Summary - The Function of Sound in Games

Paul Arnold, Andrew Barnabas
The Function of Sound in Games
(retrieved from http://www.bobandbarn.com/articles/index.php?page=soundgames)

The article briefly describes the different areas of sound in games and its subregions. Generally, audio in games can be divided into:

- Music
o Linear
o Non-linear
o Source Music
- Sound Effects
o Spot Effects
o Ambience
o Foley
- Dialogue
o In-game
o Non-linear

Music
Music generally enhances the mood and atmosphere, and can help the pace. Linear music is used for things such as ambient music, in cutscenes, or any other place where the music doesn’t have to change. Interactive music is reasonably unexplored because music is linear in nature. This is either negated by trying to establish the general mood of a scene and scoring that, or writing specific cues for specific situations, like in film music. However, due to the vast number of situations possible in a game, this poses issues in terms of development time and budget.
A final variant of music in games is source music, which is music that characters in the game can hear or respond to, whether the actual source of the sound within the world is clear or not.

Sound Effects
Sound FX generally reinforce the current actions. Spot Effects account for most of the effects synchronised with entities in the game. Ambience sounds include all diegetic continuous sounds in the background. Foley sounds are all sounds that don’t convey important gameplay information, but instead are small sounds that can greatly enhance the sense of realism in the game.

Dialogue
Dialogue brings characters to life and can convey more complex and specific information, as well as create emotional bindings with characters. Linear dialogue is text that does not need to take into account the various possibilities of a game, for instance during cutscenes and events that will always occur. In-game dialogue however can depend on how a player has played the game thus far and will have to take into account many situations.

My first impression to online matchmaking

The last few days I have been busy playing most of the games from the list I posted before. I didn’t have the time to play them all yet, so let me start with the genre I had the most difficulties with. Namely RTS(Real Time Strategy)Although I’m not unfamiliar with RTS,(It’s actually one of my favourite genres in pc gaming) I have to admit that playing online isn’t one of my strong points. I normally enjoy myself by building op a defensive base or a huge army before engaging the enemy. This is considered “turteling” by experienced online gamers, because it takes quite some time to establish this. The trick in most online RTS games is that offence is the best defence, and you need to be quick about it. So spam as much units as possible as soon as possible and attack your antagonists before they attack you.

Starcraft 2
Last Friday I received an emailed notification by Blizzard that I have been invited to participate the Beta test of Starcraft 2. I originally wanted to use Warcraft 3 as a case study but since Starcraft 2 is a more recent title and it’s as conventional as RTS games get, I choose to use Starcraft 2 instead.

The first thing I noticed after creating a profile was the ability to indicate a skill level:

• New player - little rts experience
• Experienced – played rts in the past
• Veteran – you played rts for competitively years

Although I don’t really know what the consequence of this choice is, I presume it predetermines a rank. During my test u choose the experience option what plces me in the copper league, which should be one of the lower leagues in the games ladder system. Since the game doesn’t provide in depth information about its ranking system, I looked up it on Blizzard Battle.net page and its supporting forums.

The system should work like this:
Starcraft 2 places its players in a ladder system with ranks and leagues. Each league represents a certain skill level and each rank represent the player’s position within that league. Winning matches raises ranks and should eventually raise leagues. Unfortunately it’s unknown what specific statistics are needed to climb up a league. Losing matches Obviously does the opposite of winning the players ranks.
Rob Pardo (Executive Vice President of Game Design at Blizzard Entertainment) even claims that Starcraft’s matchmaking is to good, which result into exhausting matches. A form of randomness is added to create more diversity.

Encouraged by this system I started up the game full with hope and promise, to soon find out that I’m not by far up to the task to defeat its current players. I quit the game after five humiliating defeats, and I would have deleted the game of my hard drive if it weren’t for the optional A.I. opponent.

