• Hey Guest. Check out your NeoGAF Wrapped 2025 results here!

Chasm Kickstarter, Procedurely generated Metroidvania ARPG with demo [$150k funded]



Current and Planned Features

  • Explore expansive procedurally-generated Metroid-like dungeons in six diverse areas. Never play the same dungeon twice!
  • Enjoy smooth, responsive control, challenging retro gameplay and gorgeous authentic pixel art (320×180 native resolution)
  • Discover new special abilities and powers to reach previously inaccessible areas
  • Play how you want by equipping weapons, spellbooks, shields, and more to both hands
  • Uncover random loot drops and rare items
  • Battle massive bosses & mini-bosses
  • Avoid devious traps and platforming hazards
  • Includes Normal and Hardcore (Scored with Permadeath) Modes
  • Leaderboards for Hardcore playthroughs featuring quickest time, least damage taken, and more
  • Windows, Mac, & Linux versions with Gamepad support

Kickstarter Trailer
GDC Trailer
Kickstarter page
Steam Greenlight

Demo 1.05:

339088ddfe5498fbd489f74066b7d0e4_large.png


0aee405e3fce92ad85d62fec5f262287_large.png


d0be9615b847322b2c9c5f702aefbfb3_large.png


Stretch goals announced:

4tJZIUI.png


  • $160,000 - Extended Soundtrack - A digital double album: one of remixes, the other of demos, out-takes and unused material.
  • $175,000 - Achievements - Challenging Achievements to obtain in all versions of the game. These won't be awarding you for pressing jump!
  • $200,000 - Arena Mode - Battle waves of enemies and bosses in an arena for unique items after you complete the main game.
  • $225,000 - Crafting - Search for long-lost recipes of rare items, and the difficult to obtain ingredients to craft them.
  • $250,000 - Roguelike Mode - Challenge yourself with unidentified & cursed items, hunger, permadeath, & constantly shifting dungeons in this mode.
  • $275,000 - New Game+ - Play through the game again with your character, battle harder enemies, and find an array of new items.
  • $300,000 - Gender Swap - Play as a female soldier on the same perilous journey.
  • $325,000 - Language Translations - Professional translations and localization into Spanish, French, German, & Italian.
  • $350,000 - Extra Area - A seventh area added to the main game with its own unique challenges, treasure, and enemies.
  • $400,000 - Community Choice Mode - Backers vote for their favorite from a selection of game modes by the developers.
 
weird I saw a trailer for this a week or 2 ago, but the kickstarter just started today? Cant remember where I saw it. looks cool though.
 
Explore expansive procedurally-generated Metroid-like dungeons in six diverse areas. Never play the same dungeon twice!
i

what

Are they just throwing words together and hoping it sticks? This doesn't make any sense.
 
NOTE: You will need XNA Redistributable 4.0 installed if you do not have it installed
already! It can be easily downloaded from:

Using a dead language... :/
 
i

what

Are they just throwing words together and hoping it sticks? This doesn't make any sense.

makes sense to me, but usually with procedurally generated levels, you see a lot of repetitive "set pieces" or rooms, and get poor level design in general.
 
I'm not sure I understand the procedurally generated aspect though? These Metroidvania games typically are built around a certain amount of backtracking and item exploitation. If it's just a straight-forward 2D action-platformer, that's fine, but don't call it a Metroidvania then, right? (You wouldn't call a Mega Man game a "Metroidvania", right?) From the video it sounds like it's going after something like Zelda II, with some sort of set overworld with towns and villagers that then take you into randomly-generated dungeons. The dungeons themselves generate unique challenges, but from what I can tell they're still point-A-to-point-B-to-Boss quests. That's cool, but if the only point of exploration is lootdrops then I'm not sure it fits the definition.

Or are they saying that these worlds generate themselves for individual players, but once they're rendered for one player they're consistent and can be returned to since they also generate item-capped roadblocks? As in, the dungeons themselves still invite exploration and discovery and repeated visits because they adhere to the classic Metroid/Castlevania structure of interconnected passageways that continuously expand as you get better and more powerful with the game?

Looks fun though, whatever the terminology they use to describe it.
 
