Thanks!
Grappling hooks sure are fun, aren't they? 

Before the game gets too complex, I've decided to re-write some code.  My last physics system was really inefficient and didn't work perfectly. I've totally re-written the physics code, and it's a lot faster and perfect. There's no clipping through corners or any nonsense like that. The red blocks are the ones being checked for collisions.  SUPER EFFICIENT, YO! 


I've got a few more things I'd like to add.  Moving platforms is an important one, and I think that'll be fairly easy.  I also want to add different terrain properties.  So ice blocks could be slippery, and bouncy blocks could make objects....bounce more. This is all easy peasy, I just want to move on to gameplay stuff for now. I'm sick of physics coding for the time being 


I've also simplified a lot of movement code so I can have a little more control over the player.  Also runs better.
And I've completely re-made grappling hooks. My old ones were all springy and used some weird canned movement that I came up with on my own, using sin and cosine to try to figure out how much velocity to apply to the player. That was dumb . The new ones make the player move in a perfect circle, by limiting the player to a certain distance if going past the length, then gives the player a new x and y velocity based on where the new position is relative to the old position. If you're interested in putting something like this into your game, this article is what I used
http://gamedevelopment.tutsplus.com/tutorials/swinging-physics-for-player-movement-as-seen-in-spider-man-2-and-energy-hook--gamedev-8782
It doesn't wrap yet and the player can go through walls, but these will be easy fixes, I'll just use the same code from the old grappling hook system. 

Sorry I don't have any new content to show off. I've been doing a lot of behind-the-scenes work. I have a SUPER basic inventory system right now, and I'll be working on that from now on.

I love that majora's mask-y moon 

I LOVE TOPHATS!

The tech behind this game is ridiculous.  Seriously, you guys are geniuses! 

Thanks!  I'm really into climbing/parkour in games.  I really enjoy wall jumping and slopes in the new super mario bros games, and climbing walls in towerclimb/assassins creed.  I've never seen a game with the kind of ledge jumping I have, though. I like to think that's pretty original.  Ledges are an important part of the game.  You can practically get through a level just by jumping from ledge to ledge and climbing up walls, without touching the ground.
There will also be jump potions, to double jump, inspired by towerclimb.  Grappling hooks inspired by worms.  Chain-hook, like the Hookshot from zelda.  Portal-creating gun.  Placable ladders/ropes.  Teleport grenade.
LOOOOTS of movement options.
Not only does this give players a lot of options when encountering an obstacle/enemies, it also means I don't have to be very careful with the level generator.  It can fill a whole room with spikes, and the player could still get through it. (But I'll make sure that doesn't happen ) It all depends on what ingredients they have, and what they want to craft. Only thing that can't be crafted is the portal gun.  That you need to find in chests.
Basically, I can put traps and hazards anywhere without having to worry about them being impossible to avoid.
I am once again, making a response that is WAAAAY longer than neccessary, but I can't help myself 

sign me up when you need testers! :D

I might put out a free alpha at some point, like vagante did.  Not quite sure yet. But if I decide to do a private beta instead, I'll be sure to give you a key :D

Cool!  The game is getting more interesting! 
I think you should add an arrow to warm you where a meteor will fall.  They're really fast and hard to dodge

Wow this is old.  I hope you didn't give up on this!  The A.I is really interesting to me.  I've been thinking of ways to tackle pathfinding in my game, which has procedurally generated levels.  I'd really like to hear more about how it works.

This looks hilarious, lol.

Yeah, it seemed like such an obvious problem that I didn't bother going into it in my first response since I assumed you were probably aware of it.


It sounds like you have things under control. Best of luck to you. I do hope you are able to make the game eveything you want it to be. It seems like exactly the sort of thing I would like to play.

Thank you very much!    I hope you will like it when it's released.
By the way, I was checking out your game before I even made an account on here.  It's looking good! 

Which you will notice does not actually have a clear path from top to bottom.

Yes, I mentioned that in my response. 

When it's done, it'll make sure the exact walls of each room line up with the adjacent one, so you don't have rooms that you can't get into because the openings are off.

