Dig the LoL sprites.

Frankfort developer checking in.

Been busy refining things.  

Lucky enough to have gdrush sign up to art things for me.  Got a bit of work-in-progress tiles in-game now.  Will slowly replace Oryx's stuff as time goes on.

Think I have a dungeon gen setup that I finally like.  It's more of a standard room and corridors setup, but I actually think that lends itself better to the gameplay I have in mind here.

Have room decorators working too, so I can have room features like pillars and inner rooms pretty easily.  Can even use the decorators to add small things to rooms like mazes or pre-set mini puzzles.

Will probably look into adding extra things like magma seams, etc using some CA overlay at a later date.  This is good enough for now, though.

Also playing with lighting and a darker atmosphere.  Not sure if I like it enough to keep it, but we'll see once I start lighting more real things like sconces, fireballs, certain monsters, etc.  Could add some nice atmosphere or could just be meh.

Next things to look at are monster AI and a loot placement/score system!  Then start fleshing out player characters and their abilities/themes.

Anyway, here's my update GIF!  Sorry about the banding on the lights.

I've found someone interested in arting this game up for me!  That's great news, and I'll post more about that in the next couple of weeks once I get some things to add to the game.

I've completed the steering behaviour implementations I needed.  Entities can now seek, flee, arrive, separate and avoid walls properly while chasing the player around.  Feels and looks good to have a swarm of dudes after you while you shoot like mad.

I was unhappy with the dungeon generation above.  The maps wind up too contiguous while being too chaotic.  I feel like for this type of game, where you need to be able to aim your shots while moving, that having a bunch of terrain in the way all the time isn't great.  So, I went looking for a way to make interesting maps that predominantly feature rooms and corridors.

What I found was a nice process shown off by the TinyKeep guy over at http://tinykeep.com/dungen/.  The steps as are:

• Generate randomly sized cells within a starting boundary area.  They will be overlapping.
• Separate the cells until there are no more overlaps.  The separation steering behaviour is useful  here.
• Get the map bounding box by examining the cells' corners.
• Fill the map's remaining empty space with 1x1 cells.
• Examine each cell and pick out those who have both sides >= a certain threshold.  Mark them as rooms.
• Create a connectivity graph of all marked rooms using Delaunay triangulation.
• Find the MST of the above graph.  Those edges are your core path through the map.
• Add a certain percentage of the rest of the edges to the MST to create some loops and variety.
• Created corridors between rooms linked by an edge.  Any of the cells that weren't marked as rooms get added to the map as corridors if they are intersected during this phase.

Results are something like this: 



Much happier with this for a 'standard' dungeon feel.

Other generators will be added later (cave-likes using some sort of CA) for what I want to do next.

Dungeon names and titles as a generation grammar.

The basic idea is as follows:

• Each time the player wants to explore a new dungeon they are asked to name it.
• From the given name generated a hash code for the name, ie, 'Yserbius'.  That's our dungeon seed.  Now the player can share the dungeon by name, and go back to explore the same dungeon whenever.  I think Minecraft did this.
• Now the player is given the option to add a title to the dungeon using a preset list of nouns and adjectives in a certain order, ie 'The <map descriptor> of <modifier-descriptor0 ... n>', ie 'The Yawning Maw of Decaying Doom'.
• Use the title to define what the generated map actually is.  The map descriptor will define the overall structure of the map, ie dungeon-like (as described above), cave-like, ruins, forested, etc.  It sets the basic theme of the map.
• The modifier descriptors will be used to modify what's in the map rather than what it is.  'Decaying' could mean that the majority of the monsters are undead, and that there are putrid pools of water that are best to avoid.  'Doom' could simply mean that the difficulty is increased overall.  Monsters have more health, hit harder, and are more numerous.

That's next on my plate.  Followed by actually placing monsters and loot on the map.  Followed by lots of content generation and endless iteration on evreything.

Once I get monsters/loot in I'll begin posting playable builds for feedback.

[What:]

Devlog here.

What: At it's core I want the game to be an arcadey gauntlet-like roguelike-like. This means fast paced gameplay, permadeath, randomization in loot and maps coupled with simple character mechanics and lots of projectile weapons!  That's the work-towards goal.  Other than that I'm keeping the design fluid, and wouldn't mind a collaborator who wants input into that aspect as well.

Who: I'm looking for an artist dude and/or dudette who can create clean, original art and animations in an oblique isometric perspective.  Pixel art is fine, but not required.  I'm open to any available aesthetic as long as you're strong in it.  I'm more concerned with finding someone who can mesh well with me and wants to truly collaborate on this.

When: Now-ish!

Why: Cause I'm tired of stealing Oryx's stuff for prototyping.

Where: Our respective bedrooms, probably.  Skype is nice.