I'm not sure I understand the procedurally generated aspect though? These Metroidvania games typically are built around a certain amount of backtracking and item exploitation. If it's just a straight-forward 2D action-platformer, that's fine, but don't call it a Metroidvania then, right? (You wouldn't call a Mega Man game a "Metroidvania", right?) From the video it sounds like it's going after something like Zelda II, with some sort of set overworld with towns and villagers that then take you into randomly-generated dungeons. The dungeons themselves generate unique challenges, but from what I can tell they're still point-A-to-point-B-to-Boss quests. That's cool, but if the only point of exploration is lootdrops then I'm not sure it fits the definition.

Or are they saying that these worlds generate themselves for individual players, but once they're rendered for one player they're consistent and can be returned to since they also generate item-capped roadblocks? As in, the dungeons themselves still invite exploration and discovery and repeated visits because they adhere to the classic Metroid/Castlevania structure of interconnected passageways that continuously expand as you get better and more powerful with the game?

Looks fun though, whatever the terminology they use to describe it.

I have no idea how they are developing this game, but it is possible to procedurally generate a game that involves backtracking after you get an upgrade.

Potentially easy, in fact.

You're going along, generating the dungeon, and then insert tangent area that requires double jump to reach, and then continue the main dungeon, and eventually place the double jump item, continue generating, create a spot that requires breaking the wall to continue, then jump back to that first tangent you made and stick the wall-breaking item at the end of it.

etc.

Procedural and Metroid(vania) are antithetical to one another.

Nah. While generating, you just keep track of potential paths the player can take whether they have a specific item or not.
 
Nice. I thought I never would've wanted to pledge $25 on it, but I played the demo and it's pretty much awesome. Really thinking about it now. $25 is a bit much, but who knows, got the whole Kickstarter to decide.
 
What's up with the anti-procedurally generated attitudes? Isn't it nice being able to play the same game more than once without knowing exactly where everything's going to be?
 
Why do people keep calling these "Metroidvania" games? That term is a fusion of Metroid and Castlevania, and as such it's not its own genre, but literally just Castlevania with a Metroid layout. I mean it was a new type of Castlevania with Metroid elements so it makes sense to refer to those as a "Metroidvania," but I don't see how something like Guacamelee is a "Metroidvania" because it doesn't have Metroid+Castlevania elements, just Metroid elements.

I don't mean to act like it's a big deal, it really isn't especially compared to like, the rest of the galaxy, but I think it's a... very misused term.
 
What's up with the anti-procedurally generated attitudes? Isn't it nice being able to play the same game more than once without knowing exactly where everything's going to be?

I don't know about everybody else, but my concern is that because its procedural generated it will result in potentially boring dungeon layouts and "broken" dungeons that will be impossible to progress in. Procedurally generated stuff can be really great when done right, but more often then not it leads to a bad game.
 
Why do people keep calling these "Metroidvania" games? That term is a fusion of Metroid and Castlevania, and as such it's not its own genre, but literally just Castlevania with a Metroid layout. I mean it was a new type of Castlevania with Metroid elements so it makes sense to refer to those as a "Metroidvania," but I don't see how something like Guacamelee is a "Metroidvania" because it doesn't have Metroid+Castlevania elements, just Metroid elements.

I don't mean to act like it's a big deal, it really isn't especially compared to like, the rest of the galaxy, but I think it's a... very misused term.

Metroid elements - exploration, finding specific items that provide a new power to get past an obstacle in another part of the world, backtracking

Vania elements - primarily about melee instead of shooting, more RPG elements like leveling up and collecting currency, buying equipment

This game has the Vania with the melee fighting and RPG leveling, but you wouldn't just call it a Castlevania-type game because not all Castlevanias have the Metroid-style exploration/map/backtracking.
 
This looks kinda generic but I'm a sucker for this style of game and I love pixel graphics, so I hope it reaches its' goal in the hope of there being a Vita version a year later.
 
I don't know about everybody else, but my concern is that because its procedural generated it will result in potentially boring dungeon layouts and "broken" dungeons that will be impossible to progress in. Procedurally generated stuff can be really great when done right, but more often then not it leads to a bad game.

No, that problem is solved.

