Trionic Post Mortem – Audio/Visual Analysis

Posted by (twitter: @djfariel)
April 20th, 2016 10:51 pm

A Post Mortem




Ludum Dare 35 : The Audio Visuals

By Sondrian the Art Horse (Daniel Pulley)




For our fourth Ludum Dare Jam we made Trionic, a game about a shape shifting robot infiltrating a space installation to destroy its mainframe/powercore/core computer/Master Computer. The theme of this jam was Shapeshifting which presented us with some interesting challenges. We decided upon a sci-fi shape-shifting robot as opposed to other venues out of a desire to break our own fantasy mold a bit and explore the futuristic technological side of our imaginations. I mean, who doesn’t love giant lasers and gothic dallek enemies? (Galleks?). Throughout the process, we did a lot of things well, and we did some things poorlyl. For this post mortem, I am going to look at both of these aspects and more from the perspective of Audio/Visual development and design.


The Visuals


At the onset of the game, I thought that it would be really cool to go with vector based graphics. It would be new, exciting, and look really cool. Also, since our previous entries have been largely pixel-art retro styled games, it would be a cool way to stretch our legs a bit. And look good those assets do:


Figure 1.1 – Sweet ass shit terrain tiles


I mean, look at that sweet ass shit! A whole game done in graphics like that would be really appealing to the eye no? You’re damn right it would! But there is a bit of a catch to these stylized, tight, pro-ass graphics: They take a lot of time.


If you want to give a value to your game assets based on the amount of time and energy it takes to produce them, and label them as either “cheap” (little time and/or effort to produce) or “expensive” (lots of time and effort to produce), then vector art is very, very expensive. The above image represents about 4 hours of work. In the same amount of time, I was able to produce a pixel based tile-map of about 20 pieces that would be used to produce the entire ground mapping for the game. Does it look quite as good? Maybe not, but it took 1/10th of the time to produce, and especially where LD is concerned, you need to buy affordable assets, and vector art is too high budget for us. At least at this point anyway.


It was about 5-6 hours into the jam that we decided to change art styles and go back to pixel art. At first I was disappointed about this (I wanted to try something a bit more gorgeous), but once we got going, I find that this was really the right decision. Pixel graphics can be produced so quickly, and for that reason, we were able to provide a lot of content to the game given the short timeframe we had available to us.


As a compromise to the new art direction, I decided to make the effect animations in gimp. (I think the exception to this is the bullets, which were done in Pyxel Edit.) The result of this approach was some really nice looking animations that took the game from retro to neo-retro. I, personally, thought that it looked really neat.

charge-particle bosscube-core-charge-3 death-explosion4 rocket-flames5

I feel that these animations being done with smooth gradients really gave the game a cool, slick look. I mean, that explosion is awesome! Some of the animations that we used could have been done better, but that really comes down to practice.


The player-character himself, Trionic, was an interesting. He takes on three forms, a Mech armed with guns, a tank armed with gravity beam and shield, and a flying form that sports a boost and a minigun. Let’s be honest though, you are going to spend most of the game as a flying space fish.  The challenge here was to make not 1, not 2, but three character sprites that felt good, complete, and whole in a short amount of time. I think we did pretty well in that endeavour.


In order to limit the amount of animation that needed to be done, we tried to find shortcuts and tricks to create the illusion of animation. The tank is completely stationary but has spinning wheels that make you feel like its moving. The flyer has animations on the propulsion wings that make it feel more complete, and it has a rigging of moving parts to so that you don’t just have a stationary object flying about. The mech was the most difficult of the three in that he needs to have a walking leg animation. We actually achieved this by way of a unity plugin that creates 2d bones for the sprite. Then using the engine, we were able to actually animation the walking motion without hard animation redrawing. It worked out pretty well and gave us some tools that I know we are going to use in the future.


For transitioning between the various forms, we had to figure out something that both looked good, and didn’t take up the rest of the jam to create. Sure, we could have animated each form changing into the other form, and that probably would have looked awesome, but instead, we used a simple stage trick. That’s right, a plasma curtain. When the player changes form, a giant beam appears that completely hides the character from view, and when it clears, suddenly you are in a different from. This turned out to be a fantastic choice as it worked phenomenally well.  


What I’m not happy with, what I think we didn’t do as well as we could have, is color choice (things are a bit too saturated to look good) and better animations. A third would be to not have so many similar looking enemies, but that’s mostly a time constraint issue and has nothing to do with mistakes on the development side. Overall, I am extremely pleased with the results. The biggest enemy in the game turned out to the clock.


And since most of us will never see this bad mammie jammie in game…


The Audio…s?

The Music


Music is probably my strong suite in all of this, and I was extremely excited to work on the soundtrack for this game. Now I would have loved to just spend 72 hours writing an amazing soundtrack for this game, crafting it first in notation, and then porting it into FL Studio and making it even more amazing. But then, you’d have great music and no game. Doesn’t work. So I ended up pumping two songs out in about 5 hours and then moving on.


The musical style for the menu screen was a heavy nod back to Halo: Combat Evolved with the spacey choral. I thought it would be fun to do the melody in D Dorian, really just giggles, and I like it. It’s not amazing, it needs polish and added content, but for the time I spent on it, its nice.


The second piece is set in Dm, and using a drum and bass drum pattern to give it a sense of speed and intensity. Add some Harmor effected synths and it basically wrote itself. I would have liked to make the content longer as it can get redundant, but overall, it’s a nice back track for the game. I didn’t get the chance to write a boss battle theme, and that made me sad.


The Effects


If music is my strong suite, then sound effects are my achilles heel. I probably spent half of a day on the sound effects, either generating them purely from scratch, mashing them out of some basic royalty free samples as a basis for things I didn’t have access too. (I don’t have a helicopter lying around y’know?) or recording them in-house. I think that I did a better job than in past projects in terms of making the sounds suitable for the purpose.


In “The Darkness”, we have a pretty big problem with sound effects being absolutely too loud. Granted that game was producing thousands of sounds per second at times, but it was a major issue that I wanted to be sure was corrected in Trionic. Stereo shifting the sounds to distant helped take them out of the foreground and push them back into the main game mix. The result was sound effects that could be heard, could overlap one another, and keep the mix from “Exploding” in the player’s ears. Which I call a big win. I noticed this while watching footage from League of Legends. The Sound designers make the sounds distinct, but suited for the background of the soundscape of the game. Following this example, I found a way to improve our balance of music to sound effects without dampening the volume on either too much. Some of the guns still sound like dolphins, but I’ll just have to live with that. 😀


Going forward, I want to include a wider variety of sounds, and recycle elements for the sound design a lot less. The goal is to produce sounds that are unremarkable in that you just expected that sound to be there, sound like that, and produce that sense of feedback and immersion. I have heard it said and it is true; Good sound design is not something that you notice, but remove anything and suddenly the player can tell that something is missing.


If you would to give Trionic a try for yourself, click HERE


Thanks guys!

Leave a Reply

You must be logged in to post a comment.

[cache: storing page]