Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
November 1, 2014
arrowPress Releases
November 1, 2014
PR Newswire
View All
View All     Submit Event

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

Grinding and the Burden of Optimal Play
by E McNeill on 07/21/14 01:29:00 am   Expert Blogs   Featured Blogs

The following blog post, unless otherwise noted, was written by a member of Gamasutraís community.
The thoughts and opinions expressed are those of the writer and not Gamasutra or its parent company.

Just before I announced Darknet, I sent an early preview build to Ben Kuchera, a writer at Polygon. The resulting article was generally positive, but there was one negative point that came up right away:
“It sounds grindy”
This was a bad sign. (For those who don’t recognize the term, “grinding” refers to mindless repetition in games.) I hate grinding. I’m fundamentally opposed to it, and I’ve always made an effort to keep it out of all my games. And yet, Ben was right. Somehow I had made a game in which you repeat the same task (solving a hacking puzzle) over and over again. How did it come to this?
The grinding problem originated in the decisions I made at the very beginning of Darknet’s design. I decided to build a hacking game, and I immediately knew that I wanted the player to be hacking a network composed of lots of interconnected nodes. The player should be able to capture individual nodes and thereby influence the wider network, which I figured would provide a strong basis for some interesting strategic dynamics. I also wanted the game to feature a wide range of difficulty (to support newbies and experts alike), and for various reasons I wanted to give the player lots of freedom in choosing which level of difficulty they wanted to tackle.
Together, these decisions led to an unintended consequence. First, the primary action in the game is always to capture a node, which naturally lends itself to repetition. Second, the player can freely choose what difficulty of node they want to capture. Thus, the player is incentivized to go after the easiest possible nodes, over and over again, ad nauseum. Grinding. Ugh.
Let’s zoom out for a moment. You might ask: isn’t grinding ultimately the player’s choice? Why would players choose to grind if it’s not any fun? I think the answer lies in a concept called the Burden of Optimal Play, a term introduced in a great GDC talk about Diablo 3. The speaker, Wyatt Cheng, defined it as “the idea that the player can increase their power level, but that they do so in a way that’s not necessarily enjoyable.” 
Here’s how I think of it: Players will tend to make the choices that lead most directly and surely to victory. If you set up a goal that players care about, you should expect them to try to reach it as efficiently as possible. If the most efficient way to win (i.e. optimal play) is a pain in the ass, then you’re asking players to accept a worse experience in exchange for victory, and that’s a crappy way to treat your players. (There are lots of caveats that I could make here, but let’s stick with this simplistic version of the concept for now.)
The Burden of Optimal Play suggests a guideline for game designers: if you want the player to have fun, then the most effective path to victory should also be the most fun. Playing well should never result in a worse experience.
This guideline helped point toward the solution to Darknet’s grinding problems: I needed to make it clear that grinding is not an effective strategy.
Keep in mind that it’s not necessarily repetition that’s bad. (Most great games are repetitive in one way or another.) Rather, the problem is mindless repetition. The issue in Darknet was that players felt incentivized to seek out the easiest puzzles available, which was never an interesting challenge. All the hacking puzzles in Darknet are randomly generated, and since each network includes very difficult levels as well as very easy ones, all I really needed to do was incentivize the difficult levels more than the easy ones.
With that in mind, I made sure that the reward for completing a puzzle grew drastically with difficulty. Then, I added a time limit to each level. (This was actually planned from the beginning, as Ben noted in the Polygon article.) Together, these mechanics incentivize players to tackle the hardest puzzles that they can handle, lest they waste time on the less-valuable easy ones. Optimal play now requires players to overcome interesting challenges, rather than forcing them to accept the burden of the grind.
This has proven to be difficult to communicate to new players, however. When you enter your first game of Darknet, you don’t know how to judge a puzzle’s difficulty against its potential rewards, and you don’t realize right away that the more aggressive strategies tend to be more efficient. All you know is that some puzzles seem pretty much impossible, whereas tackling the easy ones seems to actually be getting you somewhere (albeit slowly). If you pursue that cautious strategy, you could easily come to the conclusion that the whole game is one big grind.
So, although I think I’ve dealt with the grinding at a mechanical level, I’ve still got a serious communication problem. Sure, the game isn’t meant to be played in a grindy way, but that doesn’t mean much if I don’t point players toward an alternative. I really value the freedom in Darknet’s design, and I don’t want to do anything heavy-handed like telling players how to play, but somehow I need to make the good experiences in Darknet more accessible than the bad ones.

