CellarDoorGames · @CellarDoorGames
20th Apr 2018 from TwitLonger
Full Rock Paper Shotgun response
[𝐏𝐞𝐫𝐬𝐢𝐬𝐭𝐞𝐧𝐭 𝐮𝐧𝐥𝐨𝐜𝐤𝐬, 𝐚𝐧𝐝 𝐞𝐥𝐞𝐦𝐞𝐧𝐭𝐬 𝐭𝐡𝐚𝐭 𝐜𝐚𝐫𝐫𝐲 𝐚𝐜𝐫𝐨𝐬𝐬 𝐛𝐞𝐭𝐰𝐞𝐞𝐧 𝐫𝐮𝐧𝐬]... 𝐂𝐚𝐧 𝐲𝐨𝐮 𝐞𝐱𝐩𝐥𝐚𝐢𝐧, 𝐟𝐨𝐫 𝐭𝐡𝐞 𝐩𝐞𝐨𝐩𝐥𝐞 𝐚𝐭 𝐡𝐨𝐦𝐞, 𝐡𝐨𝐰 𝐑𝐨𝐠𝐮𝐞 𝐋𝐞𝐠𝐚𝐜𝐲 𝐜𝐨𝐦𝐛𝐢𝐧𝐞𝐝 𝐭𝐡𝐞𝐬𝐞?
Rogue Legacy combined Permadeath and persistence by allowing players to spend whatever gold they had accrued in their previous run, to permanently bolster their stats for subsequent runs. Whatever gold the player had left over, was lost forever. We wrapped this all up thematically via a lineage system. The money your ancestors made directly went to their children.
[𝐏𝐞𝐫𝐬𝐢𝐬𝐭𝐞𝐧𝐜𝐞 𝐢𝐧 𝐚 𝐠𝐚𝐦𝐞] ... 𝐬𝐨 𝐰𝐡𝐚𝐭 𝐦𝐚𝐝𝐞 𝐲𝐨𝐮 𝐝𝐞𝐜𝐢𝐝𝐞 𝐭𝐨 𝐧𝐨𝐭 𝐨𝐧𝐥𝐲 𝐢𝐧𝐜𝐥𝐮𝐝𝐞 𝐢𝐭, 𝐛𝐮𝐭 𝐦𝐚𝐤𝐞 𝐢𝐭 𝐚 𝐩𝐢𝐯𝐨𝐭𝐚𝐥 𝐩𝐚𝐫𝐭 𝐨𝐟 𝐭𝐡𝐞 𝐠𝐚𝐦𝐞?
We wanted to add persistence for a bunch of reasons, but two stick out in our mind.
First, we wanted to reduce punishment, and remove "time-sinks". By reducing punishments, we specifically mean having a play session and gaining nothing extrinsic from it. This is one of our more contentious decisions, and a lot of people say this is what makes our game grindy, but removing that artificial punishment opened a lot of avenues to add depth and difficulty that would be harder to offer in other roguelikes. For example, in Rogue Legacy, when you beat a boss, it's dead forever. You never get to play them again. Because of that, it allows us to make those encounters more memorable. And then, we make it even easier, by giving players the choice of immediately re-challenging if you fail, so you don't have to redo the castle to fight them. Because our level of punishment is so low, we can increase the compexity of major encounters without artificially inflating playtime. For example, an average player in Rogue Legacy would take 5 or so tries to beat Khidr, the first boss of the game. He's the simplest enemy, and he has a custom arena that can harm the player, and multiple different attack patterns which force the player to engage with the arena layout. Most other roguelikes are built with the expectation of the player defeating all the bosses in one try, and as such have to make these encounters much more basic.
