My Adventure Game Engine - Making Way For Adventure Engine 2

[Chucknorrisaurrus]

I'm at a complete loss of words for how in God's name you made this game in QBASIC. I mean, seriously, what? How is that even possible?!

[Jasoco]

I'm at a complete loss of words for how in God's name you made this game in QBASIC. I mean, seriously, what? How is that even possible?!
http://popewatchnow.files.wordpress.com ... ied-21.jpg

I am a GOD. No seriously though. I don't know. Impressed? Well, you should be. It was a lot of code. Sadly I ran into the memory wall and had to find an alternate language and platform to develop it on. I went to JavaScript last January. Then discovered Löve and switched over to it a few months later. I have three different versions of my engine on three different coding languages right now. The Löve version is the only one I'm actively developing though.

The QBASIC version was going to be an RPG with random battles due to the speed limits, even with DOSBox turned up to 50 times the speed. The JavaScript version had live battling like Zelda. And with Löve I was able to go even further. Thank god for Löve.

[thelinx]

Can you re-upload the demo?

[Chucknorrisaurrus]

Well, I hope to develop something half as good as what you've made. And you should totally reupload the demo xP

[Jasoco]

I'm just going to leave this here.
REMOVED, SEE OP FOR NEW DEMO!
It's a snapshot of where I am currently. Lately I've mostly been revamping the UI and trying to fix some stuff, optimize some code. There's still leftover stuff that isn't needed that I need to clean up. I am only putting it up now to get it out there. There will probably be a newer version in a few weeks too.

There is no audio to save space. I mean I could include it, but it's just sounds ripped from Zelda anyway.

I'm still trying to figure out the best presentation for the menu system. I envision when it's finished that the game will go right from the intro logo into the game. If it's a new game it'll just start right away, and if a saved game exists it'll open the menu asking if you want to continue, etc. It's a work in progress. The actual menu is gone. It's all a game overlay now. So you see the game in the background under the menu screen. I'm more inspired by Braid. In Braid the game started right into the game. The title screen was part of the game. It would load your saved game if it existed. The logo, a floating "Braid" would fade off when you started moving. I'm picturing something like that.

For an example, here is Braid's Xbox 360 introduction:
http://www.youtube.com/watch?v=Urv-UGogWxo (Watch the first 15 seconds)
Note the floating smoky "Braid". That could probably be done easily with Particles. Of course since I don't know what the game is called, I don't know what to spell.

The intro logo is Box2D. Just something I was playing with.

I believe presentation is the greatest key to a game's success. Braid did it very well. Cave Story also did it very well with its SNES-like feel.

Edited: Oh, by the way, if you ran any of my demos before, DELETE THE PREFERENCE FILES first. That error is because I had Physics turned off before and it saved a config file with Physics set to false. Now it uses them so it sets it to true. My mistake. Just delete the config file.

[Robin]

Nice. Some feedback:
- The conversation animation is a bit slow, which makes it annoying.
- The bridge can be hard to cross if you're not on exactly the right spot.
- You can go down to the trees and get stuck, at which point you can only move down. (It requires some jiggling of controls to do that, but still.)
- I liked the Jasoco “title-card”.

[Jasoco]

Nice. Some feedback:
- The conversation animation is a bit slow, which makes it annoying.
- The bridge can be hard to cross if you're not on exactly the right spot.
- You can go down to the trees and get stuck, at which point you can only move down. (It requires some jiggling of controls to do that, but still.)
- I liked the Jasoco “title-card”.

- You mean the whole thing? Well yeah, it's just for show. To show how stable the scripting is that it can handle going for a while.
- The player's hittable area is only 16 pixels high. The bridge is 32. Just move up a few pixels. You have 16 pixels plus tolerance to play with.
- That only happens when your computer skips a lot of frames. I get these hiccups too and I can't figure out how to stop them. It's only occasionally but it's annoying. I'm looking into it.
- Bouncy bouncy. Will it remain? I dunno. Maybe it's more suited to a physics based puzzler. We'll see how the game ends up when it's finished... in about 5-10 years. (Being optimistic)

I'd originally envisioned the logo and menu to all be physics based. The logo would fall onto screen then fall off, then the title and menu options would swing onto screen dangling from ropes. But as I said, that'll probably work better in a game that it feels at home in like a physics puzzler.

[Luiji]

You could always have it start a new game and have people press menu and load a game from there. You could also make the main menu a level, i.e. with a door per save, one for new, and one for exit.

[Jasoco]

I had thought of that, since it's exactly like what Spelunky does:
http://www.youtube.com/watch?v=NJPIFKSkuT8

But it will all come down what the game requires. If it's a new game I can have it go right into the opening script, introducing the characters, telling the story of why the player is who he is, then take you right into the game. And if a saved game exists it would simply load your game and show the menu (Sort of a pause in case you are saved at a point where you need to be ready) where you can choose to continue that saved game or start again.

I don't think the Spelunky method would work for the idea I have at this point.

I'm also thinking that it might not be a retro styled game when it's done either. Right now it uses 16x16 sprites doubled up, but I want to actually put detail into the game, so I'll make them 32x32 again. And I'm planning on removing tiles for the most part and switching to object based worlds where you can plop things onto the map as a whole. Trees, houses, rocks and flowers, etc, then detail it all with accessories. It might actually save FPS ticks this way. Right now each screen has an average of 600 tiles per frame depending on how much over and under detail you are using. It might end up being easier to put all the objects into a pool then draw all the visible ones and discard the off-screen ones. I haven't put in a framework yet but it should be easy to get an ObjectMap framework up and running as a prototype soon.

I wish I could hire an artist to paint me some scenery. That would be much more awesome. Braid is just so amazing looking and Aquaria is really nice. These small-time games developed by tiny Indie developers can look so good because of how much talent the people have. But I guess I should have a game story first. Also, I could probably make my own style.

If I did do the Spelunky method, I'd also have a house for Options. With pushable crates or flippable switches for turning stuff on and off. Or maybe people you'd talk to who control the options for you.


Side note on a Tangent ramblings: And even though I won't use it, I was just playing around with time controlling. You know how Braid can control time? Well, in order to do that, you'd have to record a LOT of data every frame, or every few ticks or milliseconds at least. And if it remembers everything back to entering a room/map, you end up with a lot of data in memory. I just did a test with my rudimentary time recording functions and even though I forgot to check the memory usage, simply recording just the Player's X and Y coordinates and the current game time into a multi-dimentional array will consume a text file ~450KB for just 10 minutes of recording every .05 seconds. I probably won't use time controlling in my game as that would be very complex, but it might come in handy one day. Only recording the player's position is nothing. If you truly want to save time you'd have to record all the player's stats, how much stuff he holds and every enemy, NPC, everything that exists every single increment. Once you have all the data in memory, if you want to rewind you would simply run backwards through the array placing all the actors and objects in their recorded locations as you go. So complicated. Just in case anyone ever tries to make a time manipulation game.

[Luiji]

You could try saving a limit of 20 seconds or something.

Also, I could work on some scenery graphics, too.