Dialing It In

It had been a scant seven or eight turns since a cadre of my battle-hardened musketmen stormed through the streets of London. The cannons, which had camped ominously on the heavily mined hills outside the city limits, were now coming into position outside the final bastion of the crippled nation. There was no longer any army to speak of, and the defense of Birmingham was tepid to say the least. A few volleys of cannon fire, and then the ground forces surged in, unhindered.

With that, England fell, just as the Chinese had done a few hundred years before, and my Celtic eye of Sauron turned its burning gaze to the Americans. A small, simpering nation that had been bottled into a tight but resource-rich corner of the continent. There’s no real reason to steamroll, but the war machine is already active, and the annihilation of the English was not as taxing either to my economy or my army as I had initially planned. I call it Manifest Destiny, as great swaths of units swivel eastward out of the conquered lands of burning cities.

And over the course of this conquest, my game of Civilization V now begins to feel too easy. What had been challenging and tense as I juggled relationships with demanding countries to my north, east and south, is now a somewhat mechanical effort of maneuvering a juggernaut army counter-clockwise across the continent. I already begin to daydream of playing through with different parameters, different countries, different goals, and I know that barely into the Renaissance I will never actually finish this game.

Just as I would likely not have finished it had the game been too hard. No, for today the range of difficulty under which I am engaged is here at point x, and this game is providing the area y of difficulty. Since x does not fall within the bounds of y, I’ll just go play Diablo.

Another way to think about it is this: Imagine the spectrum of difficulties in the gaming universe to be like the electromagnetic spectrum. Here, in what would be the realm of gamma rays and x-rays, are games like Super Meat Boy, Ninja Gaiden and most ‘shmups. On the other side of the spectrum, over there by those long radio waves, are Kirby’s Epic Yarn and arm wrestling my three-year-old. And somewhere in a compact realm, barely a sliver of the grand spectrum, is what we think of as visible light. That tiny segment is a good representation as the sweet spot for me, the narrow window of difficulty that describes games I like to play.

Of course, difficulty is like any other factor in games, and it can trump or be trumped by other considerations. In fact, I played the snot out of Kirby’s Epic Yarn in part because it was so damned delightful, and equally in part because it was a great time for my son and I, despite being criminally easy.

Similarly, I am currently working my way into Act 2 of Diablo 3 on Inferno mode, which at times is about as easy and fun as choking down a spoonful of cinnamon. But I find myself persisting because I really like it when stuff pops out of other stuff that’s been hit with a sword, at least in videogame terms. In both cases, these games fall well beyond the boundary of my normal spectrum, but if I’m honest with myself these are also cases that represent the exceptions, and most of the time I’m pretty rigid about difficulty.

I would say that being someone responsible for balancing difficulty in a video game is a job I would never want to have, but these days I’m not sure I’d ever want any kind of job making video games. However, if forced at gunpoint to work for a developer of a AAA game title — which I assume is not the standard corporate recruiting process — one of the jobs I would least want is Guy Who’s Responsible For Figuring Out How Hard the Game Should Be. One, because there feels like such slim margins for success, and it is extraordinarily likely that a lot of people are going to yell at you. Two, that’s a terrible job title to have on a resume.

Take a game like Portal, which in my mind is at the exact, perfect center of my difficulty spectrum. Playing that game, it is patently evident how much work was spent balancing, tuning and engineering the experience. You almost can’t help but notice how dialed-in the difficulty level is, and whether it falls in or out of your personal spectrum, I have zero doubt that Portal is exactly where Valve wanted it to be.

The thing is, what I really remember after playing a game is the experience. I may remember on a cognitive level whether a game had good graphics, decent sound design, cool weapons or an interesting setting. But the feel of playing the game — that emotional sense evoked by just remembering a game — is hard to define. It is a sum of numerous parts, and in many ways feels held together by the framework of how challenging the game was. And so many things, so many variables impact that.

For example, if I know what to do to succeed and I have the in-game tools to accomplish that task, but I fail because of some mechanic that seems arbitrary, then I’m going to get frustrated. Or if the camera doesn’t work. Or if some small segment seems disproportionately and arbitrarily difficult. These are all things that become the experiential memory.

When I remember games, I remember how I felt playing them, and one of the most powerful feelings I can experience while playing games is frustration. It is the experiential memory I am most likely to conjure first, because in many ways it is usually the last thing I experience with a game before I quit. This is my playtime, my brief but critical recreation, and I don’t want to spend that time feeling frustrated. It is, of course, possible to create experiences that are difficult and rewarding at the same time, but I think it is a hair’s width tightrope to walk.