Related Jobs

Rensselaer Polytechnic Institute
Rensselaer Polytechnic Institute — Troy, New York, United States

Assistant Professor in Music and Media
The College of New Jersey
The College of New Jersey — Ewing, New Jersey, United States

Assistant Professor - Interactive Multi Media - Tenure Track
Next Games
Next Games — Helsinki, Finland

Senior Level Designer
Magic Leap, Inc.
Magic Leap, Inc. — Wellington, New Zealand

Level Designer


Kayne Ruse
profile image
"The most effective path to victory should also be the most fun." I love finding quotable little tidbits. Actually, I'm making a roguelike RPG thingy (don't ask, it's hell on wheels right now) where different areas have different levels of difficulty. This article makes me think that the monsters shouldn't appear or should run away from players that are too strong for them. Thanks!!

Colin Cronin
profile image
Monsters running away from players too strong works great. This was one of my favorite little touches in Ni No Kuni. Although, there were a couple times I ended up chasing weak enemies for one reason or another.

profile image
I like to read about the never ending quest for improvement! thank you.
I might suggest you investigate what is meant by "Fun" in your context. It can be an incomplete concept and you may be better suited to identifying the points of engagement in your game and showcasing them on your path of "dominate strategy" for the user. (
Just some light reading and easy thought experiments ... that man and game designers have been chewing on for eons.
Good Luck!

Brandon Shelton
profile image
We mitigated grinding in our RPG by decoupling the traditional benefit of gaining experience levels, meaning we removed major statistical upgrades on level up entirely. Instead we made it so gaining a level lets the player learn a new skill and passive effect, because thatís the core of what our gameplay is in the first place. I initially wanted to set hard limits on how much the player could grind against low level enemies, but ultimately we didnít want to eliminate easy grinding completely because some players still enjoy doing it (among a few other reasons). I feel a lot of game designers donít really think about Optimal Play and all its implications, or if they do they canít/donít change anything to fix it being boring. Of course this also depends on the genre and seems to get addressed much more frequently in RPGs and strategy games.

Ian Morrison
profile image
This was actually a problem I felt the latest XCOM had (and I adore the game). Optimal play was extremely cautious, taking turns slowly, minimizing your risks, trying not to trigger multiple pods of aliens and not even giving aliens a chance to take a shot, etc. On the face of it, this is what you`d expect, as XCOM is in many ways a game about managing your risks, but the greatest moments in the game came when the shit hit the fan, your plan was disrupted, and all the safe bets were out, forcing you to take bold, potentially costly decisions. In this way, the game was incentivizing a less engaging way to play the game.

The Enemy Within expansion (and certain special missions in the original like terror and bomb defusal missions) addressed the issue somewhat with time-sensitive objectives, but these were bandaids. Scripted missions aren`t the focus of XCOM`s emergent gameplay and timed objectives like civilians or meld were a temporary element. Once the meld canisters were secured (or lost) you had no incentive not to go back to the slow, defensive style of play.

Jonathan White
profile image
"The most effective path to victory should also be the most fun."

I would challenge that statement and say, all paths to victory should be fun and not necessarily the most effective (whatever that means) should be the most fun.

Jonathan Jou
profile image
I think it's meaningful to note that "sounds grindy" is more easily interpreted as "the central gameplay mechanic doesn't sound rewarding enough to me."

This should lead to not one, but several considerations:
1. Who is the target demographic, and why will this game be fun to them? If you're making a game about hacking puzzles, it's likely you think this game will appeal to puzzle enthusiasts, who do indeed want a greater challenge to go with their certain victory. This would mean avoiding penalizing players for choosing harder puzzles, by increasing the relative rewards of the harder puzzles accordingly.

