Video game design: you need to understand that you inherit
- Transfer
Last year, I made a short presentation at the GDC on the importance of understanding what legacy your game is based on. The video of the report is published here (my part starts at 28:50):
Link
Three strikes - and you retire.
This phrase is so popular that it has almost become an idiom. Indeed, even though some of our non-American friends may be confused by baseball rules, they probably know this rule. But…
... actually this is not true. The batter does not drop out after the third strike. He drops out only when the catcher catches the ball.
If the catcher misses or drops the ball, then the batter does not drop out and has a chance to get to first base. It almost always ends with an out, because the catcher just picks up the ball and makes a simple throw, but sometimes this little-known rule becomes very important, as happened in the last game of the playoff series between Chicago Cubs and Washington Nationals in 2017. Max Scherzer threw a third strike, flying past striker Javier Baez, but look what happened:
Nationals catcher Matt Witers missed the ball between his legs, allowing Baez to calmly reach first base. This was supposed to be the third inning out. But instead, the Cubs earned two more wounds, and later won the game, having an advantage of just one wound, and moved on to the next round.
So this little-known rule kicked the Nationals out of the playoffs.
Where did this rule come from?
It goes back to the first print of the baseball rules issued by German Johann Christoph Friedrich Guts-Muts.
He described a game called "English baseball," in which there were innings, hitters, fielders, bases, and points for touching the house. However, there were no strikes and pains yet. The pitcher stood close to the batter and practically “served” the ball for a convenient hit. The pitcher did not try to compete with the batter; the game was to catch the ball in the field after it was hit.
However, what was to happen if the batter was so terrible that he could not hit the ball? There was a special rule in the Guts-Muts edition game for such a situation: the striker had the opportunity to deliver only three hits. On the third hit, the ball automatically entered the game, even if it was not hit. Therefore, the batter should have fled to first base by hitting the ball or having missed a third time. Indeed, there was no catcher to catch the ball; therefore, the pitcher had to run home to raise the ball and throw at first base.
In 1845, the American Knickerbocker Base Ball Club wrote down its own rules of the game, and something changed about them.
The pitcher was now much further from the batter and threw the ball horizontally, so a new position for the catcher was needed. However, they retained the logic of the old rule - the ball remains in the game after the third missed hit, like an old legacy code.
The strike-out (knockout after three goals conceded) essentially arose on its own (that is, it became an “emergent gameplay”), because after the third miss, the ball now theoretically remained in the game, and the catcher turned it into an out catching feed. That is, now there was no difference between the fact that the catcher makes an out, catching a goal and the catcher makes an out, catching the serve after the third missed shot. In both cases, the ball now remains “alive” and the catcher makes an outfit catching the ball before it touches the ground.
However, later they had to “patch” the game, because they did not bother to make the strike-out an official rule. Since the ball would be considered “live” after the third strike, there was the possibility of an ugly double or triple.
For example, if the bases are occupied by runners, then the catcher can intentionally drop the ball, raise it again, stand at home base to make an easy out, and then throw the ball to third base, and then to second. Therefore, in 1887, a new rule was added, according to which the batter automatically makes an out if the runner was on the first base. And there were less than two outs.
Consequently, the rule of “three strikes, and you retire,” which, according to the general opinion, is used in baseball, is “true”, but only under very specific conditions.
The creators of the rules decided to use a clumsy patch, instead of just rewriting the rules according to how they actually play the game!
Think about the situation with Javier Baez. He was a runner at first base, so even though the catcher dropped the ball, it should have been considered a strike out ... but there were already two outs, so we return to the original rule about losing the third strike.
It would be possible to rewrite the rules so that the principle of “three strikes and you retire” is always applied . Wouldn't that be easier? More intuitive? Why bother trying to fix one obvious problem with a catcher intentionally dropping the ball, rather than getting rid of this old rudimentary rule?
The reason is that we inherit the design of the game from everything that came before us.
Sometimes this inheritance is obvious - Civ 6 inherits from Civ 5 , which inherits from Civ 4 , and so on.
Sometimes a designer inherits from those games that he played in his childhood ( Mario -> Braid , Myst -> The Witness )
Sometimes games borrow from themselves. Here's what the development process of our economic RTS Offworld Trading Company looked like .
At the early stages of development, you can add simplifications or hacks so that the prototype is playable, but all these assumptions are now fixed in the design, whether you want it or not. You need to remember that this was a random or arbitrary choice.
However, game mechanics are most often inherited, usually from games of the same genre.
For example, although the Offworld Trading Company is RTS, it is notable for the fact that it has no units.
But we didn’t start with this, because we inherited from other RTS that existed before us - from StarCraft , Age of Empires , etc. Therefore, we had scouts, builders, transports, pirate ships, police ships, and the like.
Over time, we found that inheritance pulls the game down, forcing the player to spend time taming units, rather than playing in the market. Gradually, one by one, we got rid of these units: first from transports, then from combat units, then from builders, and, finally, from scouts. The game now looks like a complete break with the past, but for this we had to go a long way.
The problem is that iterative design development can become a trap - you can no longer see which parts of the game prevent you from creating a much better design. It’s easier to make small changes that fix the most obvious problems, rather than reviewing the design as a whole.
Sometimes the problem of the game can be at a conceptual level. Take for example Spore ...
... which was perceived as a game, varying in powers of tens - from the scale of cells to the galactic level. This was the bait, the reason for creating the game.