This will be fairly simple.  I'll just compare the two rooms and make sure at least one open block on the bottom of the top room has the same x coordinate as an open block on the top of the bottom room. Once again, you could say this is broken because of this situation:

top room (connects right and down)
XXXXXX
XXX000
XXX0XX
X0X00X
X0XX0X

bottom room(connects up and down)
X0XXXX
X0XXXX
X00XXX
XX0XXX
XX0XXX

put together
XXXXXX
XXX000
XXX0XX
X0X00X
X0XX0X
X0XXXX
X0XXXX
X00XXX
XX0XXX
XX0XXX
Here the openings connect, but the actual path doesn't connect.  I'll solve this by marking the tiles of the open path with a 'path' variable, or something.  So the path tiles have to align, not just any open tile.  Finding which tiles are part of the path is just a matter of flood filling, mentioned below.

There's also a few other issues I will address. Such as, if it's carved 30 times, and there's still no open path, just make the whole room open blocks, or make a simple L,T,or I shaped carve through the whole thing to force it to be opened on the required sides.

There's also this situation, for a room that needs to be open on the top and left:
XX0XXX
XX0XXX
XXXXXX
00000X
XXXXXX

As you can see, both sides are open, but they're not connected, and the player can't navigate through the room without using bombs to blow up blocks.  The solution to this will be flood filling.  I'll start at one of the openings on a side, and flood fill to see if all open-sides are connected together, if not, keep carving.  If the flood fill pours out of every required side, I'll mark those tiles with a 'path' variable, check if the paths line up with previous room's paths, and move on to the next room.

What I'm trying to show here, is that I'm totally aware of how much work this will take.  My examples were just to show the basic idea of it. I guess I should have more in-depth, complicated examples, considering I'm on a forum with a bunch of fellow programmers/designers. 

it might be worth taking a look at what spelunky does for level generation http://tinysubversions.com/spelunkyGen/

Yeah I looked at this while researching a while ago.  It actually helped me develop the method I'm using. 

Edit: here's some new music https://dl.dropbox.com/s/dmn7l8u8f6caik9/Rainy%20day.mp3?dl=0
This is when it rains in your town.  Tried to go for an animal crossing-y feel.

I think you might be underestimating the difficulties you will face in getting the procedurally generated rooms to A: not break the game, and B: be interesting and distinct.

I can already see some ways that your room code is going to break unless you refine it further, but I realize that you were just giving a basic overview, so maybe you already solved it.

More problematic is the question 'How are you going to make randomly generated rooms feel distinct and interesting?'. It basically boils down to this  - if you only use the method you just described to create your roooms, then all your rooms will feel pretty much the same.

For example, here is a room layout that your proposed method will practically never generate:

Code:

XXXXXXXXOX
XOOOXXXOOX
OOXOOXOOXX
XXXXOOOXXX
XXXXXXXXXX

To get that shape you would need to 'carve away' at least seven times in a very precise sequence. Sure it's theoretically possible for it to happen, but but it's vastly more likely that you will get a shape defined by a few large 'carved out' rectangles.
It's like rolling dice. If you roll two dice you can say that there are 11 different possible totals, but it is much more likely you will get a total of 7 rather than 12. It's the same for generating terrain procedurally - your code might be able to generate any 20 x 20 shaped block in theory, but in practice it will tend towards a particular type of layout.

I have thought about this in the past... making more interesting room designs. I simplified my carving example to only 4 phases, but the rooms will have a higher resolution, and the carvings will be smaller in tighter dungeons. I also updated my previous post to explain that the rooms aren't really....rooms.  They're more like 'chunks' from minecraft. A series of regions of blocks generated to produce a full level.
I could also add special rooms that are generated differently from the rest.  For example, I could make a room that generates in a snake-like pattern, like you mentioned above.  Combining these different room generation methods together in one dungeon could make some really interesting levels.  But it WILL be hard, no denying that.
This game is the biggest project I've ever worked on, and I want it to be as good as possible.  I don't care how many years it takes, I want my vision for this game to be a reality. I've been working on this game for almost a year, and I'm not stopping till it's awesomesauce! 

Also, could you explain what you mean by my code will break?

Thank you for the response.
This was the very first pass of level generation, and I'm actually working on it this very minute.  It literally just makes a grid, and each spot has a 1 in 2 chance of spawning a block.  This is only to prototype the idea of random generation, to test things like procedural tile decorations, performance, and size of the dungeons.
The final generation system will be much better, and make much more appealing results.  I didn't want to talk about how it'll work, because I worry people will copy my idea, but I'll just give a general rundown of how it will work.
Think of spelunky, that game uses premade rooms and stitches them together to make a full level.  My game will work on a similar idea, but the individual rooms will be randomly generated. First the game creates the room path, which is a path of rooms linked together.  Other, possibly non-connected rooms, will be created to fill in the empty spaces, so here's what that would look like:
X-rooms that are linked to make a path
0-rooms that may or may not be linked to adjacent rooms
E-dungeon entrance
D-dungeon exit

00E00000
00XXX000
0000X000
0000XXX0
000000X0
000DXXX0

So there will always be a path from the entrance to the exit, and other rooms might require you to bomb a wall to get to, or they might just be open. Certain rooms like 'locked rooms' will be surrounded by indestructible blocks with one 'door' block, that you need a key to open.

