It was the third time I taught this class, I enjoyed it a lot!
I hope you enjoyed taking it, too
Also check out my master's/PhD level class on Machine Learning and Statistical Learning!
Y'all were a great (and patient) audience!
I hope you all had fun with your projects
If you have any intentions of publishing them in any form, I'd appreciate it if I got a link or something
Please participate in the official evaluation
You can also let me know if there is anything you particularly liked/disliked (for example by posting anonymously and/or privately on Piazza)
Beyond just "classical" software bugs, games, as complex systems, also have many other sources for errors
Some of these might be hard to find or judge
For example, how do you determine if a mechanic is "fun"?
How do you judge if a game is "easy" or "hard"
Companies are trying to get players involved earlier
When you get to this room, you have to talk to Shad before you save and quit. He will not be there on reload, and is necessary to progress.
Pool Of Radiance: Ruins of Myth Drannor was a mediocre RPG, but when you uninstalled it, it would also delete some important Windows system files and your computer wouldn't boot anymore.
We already discussed "easy" vs. "hard"
But what can you do to change that?
Also, what if players actually play against each other?
And which rewards do you give players?
The best basis for balancing is math
You want to know the relative strength of every item/unit in your game
Build an Excel sheet (or code) that can tell you how good any combination of items/units is
Tweak numbers in the Excel sheet until the relative strengths are what you want
And then playtest!
Many players hate Random Number Generation (RNG) in games
But it actually serves important purposes
Randomization helps with replayability
It also helps with player engagement and retention
The problem is "too much" randomness
Many games use randomized levels to keep things fresh
Card games typically shuffle the deck
Bobby Fischer suggested Chess960 because he felt chess got stale from people using memorized openings
If the outcome of a match is less certain it is also more likely that you'll try again if you lose
Conversely: If you lose, you can always just blame "bad luck", rather than admitting that you made mistakes
Studies in rats: If a lever gives a reward every time, or consistently every n times, the rats actually get bored and stop using the lever. If it gives rewards randomly instead, they will keep trying.
If you give your players a legendary item every 10 bosses they are more likely to get bored than if they get it randomly 10% of the time
People have trouble with perceiving randomness, though
If a player kills 20 bosses and doesn't get a legendary, they will claim the game is broken, even though that should happen in 12% of the cases
1.5% of players would even have to kill 40 bosses to get a legendary
Now we have a problem: RNG is good for the game, but too much of it makes players feel bad
What if we "fix" our randomness if it gets too bad
You still want "some" randomness, but it should be much less likely to get extreme results
This actually applies to game play as well as rewards!
Consider a game with a percentage chance to get a critical hit ("crit") which deals double damage
Say you have a 35% chance of dealing a critical hit
If a player is "really lucky" they might deal 5 crits in a row
Conversely, a player that is really unlucky might not get any crits in 10 attacks
What if we actually start with lower odds, but if you miss, you get a bonus on your chances?
![]() |
|
With crit smoothing:
Without crit smoothing:
Games do this all the time!
Diablo 3: If you don't get a legendary item for a long time, your odds are increased over time
Crusaders of the Lost Idols: Every 4th chest is guaranteed to be a "golden chest"
Smite: Crit-smoothing items
League of Legends: Critical Strike Smoothing Algorithm
AAA games usually sell around $40-$80 (25 000 - 50 000 CRC)
The average price for indie games on Steam is $9, with 21 000 units sold on average
Stardew Valley was $15, and sold millions of copies
Mobile Games are often free to play, or sell for less than $5
You can also make your game "free"
How do you make money? Sell in-game items
The idea is that everyone can play your game, and spend money according to how much money they have
There are several pitfalls, though
Which items do you sell?
Say you have a fighting game
Players can spend 500h to craft a legendary sword
Or they pay $20
This is not necessarily a problem
However, if the players compete in the game it might be, and/or if they need a large number of such items
A better (?) way to make money is to sell cosmetic (and convenience) items
For example, new hats for characters, additional character slots
It can be beneficial for player engagement to allow them to "grind" for these items
If there are 20 character skins for $20 each, and players can get one in 500h, a dedicated player can unlock a few of these
Loot Boxes have become very controversial
They can become part of a compulsion loop/addiction
Players also don't like that they may not get the skin/item they really want
Several countries have also declared them to be gambling and falling under gambling laws
Games usually display the odds of getting every item (Chinese law)
Games usually don't let you buy items (or loot boxes) directly
Instead you buy some made-up currency like gems and then spend that currency on the items
This gets around some gambling laws
It also makes it harder for players to judge how much they are spending
Finally, if you sell 400 gems at a time, but your items cost 350 gems, the players have left over gems and are incentivized to spend more
You can also give your players 20 gems per week for free
Whales are the players that invest "a lot" of money into your game, often multiple hundreds of USD
Dolphins are players that spend a little bit of money, by making maybe one or two small purchases
Minnows are players that spend (almost) no money on your game (sometimes "Freeloaders" are used as a distinct group that spends absolutely no money)
Some people argue that there should also be a category like "Megalodons" or "Mega-Whales"
However, none of the players typically spend a lot of money at once, Whales usually make many small purchases that add up.
The "conversion rate" of a free-to-play game is the percentage of players that become payers. What's your estimate?
The "conversion rate" of a free-to-play game is the percentage of players that become payers. What's your estimate?
For a typical free-to-play game the conversion rate is less than 5%. Candy Crush has about 8%. How many Whales?
