Charas-Project
Game Creation => Other Makers => Topic started by: fruckert on November 18, 2012, 10:01:25 PM
-
So, yeah, I'm making a roguelike.
It's not even a prototype yet, but I'm making a topic right now because I'm sick of not having one.
This topic is going to be more like a dev diary than anything else, I probably won't post builds THAT much, due to focusing on getting the damn thing working.
Anyways, the plan at the moment is to get stats working.
You can run around and shoot **** (https://dl.dropbox.com/u/3104371/Projects/RogueFPS/build.zip) but that's all at the moment.
Things don't really have health yet, and things don't respond to gunshots besides bouncing around.
Bullets simulate actual physics, kinda sorta, and move at a realistic speed. This means they DO drop a little bit, but they're moving so fast at a short range you won't be able to tell at the moment.
The crosshair is also a little...obnoxious at the moment, so you probably won't be able to see very well.
I don't have any screenshots because I can't figure out how to do it within Unity, and I don't feel like cropping a Print Screen.
It doesn't matter anyways, seeing as there's only two textures.
-
Well, it's a start. God, that Wii-esque aiming. So bad. Can't wait until you fix that. You know what? **** stats, make that your first priority.
-
The control scheme is intentional, but the speed jump after you leave the minimum radius is a bug.
Currently fixing that.
ed: Aand fixed.
It now smoothly goes from zero movement to max speed the closer you get to the corners.
-
I think a better system would be to make a rather large deadzone in the center than doesn't move the camera at all. Then at the edges, you can make it have scaling scroll speeds. On the Wii, it works with a smaller deadzone, but with a mouse, it's just really awkward.
-
I'm also experimenting with having the mouse lock at a certain range, which might help with the whole muscle memory problem, but it feels kinda funky at the moment.
I could also have it move backwards, but that might screw with people even worse than it does now.
The scale is going to be customizable, though. That's already wired up, but has no user interface yet.
-
Roguelikes are cool. I ran into some real design issues when I tried to make a roguelike platformer, and in the end just decided that randomness couldn't deliver on that sort of fun for me. Hope you don't run into the same problem.
-
I'm having a couple of technical design problems with the inventory, but I'm muscling through that right now.
Anyways, time to talk about the game itself!
There isn't a story, it's just you vs a dungeon versus and a final boss.
You are a robot, and you kill other robots in a gigantic junkyard, working your way to the big city.
Your right hand turns into guns (both bullets and lasers), and your left hand spawns thrown weapons, like grenades and shuriken.
Kill things, find weapons, rinse and repeat until you die...which may be sooner than you think.
Currently working on how guns work, and how their stats work.
It's very boring.
-
Ah, the tedious stat calculations. Hate those. Keep at it man.
-
That and storage.
I'm trying to figure out how exactly I add effects to the weapons, and how the weapon design itself should work, because of the random.
-
So, let's talk game design!
Probably the heart and soul of this project at the moment is the weapons.
The weapons are why I started making this game, and it's mostly what I'm focusing on right now.
They're also probably the single most difficult thing I've ever tried doing in game programming, due to various technical problems (i.e. I'm Not Very Good At This)
So, what I guess I'm leading into is I'm going to list off the various designs I've done regarding the guns and go through the pros/cons of them.
Because why not.
So, here we go!
The first system I had in mind at the beginning was pretty much Borderlands, down to individual weapons manufacturers.
The weapons manufacturers got dropped really quick, though, because it was unnecessary bulk for a randomized game, and at that point it was basically copyright infringement.
This system works by taking a big database of gun parts, like individual barrels, chambers, stocks, sights, etc. and combining them into one finished piece, each part contributing in different ways to the stats.
I introduced a little twist, though, with something that I just call "effects", but are more accurately called "enchantments".
Basically, weapons will also have a chance to be "enchanted", and do special stuff...like bullets exploding upon impact, or leaning the projectile towards the target mid-flight. Or exploding.
I figured this'd make the weapons be pretty interesting, and it would.
...but it would also make it really frustrating, because you're probably not going to beat the game.
And losing all your really awesome random loot is not cool.
It works better for a permanent game, like Borderlands, but not for a roguelike, in which you're not going to keep your save for long.
So, I decided to make the weapons static, while keeping the random effects.
This means that you'd find specific weapons, like "heavy pistol" or "charge laser" or "arc thrower", whatever.
But they'd also have randomized enchantments on them, in some cases, changing them ever so slightly.
I decided that that would be frustrating, because the super awesome endgame weapon you found in level one might end up literally exploding in your face. It takes way too much control from the player.
Which is bad.
So, how do I keep my interesting weapon effects, and randomize my weapons?
Mix the two different weapon systems, and introduce the weapon effects as some kind of skill system.
You find weapon parts (just barrels, chambers, and frames likely) which you use to build your own weapons.
You also buy skills/effects (interesting things like homing bullets, explosives, etc.) and apply them to the weapon.
This brings in the strategic play of the second system while combining it with the random play of the first.
Which is good.
...I can't think of a way to segue out of this, so yeah.
That was the design process in a nutshell.
-
What if the likelihood of certain modifiers being integrated into a weapon changes based on the point of the game that you're in? For example, let's assume the game is separated into 5 acts, and all weapon effects are split into low, medium, and high level groups. Act 1 has a very high rate of a weapon spawning in low-level groups, low rates for medium-level groups, and no chance for high-level groups. Act 2 introduces a rare possibility for high-level groups, and reduces the rate for low-level groups to medium-high. Act 3 continues the pattern, up until act 5, where high-level groups are suddenly common, reversing the pattern seen in act 1. This would help ensure that the weapons progress naturally with the game, allowing the user to encounter some really cool guns without having that feeling of never seeing something like it ever again and making the loss of it a disappointment. You could always introduce a super-rare group that has an extremely-low drop rate that exists across all acts if you want to make the idea of a significant loot drop part of the formula.