When designing the GreatForge RPG system, on two different occasions that I can remember (and probably a few that I can't), we ran into enormous tangles of problems so bad that we had to scrap everything and start over. A system would throw out an issue. We'd try to correct that, and a new problem would pop up, and that solution would show two more problems, and we'd figure out that the system itself had to be taken out. That would unravel the systems that relied on it, and we'd end up tossing the whole thing and started over. We had the luxury of doing that, because we were purely a hobby group, doing this because we loved doing it.
In economics, the sunk cost fallacy (also known in game theory as the Concorde fallacy) is (roughly paraphrased) the impulse to continue down a dead-end path for fear of losing the resources already sunk into it. You can't just drop it all; you put too much time and money into this to just waste it by starting over. Maybe, if you sink more into it, you can salvage it all. But it just digs you deeper. You've thrown good money after bad.
Sometimes, this is unavoidable. The ones signing your paycheck aren't going to be happy if you tell them that you're throwing away the past few months of work. It's easy to understand why they might see it as throwing away those months of money (it isn't, really; time spent screwing up is time spent learning what does and doesn't work, but that's a whole different discussion). But it doesn't mean that nothing can get thrown away.
Riot Games, the studio behind League of Legends, is not shy about throwing out bad ideas. The champion Evelynn hinged on a binary stealth mechanic; either her opponents could see her and she had no fun, or they couldn't and they had no fun. Riot removed her from the rotation, and then completely redesigned her. She kept the theme of an assassin who could sneak around, but stealth stopped being an integral part of her kit and turned into a nice bonus on top of her burst damage.
With their big tournament over, they're starting a new Season and shaking up everything. Items that were too bad or too good have been reworked or trashed. Masteries have been completely redone, radically changing how certain character types play. The Support role is now more dynamic and exciting, with new items and sources of gold.
I've always liked Riot for their open communication and dedication to quality and detail. This just makes me like them even more.
Wednesday, December 5, 2012
The sun never sets on Praetoria
On November 30, 2012, City of Heroes, my favorite MMO and among favorite games period, was shut down. It was an upsetting day, for quite a few reasons. The game that had pushed me through college, the game that had led to me meeting my friends and eventually forming GreatForge Studios, the game that had inspired me to become a game designer was gone. The developers at Paragon Studios, who were some of the most talented, most friendly, and most openly communicative I had ever seen, were scattered.
But possibly the most dismaying is that future designers will have lost one of the best object lessons in game design they could have had.
City of Heroes pioneered several brilliant ideas in the MMO industry, primarily due to its focus on playability. Sidekicking, and its upgrade to Super Sidekicking, made teaming between players of different levels easy. Travel powers and other travel methods made traversing the city easy. The Mission Architect let players create their own stories and share them with the world.
Those are the most obvious things that made City of Heroes stand out, but it's the little things that I'll miss most. Thanks to fantastic animations and sound effects, powers had a weight to them that most MMO combat doesn't seem to grasp. Every power, even the wimpiest of blasts, felt like had impact. With the comprehensive toolkits spread around, every archetype was useful while none was necessary. No Tanker on your team? The Controller's lockdown potential and the Defender's debuff capabilities can keep the team safe. No Blaster to nuke mobs? Everybody can contribute damage to the team.
I won't be able to point out City of Heroes and all of its successes and flaws to the generation of game designers, and that's what saddens me the most.
But possibly the most dismaying is that future designers will have lost one of the best object lessons in game design they could have had.
City of Heroes pioneered several brilliant ideas in the MMO industry, primarily due to its focus on playability. Sidekicking, and its upgrade to Super Sidekicking, made teaming between players of different levels easy. Travel powers and other travel methods made traversing the city easy. The Mission Architect let players create their own stories and share them with the world.
Those are the most obvious things that made City of Heroes stand out, but it's the little things that I'll miss most. Thanks to fantastic animations and sound effects, powers had a weight to them that most MMO combat doesn't seem to grasp. Every power, even the wimpiest of blasts, felt like had impact. With the comprehensive toolkits spread around, every archetype was useful while none was necessary. No Tanker on your team? The Controller's lockdown potential and the Defender's debuff capabilities can keep the team safe. No Blaster to nuke mobs? Everybody can contribute damage to the team.
I won't be able to point out City of Heroes and all of its successes and flaws to the generation of game designers, and that's what saddens me the most.