The "conversion rate" of a free-to-play game is the percentage of players that become payers. What's your estimate?
For a typical free-to-play game the conversion rate is less than 5%. Candy Crush has about 8%. How many Whales?
Whales make up less than 1% of your player base, often less than 0.1%. In other words, 0.1% of players pay for the other 99.9%.
Treat your whales well, give them special offers, don't show them ads, keep them engaged.
On the other hand, also treat the freeloaders well, or your game will die out and then the whales leave, too.
As we discussed, it is hard to define what a "game" is
Over the last few decades, several genres and kinds of games have established themselves
However, there are many less explored ways to make games
Let's look at a few of those!
We said that (most) games have rules and goals
Sometimes all you want is to watch something grow (or burn)
Maybe with some minor interaction component
There may be some progression, for example to unlock new items
"Bohm is a game where you control a tree. Slow gameplay lets you discover the simple beauty of a growing tree. A zen-like, soothing experience that isn't about winning or losing."
Remember that games are often used for fantasy/wish fulfillment
Some of these wishes might just be "what if I had a different job?"
Especially recently, the "_ Simulator" games have exploded in popularity
Some of these are more realistic than others
Just as the definition of what a "game" is can be pushed, so can the environments they are implemented in
For the project we used Unity
But any interactive medium can serve as a platform for games, with some creativity
Note that these games are not necessarily "practical", but merely technology demonstrations (or pranks)
In 2012 three (then) PhD students wanted to play an action RPG on Twitter
So they implemented a bot that you could tweet at to participate in dungeon runs and slay monsters (including bosses)
Unfortunately the bot is no longer active :(
@sgware smashes a fancy pirate with a old crossbow of vileness and does 50 damage!
— World of Tweetcraft (@worldtweetcraft) October 16, 2012
Games are often not the most accessible software
Nowadays they often at least include a colorblind mode and subtitles
But what about other disabilities?
For example, how do you make a "video" game without video?
Audio games!
"A Blind Legend is a collaborative project of an audio-only action/adventure game for mobile phones. It's based on a very innovative technology: binaural sound. In this game, the players are guided only by 3D sound and live the adventure by controlling their hero with multi-point tactile gestures. A Blind Legend is above all a story: an adventure from the Age of Chivalry, to which we want to give epic scope."
A review:
"The combat is very easy at first, but gets harder after a while: enemies will move during combat, and depending on where they are located, you must press the corresponding arrow key. When there are more enemies, it gets even more fun. There are only a handeful of fights that kinda frustrated me, but the game balances the difficulty very well."
Some games are very story-driven
Movies are stories without a game
What if we add some interactivity?
Where does the concept of "movie" end, and that of "game" start?
Humans are not the only species that likes to play
When we think of pets playing, we usually imagine physical activity
However, there is no reason this couldn't involve something like a "video game"
The score/reward should still be something tangible, like a treat
Jack has a very intense love/hate relationship with @getcleverpet
— PJ (@OtherKnave) December 7, 2018
He LOVES that it gives him food... so much so that he can't think straight.
Then he gets mad at it when he doesn't do the sequence right (as evidenced by his floofy tail). pic.twitter.com/ZyOQ4FIfij
$100 to submit a game for publication
70% revenue goes to the developer (75% if the game makes more than $10 million, 80% if it makes more than $50 million)
Over 100 Million users
There are several tools for promotions, including weekly discounts, discoverability ("see more like this"), etc.
Players can request a refund for your game for 14 days, if they played for less than 2h
88% revenue goes to the developer, you can also use Unreal Engine without paying royalties
Newer store with fewer customers (but Fortnite is a big game!), and submissions are currently strongly selected
Epic (still) says the store would open to all developers in "the second half of 2019"
Two no-questions-asked refunds per player within two weeks of purchase
Originally sold bundles of indie games
Now also has a store for their 12 million customers with 75% revenue for the developer (10% go to charity!)
Also offers two other services:
Humble Gamepage: Host a website for your game for free and sell it, 95% revenue goes to the developer
Humble Widget: Provides a payment widget for your own website to sell your game with, with 95% going to the developer
Smaller, indie-focused store
Pricing is generally "pay what you want", but you can set a minimum price
Revenue sharing is also "pay what you want", you can give itch.io 0-30% of your revenue
You can upload anything (that is legal): games, art assets, books, comics, etc.
You have to register as a developer (once) for $25
70% revenue goes to the developer
If you make your game free you can not change it to paid later
Google also offers ads
You have to register as a developer for $99/year
70% revenue goes to the developer (85% after the first year, if you have a subscription-based app)
All apps are put through strict review (and sometime rejected rather arbitrarily)
To publish on (major) consoles, you need a DevKit
It's usually a bit of a process to register as a developer and get these
Nintendo and Sony actually loan them out for free, while Microsoft gives them away (but there's a waiting list)
You need to show that you are registered as a company and have an actual game in development
Keyboard shortcuts
↑, ←, Pg Up, k | Go to previous slide |
↓, →, Pg Dn, Space, j | Go to next slide |
Home | Go to first slide |
End | Go to last slide |
Number + Return | Go to specific slide |
b / m / f | Toggle blackout / mirrored / fullscreen mode |
c | Clone slideshow |
p | Toggle presenter mode |
t | Restart the presentation timer |
?, h | Toggle this help |
Esc | Back to slideshow |