Saturday, March 21, 2009

The challenge of parallel exp curves

Tobold proposes that players prefer to solo because the exp is better, i.e. that players choosing not to group while leveling, sometimes to the detriment of their own gaming experience, is an incentive issue which can be solved by throwing more exp at the group players. Why haven't studios tried this?

The group/solo decision is not purely an incentive problem

First of all, the suggestion that players will group more if presented with better rewards is, at best, an oversimplification of the issue. A player who cannot commit the time needed to find a group and complete the group content will not be able to change that reality simply because someone has doubled the rate of exp gain per time for dungeon content. (In WoW, the bare minimum time to clear a 5-man heroic dungeon is somewhere between 30-60 minutes, depending on the dungeon, but that time can easily double in a group that needs to practice the fights, and triple if you just can't find a healer.)

This added overhead to attempting and completing group content can be improved via accessibility - this was what made Warhammer's Public Quest feature sound so exciting - but it cannot be entirely eliminated. One point that Tobold DID catch is that we can't be talking about SLIGHTLY more incentive to group. Because of the time investment required to find a group, players probably aren't going to bother for a small exp boost. You need to be offering LARGE amounts of exp and gear before it is worth the hassle.

Scarcity of Content

Looking beyond that, for the sake of argument, Tobold is effectively proposing that a game be designed with two separate exp curves - one for solo players and one for groups. This is not a bad idea in theory, but it runs into the 800 lb content gorilla in the room - developers simply cannot create content as quickly as players can clear it.

For the most part, the content in WoW is set up so that players solo all the way to the endgame and only then begin the group content. This setup causes various problems - players either do not want to make the switch or turn out not to be very good at their new group role for lack of practice, and, as Tobold complains, are not very community/group minded because they simply haven't had to be. Tobold suggests that the problem would get better if Blizzard simply went through and added a parallel set of content for groups that offered double the exp, but this plan cannot get past the gorilla.

The current setup in WoW is not accidental. Due to the time requirements for forming groups, the players who would be able to take Tobold up on this offer are the ones who are spending the most time online. After we're done adding the parallel content, effectively 1/3 of the game would be group content and 2/3 would be solo content. It simply does not make sense to take the most active players, who are going to run out of content the fastest, and have them completely skip 2/3 of the content in the game. It makes even less sense if you consider that Tobold's ultimate goal is to get the majority of the community participating in groups, at which point the majority of the community would be skipping the majority of the content.

Under Blizzard's current model, the players who can do small groups must first use up most of the solo content, the entry-level raiders must use up most of the small group content, and the elite raiders must complete the entry level raids. There is simply no way for Blizzard to produce equal amounts of all of this content in parallel. (Even if they did somehow manage it, players would promptly start poking holes in the incentive structures - when WoW launched, players overwhelmingly preferred to run the 5-man level 60 dungeons with 10 players, even though this drastically reduced the challenge of the content.)

A niche for scaling content
Tobold thinks there is a market niche for a game that maintains accessibility without sacrificing community. I think he's right, I just disagree that having a parallel solo content track is the way to accomplish this. The "every class must be able to solo" paradigm is not the only route to accessibility, it's just far and away the easiest to implement since it does not rely on any assumptions about the player having the appropriate number of willing group-mates of the appropriate class mix to beat the content.

The approach that might work would involve scaling content. Content would need to scale with the number of players available (from some minimum, perhaps 3-5 to start, going upwards to whatever limit the server can handle) to clear it and, more importantly, needs to scale with the ABILITIES of the group. The accessibility of current group content is hamstrung by the requirement that a specific proportion of the players take on specialized tanking and healing roles - activities the devs have not been able to make sufficiently interesting to attract a sufficient proportion of players to fill those roles. If dedicated healbots simply aren't enough fun, the devs could shift towards a game with no healers and some mix of self-heals (e.g. one class has a large instant self-heal on a cooldown, another class has a medium self-heal they can cast as needed, a third class has a small passive heal for dealing damage) to maintain the challenge level.

You would also need to offer options for shorter content, or some variation on the Warhammer "come whenever, stay as long as you can" public quest to maintain accessibility. And you would be critically dependent on having enough population to support the system (i.e. if a player was the only one online, they'd have literally nothing to do). Alternately, you could implement crafting with an EQ2-like separate exp bar and have that be your solo content.

Designing, implementing, and balancing such a system would be tons of work. However, the potential payoff would be huge. You would no longer to worry about parallel content - everyone playing the game would have access to all the content. You might lose some people who absolutely cannot play uninterrupted for more than 5 minutes at a time, but that's a given when you're talking about a niche product. And, most importantly, you wouldn't need to make enough content to go head to head with World of Warcraft. The WoW approach requires a volume of content production so large that even Blizzard can't keep up with it, and anyone else making the attempt will not have Blizzard's dollars to pay the development bills.

Anyway, the point being that sometimes the game, not the incentive curve, is the actual cause of the problem.


  1. I don't agree with much in that post and I often agree with you. First, I don't agree that WoW is designed to be soloed until the end game and then you move on from there. Certainly that is one possible approach, I just don't think the game was designed to be that way. I think the game was designed so that people can take whatever route they want to the top. I know people who what gotten the vast majority of their XP via instances and others who have not. For myself, I have PUGed every old world instance at the appropriate content level.

    Second, I seriously disagree with the assumption that instances give less XP than soloing. In fact, this was something I actually ran the numbers on quite early in more WoW career. If you do instances at the correct level, the XP you get from the instance is effectively the same as if you were to spend that same time soloing. Where instances really shine, however, is in the XP you get from the quests. Because so many quests are concentrated in such a small area, the XP from the quests alone equal the XP from the monsters. And given that quests can be shared among group members, running instances is by far and away the most time-efficient method of gaining XP in the game. It kicks soloing in the rear end and tosses it off the cliff.