The Importance of Balance
One of my two favorite game genres is the RPG. No other games offer the same level of customization and attention to story. Unfortunately, I find that far too many of them aren't very well balanced. On many of the message boards I frequent, I love to discuss this, and often I get the same disheartening response:
"Why does it have to be balanced? The AI isn't going to complain."
For a while, I didn't have a response to this. Of course games have to be balanced; that's what makes them good. The more I thought about it, the less sense the question made. It was tantamount to asking why games have to have control schemes that are usable - otherwise, you have a bad game.
It wasn't until we were facing down the deadline for the GreatForge system that I found an answer. We had a system called the Perk Grid; every five levels, a player could choose to have more health, more Endurance (the fuel for their abilities), more Initiative (letting them act sooner in the turn order), or a chance to shake off a control effect or a debuff. This was a nightmare to balance; having more Endurance meant you could use your flashier, fun abilities more often. All having more Initiative did was slightly tweak the turn order, and it didn't matter if you rolled well enough anyway.
After much deliberation, we decided that it had to be removed. It just couldn't be salvaged in a satisfactory way. Scott, our beta tester and my devil's advocate, was upset about this. He liked the extra customization that the system added. I told him that we couldn't balance it properly, especially so close to the deadline. He asked why it had to be balanced. Wasn't the extra customization worth the imbalance? And that's when I found my answer.
Games have to balanced so that player choice matters.
An imbalanced game offers the Illusion of Choice. Informed players who are interested in getting maximum efficiency are always going to go after the option that gives them the best returns, making multiple playthroughs pointless. They are always going to build the same maximally efficient character, make the same maximally efficient choices, and have the same exact experience they did the last time. Players who don't have the same information or impulse for efficiency are essentially paying a "not min-maxing tax".
Some players explicitly want to be weaker for extra challenge. I have no problem allowing them to do that, but it should be a conscious choice. Players should make the game harder because they want to, not because I screwed up my design.
A balanced game means meaningful choices. Meaningful choices mean happy players. Happy players keep buying and playing my games, which means I get to keep doing what I love: balancing games.
"Why does it have to be balanced? The AI isn't going to complain."
For a while, I didn't have a response to this. Of course games have to be balanced; that's what makes them good. The more I thought about it, the less sense the question made. It was tantamount to asking why games have to have control schemes that are usable - otherwise, you have a bad game.
It wasn't until we were facing down the deadline for the GreatForge system that I found an answer. We had a system called the Perk Grid; every five levels, a player could choose to have more health, more Endurance (the fuel for their abilities), more Initiative (letting them act sooner in the turn order), or a chance to shake off a control effect or a debuff. This was a nightmare to balance; having more Endurance meant you could use your flashier, fun abilities more often. All having more Initiative did was slightly tweak the turn order, and it didn't matter if you rolled well enough anyway.
After much deliberation, we decided that it had to be removed. It just couldn't be salvaged in a satisfactory way. Scott, our beta tester and my devil's advocate, was upset about this. He liked the extra customization that the system added. I told him that we couldn't balance it properly, especially so close to the deadline. He asked why it had to be balanced. Wasn't the extra customization worth the imbalance? And that's when I found my answer.
Games have to balanced so that player choice matters.
An imbalanced game offers the Illusion of Choice. Informed players who are interested in getting maximum efficiency are always going to go after the option that gives them the best returns, making multiple playthroughs pointless. They are always going to build the same maximally efficient character, make the same maximally efficient choices, and have the same exact experience they did the last time. Players who don't have the same information or impulse for efficiency are essentially paying a "not min-maxing tax".
Some players explicitly want to be weaker for extra challenge. I have no problem allowing them to do that, but it should be a conscious choice. Players should make the game harder because they want to, not because I screwed up my design.
A balanced game means meaningful choices. Meaningful choices mean happy players. Happy players keep buying and playing my games, which means I get to keep doing what I love: balancing games.
No, but there's still a chance you could save me.
Woo! After two and a half years, the GreatForge RPG system is finally complete! Well, complete enough to enter open beta. There's still a ton of work I want to do on it - filling in pages of flavor text, adding systems that needed more time to percolate, writing sample campaigns - but it's finished enough that I'm comfortable showing it to the public.
That means I can get back to this blog! I can talk about all that I've learned from designing a tabletop game, along with other design thoughts I want to explore. So,
With bony hands I hold my partner
On soulless feet we cross the floor
The music stops as if to answer
An empty knocking at the door
It seems his skin was sweet as mango
When last I held him to my breast
But now we dance this grim fandango
And will four years before we rest.