I realize, of course, that my capacity for difficulty is my own, and it probably exists at a much lower threshold than other core gamers. But the more I think about it, the more I believe that it is one of the most impactful traits in not only determining whether I will play or continue to play a game, but also that it is one of the biggest deciders in whether I think a game is good or not. Because, ultimately difficulty isn’t just about a game mechanic, it is about your accessibility to the game itself.

A poorly tuned game, particularly one that is too challenging for its audience, has a flaw as fundamental as a showstopping bug. What’s the difference between reaching a point in the game you can’t get past because it crashes, or one where you can’t proceed because you can’t execute. In the end, the result is the same.

Comments

What makes it even more, um ... difficult, is that I cannot reliably tell what is the best spot for difficulty for me. I played through and "finished" Dark Souls and Dragon's Dogma, even though I gave up on both Demons Souls and Oblivion after two separate attempts. A huge difficulty spike will certainly sour me on a game (I'm looking at you Metroid Prime 2), but if a title captures my imagination, I will barrel my way though it. However, I fell I cannot tell if that will happen or not until I play the game for several hours.

TL; DR- Difficulty is hard.

This is definitely something that I've experienced. For that matter, this is something that Flow tried to address. In linear games, only Valve seems to really avoid this. For non-linear games, like Civ, I find that they're often easier to self-manage the difficulty, not taking on more than I can handle at once. Of course, in either case, too much mastery of the system shortcuts the carefully balanced challenge. On top of that, difficulty levels often tweak the wrong thing to adjust (e.g. adding more enemies is easy but doesn't usually make things strategically more complex).

I think this might be one reason I dislike strategy games that boil down to puzzle missions; it feels like there's only one viable solution which is either way too easy or way too hard. For more open games, like Civ and Crusader Kings, there's much less hard gating and feeling like I'm falling behind with every wrong move.

And, like Aristophan said, it can be really hard to judge the best degree of difficulty until after it's over. Which makes difficulty levels doubly useless, since they're set at the start.

Elysium wrote:

the defense of Birmingham was tepid to say the least

To be expected, really.

A poorly tuned game, particularly one that is too challenging for its audience, has a flaw as fundamental as a showstopping bug. What’s the difference between reaching a point in the game you can’t get past because it crashes, or one where you can’t proceed because you can’t execute. In the end, the result is the same.

Not really. If the game crashes (and it was really good up to that point), I'm left with a feeling of "missed opportunity" and curiosity. If it's because of difficulty, I leave frustrated and, later, accepting my limitations to finish it.

Great read as usual, Mr Sands.

Excellent article Sean. Very much sums up exactly how I feel on the same subject. The paragraph on frustration should be required reading for anyone developing games.

I think a big part of it is whether I feel that I can improve my playing and figure it out. If it's just a matter of trial and error, or if the answer/solution/movement is so obtuse that I can't figure it out, it won't be fun. If after dying (even if it takes a few deaths) I say, "OH! That's what I need to do." then that works, that's a success to me. If after dying I say, "Ummmm...huh?" then it doesn't work.

I think that one of the worst offenses when it comes to difficulty in games is suddenly introducing a new mechanic that's totally different than what came before.

I'm looking at you, all you platformers that decide they need a random kart/rail racing segment in the middle of the game where you HAVE to get first place.

Oooo - my specialist subject! The reason why Portal has such great difficulty level balancing? Valve are mentally strong enough/have enough money and time to user and playtest their games to death.

This means they can get the difficulty curve just right, and also get to learn loads about the people playing the game - their expectations and abilities.

I genuinely believe/know there isn't a single game that couldn't be improved with pure user testing (rather than the developers getting their friends to test).

*Climbs off podium*

To be fair, I've probably started 100s of Civ games, finished less than a dozen, and still count the Civ series as among my favorite games EVAR.

Same with SimCity, actually. Can't wait for next year, to leave countless cities abandoned and lonely in my wake.

My brother-in-law and I were talking about difficulty just the other night. Because I was involved in the conversation, we started talking about the Shin Megami Tensei games, specifically Strange Journey and Persona 3, and just how easy it is to die in those games. In one play session of Persona 3, I was randomly separated from my party when we progressed to the next level of the dungeon. I encountered a single low-level enemy who got the jump on me, hit my character's elemental weakness to stagger her, hit it again to incapacitate her for a turn, then took a couple free turns to kill her. I didn't even get a turn.

I lost about an hour of progress, maybe more, and it didn't bother me. I didn't get upset, and I didn't think it was unfair. The battle was over as soon as I let the enemy get the jump on me, and I knew it. But I also knew there were things I could have done differently to prevent getting killed: I could have equipped a different Persona so that I had a different elemental weakness that was less likely to be exploited by the enemies on that floor, and I could have been more cautious in my exploration so that I surprised the enemy instead of the other way around. I fired my save right back up and kept playing.

