Okay, I’ve held off publishing this post for two days now, time to stop writing and actually do it.
(But there just this tiny little thing I want to ad-)
Nope. We’ve live. Wait, who are “we”? I’m talking to myself again. Anyway…
On this episode of Where Is That Thing You Were Working On Several Weeks Ago Now?, or W.I.T.T.Y.W.W.O.S.W.A.N.? (“Witty Woss Wan”?), we have a gander at Rise & Fall. You know, the thing I worked on as part of the team Robots Mashing Keyboards for the ROM Game Jam. (Oooh, that thing.)
Quite a few things have been added or changed since the ROM Game Jam prototype. The biggest change is the inclusion of jump-through platforms (i.e. platforms that can be passed from below but not above). The original prototype only had completely solid platforms, but to have a greater number and density of platforms in the level, my team decided that we need to have jump-through platforms as well. Seeing how something like this should be Game Programming 101, I should have learned how to do a long time ago, but never did until now. Still, I had to learn it from the source code of this demo by Bill23. Even then, it took me two attempts to get it working right. Other changes are relatively minor, such as adding mercy invincibility, or bug fixes. We will also be changing up some the graphics and adding more levels before the playtest session at the ROM on October 19.
During the development of Rise & Fall, I had encountered the weirdest glitch I’ve ever seen in all my time using GameMaker, even weirder than the ones I’ve seen while working on Gonna Catcha. Furthermore, I don’t think it was (entirely) my fault:
After some thought, I’ve decided to make the Gonna Catcha playable demo a closed alpha for testing purposes, as I think it’s still too early for me to throw the game up on the internet for open testing. As a consolation prize, here is a video of the playable demo in action:
- More rounds. Which maze the round takes place in and the number of spirits, rocks, preta and bonus items (in bonus rounds only) in it are manual determined by me for each round, rather than determined by some algorithm. Even though the game will have 256 rounds, I’m only going to create round data up until a certain point, then either start repeating the last few rounds or randomly generate round data to fill in the rest.
- Co-op mode. It might have modified round data compared to single-player mode to balance the game.
- Attract mode. How else will you know how to play the game?
- High score table
- Controller support
- Bug fixes and code optimization. Even when I thought I’ve squashed most of the major bugs in the game, a new challenger appears:
- CRT shader. The game is setup up to use it, but it’s no longer a required feature to me anymore.
- Demonstration cutscenes. Whether or not I need these will depend on how much info I can cram into the attract mode.
- Bezel art. Artwork to put on my virtual arcade cabinet.
|Babby’s first shader|
Here are the results of my shader experiment:
Once you’ve regained your composure from watching the trippy and completely unnecessary graphical effects, you may notice something new in the video. The game randomly places rocks in the maze that act as destructible walls. (Rocks?! I thought they were just blobs!) This is to slow the player down and allow the spirits to better disperse throughout the maze. However, the rock impede the spirits as well, making the whole thing kind of pointless. Of course, they’re spirits; so my plan B would just have them, you know, pass through the rocks unimpeded. Preta can stay impeded by rocks, being corporal beings like the players.
One other big change I made to the game that might not be obvious in the video is the player movement code/rules. Previously, the players had free movement; now their movement is restricted to the grid much like the spirits and preta are. I did this to make turning around corners easier for the players. My previous solution had the players snap to a corridor if they were “close enough” to turn into one; it looked a bit weird and was a bit finicky.
Re-writing and messing around with the movement code did produce some amusing bugs in the process, something that hasn’t happened in a while.
Here’s some in-game test footage of Gonna Catcha, featuring Donum Dono as the test subject:
- Gun out or holstered: 2 states
- Walking or standing: 2 states
- Facing direction: 4 states
- Total combinations: 16 states
Outside of coding and pushing pixels, I’ve cleaned up and coloured in the sketch of Donum from my last post. Now he’s no longer confined in a 64-pixel prison.
|“*gasp* I’ve been vectored and coloured!”|
I also made him the temporary face of my YouTube channel and Twitter:
|“Uh, I don’t feel right about accepting this position without asking Pohena first.”|
Enjoy it while it lasts; I might have other ideas about branding in the future.
I didn’t expect myself to post so much in such a short time frame, but this is too amusing to pass up. I was just messing around with Stick and Stones for a little bit when suddenly this happens:
|Have you been hanging out with Zalgo?|
|“Somebody please put me out of my misery.”
(Note: Black background added for clarity)
As you can see, this new method still has some bugs to work out, and for my project that’s entirely text-based, this is an obvious problem that needs to be addressed. Fortunately, the devs have taken note of this and are working to fix it. The workaround they suggested is to recreate the font resource and delete the old one every time you make a any change to it, which is a bit of a bother.