"Balance Isn't Boring" demos on YouTube

Hey, figured I should showcase my mods a little on YouTube for people who want to see what they're about without bothering to test them for themselves.  Just two videos adding up to an hour or so at the moment.  I'll probably do at least one for each class eventually, when I feel like it.  Melee Engineer and lightning Embermage (+ frost utility skills) for now:






Balance Isn't Boring rant

Anti-one-shotting mod
Potions overhaul mod - diluted and extended
Cooldown and mana mod - burst vs DPS distinctions clearly differentiated

Balance Still Isn't Boring - What's Not To Roguelike?

Way back in the beginning of this blog, I made a post titled Balance Isn't Boring.  The idea that balance can actually mean things like tension, risk, reward, challenge, and ultimately gameplay has returned to the front of my mind in recent months, as I've gotten hooked on a couple of roguelike titles - Desktop Dungeons and Sword of the Stars: The Pit.  (While I have played and enjoyed FTL: Faster Than Light, Dungeons of Dredmor, and Don't Starve, I'm going to just leave them in the "different strokes/folks" bucket after their brief mention here.)

In a roguelike game, resources are ultimately finite and non-renewable.  The player explores some randomly-generated dungeon labyrinth which spawns both enemies and resources.  Death is a one-way ticket back to the title screen.  Small mistakes accumulate, and large mistakes can end a run instantly.  Luck can play a big role in individual runs, but flawless play will result in an impressive win rate.  Classic games most strictly following the formula are Rogue itself and Dungeon Crawl Stone Soup.

Casinos and Casuals

The main idea of roguelikes is that games are supposed to pose a challenge.  There are two genres I can think of that shun this principle.

The first is the electronic casino game, in which the player is not allowed to make meaningful decisions.  Because real money is being exchanged, losing is the point.  I mostly avoid casinos if I can help it, because they are weirdly absorbing in a way that isn't actually fun, and they burn money as quickly and surely as an open flame.

In some sense the opposite of the play-to-lose game, the casual farming game - or lose-to-win game - is the mindless trip to the casino played backward with virtual money.  The Learn To Fly series more or less captures the feel of it.  Every time you play, you lose, gaining money you can use to buy stuff for your dude, repeating as needed until you win.  Lose-to-win games usually include some degree of interactivity that is actually skill-based to some extent, reducing the number of times the player must lose to win if they're playing well.  (If gains are kept in check by diminishing returns or hard caps, then the same mechanics might not lead to a lose-to-win game.  Rogue Legacy has farm-and-die mechanics, but it's impossible to make any progress off of bad runs.)

Either play-to-lose, or lose-to-win - the final outcome is predestined.  These games are possessed with an inertia which plays out inevitably.  They are the tree which falls in the forest so long as someone is there to hear it - these games are the opposite of balance.

Desktop Dungeons - Hard Casual

Not all casual games are lose-to-win.  In some sense, any game which can be played on a workday to a satisfying conclusion without staying up late can can reasonably be called "casual."  This includes Chess, for example, as well as Tetris and Minesweeper.

Desktop Dungeons starts out looking like a dungeon crawling role-playing adventure.  You start by choosing a race and a class for a hero, and then you fight your way through a somewhat small dungeon - possibly stopping to worship a local deity along the way.  The game does not explicitly seem like a puzzle until you play more and learn to appreciate the huge role played by all kinds of "gimmicks" such as level-up heals and "piety storing" to front-load some forgiveness from your draconian deity - which you will need if you anticipate the need to do something Dracul won't approve of, such as drink a health potion or slay a zombie.

As the finish line is never far away, you can always incrementally build your plan as you go along.  "Converting" items from your inventory destroys them and triggers race-specific bonuses.  In more long-term games, the tendency to play the packrat because "you never know" can lead to huge end-game surpluses of consumables.  Not so in Desktop Dungeons, where even your most useful permanent items will be considered for conversion going into the boss fight.  It's a bit of a rush the first time you cannibalize your entire inventory incrementally at just the right times and barely finish the boss with nothing to spare.

This game offers a ton of replayability in the form of race-class-religion combinations.  For the harder levels, you may find that one size does NOT fit all, and that hammering away at your favorite setup won't work all the time.  (The downside is that preparing for a run adequately requires spending accumulated gold from the town's coffer, and it often takes experimentation to learn what will and won't work with a given level's minions and bosses.  A dry run with no preparation is typically in order if you are trying to save gold.)