That means I can get back to this blog! I can talk about all that I've learned from designing a tabletop game, along with other design thoughts I want to explore. So,
With bony hands I hold my partner
On soulless feet we cross the floor
The music stops as if to answer
An empty knocking at the door
It seems his skin was sweet as mango
When last I held him to my breast
But now we dance this grim fandango
And will four years before we rest.
Wednesday, November 23, 2011
Learning from Everywhere
I just returned from vacation in Orlando. My family and I visited Universal Studios, Universal Islands of Adventure, and Disney’s Magic Kingdom. As we walked throughout the parks and waited in line for rides, I couldn’t help but notice that theme parks have a lot of lessons that game designers can learn.
Both Disney and Universal spend considerable effort to ensure that every part of a ride works toward the overall experience. Placed throughout the queue area are television screens, posters, and animatronics to set up the story of the ride. Waiting in line for Universal Studios’ Jaws ride, you watch television interviews with the survivors of Amity, discussing the way they’ve coped with the trauma of a shark attack. Outside the ride, the park is set up to mimic Amity, complete with tacky shark souvenirs. At Disney, the path to the Haunted Mansion is littered with tombstones inscribed with cheesy jokes. The ride operators wear funeral attire and make terrible puns in somber tones. Once you get off Space Mountain, you take a walkway back to the entrance. The ceiling is covered in stars, there are alien landscapes built around the conveyor; the ride is over, but the experience isn’t.
Game designers need to remember that every part of a game should work towards enhancing the experience. The opening menu, the pause interface, everything needs to contribute towards the atmosphere the game is trying to achieve. I’ve played a number of games that shattered my immersion because the interface wasn’t as streamlined as the rest of the game, throwing the fantastic pacing down the toilet.
There’s more to establishing tone for a game than just the gameplay and the visuals. Game designers need to remember that they have lots of tools at their disposal, and they need to use all of them.
Wednesday, November 9, 2011
Pacing in Alan Wake
Lately, I've been replaying Alan Wake. It's a survival horror game heavily based around contrasting light and dark. Enemies are surrounded by darkness, which needs to be burned away with the beam of a flashlight before they can be hurt. I could talk for days about all the things Alan Wake does right, but I'm going to focus on one thing: pacing.
In order for a survival horror game to work, it needs to be scary. And in order for the player to feel scared, they have to feel threatened. Previous survival horror games, like the original Resident Evil and Silent Hill, did this by crippling the player's supplies. The player never really had the tools to fight back, so the best course of action was always to run. More modern games, like Dead Space, have tighter controls, but they still rely on giving the player too little ammo to properly protect themselves.
Alan Wake - I assume through incredibly strenuous playtesting - manages to give exactly enough for the player to fight their foes while still feeling as though they're lacking. Every time I run out of ammo, I turn the corner to find a checkpoint with an ammo cache. And while it was predictable, it didn't lessen the tension. I still felt like I could run empty and be left helpless.
Part of this is due to brilliant level design. Enemies generally require a very specific amount of ammo to dispatch; standard foes take three pistol shots, more difficult foes can take a whole clip. So when the designers build an encounter, they can know exactly how much ammo a player will need to face it. By stringing these encounters together cleverly, they can ensure that the player has just enough ammo to deal with the dangers they'll face. There's still enough wiggle room that the player doesn't feel that they have more ammo than necessary, but they never feel that they need to run rather than fight (unless the encounter is specifically designed that way).
Proper pacing makes or breaks a game. I wish more developers would play Alan Wake and see that.
In order for a survival horror game to work, it needs to be scary. And in order for the player to feel scared, they have to feel threatened. Previous survival horror games, like the original Resident Evil and Silent Hill, did this by crippling the player's supplies. The player never really had the tools to fight back, so the best course of action was always to run. More modern games, like Dead Space, have tighter controls, but they still rely on giving the player too little ammo to properly protect themselves.
Alan Wake - I assume through incredibly strenuous playtesting - manages to give exactly enough for the player to fight their foes while still feeling as though they're lacking. Every time I run out of ammo, I turn the corner to find a checkpoint with an ammo cache. And while it was predictable, it didn't lessen the tension. I still felt like I could run empty and be left helpless.
Part of this is due to brilliant level design. Enemies generally require a very specific amount of ammo to dispatch; standard foes take three pistol shots, more difficult foes can take a whole clip. So when the designers build an encounter, they can know exactly how much ammo a player will need to face it. By stringing these encounters together cleverly, they can ensure that the player has just enough ammo to deal with the dangers they'll face. There's still enough wiggle room that the player doesn't feel that they have more ammo than necessary, but they never feel that they need to run rather than fight (unless the encounter is specifically designed that way).