Luckily there is still hope, Starcraft 2 is still in beta and is not working in optimal conditions. I can only hope that the final product will be more friendly to less experienced online gamers .
Lets move on to another game

Dawn of War 2 (+expansion)

As some of you might know, Dawn of War 2 took a step away from the conventional RTS style of gameplay by taking away base management. There is no base building at all. Instead the game focuses more abilities of the units itself and on character progression trough a RPG system. Note: the RPG system is available in the single player component. This system is replaces in the multiplayer component by optional gear and a tech tree.

The first thing I found out after starting up a multiplayer game is that Dawn of War uses TrueSkill, a ranking system that is also used for xbox live matchmaking. TrueSkill is developed by Microsoft Research. From what I understand, TrueSkill distinguishes itself from other ranking system by keeping track of not one but two values. Average skill and the degree of uncertainty. The system then compares these values with that of the other players to make a match. TrueSkill players that are most likely to draw each other. Off course there is complex underlying mathematic system that is used to make these comparisons, but that’s not the point of interest to my research.

What is important to know is that is system like this works best when there is a low level of uncertainty. The only way to gain is to have a lot of data on the player. In other words, the players needs to player as many ranked matches as possible to get a good match. And that’s where the system flaws, at least to the appeal of new players.
To test the system I played an online ranked matches of Dawn of War 2. I’m still a level 1 player en got teamed with a level 3 players. The level of my opponents on the other hand where dangerously high, 37 and 48. Obviously the game ended up in a big humiliating defeat. The other matches that I played where very similar to this one and also ended up lost. Note: being defeated like this does not encourage to play another game is not a warm welcome.

The flaw is that new players or not likely to continue after being defeated in a game in which they have no chance of winning. By this the matchmaking system never gets a change to get enough data..

Introduction to Tom's research

Right, high time to put my info on this blog:

I'm Tom van Kruijsbergen, Django's team mate in creating what will hopefully become a great game, as well as fellow student at the Utrecht School of the Arts. I tend to be drawn towards the systems behind games, and how they can be used to create an experience for its players.

My "skills" lie in areas such as game prototyping, designing, tweaking, and thinking about stuff™. I enjoy testing out a game in small concepts, and watching it grow into something bigger and better over time. I strive to get my games playable as soon as possible, because I believe that this core experience is ultimately what it's all about.

My research, then, is about game sound, and music in particular. Game audio is a field that is becoming more and more interesting as of late, and particularly because I think more and more developers are starting to see its potential.

Generally, sound in a game serves to support what is happening in a game; roughly, music articulates setting and story, ambient music enriches the environment, sound effects bring small events to life, and dialogue communicates all sorts of other things (multiple models exist for this division, examples to follow. This is obviously a very rough outline).

So sound generally supports game events. The problem lies in what is actually being supported. The impact of any piece of art on you as a viewer is very great if all the source's elements work together in communicating its meaning as precisely as possible. If a painter wants a painting to make you feel happy, he will make sure he uses the right colour and lighting in his painting. If a film maker wants you to feel sad when the main character's sidekick's aunt dies, he'll use different colours, lighting and camera angles. There'll be a sad sound playing to further prove to you that this is definitely a sad scene, and so on.

In games, however, this is harder to achieve. This is because you can't know when a character will die, or if he even dies at all! You also can only vaguely estimate what is going to be happening in the next five minutes; it's up to the player to make the next five minutes happen. Because of this uncertainty, it has traditionally been hard to closely score events. Cutscenes are very often used for this purpose, because you as a designer do know what will happen in the cutscene, or games are designed with the idea that the level will lead up to defeating the dragon anyway.

But what if that outcome is more uncertain? What if you're playing a game where it's not sure whether you'll win or lose at the end? How can you make sound support a game, if what happens during that game is unknown at the start?

That, in a (pretty long) nutshell, is what my research is going to be about. I'm going to be looking more closely at that problem and particularly solutions, and then incorporate the solution into a game.

Blizzard.net Matchmaking

Just found out about Blizzards new matchmaking system.