The game has grown and improved a lot since the free-to-play alpha, both in graphics and in game balance and presentation, but the overall game flow is similar enough that the alpha serves as an informative sample of the finished product without spoiling the whole thing.  It's the ideal "demo" from both a player's and a studio's perspective.

Sword of the Stars: The Pit

Unlike Desktop Dungeons, Sword of the Stars: The Pit is a straight-up science fiction dungeon crawl, with the emphasis on crawl - while the original Pit was originally stocked with 30 floors, the first expansion increased this to a total of 40.  Considering that it often takes half an hour or more to complete a floor carefully, the audience is absolutely assumed to be deeply invested in the outcome of a run.  So The Pit takes the opposite approach - instead of Desktop Dungeon's quick runs with a tight puzzle feel designed to draw in a broad audience, The Pit aims to be a cult classic.  (I strongly suggest starting with Gold Edition.  It is the second of three DLCs, but it is also sold in the bundle consisting of The Pit + Mindgames + Gold.  Mindgames is where the core gameplay reaches maturity with the introduction of psionic powers, but it's just as easy to get Gold from the beginning.  The third DLC, The Pilgrim, is a nice expansion if you own and like Gold, in which case you will naturally want more.)

Easily a week or more's gaming time can be devoted to a single run, and each minute is full of tough assessments of risk, reward, and resources.  Food, ammo, health, armor, durability, and psionic energy reserves are constantly in a state of flux as the player continuously spends one or two in hopes of either conserving the others or finding more of it.  Am I so hungry that I'll waste a few rounds of ammo to get this over quickly?  Or is my health high enough that I want to get in melee range?  Can I just run away, or will I probably just get surrounded later?

Aside from depth of gameplay, there is an odd and paradoxical depth to The Pit as a work of fiction.   The overall plot of The Pit is hokey to the point where Kerberos Productions goes so far to break the third wall.  In a random dungeon run, you may decode a random message from The Scout's sister which reads (paraphrasing), "Hey Toshiko, remember the game we used to play with the secret laboratory and the mad scientists and the monsters?  I think the monsters are real this time..."  Incidentally, the overall plot is that you're diving into 40 floors of corrupted science unleashed because a cure to the Xombie virus is at the bottom, no more and no less.

The depth emerges incrementally with each run and with each new playable character you spend time with, whether it's The Engineer who jubilantly declares that "it can't be worse than calculus!" or the brooding and intense Liir, perhaps best described for outsiders as a "psionic space dolphin".  You'll also meet a fraternal Marine, a psionic soldier who was apparently trained on a strict regimen of watching 90s films on VHS, and the avaricious Morrigi - a sort of six-limbed red-and-teal dragon-bird.

Just getting to know your enemies is a bit of a journey.  For example, you will enter mortal combat against members of the Zuul race at various life stages.  In the early floors, you will find a quiet and seemingly deserted ruins of a research facility where you'll run across a few bats, some rats, and some hairy grub-like fanged Zuul infants, but as you go further, you will find giant rat-like Zuul pups before you eventually encounter the adult female, looking like an apish clawed wolf, and the adult Zuul male, a sort of psionic mastermind and all-around bad guy.

The reason that The Pit bursts with so much fertile imagination is that it is a spin-off of the expand-and-conquer-in-space RTS game Sword of the Stars, belonging to Arinn Dembo's expanding sci fi universe set in the 2400s, including one complete novel to date.  If you ever wondered what games would be like if real writer(-anthropologist)s wrote for them, then this is one of the best places you could look.

The Pit is a challenging mad plummet into the depths of a deranged Sword of the Stars buffet of death.  Have one of everything, but they're not responsible if it doesn't agree with you.  It's a pretty good mix for a hardcore game meant to be played repeatedly - don't bog the player down with too much plot, but do present a rich world and quirky characters with well-controlled voice acting that somehow manages not to be annoying months later.

I do have some quibbles with the game, especially with the inadequacy of the tutorial to unveil the arcane game mechanics one must master to go far in the game.  I also feel like the level-up heal breaks the fourth wall in a way that works for Desktop Dungeons but not for The Pit, and some minor control/interface issues get in the way.  But there's nothing that can't be overcome by Let's Play videos, the wiki, the forums, and old-fashioned dying from different things until you learn how to play better.  I'd like to make suggestions for improvement at length in a separate post sometime, but for I'd just like to recommend The Pit as a modern cult classic hardcore rogue-like that is very slow-paced and takes a weekend or two to warm up to.