My brother-in-law, on the other hand, said he would have quit playing entirely. That kind of setback punishment wasn't interesting to him, regardless of whether or not there were steps he knew he could take to keep it from happening. One or two small mistakes costing you an hour of progress? That was simply out of line.

He likes his difficulty tuned a little differently than I do. Combat should be challenging, but if he has to repeat a boss fight more than a couple times, he'll lose interest. Repeating a fight against a random mob? Forget about it. And death should set him back, at most, to right before that last encounter. Me? If each and every fight isn't an opportunity to die, I'll get bored. If death just sets you back to the beginning of an encounter, it robs your progress of tension.

And you know what? I don't think either of us is right. I'm just happy there are games on the market that cater to both of us.

I don't know. I think I've got the bar set a bit differently than Elysium. I voluntarily play Rogue-likes and consider perma-death a feature.

Part of it is because for me, the struggle is with the ability to accomplish the given action rather than trying to figure out what is needed to be done. My age, vision, hands, and schedule are more of an opponent than that boss, so the failures feel like they're mine instead of the game's. Even if it is some sort of wonkiness in the controls I'm more likely to blame myself. Particularly when I see the kids play it and breeze through.

That, combined with a wide streak of stubborn old bat will keep me beating my head against something to the point of risibility.

Gremlin wrote:

And, like Aristophan said, it can be really hard to judge the best degree of difficulty until after it's over. Which makes difficulty levels doubly useless, since they're set at the start.

That's one of the biggest mistakes I still see games making today. Locking in the difficulty for the entire game. To change it having to start over. Sometimes there's just one level or one boss or one something that I can't get past on Medium/Hard/whatever and just need to turn it down for 5 minutes. But the game won't let me. And instead I quit and never finish the game.

Ghostbusters was the most recent title to do that to me, when I played it last summer. I should have put it on Easy to start, just to enjoy the experience, but I'm stubborn and I don't need to play on Easy and blah blah... somewhere on the last or next-to-last chapter of the game I had to bail after a bunch of frustration.

And as the article points out... overall it was a great game, but all I really remember is quitting it in frustration.

LordSpaz wrote:

I think a big part of it is whether I feel that I can improve my playing and figure it out. If it's just a matter of trial and error, or if the answer/solution/movement is so obtuse that I can't figure it out, it won't be fun.

This is pretty much how I feel too. I'm not a huge fan of trial and error to figure out the ONE solution to a problem, and if I don't feel like the solution is anything skill based (i.e. you need to find the superweapon) then I'll quickly get frustrated and either shelve the game or, if the rest of the game is good, I'll swallow my pride and look at a FAQ. Which ultimately will cause me to spew forth all kinds of obscenities with the same theme: "How the heck is THAT the answer?"

However I'm also a hypocrite since a game like Persona 3 is basically trial-and-error as you try to figure out the enemies weaknesses, but I freaking LOVE that game as a whole. So while I can agree that the difficulty can be a big factor, I think having a solid game (mechanics or narrative) around that difficulty can go a long way to offsetting any balance issues that may be caused once a gamer figures out the system.

Cassius wrote:

I genuinely believe/know there isn't a single game that couldn't be improved with pure user testing (rather than the developers getting their friends to test).

I think there are a few shops that over-test their games, to the point that the game becomes too slick for me, too polished to be interesting.

Playing through Catherine on Easy Mode right now, because I want to see the story more than I want to pull blocks. I'll probably give it a try on normal/hard at some point, but the Q-bert aspect is less interesting than the characterization.

I play some games for their storyline, some games for their battle system, and some for both. Catherine is a game I play for the story, so I stick it on Easy mode.

Of course, the games I play for their battle system don't necessarily need to be hard. They just need to be interesting in the way that you win. Grandia 1-3, FF13 and 13-2, and the Mana Khemia series are my best examples of games with interesting battle systems with uneven difficulty. Even though they're hard at times and easy at times, even when they're easy it's interesting to find a really good way to take the enemies apart as fast as possible.

I don't consider excessive difficulty to be a problem as long as you can go find a solution to it online after trying for 20-30 minutes to beat it yourself.

GalCiv2 is another one where the challenge isn't necessarily the reason to play, but the satisfaction of spending 300 turns setting up an end game where you completely roll over the enemies. I recently played one where there were 2 other major races and 3 minor races where for 270 turns none of the major races created any ships with weapons on them, and most of the planets around the map were extreme colonization ones. Most of the early part of the game was a contest for the few planets that weren't extreme colonization, and picking the ideal tech to get the best planets by scouting quickly to figure out which type had the best candidates. The Altarians and I had about the same number of colonies, but I had cherry picked most of the 17-25 quality planets in the galaxy with my careful research, scouting, and faster colony ships while they had far more 4-7 quality planets.