As for the actual room generation, it will be fairly simple.  Each room starts full of tiles, just a X by X block of tiles.  Then the game carves out the room until it links with all adjacent rooms.  It will carve it in rectangular sections with random dimensions, so it won't just be chaotic noise.  Here's the best example I can come up with it, with a 10x5 region.
So here is a room that has to link with a room above it, and a room to the right of it.

X-block
0-empty space

1ST phase
XXXXXXXXXX
XXXXXXXXXX
XXXXXXXXXX
XXXXXXXXXX
XXXXXXXXXX

2ND (start carving, random result will be 3,2, at the coordinate 1,2)
XXXXXXXXXX
000XXXXXXX
000XXXXXXX
XXXXXXXXXX
XXXXXXXXXX

3RD (random result will be 1,4, at the coordinate 6,1)
XXXXX0XXXX
000XX0XXXX
000XX0XXXX
XXXXX0XXXX
XXXXXXXXXX

4TH (random result will be 7,2, at the coordinate 4,3)
XXXXX0XXXX
000XX0XXXX
0000000000
XXX0000000
XXXXXXXXXX

At this point, this room has open space at the top, and at the right, so it will stop carving and move on to the next room.

This is a simplified version of it, just to explain the general idea of the system.  When it's done, it'll make sure the exact walls of each room line up with the adjacent one, so you don't have rooms that you can't get into because the openings are off.

Also, each dungeon has different generation parameters.  So the jungle might be more open, and the crypt will be more tight.
The parameters will be something like this: Room size X, Room size Y, max carve X, max carve Y, min carve X, min carve Y, dungeon dimension X, dungeon dimension Y.  Dungeon dimension will be based on rooms.  So the jungle will be like 10, 2, which is 10 rooms by 2 rooms. So the jungle will be very flat, but wide.  There will be additional parameters like 'roof open', which will basically cut the whole dungeon in half longways, so there will be a sky, like in terraria, as opposed to spelunky.

It's going to be hard, but I'm doing it.

Also worth mentioning.  The levels won't look like they're made up of rooms.  I'm just using rooms as a way of making it less random-feeling.  You probably won't be able to see the borders of the rooms, it'll just look like the underground areas of terraria, with a little more structure.
Of course, this is what I'm imagining.  Once I implement it, I'll post some giffy gifs.

Got a super basic level generator in place, just to test things out.

Honestly, I think if you add a lot of stuff to this game, it could be really cool.  Take avalanche for example.  The original flash game was just "climb up".  The sequel takes that super simple idea and adds a lot.  There's powerups, coins, enemies, boss battles, shops, co-op etc. I feel like you could do the same with this game. Just go crazy and add a lot of stuff.  Make it something special! 

It ran super shitty on my computer, lol, but I liked it.  I couldn't get very far, at points it was running at like 1 fps, so I couldn't do much and just got killed.  The guns feel really satisfying, especially the sound effects.  As long as the final version runs well, I'll totally buy it.

This looks great.  I feel like there could be enemies, or something to make it more interesting. Something that you could crash into, to get some sort of bonus.  The idea of "move to the right before the clock runs out" seems like the easiest way of making a game out of something that's just a toy.
That sounds so mean... but I hope you understand what I'm talking about.  There should be more goals/objectives. And the game looks really awesome, btw. Like, I'd buy it right now, cause it looks fun, but I just feel like it could be drastically better if there was more purpose.
Also, if there already are enemies/other things to do besides move right, I'm sorry.  I haven't played it, so I can't really tell 
Keep up the good work 

I literally had this exact idea recently. But now I don't have to make it to see if it works. 
Hope it works out for you!

Congrats!  You now have LOTS OF MONEY! 

[Wood, ore, gears, stingers, string, tech, mushrooms, blastroot, honey, beans, metal, glass, sand.]

Here's the first "actual" art for the game.  These are the materials for crafting.
From left to right. Wood, ore, gears, stingers, string, tech, mushrooms, blastroot, honey, beans, metal, glass, sand.
Please let me know if anything doesn't look right
I think they turned out pretty good