This ties into our other reason, and that is we wanted to implement an RPG system to act as a great equalizer in terms of player skill. A lot of people think the world of Rogue Legacy scales with the player. It absolutely does not. If you want to grind, you can outscale the content in the game by an order of magnitudes. People who are not good at games or don't have a natural knack for RL's combat system can "grind" their way to the end. If you put in the time, you'll get the reward. We don't think it's necessary for a roguelike game to say “you MUST be this good in order to see content”. Rather than implement a difficulty system, we ensured the RPG mechanics allowed for leeway. This doesn't automatically imply Rogue Legacy is a grind though (which many people believed). If you're good, you can beat the game in a single life, and speed runners have beaten the game in less than 20 minutes.
*I didn't explain time-sinks very well, but when you play a roguelike which is 10 floors or whatever, and you beat a floor, chances are you’re probably good enough to continually beat it. But many roguelikes require you to beat it over and over again to reach the later content. That means all the time you're wasting playing previous floors are just that, wasted time.
𝐖𝐞𝐫𝐞 𝐭𝐡𝐞𝐫𝐞 𝐨𝐭𝐡𝐞𝐫 𝐠𝐚𝐦𝐞𝐬 𝐲𝐨𝐮 𝐥𝐨𝐨𝐤𝐞𝐝 𝐚𝐭 𝐟𝐨𝐫 𝐢𝐧𝐬𝐩𝐢𝐫𝐚𝐭𝐢𝐨𝐧 𝐨𝐧 𝐭𝐡𝐢𝐬 𝐩𝐚𝐫𝐭𝐢𝐜𝐮𝐥𝐚𝐫 𝐚𝐬𝐩𝐞𝐜𝐭 𝐨𝐟 𝐭𝐡𝐞 𝐝𝐞𝐬𝐢𝐠𝐧? 𝐖𝐚𝐬 𝐢𝐭 𝐚𝐧 𝐚𝐭𝐭𝐞𝐦𝐩𝐭 𝐭𝐨 𝐛𝐨𝐫𝐫𝐨𝐰 𝐬𝐨𝐦𝐞𝐭𝐡𝐢𝐧𝐠 𝐭𝐡𝐚𝐭 𝐰𝐨𝐫𝐤𝐞𝐝 𝐟𝐫𝐨𝐦 𝐠𝐚𝐦𝐞𝐬 𝐲𝐨𝐮 𝐥𝐢𝐤𝐞𝐝, 𝐨𝐫 𝐟𝐢𝐱 𝐬𝐨𝐦𝐞𝐭𝐡𝐢𝐧𝐠 𝐲𝐨𝐮 𝐝𝐢𝐝𝐧'𝐭 𝐥𝐢𝐤𝐞 𝐢𝐧 𝐫𝐨𝐠𝐮𝐞𝐥𝐢𝐤𝐞𝐬?
The only true "roguelike" game we were basing this game off of was Dungeons of Dredmore. We never really dipped into roguelikes until that point. We ended up playing Spelunky and Binding of Isaac later on, but at that point we were too far into development for them to have a deep effect on where Rogue Legacy was going to go.
Four years ago we wrote a talk for GDC that listed in excruciating detail how we designed and developed Rogue Legacy. It was 80 slides long and took over 2 hours to go through, and we only had a 25 minute time slot so we just scrapped it and wrote a new one. But it listed in a roundabout manner the way we make games, and a lot of our core design philosophy really stems from fixing things we don't like. I’m not referring to surface level technical issues like, the hitboxes for a game feel bad so let’s make it better. It’s usually for more fundamental ideas that games use. For a more concrete example, one of the major things we wanted to change from Dungeons of Dredmore (and rogue-likes in general) was what we internally called "Room Clearing".
In a lot of roguelikes, all the "stages" are built to be mini-labyrinths. This means that a run could be very long (if you take a lot of false paths), or very short (if you go the optimal route). Room clearing is the idea that you clear every room on a floor before progressing to the next one, regardless of when you find the exit. In reality, the correct path is to always clear the floor, because the game is built with that power-curve in mind. If you don’t do that, you’ll be severely underleveled for the next area. So when we decided to try and fix that issue, it becomes a basis from which we grow an idea. No room clearing meant we had to remove slow power scaling, which meant we had to upfront change, which meant we had to design classes and so on and so forth.