Proper pacing makes or breaks a game. I wish more developers would play Alan Wake and see that.
Tuesday, November 1, 2011
Games I Love: Team Fortress 2
With Battlefield 3 just released and Modern Warfare 3 on the horizon, I'm often asked which I prefer. And my answer surprises more often than not: neither.
My favorite shooter - and the only one I'll play - is Team Fortress 2.
I don't like military style shooters. I find that battles between two players end in a second or two at most; either someone catches the other by surprise and kills them instantly, or the player with the better aim gets a headshot and kills the other instantly. The guns in military shooters all tend to blur together; there are the assault rifles with slightly different stats, the shotguns with slightly different stats, and the sniper rifles with slightly different stats.
Listening to the developer commentary in TF2 (Valve commentary is ALWAYS worth listening to) reveals a set of brilliant mantras.
At a glance, a player should be able to tell a.) what team someone is on, b.) what class they are, and c.) what weapon they are using. In military shooters, everyone is the same "soldier in camo", and I can never tell if someone is on my team or if I should be shooting them.
Only two classes have the ability to kill in one shot: the Sniper and the Spy. The Sniper has to get a headshot (no other class has that ability), and his laser sight will alert his prey that they need to beware. The Spy has to get a backstab with his knife, which requires him to be in melee range and sneak up on his target. Both classes have the least health in the game, so they pay for their power with fragility.
Balance between classes is, while not perfect, incredibly well done. All classes fit into a rock-paper-scissors configuration, but skill can trump class counters. Nearly all unlocked weapons are sidegrades, gaining strength in one area at the cost of a weakness elsewhere. One of my biggest complaints with military shooters is that they all have weapon unlocks that are clear upgrades to the defaults, making the barrier to entry incredibly high.
Most of all, Team Fortress 2 gets the pacing of combat perfectly. When combat starts, I have long enough to decide if I should engage or if the fight is a guaranteed loss. Once I start fighting, I know it's going to take more than one shot to kill either of us, so there can actually be a fight. And if it's clear I'm losing, I have the ability to escape while simultaneously allowing my opponent the chance to finish me off. Fights are (almost) never over too quickly or too slowly.
I've come to expect a lot from my shooters. I want fights that last, balanced weapons, and game modes that reward teamwork rather than mere deathmatching. And nothing has offered what I want like Team Fortress 2.
My favorite shooter - and the only one I'll play - is Team Fortress 2.
I don't like military style shooters. I find that battles between two players end in a second or two at most; either someone catches the other by surprise and kills them instantly, or the player with the better aim gets a headshot and kills the other instantly. The guns in military shooters all tend to blur together; there are the assault rifles with slightly different stats, the shotguns with slightly different stats, and the sniper rifles with slightly different stats.
Listening to the developer commentary in TF2 (Valve commentary is ALWAYS worth listening to) reveals a set of brilliant mantras.
At a glance, a player should be able to tell a.) what team someone is on, b.) what class they are, and c.) what weapon they are using. In military shooters, everyone is the same "soldier in camo", and I can never tell if someone is on my team or if I should be shooting them.
Only two classes have the ability to kill in one shot: the Sniper and the Spy. The Sniper has to get a headshot (no other class has that ability), and his laser sight will alert his prey that they need to beware. The Spy has to get a backstab with his knife, which requires him to be in melee range and sneak up on his target. Both classes have the least health in the game, so they pay for their power with fragility.
Balance between classes is, while not perfect, incredibly well done. All classes fit into a rock-paper-scissors configuration, but skill can trump class counters. Nearly all unlocked weapons are sidegrades, gaining strength in one area at the cost of a weakness elsewhere. One of my biggest complaints with military shooters is that they all have weapon unlocks that are clear upgrades to the defaults, making the barrier to entry incredibly high.
Most of all, Team Fortress 2 gets the pacing of combat perfectly. When combat starts, I have long enough to decide if I should engage or if the fight is a guaranteed loss. Once I start fighting, I know it's going to take more than one shot to kill either of us, so there can actually be a fight. And if it's clear I'm losing, I have the ability to escape while simultaneously allowing my opponent the chance to finish me off. Fights are (almost) never over too quickly or too slowly.
I've come to expect a lot from my shooters. I want fights that last, balanced weapons, and game modes that reward teamwork rather than mere deathmatching. And nothing has offered what I want like Team Fortress 2.
Subscribe to:
Posts (Atom)