Achievement Spread

Games with achievements do better. I’ve lost my citation of the Microsoft data analysis that showed this, and it would be a stronger effect there because you have an overall score for total poundage of achievements there, but at least grant it for the sake of argument here because the point lies beyond this. Players are more likely to buy a game, play it more hours, and rate it more highly when it has achievements.

My question is how you set those up. Take two rather different MMOS: DC Universe Online and Yohoho! Puzzle Pirates. DCUO has hundreds of in-game “feats,” of which 23 are Steam achievements. Most of those are “beat the game as x” or items from the DLC. Puzzle Pirates has 220 Steam achievements, many of which are the ranks of the various mini-games. I haven’t played Puzzle Pirates in a long while, but I’m guessing it is not 9.6 times as much game as DCUO. DCUO gives you an in-game feat after any story arc or minor accomplishment, but outside the game you see one at the start and then come back for a few shinies when you beat the game. Puzzle Pirates thinks it better to give Steam achievements to you constantly.

I’m wondering to what extent these are business or game design decisions, or perhaps very little thought goes into them at most shops. Achievements seem used to note progress, to highlight nice touches, to reward people for doing difficult or poorly designed content, to incentivize perverse behavior in team games, or to reward very long term play of the “collect 1 billion x” sort. See Torchlight for examples of most, from “Find the entrance to the mine” through the course of the game, over the game’s various difficulties, and into long hauls (100 levels) and the WTF of “talk to the horse 100 times.” Alternately, see Grotesque Tactics with just 10 achievements, 4 of which you can/must complete in the tutorial, and I assume the game slows down that pace or else the whole thing must be about two hours. (I and many others must have picked this up in a sale pack, because 62% of owners never made it as far as the first fight.)

There must be some optimal system of achievements that serves as verbal praise to encourage and reward the player. (I’m also fond of games that give bonuses for them, like DCUO’s feats that grant skill points.) It’s strange how rarely achievements are treated as a serious development subject, since they affect how players play games and feel about them. Like drugs and other things that affect your meat brain, psychological tricks can still be quite effective even if you know they’re there.

: Zubon

If you don’t care about achievements, you don’t need to comment to tell us that again. Really.

Psychonauts: Achievements

The Psychonauts achievement list is an exciting mix of things you gain through normal play, pointers to little gems, stretch goals for achievers and completionists, and “dude, seriously?” The normal play ones are what you might expect: beat each level, level up, complete the game. The completionist goals are exactly what you’d expect, although actually completing everything turns into “dude, seriously?”

Pointing out little gems is one of the best things about having achievements in games. It gives developers a way to encourage and trivially reward players for seeing the best that they have to offer. “I’m Sure She’s Over It” is a wonderfully compact bit of storytelling, and it explains what might otherwise be a BLAM in a later level. “Self Aware” directs you to some subtle content you might not otherwise try, and it includes some foreshadowing for the aware player. The achievements also include some direction on what to do next if you are stuck and a few things you may not have realized you could do (like “I Think They Were Impressed” and “Stump Speech”).

Some of the achievements are unfortunate, or perhaps just unfortunate in how they interact with the game design. I still cite WoW’s The Green Hills of Stranglethorn as the worst achievement ever. It presumably was meant as a reward for completing some arduous content, but its effect is (was pre-Cataclysm?) to encourage players to slog through some of the worst content in the game. Don’t encourage your players to punch themselves in the face, and don’t be surprised when they complain about your game after they react to the incentives you gave them. For Psychonauts, those achievements are the ones that become unachievable at a later point in the game and 100% completion, which is painful here. Out of 35 achievements, 6 become unavailable unless you did them at the right time (or saved immediately before) and another 6 demand a separate “point of no return” save game; there is no post-victory wandering around the map and re-visiting levels. These stack on each other: the most painful are gathering all the figments (which are 2-D, translucent, sometimes mobile, and can be half-inside objects or entirely inside them while moving) and the resulting max level of 101 that calls for several “gather all”s. This is more of a criticism of the difficulty of 100% completion in this game, but there is an achievement encouraging you to go for it.

