the puzzle pieces. Completing the puzzles is the objective of the game.



Note how the achievements/trophies distinguish between traversing a level and solving it.

Germain Couët
profile image
I agree with shay on the issue of letting the player go on without solving a puzzle. I think the strength of Braid was the mystery of knowing that you've seen every puzzle but can't make sense of it all because you don't have every puzzle pieces. It makes the ending all the more mysterious and compels you to keep working at it until you gain access to the last/first world. The structure might be simple and predictable, but adding more complexity might have introduced a plethora of other problems.



Bottom line, I think there's room for improvement, but why fix something that's not broken?



On another note, another great article, thanks adam! I don't agree with everything said, but it's great food for thought.

David Fried
profile image
Braid for me, is perfect as it is. Each set of puzzles is teaching you something about time, or relationships, etc. You need to finish (many of) them in a specific way in order to grok the meaning of the lesson. The end level is meaningless without having learned all those lessons, and quite frankly, the end level itself was such a stunning and perfect end to what I had experienced that had I been able to go to it before finishing all the other puzzles, I would have felt cheated.



In other words, that final level is the game designer's equivalent of an epic cinematic cut scene that wraps up the story...



Pure genius.


none
 
Comment:
 




 
UBM Tech