I’ve been working a new project lately, which is called The Chinese Business Massacre. It’s a top-down role-playing game focused on melee combat from the player’s perspective. With the graphics I aimed at contrasting some atari 2600ish style more modern effects, such as particle emitters (blood; explosion).
The main difficult I’m having is to create a enjoyable combat system. I’m spawning enemies randomly in this chapter, but originally I was planning to have a previously set number of enemies in each chapter. Since the focus is on the “massacre” aspect, I thought: the more, the better, but I’m not feeling very confident about it. Maybe if I come with a better way to program the AI, it could be greater.
The game starts lagging when there are several enemies, there are a few minor camera issues, and walk animation doesn’t turn unless I release the key before pressing another.
Aside from that, it’s a good start
The enemies often come together, maybe you could make them coming at different speeds, from different direction, at different times?
Yes, I’m aware of the lagging problem, it sometimes happens, sometimes doesn’t, at least with me. I’m working on it.
What do you mean by camera issues? Do you mean like when you try to stab imovable objects, such walls and trees, or walk towards them?
The enemies being together is also a problem I’m trying to figure it out. When I tried to change the speed based on the proximity with the player, the game bugged hard, I don’t know why. At the moment, I’m just using pathfinder, and writing the enemies to go to player.X/Y() + Random (50). It helped a bit (compared with only going to player.X/Y(), which they completaly overlaps), but they’re still overlapping a ‘little’.
Yes, there are some camera sudden movements now and then, when close to obstacles.
Why don’t you set a different speed for each enemy?
To avoid overlapping, try the pathfinding obstacle behavior, and try the separate objects action.
How would you set different speeds for each enemy, considering that in this chapter there’s only a lot of copies of the same enemy? I tried to pick the closest to the player.X/Y() and subtract -10 in its velocity, but it bugged.
I can’t set both pathfinding and pathfinding obstacle behavior in the same object. I tried, and the enemy just stands still with no movement.
Regarding the separete objects action, I feel this is not the best solution, because if there’s two enemies close to a ‘pathfinding obstacle’ and the code separete them, one of them can ‘enter’ or pass through the obstacle and get stuck. Maybe I could delete the enemy that gets stuck, this could be a solution…Gonna think about that.
hey just played it, Gruk, hit every mayor point you have to fix, I only can add 3 things to the gaming experience
Something I noticed was that "slashing" a wall will throw you backwards, I wonder if you are using the same hitbox for movement and attack, (this may also have something to be to how you are handling collisions)
-you can’t skip the cinematic (I’m inpatient)
You can’t attack while moving
Now for gameplay, the deermens are not connected to the core gameplay, let me explain myself.
The explosive suitcase has 2 functions, open doors and killing the deermen, so they are like a key and I got 47 keys just walking around.
If instead of laying around, you got to kill x amount of deermen for them dropping a explosive suitcase, having them respawn eternally would make sense.
Hey man, thanks so much for the effort you put in your feedback. I really liked your approach regarding the objective of the game and I’m going to work towards it.
The hitbox is one of the main issues I’m having. Yes, every animation (of attacking) has a different hitbox for every single frame of the sprite, which is causing the “camera bugs” - at first I thought this was ‘kinda cool’, because you can enter certain parts of the game by glitching, and I would try to explore that, but it sure looks unpolished that way. I believe the best solution would be the weapons being separete objects with its own hitboxes, and the player having only one hitbox. Isnt it?
I see, and agree the idea of taking advantage from a glitch sounds interesting, but the common practice is to create a separe object and attach it to your player, this video sums it up pretty well.
Copies are not a problem. You just need to tell GDevelop clearly who does what how.
An example: If player distance to spawner < 200 px - create enemy at spawner.X/Y - move enemy to player.XY - change maximum speed of enemy to Random(200)
I’m pretty sure you can. Although, a problem is that paths are only computed before moving, so two moving pathfinders may cross paths.
This is an issue that I’m suffering, specially because some tile objects are pathfinding obstacles (like a group of trees). The same can happen with the “player” as well. I’m trying to fix it now. Since I’m working both in this game and Jurassic Fantasy, it’s beeing really hard. I’m never making 2 games at the same time again.
The biggest one so far, now you move with AWSD and shoot with your mouse - you can see the little crosshair at the center of the screen. The only thing that changes from one weapon to other (all melee) are the range. So you start with a knife, which gives you a “X” range. If you get, let’s say, a Katana, the range is X+150 and so on… I believe the gameplay is much more enjoyable that way, because it kinda feels like an third person shooter or top-down shooter.
Added a ‘type-writer effect’ extension that I found on GDevelop. I don’t know how did that, but it’s so great!
Changed the HUD/UI and the design of the first enemy.
Well, I got back into this game lately after a few months of bed-depression.
I decided to change the name of the game to METROPOCRE and I advanced really well the past weeks with the enemy AI.
This is a screenshot with the new HUD/UI design, sprites and the “blood bath” I’m trying to create with particles.
I believe I’m going to create a new topic with the new name/informations when I make some gifs and videos to not generate any confusions down the road. Any feedback? XD
The walls are being a bit of a problem to me. I’m not sure if I’d preffer those walls, or just the ‘top’ of the wall, like a completely top-down game.
I’m aiming at creating something that vaguely reminds the Atari 2600, but at the same time it stands as “original” (I know this is not easy, and I could be aiming at that, but resulting in something desastrous lol). The way I’m trying to do that is 1) not using a black sillhouette on the sprites but only their colors and 2) using the blood as particles to contrast with the “rigidness” of the design. With that in mind, I kinda want to leave the walls only with a collor and no addons. But I though adding some windows or paintings could be a solution.
Anyway, these are some pictures of the recent state of the game. I’m using 800x600, with a minimium 3 pixel size. What do you think? I believe I’m going to add more textures to the ground maybe.
Thanks for the suggestion. Indeed it looks better the way you said.
This game has changed A LOT since I began its development. Pretty much everything changed, from the name to the protagonist and the plot lol.
I’m trying my best to finish a playable demo by October so I can participate on the Steam’s demo festival.
Soon I’m going to create a new topic with the “new” game, and also to post regular devlogs.
Although, a problem is that paths are only computed before moving, so two moving pathfinders may cross paths. 
