Here’s the time lapse for my 72 hours of development on Stop That Hero!
Category: Game Design
I apparently needed sleep and a third day to get this game completed, but here it is!
So far, the Stop That Hero! Linux version (1.6MB) is all I have. I just started using CMake for my build scripts, so please let me know if the game won’t run on your computer.
Get the Stop That Hero! source
I’ll work on a Windows port, but for now, I’m going to relax! It’s been a grueling 72 hours!
UPDATE: Windows port (2.2MB) created and available for you!
I also updated the Linux build. It’s the same code, but the build was done on an older Linux-based system, so it should run on more systems without a problem.
I’m at the final iteration!
Basically, the tasks I have left:
- create instructions screen
- add sound effects
- package this baby up!
In the last iteration, I was play testing and balancing as best as I could. I decided that the bat was too expensive, so I moved it down in front of the slime in terms of costs. I lowered the starting health of the Hero and gave him fewer lives. I also tweaked the level design a bit to accommodate the AI’s inability to see too far ahead. There are also victory and defeat screens now.
I finally got the AI working well enough that I realized that I wasn’t just debugging the program anymore. I was playing the game! And it’s actually not that lame! B-)
I missed the deadline for the main compo, but I can still submit my game to the Ludum Dare Jam. The deadline is 24 hours after the end of the compo, so I have until 9PM tonight to finish this game.
After a good night’s rest, I woke up with bug fixes. Seriously, I was lying in bed, slowly waking up, and I thought, “Oh, yeah, I implemented cooldown for the entities after they attack, but I forgot to ensure that cooldown ends.” I fixed a crash bug which also prevents the creatures from trying to leave the world map. And I improved the exploration mode of the AI so that the entities should no longer wiggle or fidget. They pick a direction and go until they reach it. I found that the reason why the hero was getting stuck in place. It was due to the fact that I was checking if he had reached a very exact location, and with his speed, he sometimes overshoots it. When he tries to go back to it, he overshoots again, forever. I enlarged the collision detection box to compensate.
When I finished Iteration 4, I found out that accidentally implemented some stories from Iteration which deal with entities attacking each other. I’ve simplified combat so if entities are not in cooldown and are touching, they’re attacking. That implementation left Iteration 5 fairly moot. In fact, I decided to skip the remaining story in Iteration 5 which dealt with some nuanced AI that I’m not even going to look at, so I’m on to Iteration 6. You can see the backlog of skipped story cards under the Iteration card.
I’m pretty excited. It’s only 10AM, and I’m on the last two iterations. Iteration 7 is basically sound effects and packaging the game up, and frankly (and sadly), sound is optional at this point. B-)
I got to Iteration 3 by scrapping some of the art and animations I had planned for Iteration 2. Those story points go back into the backlog.
Iteration 3 was a bear. It was where a lot of the advanced AI had to be implemented. I’ve never done A*. Most of my AI was very simplistic, which went with the simplistic games I’ve done before. This game was going to be different.
Or at least I hoped it would. Entities would bumble about, get stuck, and fly off the world map even though I explicitly told them not to!
I had a quick lunch which was easy to make while I continued to work.
Unfortunately, the deadline for the main Ludum Dare compo passed when I finally figured out the A* algorithm (it turned out that there was a greater-than sign when there should have been a less-than sign, which is why the entities were moving so strangely). After the pathfinding, the entities still had to interact.
I had the Hero moving toward targets it sees nearby, although every so often I see that he gets stuck in a certain tile for some reason. Enemies seem to have trouble moving to the Hero like I expect, but they find him soon enough.
But with Iteration 4, they fight!
Do you see how the Hero’s health is down?
That’s the first time it has been like that outside of arbitrary tests being conducted. One of the dragons did that! Good job, my babies!
Since the main compo is over, the only option I have left is to continue on. I can still try to enter the Ludum Dare Jam, which gives me an extra 24 hours before the submission deadline.
For now, though, I’m going to bed. These last 48 hours have been grueling, frustrating, and exciting.
I went to bed after 5:30AM after getting some rudimentary AI in. I don’t know why the ground-based characters have no second-thoughts about going through the trees and mountains. If anything, they should want to avoid them. I’m sure there’s a bug in the implementation.
I woke up feeling fairly crappy, but I know what will fix that!
Yeah, OJ! Oh, and, yeah, sure, I suppose Banana can come along, too.
What you see is the game’s rendering of a small 50×32 image, blown up below:
Besides creating a world tile map, it also figures out where to place towers, chests, and the castle. The black spot is the hero’s spawn point. Red squares represent chests with health in them, and yellow ones represent chests with weapon upgrades.
The tile map will also help the AI figure where and how it can move. I need to do some trickery to make sure it deals with towers and castles correctly. They’re larger than a single 16×16 tile, but the AI should act like there is an impassable wall where it looks like there is one. Also, I need to make sure the hero knows how to get to the entrance of a tower or castle.
UPDATE:
Iteration 2 of 7 started with a little over 20 hours left in the compo.
I had some fig bars as a snack as the evening wore on.
My fantastic and amazing girlfriend made me a taco dinner, which was so delicious that my taste buds were fist-bumping each other.
I went to a party my friends were hosting so I could take a short break and get some much-needed social time, but I came back home shortly.
Looking at The Big (Cork) Board, I can see that I’m almost done with the first iteration of work.
Unfortunately, I still have 6 more big iterations to do, and a lot of it is doing AI path-finding and other AI things I’ve never done before.
And the deadline is looming.
I’m feeling pretty screwed.
Wait a minute. I think I have some Red Bull shots in my fridge…
I don’t drink caffeine usually, but desperate times call for desperate measures.
Let’s roll.
What you see is a working UI. The hero’s status is running off of live data, which means that he’s at full health, has 10 lives left, and has no weapon.
Also, the number of resources you have is running on live data.
The world is rendering 16×16 tiles of grass and tree right now, but I don’t have real data for it yet. I’m going to have to create the world, and at first, I’m fine with it being a bit smaller just so I can test out the AI when I finally get to it.
We’re nearing the 24-hour mark, though. I feel like I’ve been working non-stop and have been quite focused, but so many things still need to be finished. I might not sleep well tonight. B-\
My game has a title screen with working GUI elements. And with a title screen comes a title.
It’s “Stop That Hero!”, the game where you throw everything you have at that pesky heroic figure bent on stopping you from doing whatever evil things you are doing. What are you doing, anyway? Kidnapping princesses? Destroying the world? Littering?