We recently released Camp Keepalive, a turn-based strategy game for Mac and PC. We've been working on it for about a year, and we're really happy to let out into the world. The game borrows its theme from basically every horror movie released in the 80's-- your goal is to guide campers lost in the woods back to the cabin before they're eaten by monsters.

• Main site with free demo
• 
  
  

We just released a demo of Camp Keepalive, a turn-based strategy game themed around 80's horror movies. The goal is to rescue campers before they're eaten by a whole host of monsters. You control a team of counselors, each one with a special power.

Win and Mac versions are here: http://twofoldsecret.com/campkeepalive

The plan is to build out a full game with missions that have different goals, i.e. so you're not necessarily doing the ferry-the-campers-to-the-cabin thing all of the time, and also add in more counselor, monster, and terrain types.

Two things I'm wondering about:

• Any suggestions on streamlining the tutorial? We've already cut it down a lot from what we originally had, but I still think it can get smaller.
• Are there anything annoying or confusing about the UI? For example, a lot of early feedback we got from people was that they had trouble getting a sense of the entire strategic situation, so we added the minimap.

We have a Mercurial repository on Dropbox that we push/pull to, and a separate folder on Dropbox as a working area (images/sounds that aren't ready to be plugged in, that kind of thing). It works great except when sometimes there are slight delays in Dropbox syncing -- only really an issue when we are working at the same time.

And Linux, right?

Whoops, yeah . I think Linux support is a bit tricky if you don't want to compile LOVE from source... there are official packages for Ubuntu, though, and I know there is a statically-linked x86 executable, but the latter was created in a bit of a voodoo-y process from what I could understand.

Here's the quickie summary:

• creating classes, subclassing existing ones, reusing objects
• animated sprites, including alpha fading, color tinting, and rotation
• user input via keyboard, mouse, and gamepads
• framerate-independent physics (velocity, acceleration, and drag)
• collision checking
• static and animated tilemaps, loaded from either CSV or Tiled files
• tweening object properties
• recording user input and debugging

Nothing superastounding -- in a lot of ways I am aiming at feature equivalence with Flixel and FlashPunk but for Mac/Windows platforms. The goal I have for it generally is to be basic but comprehensive.

[What is LOVE?]

Hi, wanted to share Zoetrope, a library I have been working on for LOVE.

What is LOVE?



LOVE is a 2D game engine that's scriptable in Lua. Right now it runs on Windows, Mac, and Linux, and there are also ports in progress for Android and WebGL. It's zero-compile -- you drag your folder of source code onto the engine and it runs immediately -- and is licensed so permissively, you can use it for commercial works without even crediting it (but don't be a jerk).

LOVE was used to create Mari0, and some crazy people on Reddit are using it to make an actual game out of that one episode of Community (seriously!).

So ok what's Zoetrope?

LOVE is relatively low-level -- it has functions to draw images and play sounds, but doesn't have anything to handle sprites, collisions, and so on. I wrote Zoetrope to get all that. I modelled it a lot after Flixel, because that's what I used to use, so if you've used Flixel before, a lot should seem familiar.

That's cool and all but why should I use this instead of the five million other frameworks out there?

• Testing while you're developing is way fast. Change a variable and hit Ctrl-Alt-R and all of your code instantly reloads from on disk. There's also an interactive console that lets you execute any Lua command on the fly. When If your game crashes, the debugger pops up immediately so you can inspect the state of things and try to find out what went wrong. You can also record and play back keyboard and mouse input, to save time and have something beta testers can give you to highlight bugs.
  
• You can scaffold out full levels in Tiled; Zoetrope will load them with a single method call and substitute placeholders for classes you haven't defined yet. If you create a class whose name matches a Tiled object's, Zoetrope will automatically connect the two and pass in all properties you define in Tiled. Again, make an edit in Tiled, export, and hit Ctrl-Alt-R to immediately see what it looks like in-game.
• It's reasonably fast (I get 1,000 simultaneous sprites at 60 FPS on a GeForce GT 525M) and Lua is generally a pleasure to use. It's also totally free to use.

Anyway, feedback/bug reports/pull requests are seriously welcome! I just put out version 1.0.1 so the tires could use some serious kicking.

• Zoetrope project page - demos, tutorials, downloads
  
• LOVE homepage
  

We've made the leap from Flash to downloadable games with Sought, our new game. It's a maze game with alternating controllable characters, some lo-fi graphics, and a story to puzzle out. It's available on Indievania here: http://indievania.com/games/sought



We've released it under a pay-what-you-want model, which includes $0. Hope you like it! Would love to hear thoughts.