(As I mentioned in earlier posts, I started a fan comic about the characters and plot of The Pit, but with a multiplayer twist speculating on how they'd interact with each other.

Also, I finally beat the game on Normal.  I was playing as The Scout and I had really good early luck.)

Balance Isn't Boring

I suppose I should apologize for being a year late to this particular party, but Erich Schaefer's odd remarks about balance rub me the wrong way. Today the "balance is boring" speech continues to serve as the ideal landmark identifying the flaws of Torchlight which Runic "didn't even bother" to address in Torchlight 2.  As you can see from the title, someone needs to play Danica McKellar to Schaefer's Gucci-obsessed Paris Hilton.

It only takes a few afternoons with Torchlight 2 to see that it's a fine loot-centric action RPG with a lot going for it, in general and compared to Torclight 1: great aesthetics with smooth-running graphics that look great on low settings on mildly outdated machines, responsive controls, vastly improved style and variation in level design, more and better bosses, online and LAN multiplayer capability not present in Torchlight 1, and the return of the trademark pets who run your shopping errands for you. But what may turn out to be its saving grace is its extensive modder friendliness.

Indeed, Torchlight 2 comes with the Guts mod editor (so called because it lets you modify the "guts" of the game to your liking). Mods are kept in a folder kept separate from the canonical patch version of the game, and any number of them may be chosen for inclusion whenever the game is launched.

The reason I call this a potential "saving grace" of Torchlight 2 is because of, well, balance issues. Oddly enough, the balance between the player and the enemies is good, but the overall play experience is currently a bit of a spam fest fueled by a never-ending supply of overpowered health and mana potions.

It is a compliment to the game that I think it's even worth modding in the first place, though!

The potions problem

I was only in Act 2 before I broke down and started working on a potions nerf mod, and I'm finding it to be a vast improvement so far. It is difficult to exaggerate how overpowered potions are in Torchlight 2 and the damage it does to the game experience. They are both potent and spammable, restoring health or mana at a fast rate and having little downtime before becoming available again. Mana costs are way too high, but it literally doesn't matter. And health potions? The player has to walk into the proverbial Mordor if he or she is expected to die while under the effects of a health potion - and the game tempts the player to do just that, since potions bolster the player's cockiness up until the critical point at which enemies are overleveled enough to burst the player from 100 to 0 in half of a second, catastrophically breaking the sustain barrier. Barring this circumstance, the net effect of health potions is that the player is unkillable as long as she doesn't play too far ahead and is quick on the potion keys. Runic may as well cut the middleman and simply put up a big golden bar representing the player's wealth which fills up as she gains loot and empties as she casts spells and/or takes damage. Hopefully the reader is as bored by the prospect as I am!

I would go so far as to say that the unbalanced state of potions makes the game more boring. In other words, sometimes unbalance is boring. Anyone who's ever played a first-person shooter with cheat codes for more than ten minutes is likely to agree. Schaefer "doesn't even try" to balance Torchlight 2? Maybe he should try. It's somewhat astounding that potion balance has been totally overlooked when most modern games (see World of Warcraft's cooldown timers, or League of Legend's gradual potion effects) keep them in check. Even Baldur's Gate had this under control!

Let's talk about cooldown timers

Not satisfied with a simple potions nerf, next I set out to break up the monotony of skill use by balancing around cooldown timers. As my examples of the problems I am faced with, I am using my Torchlight 2 remakes of two League of Legends champions - Anivia, recreated as an ice Embermage, and Leona, recreated as a sword-and-shield Engineer.

My Torchlight 2 Anivia has a right-click attack that shoots several bolts of ice in a cone, sometimes (but not always) stunning or immoblizing enemies it hits. She also has a hotkey skill which drops a comparable AoE of ice on the cursor, doing about the same amount of damage and applying roughly the same crowd control effects with similar likelihood. This hotkey skill has a measly four-second cooldown. In other words, I have a second copy of my right click on a short cooldown that adds almost nothing to the game.

I hope this looks like a problem to you too. Imagine if the hotkey skill did several times as much damage, always applied crowd control effects, and had a cooldown timer of something like 10-20 seconds. The result would be that I would have to make a meaningful decision about when to use the hotkey skill. If it also cost a sizeable chunk of non-refillable mana, then things really could get complicated - why, Anivia would become a character having skills different from each other who depends on the precious resources of health and mana! In other words, Torchlight 2 would be a fully-fledged action RPG as advertised, rather than Farmville with mild fantasy violence.