Spent most of the day implementing dungeon generation.  It's pretty simplistic in nature at the moment, but here's a breakdown of the process:

• Create a randomized dungeon map layout using Diffusion-limited Aggregation
• Create paths between the generated map layout using a DFS search.  This gives us a basic connected layout like so:
• For each dungeon map cell generate the tile map using pre-made herringbone wang tiles which gives us something like

And that gives me a basic generated dungeon.  Steps from here are to flesh out the herringbone tiles themselves to place dungeon features (aesthetic and functional), and to setup doors to travel from map to map.

More progress today.  Lots more code refinement and improvements all around, but I also implemented 'scripting' (really just Java state machines) and events/listeners so interactive gameplay stuff can start happening as evidenced below.  Animations, and death, and spawning, oh my.

Been hard at work putting more systems into place.  Nothing that I can really show in lovely GIF form yet, but I am quite happy with the entity/component system I have going. 

It's basically just the Artemis Entity Framework with some tweaks. First, I've made entities and components pool properly so there's no needless garbage creation.  Second, and more important, I've implemented an entity template system.  What that means is that I can define, in external JSON files, what kind of entities I want in my game and what their respective components (and parameters thereof) should be.  On top of that multiple templates can be applied to a single entity which gives me a basic sort of inheritance system to work with. 

For example my current 'entities.json' looks like this:

Code:

{
    "player" : {
        components: {
            player: null,
            transform: null,
            movement: null,
            size: { w: 1, h: 1 },
            group: { group: "PLAYERS" },
            stats: null,
            ranged: null,
            collision: {
                group: PLAYER,
                collision: [ ENEMY, HAZARD, PICKUP ],
                separation: [ PLAYER, DESTRUCTIBLE ],
                boundingBoxes: {
                    "world" : { w: 1, h: 1 },
                },
            }
        },
    },
    "player-wizard" : {
        inherits: [ "player" ],
        components: {
            movement: { maxSpeed: 3 },
            health: { "numHitDice" : 2, "hitDiceValue" : 4 },
            stats: { atk: 3, spd: 5, def: 3, mdef: 6 },           
            ranged: { ammoType: "fireball", interval: 0.33 },
            animated: {
                pack: "player-wizard",
                animations: {
                    "idle-north" : { fps: 0.1, playMode: "loop-pingpong" },
                    "idle-east" : { fps: 0.1, playMode: "loop-pingpong" },
                    "idle-south" : { fps: 0.1, playMode: "loop-pingpong" },
                    "idle-west" : { fps: 0.1, playMode: "loop-pingpong" },
                    "walk-north" : { fps: 0.1, playMode: "loop" },
                    "walk-east" : { fps: 0.1, playMode: "loop" },
                    "walk-south" : { fps: 0.1, playMode: "loop" },
                    "walk-west" : { fps: 0.1, playMode: "loop" },
                    "die" : { fps: 0.1, playMode: "normal" },
                },
            },
        },
    },
    "ammo" : {
        components: {
            transform: null,
            movement: null,
            size: { w: 0.5, h: 0.5 },
            collision: {
                group: BULLET,
                collision: [ ENEMY, DESTRUCTIBLE ],
                boundingBoxes: {
                    "world" : { w: 0.5, h: 0.5 },
                },
            },
            damage: null,
        },
    },
    "fireball" : {
        inherits: [ "ammo" ],
        components: {
            movement: { maxSpeed: 6 },
            damage: { min: 2, max: 4 },
            animated: {
                pack: "ammo",
                animations: {
                    "fireball-firing" : { fps: 0.2, playMode: "loop-pingpong" },
                    "fireball-explode" : { fps: 0.2, playMode: "normal" },
                },
            },
        },
    },
}

This lets me quickly create different player entity types and ammo types without needlessly redefining the same thing over and over.   

But, more work to be done this weekend!  My goal is to have dungeon generation up and going by Monday evening (the last day of my weekend) at the latest.  So far I'm on track for that.

The building scene gave me nostalgia goosebumps. I remember playing this old DOS game where you had to build a castle by placing individual wall segments, and you could actually see the little peasants walking around it as they built it. I can't remember the name of it but I wish I did; It was pretty fun. xD (I know that it had an option to choose between fantasy/reality, but I think the only difference it made was whether or not humans or orcs attacked your castle.)

Anyway, looks really charming. I agree with stick on a slight contrast bump for the castle.

Castles by Interplay

The game isn't really tiled based, just the map.  So all the entities are able to freely move around and collide with the terrain fine.  Pushing a barrel into corner is pretty easy.

The size of the room here isn't indicative of the size of the levels.  It's just a test area.  Randomized dungeon layouts will be a focus soon.  Just laying the groundwork first.

And, ya, the graphics are hodgepodge re-sized messes just so I can have something on the screen.  Worrying about pixel sizes and aesthetic isn't really on the table yet.

Any number of players greater than one will have to require a dual axis controller.  I want 360 degree shooting and don't want to settle for 8-axis or 4-axis or auto aim or anything.  So gamepads it is.  The game will be single player as well.

Just something I'm working on.  The name is kinda placeholder, but the game is a Gauntlet-like, roguelike-like, Robotron-like thing.  So that's where that comes from.  The aim is to have a 4 player local co-op arcade-y fast paced dungeon romp.

