next »

[sandcrab]

Some games already do this already, or at least the players can!  As the player starts out, he's overwhelmed with options and different weapons or abilities, and as soon as you start to find one that works, you begin to ignore all those other ones, and become increasingly focused on specific abilities.  In a game where you level up, you might find that you have so much invested in some areas that it's not even worth it to power up those areas that you ignored because 1. it'll take forever and 2. you're doing fine without magic or swords by just punching guys to death in two hits.  Or some games "regress" by giving you something so awesome near the end that you don't ever use anything else because it's just so stupidly powerful.

[Gimym TILBERT]

Organic and logic character regression

You start as a character with a lot of potential energy, but as you progress in the game, this energy is being divide and distribute equally between all the skill you get (health too), you got increasing gameplay options but decreasing power. Would be neat with a megaman like collecting progression.

[Ninja Disguise]

I feel like linear regression is really gimmicky. A neat idea once (maybe not even that long), but the novelty of the concept wears thin quickly.

I think the best way to handle it is the strategic balance of progression and regression - empowering you in certain moments and taking away some of that power in others. This is present in survival horror to some degree already, as conserving resources keeps your power level in check. 15 shotgun shells make you a virtual god now, but when you only have 2, you're in a much different situation. The Siren series is an especially good example: only one or two characters have guns and only a handful more can use weapons of any kind. You switch between these characters often, so you never get too comfortable.

[st33d]

This thread gives me a great idea for a shoot em up:

You've got auto fire streaming backwards in to you that makes explosions and then enemies appear. Then then enemies suck up bullets which you have to dodge otherwise you get lives. A full rack of lives and then you're booted to the start screen.

You'd start at the credits / play again screen and get launched into a reverse dying boss that you're sucking bullets out of.

It'd be bloody hard to make playable levels for it without thinking laterally about the challenge, but it's certainly a game I haven't seen before.

Pity I haven't got the time to have a crack at it...

[iffi]

This thread gives me a great idea for a shoot em up:

As in a backwards shmup? I think that's the concept behind Retro/Grade, though it's more of a rhythm game disguised as a shmup.

[starsrift]

IMHO there's a lot of pitfalls with doing this... There are many good reasons why progression is the way to go.
Learning curve. I don't think this needs much explanation... A regressive game inverts it. Careful design could get around this, maybe.
Players like being rewarded. That's generally the motivation to play a game. Being punished(character ability loss) for playing is the opposite. You'd have to compensate with holyshit plot elements or something to keep a reward/playtime curve.
This is kind of the same as the learning curve - but regression cuts player options, and level design options. The game has to be the most complex, most ambitious, most open to possibilities to start with, and then at the end of it, you only have a couple different challenges to put at the player, restricted as they are with their abilities to overcome challenges in general.

So I'd say it's tricky, but not impossible. The first kind of regression game that comes to mind for me is a time-travel game, and regress by locking off times to travel to as the player explores them. This could be quite intricate though, because if you're going to have a meaningful time travel game, it'd be hard not to keep the plot content exponential. For instance, you could have a number of player abilities existing in a quantum state so that the player both has and doesn't have them, until their choice is confirmed in a level that they journey through and it shuts one of the abilities off.

[snowyowl]

You could design a game where although you get more powerful equipment and items later on, you still have to use the base versions from time to time. Perhaps to conserve resources?

Example: You start the game with Fireball 1. Later on, you learn Fireball 2, which deals 3 times the damage but uses 10 times the mana. When fighting a battle against enemies with high attack and low defence, you want to finish them quickly, so you use the spell with the highest damage-per-second/damage-per-turn (depending on what genre you're in) - in this case, Fireball 2. But you also face enemies with low attack and lots of health points, or perhaps large quantities of cannon fodder to weaken you before a boss. In these cases, you want to kill the enemy without running out of mana, so you spam Fireball 1.

There are variations. Perhaps Fireball 3 is even more powerful, but uses an item that can only be gotten from random drops, not bought at the shop - so you generally want to conserve your uses unless you have no other option. Maybe Dark Fireball negatively affects your karma meter, making enemies harder and preventing you from getting the best ending.

But generally, my idea is that gaining a new ability should not make you more powerful - it should make you more versatile.

(In case you're wondering, I was originally thinking of Minecraft and its diamond pickaxes. Sure, they're fast and they can dig through anything, but if you use them when you don't have to, the 1025 uses you get from each one will be gone in minutes, and good luck finding more diamonds. Stone pickaxes give you 65 uses each, but stone supplies are practically infinite, so they're what actually gets the most use.)

[Fallsburg]

IMHO there's a lot of pitfalls with doing this... There are many good reasons why progression is the way to go.
Learning curve. I don't think this needs much explanation... A regressive game inverts it. Careful design could get around this, maybe.
Players like being rewarded. That's generally the motivation to play a game. Being punished(character ability loss) for playing is the opposite. You'd have to compensate with holyshit plot elements or something to keep a reward/playtime curve.
This is kind of the same as the learning curve - but regression cuts player options, and level design options. The game has to be the most complex, most ambitious, most open to possibilities to start with, and then at the end of it, you only have a couple different challenges to put at the player, restricted as they are with their abilities to overcome challenges in general.