(Forum-goers have also reported that Prismatic Bolts, Anivia's left-click skill, is so overpowered that I could just level that up and let its over-the-top DPS and crowd control more than make up for it being a single-target spell. Much like the Ricochet skill from Torchlight 1, it's a skill that's just too good at everything, so the player can just spam it all day every day - from now until someone releases a better-balanced game, presumably.)
My Torchlight 2 Leona has Shield Bash on right click, naturally. The trouble is that it's too good at everything: zero cooldown, hits in an AoE, applies stuns about half the time, and does more damage than her sword attacks. It's an AoE attack and a DPS attack a crowd control attack all at the same time. I can just spam it all day every day and forget that I even have a left mouse button or... anything else other than my potion keys, really. Shield Bash needs a cooldown timer, and preferably, Leona would have a more appealing left-click option to compensate; the "Sword and Board" passive goes totally unused because it is utterly outscaled by Shield Bash. Boring? Very. Balanced? Not at all.

League of Legends

In League of Legends, Leona would be looking at a cooldown timer of 4-11 seconds, depending on skill level and item bonuses. Anivia has two burst damage skills and a damage-over-time skill which are clearly differentiated from each other. Either of them can use potions to gradually restore health and mana, or buy slower permanent items to do this even more gradually. All game long, each player makes hundreds of decisions which have a real impact on the outcome of the game. Thanks to the balanced state of the game, countless different team compositions can and do win games at every level of PvP play, and they all play differently in meaningful ways. Balance is an important factor in making variety possible in PvP. While a PvE game can afford a little cheese and excess here and there, a PvP MOBA game with an international championship league simply can't afford to have genuinely overpowered skills or champions.

League of Legends has 100+ champs who are balanced and fun to play, but each champion is restricted to a single axis of viable play. Many of these champions have fun-but-not-viable variations that simply can't be used in competitive play. AP Master Yi is no more; AP Lucian was nerfed to the ground before released instea of made into a balanced alternative; Taric's mana-scaling basic attack damage was nerfed into afterthought teritory; Alistar was nerfed into the ground after he won a couple tournament games in mid lane a while ago and has hardly seen tournament play since; supports released after Season 2 are never viable in mid. It is still up to debate whether LoL is overzealous in role restriction, or if it really is impossible to balance along extra dimensions (but they're not there yet).

In any case, PvE games have more freedom to liberate the player from a single pre-defined path. In PvE, all that matters is whether progress is possible and fun. Fluctuations in the power curve don't make a dramatic win vs lose difference before anything else can even be considered. A wobbly game has more chances to balance out in end.

Is balance boring?

It is true that Schaefer's examples were more along the lines of saying that cheese and excess can add excitement to the game as long as it's all balanced in the end:


I could expound on the complexity forever, but it’s pretty dry to discuss and probably not that much different from any other RPG, so let me digress to where I think our methods are different, and where it gets fun, for me, at least. I don’t even try to balance the game! … There are too many systems and too much randomness for my puny brain to deal with. But the more important reason is that I think balance is boring. I specifically want you to find a weapon that’s just too good. I want you to discover a skill combo that makes killing certain monsters seem too easy.

A few levels deeper into the game, you might be struggling to find a replacement weapon, your skill combo won’t work as well against the new monster varieties and your pet will start to seem weaker. The multiple, overlapping systems and heavy randomness work to my benefit in this respect … So all my spreadsheets and assumptions become less important as we finish development, and I concentrate on playing over and over again, getting tons of feedback, and ironing out the really crazy peaks and valleys. Fun always trumps balance.


The trouble is that his remarks, on the whole, feel like a false dichotomy. He says "balance" when he means "completely homogeneous blob-like state of never-ending same-ness". I personally prefer to think of Shaefer's ideal game state as a form of dynamic balance, a balance which the player labors to tip in her favor in order to achieve progress and complete objectives, as opposed to static balance in which everything is indistinguishable from everything else and it's all just a matter of clicking enough times.

Incidentally, Schaefer's idea of "balance" describes Torchlight 2's world of never-ending potion spam and cooldown-free, mana-unrestricted AoE/DPS/CC all-in-one right-click skills.

On his own terms, maybe Schaefer is right. Maybe balance is boring. Maybe Runic needs to let fun trump balance.

I'll be working on my mods. See you around.