An easy way to deal with it is generate a dungeon and quickly simulate it internally, rejecting the layout if it's impossible. But that's open ended and could potentially take forever, it's a sort of function to be avoided in program design.

There are techniques to guarantee a valid dungeon every time, and not a boring one, an actual tree-style structure with paths that lead to upgrades that lead to other paths with upgrades etc. It's not difficult.
 
Metroid elements - exploration, finding specific items that provide a new power to get past an obstacle in another part of the world, backtracking

Vania elements - primarily about melee instead of shooting, more RPG elements like leveling up and collecting currency, buying equipment

This game has the Vania with the melee fighting and RPG leveling, but you wouldn't just call it a Castlevania-type game because not all Castlevanias have the Metroid-style exploration/map/backtracking.

I get that reasoning, I just think it's kind of weird that it invented a pseudo-genre, that it happened because SOTN took a Metroid layout and gave it RPG elements. I just think they should call it a "Metroid" type game, fundamentally speaking. You can add a bunch of new stuff to that formula but it's still taken from that, initially. Whatever, it's nothing that resembles a big deal, I was just curious.

Now I want a peanut butter and jelly sandwich for some reason.
 
A Valley Without Wind 2 is a procedurally generated ARPG with an overworld that's also a resource management strategy game. This looks like it has a better aesthetic and art design, the procedurally generated part of AVWW2 made the levels feel like they were missing the kind of, I don't know how to put it, personality(?) that you would find in a game like the metroidvania style of game. I'd want to know how the developer would be using procedurally generated dungeons that don't feel like they were slapped together roughly.
 
Graphics looks good but...
I dunno, the whole mine thing doesn't give me envy + I prefer those kind of games on handheld like 3DS or Vita.
If they could have done something with more environnements (from what I understand, it's about going inside a mine with randomly generated dungeons), then I could've been interested.
 
Graphics looks good but...
I dunno, the whole mine thing doesn't give me envy + I prefer those kind of games on handheld like 3DS or Vita.
If they could have done something with more environnements (from what I understand, it's about going inside a mine with randomly generated dungeons), then I could've been interested.

Well theres "six diverse areas" it says. It wouldn't be too un-video game like for there to be a twist about the lower levels of the mine looking radically different (Spelunky comes to mind). Hopefully they detail some of this stuff in updates.
 
No, that problem is solved.

An easy way to deal with it is generate a dungeon and quickly simulate it internally, rejecting the layout if it's impossible. But that's open ended and could potentially take forever, it's a sort of function to be avoided in program design.

There are techniques to guarantee a valid dungeon every time, and not a boring one, an actual tree-style structure with paths that lead to upgrades that lead to other paths with upgrades etc. It's not difficult.

You're absolutely right, but I have no idea what this team's plans are to avoid the problems that concern me. It makes me hesitant to fund this project without more information about the game itself.
 
Don't call it Metroidvania if that shit ain't planned out like it's for Ethan Hunt. The genre is defined by a masterwork weaving of gameplay elements and level design. Procedural generation just isn't going to do that.
 
One easy way to generate a Metroidvania:

- Generate a maze, including solution path

- Choose a point along the solution path and close it off with a wall that requires an upgrade

- Backtrack through the maze from that point and find a dead end

- Place upgrade item at dead end

- Repeat as necessary

- Go through the maze again looking for other dead ends, and close off some of them with walls that require upgrades, placing smaller rewards behind them

- Create rooms that could correspond to each tile in the maze, for example 10 varieties of room with exits at top and bottom, 10 varieties with exits left and right etc.

- Assign random rooms to each tile of maze accordingly

- Design special rooms for events or boss encounters and place them as appropriate

This might not sound incredibly fun, but a lot of that comes from knowing the mechanism behind it. In practice, it could result in something quite enjoyable, and the method can be refined a lot - this is just coming up with something quick.

One refinement you could do: make rooms of varying difficulty. Rooms encountered earlier are one of 10 easy-type rooms with simple platforming challenges, rooms encountered later are one of 10 hard-type rooms that require more finesse.
 
Procedurally generated content would be the antithesis of metroid, so that's pretty sketchy.

Funny though because it would be a lot like the hidden world glitches created in the original metroid by map overflow data or w.e.
 
Top Bottom