Blog

Cappin' Stones (11/2/2016) Part 2

With a rhythm based game, sound is fairly important. We had this pitch early on to have a sound system that dynamically brings sound in and out as you walk around a level. Something akin to the game Proteus

In Proteus, the beautiful sound and music "design" by experimental musician David Kanaga, isn't actually designed to provide a specific experience to the player. Some sounds play only in certain seasons or with certain conditions, sure, but Proteus was designed to dynamically create music with over 500 short sound samples, ranging from 1 second long, to as long as 2 minutes. Certain key moments of the game will make pre-determined "songs" that act as themes to experiences, but for the rest of the game, the procedural generated islands with randomly placed music playing objects provides the player with a completely unique musical experience. 

A small section of Proteus's sound folder, intrepid game designers and modders and built their own games in the Proteus engine by replacing sound files and model assets (also exposed) creating a new game with the same gameplay

A small section of Proteus's sound folder, intrepid game designers and modders and built their own games in the Proteus engine by replacing sound files and model assets (also exposed) creating a new game with the same gameplay

Playing around with replacing and switching around various sound files, and playing with various Proteus mods had given me the idea to try taking this dynamic route to sound design, with no game ideas to follow up on. As we started work on a rhythm game, I thought back to my experiments with Proteus and wanted to create a sound library like that. We obviously would have a beat for the level, but each security system and element would have it's own sounds that would dynamically merge in and out of the audio mix and you progress through the level, essentially creating "music" with the game. Our professor told me early on that I needed to prove this system, and so while the programmers got to work creating a sound radius that the player character can hear:

Visualization of sound.

Visualization of sound.

I got to work on creating specific sounds. Keeping to the Proteus mentality of sound design, I tried to keep everything musical. I wanted the guards to have a cartoony Bassoon melody, and after finding staccato notes on freesounds.org I pulled together a 4 note loop. Lasers were cold and mechanical and after finding what sold itself as an "online theremin", I created a punchy, bassy electronic drone. When it came time to find a sound for a security camera, I blanked. I had already used a synthesizer sound for the laser, and wanted to come up with an audio "walkthrough" for the level, so I decided on a temporary sound of a film projector. I threw all these sounds in Audacity, added a simple click track, and out came this:

It's rough around the edges, but it got the point across. The sketch posits the player walking from left to right, with sounds entering from the right and leaving on the left. If you use headphones, you can hear the sounds enter from the distance, linger, and then leave as the character walks through. I could have sold this element more with reverb, but trying to get a quick sample out the door to prove the concept worked. I later found an 808 kick, and an 808 clap on Freesounds, and sent each sound file to the programmers, each organized in their own folders in the repo with files explaining how to use them.

And the results... are mixed. The beat definitely works, though some people though the clap was another element. I wanted to have a two tone beat sound, but after feedback, a one tone beat may work better. Also the bassoon sounds sound cheesy with the rest of the sounds, and all pile on top of each other. What works as a 4 note solo above, sounds like a mushy mess. And the laser could use more work to make it more apparent. But beyond all of that, it was a great start. Once people heard the audio they were more sold on the rhythm aspect of the game than ever before. Despite some implementation issues, the first pass of audio was solid. Specific sounds need a bit more time to get the right feel, and not feel discordant, but overall I'm happy with where the sound is right now. I could have helped work on some of the implementation issues, except the weekend that they were being put in, I was doing SCRUM training...

Luca Hibbard-Curto
Cappin' Stones (11/2/2016) Part 1

Wow, A lot has happened. So much so that I haven't been able to write a blog post in a while.

We have been hard at work on the game, getting art assets in, scratch audio, QA testing the game, getting feedback, and getting out of deep dive. I figured I'd break this blog post into multiple parts. How many? As long as it takes.

Going into deep dive, we needed to reprogram the game from scratch. The original pitch prototype was in 3D, while the game was in 2D. Work on this was slow, initially, focusing a lot on the beat system and making things easy to make levels with (see last post). What we ended up with looked a lot like this:

At this point, the game was functional, it had working guards, with paths they follow (the blue "N"'s serving as a visual representation of points the guard will walk to.), a laser system, (seen with the red line coming from the wall) and a beat. At this stage, there was no audio beat for the rhythm, and only a visualizer down below. When a beat would happen, the two would meet. In addition, all the security systems moved to the beat. We took this to QA and got strange feedback.

Without audio for a beat, people were confused, but some people actually found the game perfectly playable without it. Having everything move to the beat, at a consistent rate, gave some people enough feedback to get through the game perfectly. While some were dumbfounded. We asked people what we could do to improve the rhythm aspect of the game, and we got a lot of people saying "add sound" while other people gave us much more ingenious ideas:

Our current rhythm visualizer is out of the way from gameplay, at the bottom of the screen and kind of distracting. It could be replaced by a flashing at the sides of the screen. Other people wanted "tells". While the guards and lasers move at fixed beats, the guard can feel "unpredictable" as there is no animation on the guard. A lurching forward animation or a head bob could go a long way to making the beat feel great

We also got lashed for the lack of art. Everything in this build was programmer art, due to inability to communicate with our artist. Our artist was seemingly impossible to be reached, and promised art over and over... but didn't deliver. We were frantically shuffling around to get art in, and eventually got some assets, but only a few usable ones.

With what we had, we decided to try to challenge out of deep dive, only to be told what we already knew: we need art and sound. Luckily, I had been working on sound the last week, and our artist promised to get art to us as soon as she could. Which leads to the next part.

Luca Hibbard-Curto
Cappin' Stones (10/13/2016)

Woah. I let this blog get a little rusty. Sorry about that. 