Yep, it saves your progress after every level.

We just released our newest game, Alight. It's a platformer made in Flash that leans more towards exploration than reflexes, and has some kickass music by Smiletron. We're really proud of it.

Some weirdo on the internet named Ortoslon recorded a video playthrough, too.



Check it out!

The biggest advantage to this would be the rotations. Current sprite rotation in Flixel is done using baked sprites.

I think this is not 100% accurate -- I know you can ask for Flixel to bake a rotation for you, but you can also change a sprite's angle property to whatever you like. The downside is that when you do that, I believe you go through the vector renderer instead of just blitting pixels to the screen. But this also happens if you set the color, alpha, or blend properties, too.

Is there a definitive reference page on this kind of stuff? I've seen lots of pages scattered hither and yon about this topic, but nothing that really lays out all the issues you run into when creating a projector that runs fullscreen.

And this is a good reference for all things maze related.

Holy mother of god, I wish I had seen this when I first started delving into maze-making. This is a really good read.

Skimmed over some old code and realized I was doing things in the absolute slowest/dumbest way possible. Instead of running a glow filter on a sprite once and tweaking the result's alpha, I was re-running the filter every time the alpha changed. Bad news because I was running a tween on the alpha property! Rewrote it and no more FPS drop while the tween runs.

Rock steady 60 FPS = happiness.

To me the hard, unsolvable part is those NPCs that don't really have much to do with the game but need to be around so that a town doesn't feel totally empty. It feels weird to not get anything out of trying to talk to them, but it also feels totally shallow for them to have something really nondescript and repetitive to say. I don't know, IF has a tradition where you can just say, "that isn't important" when you try to interact with something that isn't going to do anything, and people will just accept that as given.

I always imagine that there's a group of players out there who are in it strictly for the gameplay and don't really care about the story one way or the other. To be fair, a lot of video games, especially early ones, didn't care about story either. So I think a storyline *has* to be optional for a lot of game genres -- interactive fiction/visual novels being obvious exceptions. By the same token, if someone's going to disregard the storyline entirely, then the ending is allowed to make no sense to them, either.

I like Deadly Premonition's approach, where by the time you finish the required quests, you have solved the main mystery, but a couple side quests explain who the characters are and give you a better sense of why stuff happened the way it did.

I'll be a voice of dissent here and say I actually dislike fictional books getting embedded into games. I think of Deus Ex here but also Myst... I like things that feel like found documents, but when I know I'm about to hit several full pages of text, I start getting antsy.

I feel like penalties are the wrong way to look at it. If you miss a jump, you already know you messed it up -- you don't need something else to rub your nose in it, as it were. From this point of view, losing health is just a way to force the player to learn how to play the game/prepare for harder levels down the road; you can screw up only so many times before you lose a life and have to repeat the level.

I really like the idea of Sonic's multiple paths to completion as a way to alter difficulty, especially because I never thought about it consciously while I was playing. You're never on the "bad path," so it never feels like a penalty -- in fact I do remember in certain levels, going back and purposefully missing jumps so I could see what was down there.

I would give Shade a shot, I think it's an interesting contrast to a lot of these examples. You don't have a strong goal in the game, so it's not that the ending denies you anything -- but it's undeniably a dark ending.

Likewise the penultimate level of Braid reverses much of what you thought you understood about the plot. Kinda like Shadow of the Colossus, if you're paying attention to the plot you'll have a bad feeling about the direction you're going in, but it's quite a shock when it actually hits. The last level is clever in that it appears to be a "well it's not all bad" sop to the player, but only if you don't explore carefully.

Military Complex?

This isn't strictly atmosphere, but I loved the optional bits of exploring after you win.

A forest

You're supposed to be on the top of the trees, walking on branches and tree trunks.. a lot of 8-bit and 16-bit platform games had levels like that..

I really liked this one -- I could see how you'd be able to flesh it out with actual tiles. One caveat: on screen #4, the one tile above the last jump on the right side makes it really difficult to complete. At least, it took me a while and it was kind of frustrating.

Here's my try: Escape from Cavern Mountain!

Yeah, I was going to say it is possible to package a Flash SWF as an EXE, but that option still screws over Linux users. We thought about the EXE option, actually, because at least I thought fullscreen added something to the gameplay -- I guess a greater sense of immersion is the best way to put it -- but we also found we were losing 10 frames per second even running at 640x480 blown up to screen resolution. Such are the hazards of Flash game development, I guess.