Monday, July 5, 2010

A worthy challenge

Corporate Plunderer (aka Snidely) set a challenge before me last night.  He is interested in developing a game for mobile devices that would satisfy a list of conditions:

Can be played concurrently but does not have to be
Variable number of players
Highly addictive

Obviously everybody is trying to be highly addictive, so that isn't anything new.  Variable number of players isn't too big a challenge for a game on a computer, that is more of an issue that physical games have to deal with.  The main problems as I see it are that allowing people to play a game without having specific time constraints is extremely difficult unless you want to make success and failure not really possible.

For example, there are many online games where you can build your farm, set up a city, or rule a small empire within a larger game world.  In most of these games failure is not an option - you can log in whenever and if you don't log in for awhile it is ok, you are exactly where you left off.  In those sorts of games if failure *is* an option then the game is often extremely random since you can be easily smashed by others who team up and strategy is relatively useless compared to luck and politics.

This sort of situation is much easier to deal with if you have time constraints - if a single turn goes by each calendar day for example, everyone can realistically log on and give orders for a few minutes each day.  It is also possible to arrange a game where you log on and are paired against an opponent immediately as is the case in many online games, but the market for directly confrontational games is often limited and you are faced with direct competition from real FPS and RTS games on consoles and real computers.  There is also the issue of having fairly serious intensity and dedication required for 1v1 matchups that a mobile platform is not especially well suited to - people often have only a little bit of time to play and may be interrupted.

I think it might be a good idea to consider a team PVE approach to this sort of problem and a great starting model would be WOW's random dungeon finder.  What you would need for this to work is some kind of persistent 'character' that can advance over time through winning games and a social platform where people can hit a button to group up with their friends or random others to play a game together.  It would also be possible to make the advantage from each game decay rapidly such that each game played after the first on a given day awarded half as many points as the game before so a casual player won't fall too far behind a hardcore player.

That is a fine bit of structure I think, the trick of course is designing a game that a few people can play together on small screens with limited controls and easily scalable difficulty.  That, of course, is no small task.


  1. An idea I have is instead of having "solo" advantages, give players advantages that benefit the group. That would give a single player an advantage that he personally keeps (his group does x% more damage?) but a less intense player would still share in the advantages temporarily.

  2. I think something to keep in mind is that it doesn't have to be particularly challenging. If you want a cooperative game to be addictive and popular then you can't make failure too threatening because people will just get frustrated with all the bad players they have to play with. I think the WoW dungeon finder model is a really good one.

    As a game for mobile devices, is this something you'd want people to be able to play with a touch screen?

  3. Also, I'd consider a deconstructionist approach, in addition to traditional MMO-style.

    A lot of the more enjoyable, addictive and successful games in history have been completely abstract in their presentation. Think Tetris, TapTap (, Trism, or my personal favourite: boomshine ( etc. Heck, chess, go(!), backgammon also fall into that category. Traditionally these are presented as puzzle games, but when you get right down to it it's all just playing with numbers anyways.

    The really nice parts about deconstructed games are that a) the art is easy to model (and plays to computer strengths), making really beautiful & responsive games fast to build and !iterate!; and b) the gameplay becomes everything, and I suspect gameplay really is the key to long-term enjoyment and viral strength. I'm convinced that for most people, pretty pictures are really no more than embedded instructions on how to play, and we all play games as equations after the first few minutes.

    Another thought is group games, like charades. is a trivial-to-implement game which has made a few million $, essentially just automating pictionary across continents.

  4. Sthenno: I don't know. Unfortunately I do not own a device that this would play on so the parameters are not at all clear at the moment. This is obviously a pretty big obstacle to actual development! If I approach it more seriously instead of just as an exercise in curiosity I would certainly have to figure out the limitations of my format first.

    Snidely: Boomshine was great fun, and I do remember being heavily addicted to tetris at one point. I even have some simple ideas for a 4 player PVE game with different roles knocking around in my brain.