What has happened in the last 2 weeks? It turns out... a lot! We challenged, and Rhythm of the Night now has been accepted as our game we are going forward with. As you saw in the last update, the prototype was in 3D however, the game was always envisioned as a 2D game. It makes making levels easier, and the art load is a lot lower for our single artist. 

Since you last saw progress on the game, the whole of the game has been rebuilt from the ground up. It's not quite 100% there just yet, but the rhythm system and event handler that passes a beat to all other systems of the game are almost fully in place, and systems are being made to make level design easier.

Gif Courtesy of Dan Covert

Gif Courtesy of Dan Covert

We now have a system that automatically locks tiles to a grid system, meaning when it comes to making levels, all I have to do is drag and drop. We've also been working on getting proper implementation of camera and laser systems up and running, the two easiest security systems to get up and running early on. Once those are in, we can start making some test/real levels and building out the game from there. Early on, getting a pipeline that makes level design quick and easy has been our biggest bottleneck, but getting it right means that we never have to think about it again down the line.

Lasers gif by Dan Covert

Lasers gif by Dan Covert

There's been a lot of talk about the specifics of implementation, and debate on how moving to a beat actually works. After a large conversation, and some debate, how beats work has been broken down in this example:

(timing and numbers are not exact here. This is massively simplified)

Imagine a period of 10 seconds. In those 10 seconds, the player can push the movement button, and the character will move. However, the last 3 seconds of those 10 seconds count as "correct", and moves you to the beat. If you move in the "correct" period, either as early, as late, or in the middle as you can, the game registers your movement as the start of the next beat. Systems will update along a beat, audio shifts slightly, and visuals update, allowing you some leeway with the beat and not making you feel punished if you are actually off on a beat by a little bit. However, any movement in those first 7 seconds still count, though they do not count as an on beat movement. The player can move freely during this time, but there's only so many times they can move. Plus moving off beat has it's consequences, dropping your score, and alerting security systems that are designed to react to off beat movement. Now, take this imaginary period of 10 seconds, and shrink it down to what you think of as a single beat in a song. There's leeway around the correct timing of the beat, and you can still mess up and move off beat if you aren't timing it correctly.

Lots of work is being put in, in order to make beat based movement feel good, and not a chore, or something that doesn't feel off. We want the constant forward momentum to drive the player forward at all times, and think on their feet when it comes to finding the best ways around obstacles. 

On the design side, I've been hard at work touching up our Design Document, and turning it's updated values into a VDD for the team.

Lower quality version of the VDD. The full thing is a massive, scale-able PDF.

Lower quality version of the VDD. The full thing is a massive, scale-able PDF.

I'm not the best artist in the world, but luckily our systems are fairly easy to depict based on vision cones. Using highlighted tiles, and semi transparent vision cones, I was able to show how far each system can see, what it looks like rotated, how many steps does it take for a guard to turn (and where is he looking while he does?), etc.

This will probably become more cleaned up and simplified later on, if we get a chance to add more systems, but for now, it gets the job done, conveying direct metrics of everything we're working with.

 

Luca Hibbard-Curto
Cappin' Stones (9/24/2016)

A shorter update this week.

Things this week have been progressing smoothly. All of our prototypes are in a position to "challenge" and go forward with, and the majority of the week for people has been spent writing documentation or hashing out fine details in things. 

Quite possibly the biggest piece of the puzzle was a meeting the whole team had on Wednesday. In this meeting, we hashed out a plan to get everything on track by this Monday, but not to be too afraid to let it slip to next Monday. In the meeting we decided what game to go forward with (the Rhythm Stealth game, now tentatively titled Rhythm of the Night), and our team name (Zamboni Dreams).

The VR platformer prototype is now in a somewhat functional state. Dan took over the programming and was able to create a prototype that is a combination of real perspective physics, as well as fake perspective physics in order to get the prototype up and running.

Meanwhile, Dan has been working on the Zone of Chalk prototype. The Zone of Chalk prototype was another incredibly tricky one, as both it, and the VR platformer contain ideas that are very hard to program, and thus got the axe. As much as we loved both ideas, the thought of going forward with a game we can actually make and finish is more appealing than bashing our heads against trying to ray-cast physics from a VR perspective.

Zone of Chalk, in prototype state.

Zone of Chalk, in prototype state.

Going forward, we plan to work on a game, currently titled Rhythm of the Night, which I don't believe I've fully explained.

Rhythm of the Night is a top-down, 2D stealth game, based on rhythm movement mechanics. This means that you can only move from tile to tile on a beat, and cannot on off beats. You must navigate a house, bank, or other locations and steal as much stuff as you can (stealing may be done on off beats), while also going from your entrance point to the exit without being caught. In addition to you moving on Rhythm Mechanics, so does the rest of the game, guards will patrol paths, moving one tile at a time on the beat. Lasers switch on and off to the beat. Cameras swing around to the beat, so forth and so on.

The game is non-combat oriented, so being detected by any of these causes a fail state for the level, so it's imperative to sneak through a level while remaining undetected. Players have powerups, that can slow down the beat, turn it off entirely, or get away from being detected by throwing down a smoke bomb that can help them navigate through various levels of all sorts of security systems. How many security systems? It's too early to tell, but the design document currently has 2 pages worth of security systems and numbers of how they all work. This prototype gif should be able to help illustrate the point (I keep getting caught by the laser)

Speaking of design documents, every game has their own 4-5 page full design document planning out early thoughts, and things to go forward with for the vertical slice, as well as things for next semester. These documents are living, allowing me to go into them at any time and add and remove ideas so nothing is set in stone just yet.

On to challenging!

Luca Hibbard-Curto