Aiming for PC and maybe Ouya?  I don't have one of those, though.  Mobiles are unlikely.

Using Oryx's lo-fi tiles for prototyping art placeholders so that is pretty irrelevant atm.  Will try to track down an artist when I get further along in development.

Lots of good backend work is done already, though.  Here's one of them devlog GIFs people love showing collision groups, entity separation, shooting, and other stuff working.



Will post more!

Crate job so far.  Keep going!

I want to be in as I have a hard time finishing projects and feel like LD is a good way to break that habit.  But everytime I think about trying I remember how much I cannot do my own assets and it all breaks apart.  Anyone want to jam?

I just did something similiar when I made a bounding component for my entities.  I used an abstract generic class that is subclassed to implement shape-specific stuff (circles and rects atm).  Code here for examples.

No one here can offer you real legal advice.  If you feel like it would be a potential issue for you and your game you'll need to consult an actual lawyer.

I finally got around to implementing the herringbone Wang tile method to good results.

Working (slowly) on designing the 128 tile chunks needed for randomizing my dungeons using the herringbone wang tile method I posted elsewhere.  Using a simple color mapping method.  Considered making an editor specifically to design the chunks, but any extra freedom it gave me wouldn't be worth the time invested.

It's REALLY easy as a developer to assume that knowledge of your game is intuitive and universal.  Just remember that it's not, and to explain everything in clear and concise terms as you would to a five year old.

So, here's my feedback.  My tone won't be light.  Not to be insulting or to make you feel bad, but to try and help you directly and honestly.

The #1 thing sticking out for me is your About the Game page.  I'm not going to take the time to signup and play the game if it doesn't sound interesting, and unfortunately that page has a lot of text that doesn't actually tell me anything about why I would want to play it.  You use a lot of generic and ambiguous language.  

Here's my impression as I parse that page:

• Why are there quotes around "Collect"?  Am I only sorta collecting?
• No quest log, etc which I'm assuming is some kind of dig at WoW or modern RPG mechanics.  Don't tell me what your game isn't.
• Thrilling battles with spells and items?  Well, I hope so.
• Tons of items.  Ok.  What's special about items?
• Many buildings with many different abilities, etc.  Buildings for what?
• Sound.  Really?
• Stories which warm the heart.  Don't make assumptions on my emotional reaction to your game or story.
• Realtime.  This is an interesting feature in a browser game.  Good point.
• No plugins.  Also a good mention.

None of these things 'make this game unique', unfortunately.  You are failing to capture my interest, and make me have that moment of 'oh, neat, I want to try that!'  I can't throw a rock without hitting an RPG with quests, items, buildings, and sound.

Then I scroll down and see a horizontal scrollbox of oddly stretched images which I imagine are screen shots of the game.  Below that is another stretched image.  I assume a screenshot, but I can't tell.  This is in IE8.  Not a browser I use at home, but I'm sure some people do.

Next, 'A few words from the game developer'.  Normally I would skim this at best, and more than likely skip it all together.  You've already lost me on the premise of the game (as I have no idea what that is, yet) and so my interest in reading some random words from you is nil.  However, upon reading it... HOLY SHIT!  Why is all the cool stuff I can do in this game hidden at the bottom of the page?!

• It's a bright fantasy roleplaying game.  Ok, cool, whatever.  Generic but acceptable.
• You play a character thru a storyline and different quests.  Said every RPG, ever.  You can do better than that.
• Ok, so my main game goal is working on my village and what you call a cohort.  Sweet, I know what I'd be getting myself into now.  We're looking at something sim-y.
• Ok, a cohort is a group of units + my hero.  Makes me think of tactics games.  I like those.
• Recruiting dudes with different abilities, upgrades, and spells?  Awesome.  I enjoy customization.
• I get to sim-build a house for everyone I recruit?  The casual in me is giggling with joy.
• I can visit my friend's village, and show off my own?  Good, I can rub Josh's nose in my epic new fountain piece.
• Guilds? Auction house?  Social hub?  In real time?  In my browser?  Nice!

So basically it sounds like you have an RPG that pulls in elements from tactics games, sim games, and a lil bit of Pokemon.  Wrapped all up in an easy to access medium and sprinkled with some social goodies here and there.  That's fine.  That sounds like a decent game that some people would enjoy.

You just need to tell them that!  Scrap and re-write that entire page.  Everything in my second list needs to be at the forefront.  Top of the page in large impressive lettering spelling out everything awesome and interesting about what you're offering (of which, sound is not one of them).  In a browser based game that 'About the Game' page, and your front page (which also needs to highlight what people are here to see, and direct them to the About page if not summarizing it itself) IS your 30 second demo.  I'm going to read that and already be playing the game in my mind's eye.  I'm going to WANT to play.  As it reads now, though, I just want to go back to reddit.

You want to scale up without resizing the window, correct?  Then just add a scale matrix to the transform matrix you pass into your shader/draw functions.