2. How can the central gameplay mechanic be made more enjoyable? Is the fundamental task fun? Are there user interface issues that make a task harder than it seems? Is the progression slower than it needs to be? Is it hard for players to know how well they're doing? Is there anyway to give the players a clear sense of their accomplishments and reward them for completed tasks? Is it necessary to shift the focus of the game to some other gameplay elements, or combine them more organically?

3. Is there some way to add value to the game? Many people add additional gameplay elements to balance out a simplistic core mechanic, such as sidequests or power ups and challenge modes, which can confuse the original vision but can also lead to stronger ideas. I like to think about this the other way: if the game needs "extra" features to make it fun, then I'm thinking about the design the wrong way, because the core mechanics aren't fun by themselves, or I need to look at what's wrong with the core mechanics. Running and jumping in a platformer is inherently enjoyable, and I think solving puzzles can easily be their own reward. It's worth trying to figure out how to make the puzzles more interesting, or the rewards for the puzzles more noticeable.

Adding additional constraints to change the optimal strategy will certainly change the way the game is played for optimal play, but I suspect it won't be as clear to newer players how making the game more restrictive and challenging necessarily leads to the more enjoyable strategy being a more successful one. Maybe puzzles could have more eye candy, or carry more lasting benefits that persist across later levels. Maybe the puzzles could be tuned to better entice puzzle enthusiasts, or made more intuitive for less experienced puzzlers. There are certainly more than a few things you could try!

Brandon Shelton
profile image
I really like a lot of what you said and it's something I've been thinking about a lot recently. It's interesting that you mention "challenge" modes because I think those are greatly misused in a lot of modern games. This is kind of a tangent, but I wasn't satisfied with the "core" gameplay of Shovel Knight and when people suggested that I try doing challenge runs instead I was just kinda like "why should I bother?" Maybe if I felt the central gameplay was more engaging I would have felt like trying challenge runs; I certainly did exactly that in Mega Man 9 and 10.

Tielman Cheaney
profile image
I spoke with a pretty serious grinder the other day. As a teenager, this person ground his way to the level cap in Final Fantasy VII... on the first CD. He one-hit-killed Sephiroth in the final battle. I can't imagine the hours he put in. I would never, ever, do that. But he did.

Grinding isn't always the worst thing in the world. It's a clear risk/reward system. It provides players with diminishing returns so that they can realize, in their own time, "Hey! Now I'm strong enough to move up!" Skilled, aggressive players will be rewarded with fast advancement and challenging fights, and unskilled/cautious players will be rewarded with steady gains and winnable battles.

Good grinding:
Rogue Legacy: With each death, you must replay at least part of the first level... a clear grind if ever there was one. But by the time you're powerful enough to fight in the third area, you're able to zip through the first in 30 seconds, crushing all in your path and grabbing their gold. It's fun! You have total freedom about where to go and how hard to fight. You can grind or have amazing reflexes, and usually it's a combination of both that yields the most reward.

Arkham City: The game will likely be beaten long before you unlock every skill, leaving you with endless battles against similar enemies to reach the top. But in AC, the core mechanic of swooping down on frightened thugs and pummeling them is so fun, leveling up becomes an afterthought.

Bad Grinding:
World of Warcraft: Clicking "attack" and watching your character fight isn't particularly fun, and over a few thousand hours becomes a boring chore. There isn't much of a reward for real player skill in WOW... a level 50 character can't defeat a level 60 monster, no matter how great the player is. If you're chatting with your buddies while the fight is unfolding, WOW is fun. If you're not, it isn't.

Fallout 1 & 2:
"You miss rat!" "Rat misses you!" "You miss rat!" "Rat misses you!"

Fabian Fischer
profile image
"It's a clear risk/reward system."

Actually it's not. Grinding minimizes risks. The way you get your reward is by putting a lot of your life's time into it, not by taking risky actions.

Rogue Legacy: It's still super problematic. The clarity of the feedback is all messed up due to the persistent stats. You're never quite sure if it was your growing skill as a player that let you get further, or if it was just your stronger avatar.