🕑 Added 2014-05-16 06:34:09 +0000 UTC
I've been thinking about making a video game for probably as long as I've been making my comic. When I first started Dead Winter I always thought of the characters in terms of what stats or abilities they might have in a tabletop RPG setting. I never had a comic shop near me growing up so I never got to read too many comic books, but we had a video store up the street and we rented a -lot- of video games, so playing and studying how games click together and flow has definitely shaped a big part of my youth, and when I ended up creating a little world to share with the bigger, realer world it was inevitable that I'd figure out how to make a game out of it. When I got older I finally got access to a computer that could play online games and I took an interest in competitive gaming- a world where "balance" and the math and the mechanical harmony of the game becomes paramount to the user's ability to do well at a game and feel rewarded for their skill- and I started to really understand how skill-based games clicked. That's the brief history of who I am and where my background is, so what about the game? What is it? What's the plan? You probably know the general details by now- that it's a brawler, the missions are randomized, there's four characters and they're different. For my first developer's log post I'm going to expand on the basic premise and explain how the details all work, so future posts on the matter have a solid foundation to build off of. This first post is going to break down the mechanics of how a random mission is assembled, what pieces go into it and how they all click together- that's really the core of the game so it's the natural first thing to expand upon. It seems simple on its face but we've put a lot of thought into how to make the randomization meaningful, and that goes as follows: The idea of the game is to build missions out of levels, and have those procedurally-generated levels reflect the types of encounters you could expect to find in a classic pre-planned arcade cabinet. There are plans for multiple settings- background and environmental themes- to be constructed into levels, which are then chained together into missions. A level itself is divided into two halves, the ground and the background. Backgrounds are designed to assemble backdrop elements seamlessly together to create different scenery with each playthrough- while a background piece itself might become familiar, the idea is to chain them together in random ways to create the illusion of having never seen them together that way before, to help break up monotony. The ground is where all of the action happens, and depending on the setting a number of ground templates could chain together to make one level distinct from another. Things like segments with pitfalls, segments with lots of cars, segments with hazards you need to walk past or jump over, segments that act as bends in the road, as cliffs or ramps or intersections; all of these things and more can be clicked together to create levels that vary from playthrough to playthrough without being "another flat straight line" or "random, but the same as every other random output". Ground hazards will vary depending on the setting within a level to help differentiate one level from another, and another way to do that is to give each setting its own distinct visual theme. We want to capture the feeling of moving from one distinct level to another so you may start a mission on a street level filled with cars, then at the end of that you'll end up in a park level with trees and water hazards, and then a bridge level full of pitfalls, onto a rusty beam-filled warehouse, a rooftop, sewer, a snowy setting, supermarket or more. When you play a brawler each level is distinct and we'd like to capture that feeling with each playthrough- the random order and assembly of each mission would make it hard to prepare too far in advance for any given encounter and keep a player on their toes while still keeping them familiar enough for a player to say, "okay this is a street level, watch out for cars". The backbone of a brawler is in its enemy design. A level doesn't amount to much if the enemies are a stale pushover, which is why we've planned out how to populate each level with different enemy encounters from mission to mission. The basic idea is: there will be anywhere from six to ten different factions of enemies, each with their own theme. Each level would populate with two randomly-selected factions, with a chance to roll a double serving of the same one faction, and that's when something really big happens. There's a lot to say on enemies so I will save that for a future post... We want our game to be challenging, but we also want it to be flexible. One element that inspired me a long, long time ago was when I first discovered an old LucasArts game called Yoda Stories. This was essentially a randomized adventure where you could control the scale and complexity of your random puzzles before an adventure develops so you could suit it to match your taste or time restraints. Using our own assembled-mission framework we'd like to capture this same degree of control. When you create a mission you'll be able to choose a number of variables: How many levels is it? How difficult do I want the enemies to be? How forgiving do I want to be with myself? You could design a mission to be a single level, medium difficulty and three lives for a quick 10 minute game on your lunch break or you can set up a sprawling eight-mission campaign- but make it easier- so you and three friends can spend the night trying to make it to the end on three lives together. Or maybe you're one to abandon any chance at survival and play Hardcore One-Life Endless Mode and see how many increasingly-difficult levels you can survive before you inevitably fall. The last element to touch on regarding random level assembly is the Objective. It's all well and good if the only implied goal is to fight your way from the left side of the screen to the right one, but we have an idea to randomize level incentives too. When you set your settings and boot a mission you'll be greeted with one of a number of opening cinematics establishing the purpose of your mission. Anything from "your rent's due" to "There's a dog... let's pet it!" to "Channel 4 is on fire!" will give you purpose to fight, and at the end of your mission the last level chunk to load will contain the corresponding objective: a van full of food, an adorable dog to pet or a burning news building you're not entirely sure why you bothered to rush to in the first place. There's a lot of inspiration in this element beyond just having an objective tacked onto the mission: at the end of a level in Metal Slug the player to score the last hit is rewarded with a special victory animation, in Castle Crashers players are pitted directly against one another for the final reward and in The Simpsons arcade game there are occasions where players race to be the first to complete a simple minigame. Mission Objectives will be something of a combination of all three of these things: you'll want to be the first player to reach the final objective to earn a bonus little victory at the very end of your mission- the first one into the van's driver seat or the only one to actually get to pet the dog, for example. But what if two players reach the objective within a second of each other? A quick arm wrestling minigame to push yourself to that final victory! It's a little competitive spice in an otherwise-cooperative adventure, and a thematic way to keep it from simply being "start killing ---> you've killed them all!" over and over. You're not there to beat guys, you really want to pet that dog! That's the long explanation of how the core of the game is being built. Pools of random pieces clicked together into distinct levels, strung together into missions of different lengths populated by different combinations of different enemies, each with one of many objectives and rewards. We're doing our best to build the framework of the game before we can assemble all the little pieces to fill it with but once it starts assembling itself the ball will really start to roll. I hope all of this makes as much sense on paper as it does in my head- we've done the math and the math is sound! Thanks for reading and thank you for your support!