Check it out: http://eu.battle.net/matchmaking.html

A nice list of games

I composed a small list of games that should be useful for my repertoire research. As you might notice, this list is not genre specific. I kept the list cross genre to keep an open mind.

While playing these games ill be paying close attention to matchmaking systems and in game balance itself.

UPDATE: Changed Left 4 Dead to Left 4 dead 2, changed Halo 3 to Halo 3: ODST, and changed Warcraft 3 to Starcraft 2 since I'm now participating in the beta test.

Genre	Name
Firstperson shooter	Left 4 Dead 2
	Modern Warfare 2
	Halo 3:ODST
Racing	MarioKart Wii
	Gran Turismo 5: Prologue
Platform	Little Big PLanet
	Super Mario Bros WIi
Real Time Strategy	Command & Conquer 4
	Dawn of War 2
	League of Legends
	Starcraft 2
Fighter	Street Fighter 4
	Tekken 6

Introduction to my research

Nowadays it’s very common for single player games to also offer an added multiplayer component to extend the lifecycle of the game. The problem is that most multiplayer games assume that the player has experienced the ins and outs of the whole singleplayer experience.

The difference is that most singleplayer experiences often rely on a slight learning curve that slowly builds up. The result is a gray area between the singleplayer and the multiplayer component that needs to be bridged. A lot of players don’t bother to check out the multiplayer component or quit after having touched it briefly because of this problem.

Another big problem is the nature of gamers who do play multiplayer games. Experienced or hardcore multiplayer gamers expect that every player they meet has a certain level of understanding of the game, especially when teamplay is involved. Not being able to meet this level leads to frustration on both sides and ruins the experience for both types of players. The player eventually quits playing the game out of frustration. This benefits neither the player nor the developer.

To solve problems like these, game developers use matchmaking systems that divide players into different skill levels, and place the player into a lobby with players of the same skill level. Although this sounds good in theory, this solution doesn’t always work like it should. This leaves many gamers into matches that are still too hard or too easy, which discourages the player to continue playing.

Systems like this also rely on player population, and only function optimal with a huge players base that can only be found in AAA games. A smaller player base results into long waiting times to find players of appropriate skill, and most of the time, players are put into matches with very varying skill levels anyway.

The big breakthrough in online gaming should be a system that makes it possible for hardcore and casual gamers to be able to enjoy the same competitive multiplayer experience, without being divided into different individual games.

During the following months I will be researching fun and frustration in competitive multiplayer games. Conducting this research will hopefully help me to design mechanics or a system that can be incorporated into exciting multiplayer games to enhance the player experience for both new and veteran players. This leads me to the following research question:

“ What game design or level design elements can create an equal (artificial) chance to win and agency for both veteran and new players within a single level? ”

I also will be building a game to as prove of concept and to support my thesis. The be able to build this game I teamed up with Tom van Kruijsbergen who will implement a experimental audio system into the same game. More information about his research will probably be posted soon.

Let me introduce myself

My name is Django den Boer, I’m 26 years old and I am a graduation student at the Utrecht School of Arts, Game Design and Development, Master Degree.
I’m currently employed at E-semble as 3D artist. And as you might guess, most of my free time is spend playing as much games, a good designer needs to know what’s out there. Of course I also have a social life, I own a house in which I live together with my girlfriend and have a rabbit that growls.

Over the last four years I have been schooled in almost every aspect that is connected with the design and development process of videogames. Being very passionate about almost every aspect made it a little bit hard to stick to one specific specialisation, instead I picked three: game design, level design and 3D art.

During the next few months I will be conducting research as part of my graduation. This is also a good opportunity to study more in depth to get more specialized. The only question that remains is, what is the best way to go.
For me, game design and level design go hand in hand and it’s hard to take the one without the other. But to be able to come up with a strong research question, I had to pick one. Answering this question took me quite a while, and is also the reason why I am a bit delayed with my field research. So to make it final, I like to make 3D art, and I love designing 3D spaces and game levels, but I am a game designer and I will be conducting research as a game designer.