: Zubon

The percent who have completed each achievement is driven down by the late addition of achievements, although I am suspicious that even 0.5% of players have legitimately gotten a 100% completion. Contrarily, there is the Torchlight achievement page, which says that 11.8% of people who own Torchlight on Steam have presumably never installed the game.

Engi Census

Playing the Steam free game of the weekend, I have come to wonder: how many games have an Engineer that builds a turret; how many games have an Engineer that does not build a turret; and how many games have a non-Engineer that builds a turret. (I think I will avoid counting Warhammer Online’s Magus and units/classes that “summon” rather than “build.” I’m unclear whether the Raven builds, summons, or do we count “deploy”?) Was there some first game that set the standard that Engineer = build a sentry gun? It feels like engineers and self-directed turrets have become a standard game item, but perhaps exploring some examples will reverse this. I keep finding near-hits, where perhaps they consciously avoided calling the turret-builder an Engineer in recent games. I wonder if non-builder Engineers are also intentional aversions? Inventory below the break, please contribute in the comments.

Edit: let’s see what happens if we add in enemies that do the same, some of which may mirror heroes. Continue reading Engi Census

Torchlight Sets

Tesh reminds me of Torchlight’s set items. I was excited the first time a purple item dropped, which might just be conditioning from MMOs that make purple the color of awesome, but upon reflection I never found much use for sets. As Tesh says, you out-level items very quickly; in Torchlight, leveling every 15-20 minutes is taking your time.

Ah, but there is shared storage and inter-generational item transfer in Torchlight. You can gradually build up the set and make use of it every run. Well yes, maybe if I enjoyed inventory management more, but that seems like a lot of work for a small bonus. Meanwhile, there are awesome non-set items I would need to give up, and I can risk them at the enchanter all I like without threatening to gimp the set. (Given enough times, you will have items survive 10 enchants, and given enough times, you will have set items disenchanted on the first try. One of the former gives me a permanent “best in slot,” while one of the latter breaks my set.)

But this assumes that I play Torchlight enough for it to matter. How many runs would it take to get a full set of 4-6 items in the same level range? 6-10, if you are lucky? And then that set is the best for maybe an hour or two per run? If you beat Torchlight once a week, that is probably worthwhile, although after 10 runs you will also have some very nice non-set items. I would be surprised to find many people who play through Torchlight once a month.

It seems like an over-designed subsystem. It works very nicely if you play for hundreds of hours, but it is just noise if you beat the game once with each class.

: Zubon

This is the reverse of Magic: the Gathering. The original designers knew the system would have problems if players spent hundreds of dollars on cards, but if lots of players were spending hundreds of dollars on cards, great!

Oh, and read the comment from Darius at Tesh’s. Interesting notion!

Developer’s Gift to Modders

One great way to encourage your community to add to your game is to make it easy for players to get those additions. It is not enough to have tools for modders; you want it to be easy for players to use those mods, which will encourage their use, which will encourage their development, virtuous cycle ho!

Having it in-game helps. If I can access player-made mods, maps, quests, characters, etc. without searching the web and planning ahead, I am more likely to do so. The trivial inconvenience of downloading something and putting it in the right folder will keep many of your players from using much of anything beyond the standard game. The hardcore MMO player who will crawl through barbed wire for his fun is an atypical user.

One great gift to modders in Torchlight is having an achievement for using mods. People will do absurd things for useless badges and shinies that no one else will ever notice their having. A player who might never have looked at mods will notice when there is an un-checked achievement box to try them. It is a tiny thing for the developers to do, but it adds much encouragement. Yes, it also rewards “do it once to get it out of the way,” but some people will find interesting things in that try.

City of Heroes combines the two with Architect Entertainment. There are in-game sites devoted to trying player-made content. There are badges awarded for doing various things. After the initial release, the developers even toned down the badges to “try it” rather than “grind it until you hate the system and play only farms.”