So I'd say it's tricky, but not impossible. The first kind of regression game that comes to mind for me is a time-travel game, and regress by locking off times to travel to as the player explores them. This could be quite intricate though, because if you're going to have a meaningful time travel game, it'd be hard not to keep the plot content exponential. For instance, you could have a number of player abilities existing in a quantum state so that the player both has and doesn't have them, until their choice is confirmed in a level that they journey through and it shuts one of the abilities off.

I think you have a lot of this backwards.  Why does this invert the learning curve?  If anything I think it heightens the learning curve.  If I start out powerful, then I have no incentive to learn, but as I lose abilities, things get harder, and I have to adapt and learn so as to be able to survive.  I guess a point could be made that you have to learn how to use ability X when you get it, but I really don't think this inverts (is this even the proper term? reverses seems more applicable) the learning curve.

I also feel that a lot of people on this thread aren't talking about character regression.  There are a lot of "You start out with X, but you lose that and get Y instead".  That isn't regression, it's mutation.  Regression would be "You start out with X and then you lose it. Deal with it." 

[starsrift]

Learning curve. I don't think this needs much explanation... A regressive game inverts it. Careful design could get around this, maybe.

I think you have a lot of this backwards.  Why does this invert the learning curve?

I was wrong. Okay, say you have a meaningful ability suite design - that is, the abilities unlock different puzzles and the like, rather than something like just a big gun that does lots of damage, and does less as you play - that's just increasing the difficulty (by means of not increasing enemy toughness).
So, to crib some abilities, lets say we start the player out with a jump, a double jump, a sword to kill enemies with, a bow to damage enemies and trigger far away buttons, a grappling hook that will attach to special attachment points in the levels, and a bomb that destroys cracked walls.
So in Level 1, we teach the player how to use all of these things. Huge starting curve, right there.
Level 2, we take away the player's bombs. So they can't destroy cracked walls now that impede their progress. That challenge element must now be taken out of Level 2, and all ensuing levels. Since it's a properly unique ability, there's no way to replace it.
Level 3, we take out the grappling hook. Okay, so we can't use that.
Next, the bow. So now we can't put switches over long pits or whatever.
Next, the double jump. Now we have to shorten "challenging" jump distances so that the player can make it. We can taunt the player about what they've lost and put in jumps they could have otherwise made, but we're still down to jumping and swinging the sword.
Next, we take away the sword. So now we have to make sure that the player can avoid all enemies, instead of requiring them to kill some.
That would be an inverted learning curve.

If we did it so that the player could always just get through challenges with a single jump ability, it just means we've given the player meaningless training wheels and are stripping them away as the game goes on, which isn't regressive, it's just a blunt difficulty curve that is much more apparent to the player.

[Fallsburg]

Well, I think maybe we're talking past each other. Or possibly we fundamentally disagree.

Your first example, start out with everything -- things go away with time.  It doesn't have an inverted learning curve.  It has a steep ramp-up, but then it levels out.  The learning curve, at least my understanding of it, would be Knowledge (or Mastery or Ability) vs. Time.  At no point in time is knowledge taken away (at least not by the game. life, booze, head-trauma, time, those can all take away knowlege, but the game can't).  I would argue that it isn't as mind-destroying as you seem to think it is.  The last 3 Metroids, Castlevania:SotN, others I can't think of right now, all start the player out with their full (or nearly full) set of abilities and then strip them away.  Sure the regression comes early and is there to give the player a taste of what they will later be able to do, but it is still a regression and the learning curve isn't life-shattering for the player being given all of the toys at the beginning.

As for your second point, I am not understanding what you mean.  Why isn't it regression? The definition from dictionary.com that I am going by is "reversion to [a] ... less advanced state or form ..."  This would seem to be exactly what your second example is, at least in my opinion.  And I don't quite get what you mean by "blunt difficulty curve" and how it is "more apparent to the player."  I think it is more seamless than most modern difficulty curves which tend to be "You stay the same, enemy gets stronger, game gets harder" or "You get stronger, enemy gets stronger, game stays the same". I think that "You get weaker, enemy stays the same, game gets harder" is a novel difficulty curve that isn't as immersion breaking as "This enemy looks like the previous but is red and takes twice as many hits" or "You are super powerful, but so are these enemies lurking around this surprisingly sleepy looking hamlet" (yeah, I'm looking at you Final Fantasy).

I guess at the end of the day, it really depends on how the game utilizing it would be designed.  The learning curve and difficulty curve of a game are very loosely coupled in my opinion, and low number of abilities doesn't necessarily correlate highly with either.

[SirNiko]

I had an idea just like this in the Pitch Your Game thread.

