I took it easy at a family cottage this week, but I still got work done! I mostly fixed up the environment art for the fifth level, which takes place inside the warship itself. Every time I make environment art and tilesets, I create little sample test rooms to try them out, like this:
It’s okay, but I think it’s too busy and lacks contrast. It also doesn’t seem visually connected to the previous level, which is the warship’s exterior.
I want the colour scheme and the general look to be different to keep things fresh and make it clear that you’ve reached a new level, but I also want it to seem like the two areas are related.
The fourth level is done!! Woooop! It’s been a while since I showed off a boss, so let’s do that.
The boss is a drone from the warship that spams chaff everywhere, which it then manipulates to attack you. I call it… the Chaff Spammer. Yeah. These names aren’t final, don’t worry.
I very nearly cut this boss — I really like the idea of its gimmick, but it took a long time for it to come together into something pure and enjoyable. I’m showing before-and-after gifs of its attacks, but they’re numerous so I’ve put them after the break.
Personal life took some time from work, but I’m almost done level four — just the bosses left to touch up / balance. In the meantime, let’s explore render targets a bit, since I used those this week!
The fourth level takes place on the hull of the warship, and at some point it starts heating up as it breaks the Earth’s atmosphere (interesting scenarios trump scientific accuracy, obviously.) You’ll notice the issue I was running into — you can see this weird grid effect when the ship glows.
So what up with that? Well, the ship is drawn using tiles, and in this case I’m using two tilesets — one where the warship looks normal, and one where it’s all fiery. The hotter the warship gets, the more opaque the fiery tileset becomes. The problem is that tiles overlap slightly — you never notice when they’re opaque, but when they’re transparent the overlapping edges blend together and cause that nasty grid effect.
This week I continued re-doing parts of the fourth level. It’s gonna take at least another week, but I hope it doesn’t take much longer than that…! I don’t want to show much of what I was working on, but I did stumble across a forgotten cache of old stuff from the original Bleed so I thought I’d share some of that this week. The centerpiece is an alpha of the original Bleed, from 2011!
You can download the alpha here. Revel in the ridiculous number of hit points each enemy has. Laugh at my delusions of creating 25+ bosses for the game. Be angry at how insanely hard it is.
This week was the start of fixing up level 4! A year and a half ago I’d just finished the background art for it and posted this image to Twitter. Weird to think it’s been that long.
Level 4 is the one I made first to test everything out, and as a result it’s one of the roughest. I have to re-do a lot of its layout, so I thought I’d document the process!
The level layouts are created in the 2D tile editor, Tiled. I try not be concerned with visuals at the beginning, instead just boxing in a rough idea of how the level will play. I do know how the level will look and what enemies will go in it, so that helps inform the design. I just need to get something on there, because a blank canvas is really intimidating.
This week I balanced all the bosses of level 3, meaning the third level is looking pretty done!
When I say “balancing” the bosses, I’m just talking about making them challenging, fair and fun for all difficulty levels. A lot of it is adjusting values like how much damage they do, how long they give you to react before they attack, or how fast they move… among many many others. There can also be a lot more to it than changing values — if I decide an attack is unfair or not fun, I’ll re-do animations, hitboxes, AI, whatever.
This week continued the editing/tightening process for the game by moving on to level 3! A friend of mine asked me to define “editing/tightening” in more concrete terms, and I think this week’s work provides some pretty concrete examples.
Maybe it sounds tacky, but I find most games have taken on a life of their own as I’ve developed them. The more I build and play around with Bleed 2, the more I develop my understanding of what makes the game challenging and fun, and the better I can tailor it to play to its own strengths — at least as I perceive them. So “editing/tightening” is me going back over my work with this new understanding, making the whole thing stronger and more unified.
This post contains a few large gifs, so most of those are after the break. But, here’s a snippet of how Level 3 used to play:
I have a strong love of shmups like Gradius and Ikaruga. Most of Level 3 takes place in the air on the hull of a ship, so it seemed natural to pay homage to some of those games. Enemies come in familiar Gradius-esque formations, and would contain patterns of purple and yellow that mirror Ikaruga. But playing it now… it’s pointlessly cute and not too relevant to what Bleed 2 should be. The enemies are so slow and are barely threatening, there’s not a lot to reflect or dodge, it’s not very stylish and attacks come pretty much from one angle at a time. So…
Level two has finally reached a quality where I’m pretty much done with it, so hurray! I spent the week refining cinematics, tweaking enemy placement, difficulty balancing everything, and putting the finishing touches on all the bosses. I’m feeling accomplished, and am hoping the next couple levels don’t take quite as long, since I had to re-do so much of this one. I guess we’ll see.
All the changes I made last week meant a complete re-design of the second level’s layout. It’s been a fun process watching it go from an embarrassment to something I’m having fun with, and it’s got me thinking about what constitutes good level design. I’ll take you through my thought process so far, and maybe it’ll apply on a broader scale (but probably not, haha.)
First, I had to think about what purpose non-boss segments are actually serving. Bleed 2 is an action-packed boss bonanza, so it’s tempting to dismiss everything in-between as a formality of the genre, or the filler between the main events. However, I’d argue these sections actually improve boss fights because they break them up — if the game was just back-to-back bosses you’d grow numb to the thrill of fighting them, and they wouldn’t feel as special or exciting.
So in this case, the non-boss segments are a palate cleanser to get you ready for the next big fight. Once I understood that, I had to figure out the best way to create segments that perform that function. I decided on some key points: they should be short, easy to navigate, maintain interest, and above all make the player feel powerful. I’ll briefly explain how I tackled each point.
So, at this point I’d say Bleed 2 is close to an alpha state! …you know, kind of. I left the final level half-finished to create the demo for TCAF, but that aside the story mode is playable from start to near-finish. The only problem is that it kind of sucks right now, but don’t worry (yet) because that’s part of the plan.
Rather than an “alpha”, I’m considering it more of a “first draft”, like I’m writing an novel. Now I get to “edit” it, going level-by-level refining and tightening until it’s a slick, cohesive package. I’ve already finished the process for level one since I had to prepare it for TCAF (yay!) so this week was on to level number two.
The second level is where the Kitties and Lil’ Guppies make their re-appearance, but they’re mostly physical attackers. There isn’t much to dodge or reflect for the whole level and as a result it feels pretty boring… so, I redesigned some things:
The Kitties barf up big energy balls instead of hairballs now, so at least you can reflect those! They also wear bandanas so you can tell their colour before they attack.