    I do agree it is now more time consuming to find PUGs for old world content than it was before and I suspect that problem will continue to get worse as more the WoW population is concentrated in the upper levels. But having said that, it's by no means impossible. I was working on my alt this morning and within five minutes we had a PUG together to do Uldaman, we shared quests, and were done within an hour. I garnered almost a full bar of XP at level 40 from that instance. In an hour. Beat that Mr. Solo Man.

    This not to say that I don't think that Blizzard can't make things better. One of the things I think is really limiting is the 25 quest limit. If you solo and do instances, it very difficult to fit all those quests in within the 25 limit. And I think it would also be nice if there was some way that people could get XP for assisting with quests that are profession or class specific. (A warlock helping with a mage quest, for instance.)

    But all these things tweak the system. I really disagree that there is anything fundamentally wrong with the way that group and solo XP is designed in WoW right now. The complaint by Tobold seems like purely QQing.

  2. I won't speak for Tobold, but I'm certainly not arguing that instances are actually less exp/hour - you can fail to finish an instance and still earn tons of experience just from rested exp kills of trash mobs. It's more a question of whether the average amount of time you'll spend looking, traveling to the instance and then back to wherever you were before the run, etc is enough that people don't feel it's worth the bother. (Also, as at the level cap, if you've got half an hour before work, solo content is an option and finding an instance group is not.) Clearly, you win when you've got a full group in 5 minutes, but I've seen people spend over an hour LFM for old world dungeons - sometimes they're even short on DPS.

    In some ways, the low level group game today is suffering from Blizzard's decision to nerf most of the pre-60 non-instanced elite content. The ogres in Alterac, for example, provided a quest area where two DPS and an Arms or Fury War in Defensive Stance could actually get some group content in. Blizzard's decision to pull the content kind of proves my point on content scarcity - there was not enough content to go around, and Blizzard felt that the limited content they had available was better repurposed into solo content than left as seldom-used group content in the post-TBC world.

  3. I think part of the experience comparison is looking at what experience is given when grouped up for the quests OUTSIDE of instances. Many people would rather try to solo a group quest than have someone help because experience is split between players.

    So even looking at general questing, most of the time people do not see any incentive to be grouped with others... you have to split experience rather than get it all yourself (as well as money/loot). And then for gathering quests. Ugh, those are horrible when grouped with others. It takes longer than if you'd done it by yourself.

  4. The real issue it seems to me is whether a group-based model is viable.

    EQ2 when it released was very hard to solo in. I got to about 20 on most of my characters before I had to give up soloing and just hope for a group. The benefit was that groups were very good and people played and behaved pretty well. (Do you want to be the guy no one wants to group with in a group-centric game? Of course not.)

    It was pretty boring though hanging around on a dpser hoping for a group. I would sit reading a book while occasionally glancing up at the screen to see if anything had moved forwards.

    In the end most players switched to classes like Druids and Necromancers that could solo where most classes couldn't.

    Defty's point about the exp gain from instances while levelling up kinda proves Tobold wrong. WoW already offers double value for grouping as Defty explains and people still tend not to do it.

    One of the big problems is the nature of the social contact. People are scared to join a random pug because they assume that some players will either
    - be idiots
    - treat them like idiots
    - be rude
    - all of the above.

    EQ2 kinda solved that by making grouping so essential early on but it was unpopular and they adjusted the game to be more solo-friendly.

    Darkfall kinda solves it by making the environment so hostile you're scared to go around alone. I think that the social behaviour of hardcore pvp players though (impatient, rude, elitist) will limit the game's success.

    To sum up, I think the game that sells subs because of its grouping is going to be a game that cracks the problem of the way typical gamers behave towards each other which is often pretty hostile.

  5. I kinda regret my initial comment not because its wrong but because I think it misses a larger point. Namely, the reason that we want group content in a game at all. Tobold talks about "The Vision" and I think the vision is one driven by economics. The theory is that if people form groups they form bonds and those bonds lead to a greater loyalty to the game and and a stronger revenue stream to the company.

    For me, the remarkable thing about WoW is that it's vision is based upon flexibility. You can play the game however you want. If you wann a guld, guild; if you wanna PUG, PUG; if you wanna solo, solo; etc. The designers (within limits) don't force you to play the game their way. Play it how you want.

    IMHO group based theory has serious limitations. And one of those is that familiarity breeds contempt. Look at all the drama that goes on in guilds. Forcing people to play with each other is a doomed approach, either in MMORPG or in the playground. If the average guild can't last for than a year, then why would a game based upon guild last much longer. There's only so many suckers to go around. Incentives (such as a dual xp curve) may mitigate the problem to some extent but they will never eliminate it all together. A game based on an explicit dual xp curve will only serve to piss off the soloists and do nothing to help the game long term.

  6. Flexibility is clearly a strong model, it offers something for everyone thus WoW has a huge player base of people who use the software for completely different playing experiences.

    But it can't be the only model. Has EQ2 accepted second place in 2005 and retained a complex crafting system that killed you for mistakes, an intensely group oriented pve experience that was highly punitive to soloing and no pvp I think it would have retained a significant number of fans who valued it as an alternative to WoW. Instead it just tried to be WoW and did it worse.

    I really think Darkfall points the way. The games that will be interesting in the next year or two will be games that seek to channel their players by making certain things tough rather than the WOW model which does everything reasonably effectively but not exceptionally.

    You can't launch a game successfully if it tries to compete with WoW on its great strength: flexibility


Comments on posts older than 14 days are moderated and will not appear until manually approved because the overwhelming majority of such comments are spam. Anonymous commenting has unfortunately been disabled due to the sheer volume of comments that are defeating Google's spam filter.