Metroid style platformer. The player starts with an inventory filled with upgrades (high jump, morph ball, long shot, etc.) After a short tutorial stage, player finds themselves at the gates of the castle. There they have two choices: proceed to the final stage and defeat the final boss, or explore one of the side areas of the game to acquire one of the seven doohickeys that contain the boss's power.

The catch is that each doohickey weakens the player, in the form of voluntarily giving up one of your upgrades. As a result, each stage becomes progressively more difficult than the last. I'm not sure whether all stages should be completable with no upgrades, whether a strict order must be followed to win, or somewhere between the two. Metroid Zero Mission did a good job designing a world that had sneaky alternate paths for low-item runs, similar design would be required.

Multiple endings abound based on how many doohickeys you collect before fighting the final boss. The more things you collect, the weaker the boss will be.

The ultimate ending obtained by all seven doohickeys reveals the boss to be a perfectly ordinary cat, which is more than a match for your level 1 commoner player at that point. An epic battle of trading normal attacks ensues.

I agree with Tromack, the funny thing is that a lot of games that have character progression have upside-down learning curves. All those abilities equate to nuking monsters effortlessly. In Metroid: Other M the final dash through the hall containing every monster in the game is a cake walk thanks to the Screw Attack. The moment you score Knights of the Round, every boss in Final Fantasy 7 gets even more trivial than before. Symphony of the Night is hardest when you fight the first boss, because you have shitty weapons and no potions. It's the last boss that's actually easy, because you just chug Full Restores and swing your sword that fills the screen.

I think think Character Regression is a mechanic that could serve as far more than a one-time concept, it could be adopted pretty regularly into games and work in many situations.

[Xion]

Next, we take away the sword. So now we have to make sure that the player can avoid all enemies, instead of requiring them to kill some.

this last point seems like the kind of benefit people are talking about to a regressive system. Rather than the game being harder because your sword doesn't do as much damage to the enemies, the game is harder because you have no sword. You have to develop a whole new strategy for every single enemy encounter, including the foes that used to be super easy level-1 type creatures.

Also, for the other examples earlier in your post, you're assuming some kind of strange linear progression through the game involving new areas and the gradual stripping of every single ability. Not saying you couldn't do that but regression as a mechanic has far broader applications than that - such as repeatedly traversing the same space with diminished abilities, or not taking away every single damn ability a character has, but just a few - which ones and when depends on the game, that's where the whole role of the game designer comes in and shit. Not just algorithmically turning the player character into a paraplegic.

You could also have both progression and regression in a game, perhaps simultaneously to balance one another out, or as a kind of parabola where the player reaches their peak power at a mid-game climax and then begins losing abilities from there on out. And if you want to give them back their powers eventually don't do it before the final boss and make it a cakewalk. Make them fight the final boss crippled as they come, and give them the power-fantasy shit as a reward, using it to blast their way out of the collapsing facility or something.

It just ain't like progression has to be a linear Get Power or a linear Lose Power. Make it a rollercoaster or something.

edit: also none of that would "invert" the learning curve since you'd still have to learn how to deal with each of the situations without the abilities you once relied on. Figuring out how to get through a cracked wall without bombs is still an ability that can be obtained, just not as an item in the game but as a process in the mind of the player. You have to trick a ram enemy into charging into it, or find some other mobile explosive that isn't constantly on your character's person, or or or - these things are part of the learning experience as much as how to use any tool given the player. And, as Tromack said, knowledge isn't something that you can actively 'lose' due to processes of the game. Maybe it can become obsolete, but getting a new, more powerful sword renders your old sword obsolete as well. I see no reason why not to try changing the priority of things which are made obsolete in games - information, items, abilities, etc.

[SirNiko]

It also has occurred to me, this happens in many puzzle and strategy games.

For example, in Lemmings, the "Fun" levels are frequently the same levels as from the "Mayhem" difficulty, except the player is stripped of many of their powers in Mayhem. You must replay a level that before was easy but now is very difficult because you are very limited.

In Populous, early maps gave the player many powers and the enemy very few. As the game progressed, the player gradually has to cope with situations all over the spectrum, such as an equally powerful foe, the loss of certain useful abilities, and very unequal battles where you and your foe have entirely different power sets. The final mission leaves you almost powerless against a superior foe, forcing you to win via wits.

And Xion's point is critical, it's important for games with development (progression or regression) to feature the same situations at different power levels. When you're constantly forging forward the progression becomes pointless, since you could have easily just made monsters stronger instead. The player should encounter the same foes and obstacles as before so they can see how their strength has altered how they interact with the environment. The easiest way is to include doses of backtracking. The catch is, for a progressive game the backtracking poses no threat and becomes dull. In a regressive game, the backtracking presents a new and interesting challenge (unless the player did it 'the hard way' the first time).

I also feel like a reference to players who do Final Fantasy runs without gaining levels or beat Zelda swordless is appropriate here, but I'll let somebody else do that.

[Gimym TILBERT]

Actually any old scoring game like pacman or space invader or tetris, etc... is regressive, the more you progress the less you have option, the more clutter you get.