Ease of use plus a low-cost nudge: encourage modders and harvest their ideas for official releases.

: Zubon

Torchlight Difficulty Levels

The Torchlight 2 announcement came just after I started poking at Torchlight again. I had long ago started a game on the highest difficulty but wandered off after getting tired of how that slowed down the game — it is just a matter of giving all the enemies bigger numbers, which means that my pets drop faster, monsters take longer to kill, and more things one-shot me. The game’s bosses are easy except for their masses of minions, which becomes really annoying when the end boss’s big ability is summoning masses of minions (and sometimes consuming them).

I finished that run-through this weekend. One oddity of Torchlight is that the meanest attacks are area-effect. Most games work on the principle that multiple-target effects have lower damage, but the nasty things in Torchlight tend to hit entire areas. Continue reading Torchlight Difficulty Levels


Hype has become the subject of the day, and I will contribute two repeats to the discussion.

First, You Are Judged Against Your Hype. Doing something modest very well gives you Portal or perhaps Torchlight. Take your pick on “shooting for the stars and not even delivering all the features on the box.”

Second, the example that always comes to mind on “failed to meet explicit promises” is Warhammer Online, as Zoso points out. If you ask me about WAR and I just mutter, “bears bears bears,” that is what I am talking about. Not only did developers explicitly identify a problem, identify a solution, then implement the problem exactly as described, but you were reminded of it constantly. Every time a quest sent you back to where you just came from, “bears bears bears.” Every time you killed a named enemy then got a quest to kill that named enemy, “bears bears bears.” Every time you saw a kill collector, the half-arsed version of the solution, “bears bears bears.” Then later tiers had such content/leveling curve issues that they added a bunch of kill ten rats quests as an improvement, and it was an improvement. Bears bears bears.

I am ambivalent about hype. I am skeptical, but I am gullible enough to take what people say at face value. It is not as though I am hurt if they fail to meet expectations they explicitly set; I just don’t trust the company or anyone who was identifiably a factor in lying to me.

: Zubon

Note that there is a separable issue for just doing badly. Alganon is a game that delivered everything it promised [Carson says no] badly. Earth Eternal seems to have had a similar problem.


Turnaround Time

How is it that Runic Games is publishing Torchlight 2 early next year while other games languish in development Hell for years before coming out late, buggy, and poorly balanced? My compliments to whoever is in charge of production and logistics.

I understand that Torchlight is not the most ambitious project ever. It is narrowly defined and does one thing really well. It out-Diabloes Diablo. It is also a single-player game, which simplifies quite a bit. I also respect the discipline it takes to make a narrowly focused game, rather than trying to be all things to all people. We have seen that. We have seen that fail.

Torchlight 1 seems like a teaser for the real thing. Single-player, limited range of options: it is a demo expanded to something they could use as a fund-raiser. Torchlight 2 is adding multi-player and other things people thought were missing. By the time Diablo 3 comes out, are you really going to be interested in it except as “more of the same” of Torchlight? And they’re making an MMO. It is like seeing someone re-create Blizzard, if they had it to do all over again with less baggage, more experience, and no Activision.

: Zubon

Onion Headline Syndrome

I like The Onion, but I rarely find myself reading much of it because the full text rarely improves on the headlines. You might need to read the first paragraph to see where they are taking the joke, but stringing it out for 1000 words does not add much to the first 5 seconds. (I might take this as an object lesson, but look at me go, still typing.)

Syp finds the same problem with Star Trek Online, I said the same thing about LotRO skirmishes, and many of us have said the same about Borderlands and Torchlight: it is great at first, but there is not all that much improvement or variation over time. (I do credit the two single-player games for having interesting boss fights mixed into the repetition, where MMOs tend to rely on even more repetition, even in tank-and-spank bosses.) I appreciate being able to get 95% of the benefit in 5% of the time. Portal did that brilliantly and then ended.

: Zubon

Non-MMO inspiration banished to the first comment.