At the same time, I had researched large ship types, some low-end weapons, moderate miniaturization, and very fancy defensive techs ahead of time, so when the Korx finally broke the long silence of no warships being built, I was able to fully defend all of my planets within 15 turns using ships that would have taken me 50-80 turns of research to build if I started that late. One of my ships could take out a fleet of 6 of theirs without taking but maybe 1-2 points of damage. They surrendered rather quickly when the shooting started and I didn't even need to build any transport ships.

The Civ series has always had a lot of mechanics that take players that are ahead (in terms of wonders, cities, tech etc.) and put them further ahead, or if you are behind you tend fall even further behind.

Staying on that knife's edge (where the game is challenging but not impossible) for more than an hour or two is very rare, which is a problem since Civ is usually an 8-10 hour game. Add in some dumb unit AI that cancels and forgets your orders over the smallest problem and you get a long boring endgame.

The real mystery is that in spite of all that I CAN'T STOP PLAYING.

The way I keep Civ interesting after I pull ahead is similar for all the Civs, since they all have this problem. Once I pull ahead decisively, the game isn't about beating the other Civs, but in finishing the VC in as little time as possible - it becomes a timed achievement game.

Personally I have a complicated relationship with difficulty. Shooters and the like I tend to play on either Medium or Easy, mostly because I play them to unwind and I don't want obstacles in the way of my action movie. I'll usually start on Medium then turn down if I get challenged too much. I haven't done the Pile in a while, but when I do and I just want to burn through a few titles then I start on Easy.

RTS games I like to pander to me, to give me an illusion of challenge yet fold eventually to make me feel smart. So I'm Medium all the way.

In RPGs I like some challenge. People complained about the difficulty in DA:O and Witcher 2, but I thrived. I'll confess I turned down for a couple of Revenant battles in DA:O, but I really enjoyed using as many of the tools I had in the toolbox and party positioning to make progress.

Witcher 2 had some concrete walls to throw yourself against early on before they patched a lot of the difficulty out. I'm sure everyone who played then remembers the courtyard. When this happens I do get frustrated sometimes and need to step away, but I usually dwell on the game and come back with a new strategy.

CptDomano wrote:

However I'm also a hypocrite since a game like Persona 3 is basically trial-and-error as you try to figure out the enemies weaknesses, but I freaking LOVE that game as a whole.

You touch on it later on in your comment, but I just want to spell it out. It's not hypocritical, or unusual, to get different benefits from different games. Each one is (usually) a unique collection of systems in a type of dynamic equilibrium. Different systems pull you in different directions, but as long as the overall experience is giving you a benefit then you keep playing.

Civ and similar games are rough because there tends to be a tipping point, where in the beginning you are small and weak and it's too hard, but if you manage to survive, your success tends to snowball until you're crazy OP and the rest of the dominos fall like a house of cards. Checkmate.

There are ways to design around that, but many of those (rubberband AI) carry their own pitfalls.

Based on what I'm reading in other comments, I think my difficulty "sweet spot" is not so much on a "spectrum of difficulty". No, that's too linear for me.

It's more technical, boring, and bland, (and certainly not good writing material) but if I were to visualize it, I'd make it more of a 3d graph where X, Y, and Z represent the three most important factors of the five following attributes: Genre, Brand, Story, Mechanics, and Difficulty.

If a game is marketed as being supremely difficult and I'm into that kind of thing, then my X, Y, and Z would be Brand, Mechanics, and Difficulty. If I'm playing a Final Fantasy game, then it's Brand, Mechanics, and Story. If I measure these three things and the net result is in the positive, then I'm more likely (but not guaranteed) to complete the game.

I think there are a few shops that over-test their games, to the point that the game becomes too slick for me, too polished to be interesting.

That is true... but there are a lot more that I would definitely play with a smoother learning curve/clearer introduction.

Just curious Sean, what difficulty do you play Civ V on? I find that on King I can still conquer my continent but there will still be other world powers out there on other continents that I won't be able to take out until they're super formidable. Group comp stomp against Emperor level civs is also a real challenge.

Great read, as always Sean.

When I remember games, I remember how I felt playing them, and one of the most powerful feelings I can experience while playing games is frustration.

I have found I only remember the frustration when it leads to an unsatisfied end. If I work through a tough gaming situation, that has frustrated me for a while, but it turns out it was simply my approach was wrong itś one thing but, as you said, if it comes down to missed timing or forced into an arbitrary mechanic, it becomes a much different experience.

It is, of course, possible to create experiences that are difficult and rewarding at the same time, but I think it is a hair’s width tightrope to walk.

There in lies the problem, how to make the game challenging enough for the majority of the audience, without making it too easy or too hard in the process. As a developer (software, but not games...don´t want to ruin my playtime) I can understand the difficulty in getting the balance right but eventually someone is going to find it too something and you just have to get it as good as you think you can with the time and resources available.