This part of the game has disappointed many - five separate levels seem to be five different games connected by electrical tape. However , something interesting happened to Spore's failure ...
... it turned out that this is actually not a failure. That's how many people are playing Spore right now - not so bad for a game released 10 years ago.
And in fact, look at this graph, where Spore is compared to the two most popular PC games released in the same 2008. Now Spore defeats them, and do not forget that Spore did not even go out on Steam.
It turned out that the most interesting part of the game is not the concept of “powers of ten”, but in-game editors, especially the creature editor, which dynamically animated the creatures created by the player.
However, these editors were created in the middle of project development; Maxis started to make a game about one thing, but suddenly she made a game about something else. One of the unanswered questions about Spore is what would we do if we abandoned the concept of “tens of degrees” and instead focused on the editors?
Here is a classic example of bad design inheritance. Creep denial is a mechanic from the original DOTA : you kill your own units to prevent enemies from gaining gold and experience from them.
Creep denial is one of the most important elements of a high-level game in DOTA , which allows you to maximize the number of experience points relative to opponents in order to overtake them in level. However, the question is still open whether this is actually a good design.
At least creep denial is a random design because DOTA inherited it from Warcraft 3 . In fact, even the fact that Warcraft 3 allowed you to kill your own units was most likely a belated decision by designers.
DOTA inherited this rule because the game was literally built inside Warcraft 3 as a mod. Therefore, MOBA games have inherited many of the design elements and mechanics from Warcraft 3 . Perhaps the DOTA designers wanted many aspects to work differently, but they generally had no choice given the assumptions and limitations of the Warcraft 3 editor .
Of course, DOTA 2 and League of Legends inherited their design from the original DOTA mod , but their developers made different decisions about the inheritance of creep denial. In fact, the League abandoned it, and DOTA 2 retained it.
This is a screenshot from a post on Reddit about why creep denial was not left in the League . Do not worry, you can not read it; I just want to show how “RandomGuyDota” is trying to explain why creep denial is bad for design using the game mechanics themselves. This is a fairly typical rationale for what has become part of the game design heritage - you always need to prove that the element needs to be removed from the game, and not the importance of its initial addition.
However, I have a simpler rationale for why creep denial is a bad design ...
"It's stupid."
Do you want players to spend time killing their own units? Is this really a fundamental part of a MOBA device? Will the game fall apart if you don't kill your own helpers?
aahdin sums it up better than I could: "Yes, it increases the skill of playing MOBA just as it increases the skill of playing chess, burning half of the pieces."
At some point, you as a game designer need to take a step and review what you inherit. Will basic gameplay survive without this feature? This function is not intuitive, does it complicate the game or familiarity with it? Can players spend time better than doing this feature?
In the case of creep denial, the answer to all these questions suggests that the game would be better without it. In the MOBA genre, there is only one magic function that cannot be abandoned - it retains the scale and complexity of RTS, but focuses the player’s control on just one unit, which makes the game accessible to a wider audience (by an order of magnitude). All the rest,everything else is just a random inheritance that arose because the genre originally appeared as a Warcraft 3 mod .
In fact, even though there is no creep denial in the League today ... he was originally present.
Here are the notes on the very first League of Legend patches published in July 2009. Developers inherited creep denial, but got rid of it in the early stages.
That is, even though the developers used the legacy of the original mod, they still sought to critically examine the past of the game.
But for comparison, the history of the transition of creep denial from DOTA 1 to DOTA 2 . Designers obviously realized that creep denial might not be the best part of the game.
Take a look at the description of version 6.82 “Killed creeps now give less experience.” - a clear sign that they have revised this feature by changing the reward for it. However, instead of getting rid of it, designers make small changes around the edges.
In fact, they do the same thing as baseball players when they patched the third strike rule so that it does not apply in certain conditions, and did not cut out this silly rule entirely.
Remember my creep denial value questions?
Will basic gameplay survive without this feature? This function is not intuitive, does it complicate the game or familiarity with it? Can players spend time better than doing this feature? Having asked these questions about the rule of the missed third strike, we find ourselves in the same position - this is a bad random design that harms baseball.
And now here is a comparison of these two games and some other MOBA. There are many reasons that the League overtakes DOTA 2 by an order of magnitude - and a three-year handicap is quite significant among them, but I think that the Riot philosophy of re-evaluating inheritance from the original DOTA mod , which extended not only to cutting, was also very important. creep denial.
I still have thoughts on last hitting, but, fortunately, there’s no time for that. I say “fortunately” because Heroes of the Storm , the only one to give up last hitting, is less successful than DOTA 2 , not to mention League . Therefore, I can not say that last hitting is a bad design, and this is proven by the market. In addition, I don’t think it’s reasonable to wait for Riot to give up last hitting; It is too late. League is one of the most popular games in the world. In fact, they were lucky that they had abandoned creep denial so early; if this were done later, then the opinions of the community could be divided.
We do not always have the opportunity to look at the market to confirm our decisions, and therefore reassessing the inheritance of the game is such a complex and important problem.
Failure to inherit requires real courage. Sometimes, if you see a problem, you have to trust your own rational design process. Sometimes you have to rely on instinct. But ultimately, you need to strive to see your story, find out how it brought you to where you are, and have the courage to abandon the past.
(You can read more about the rule of the missed third strike in this article .)