  5. If you look at most popular Facebook games you'll see that challenging has exactly nothing to do with anything. By and large they're all about trying to have you come back as often as possible and getting as many of your friends to come back as often as possible. And then somehow getting all of those people to buy in game benefits for real money, or to click on ads.

    If something is hard in any way then you're driving away some part of your lowest common denominator market.

  6. Ziggyny: It is true that a very hard game will send some people packing but it is a great way to make a game last. Chess isn't easy, but it has lasted for centuries. WOW is a great example here since it has scaling difficulty - each person can pursue hard goals or easy goals depending on what they want to do. My idea works very much like that in that you would have an option to try really hard things or just do easy things and progression is *faster* doing hard stuff but anyone can be successful if they are capable of pushing buttons at all and have time.

  7. The "do nothing" social network games also face a very challenging market dynamic.

    (Pardon the epidemiology references, I don't know of any better models.)
    Since they're so focused on building up repeated visits and spreading virally at the expense of gameplay, they have both a very high velocity and a fairly short active period, with susceptibility to reinfection. To keep that sort of game going requires a massive leading edge (marketing), or pure chance. Just as that model suggests, for every one which succeeds, a dozen fizzle out, and most games (even the ones which have already hit like FarmVille) spend a massive amount on marketing, upwards of 40% of revenues @ Zynga.

    If you don't have the "knack" for understanding and predicting crowd behaviour (and if you do, I recommend going straight to the source and playing the market!), a much more reliable approach is to design a game which has a very long infectious period, which minimizes the need for a wide leading edge. Games like this include Starcraft, Tetris, as well as chess/go.

    Still other games have a very high infection rate (intrinsically multiplayer, with a high degree of team trust/dependence), as a method of growth. A lot of facebook games are like that, though it's very hard to do at the same time as a long infectious period. (if long investment is required, and competent friends are required, then competent friends with a similar degree of investment are required... ouch!). Warcraft is, of course, one such game. MMOs are really the first type of game with a long infectious period and high infection rate.

    Ah, rambling... I love playing with virii...

  8. Chess has lasted for centuries, but I'd argue it's not going to make anyone a fortune as a cell phone app. It also may have staying power, but I have a hard time believing a good chunk of your population is going to want to play it.

    WoW is a great game, don't get me wrong, but it only has a fraction of the users Farmville had/has. (235M according to an April article, 72M/month in June.) Not that trying to model a new game after WoW is a bad idea, but you have to realize you're cutting out a substantial chunk of your potential userbase by taking that route.

    And WoW is hard, in terms of time commitment. If I'm taking a 15 minute break at work, what can I do in WoW? Even 'quick' tasks like a random dungeon or a battleground takes longer than that in terms of time spend logging in, queueing up, and getting going. I have a hard stop time here, and if I have to AFK out of WSG before the game ends it makes me annoyed, not happy. If I'm a DPSer I'm not even going to get into a dungeon by the time I have to get back to work.

    Civ is probably a better example for an addictive game from our past that can be played in cell phone relevant chunks. Or Diablo. Or Street Fighter 2. MMOs and RTSs tend to require a continual investment of time. You can easily enjoy Starcraft playing only half an hour a day, but it pretty much has to be all in a row.

  9. I don't know that trusting the Farmville reported numbers is at all useful. They aren't reporting active paying subscriptions after all. Many free to play games report every single person who has ever signed up for an account, even if they just played for 5 minutes once.

    Thing is, I am certainly not going to try to model WOW's whole business. The ideas I really like are a persistent, improveable avatar and the dungeon finder. The avatar in WOW is a fantasy character while the avatar in the concept I have going currently is not so much a person as a set of improvements to game mechanics you can buy. I think those things are really worth grabbing, especially since my target would be a game that you can log into, find a group and complete a dungeon/level/challenge in 10 minutes and then log off again if you want.

  10. Possibly the Lineage model could apply here as well. You start in a solo game, then other players are introduced in increasing numbers as you progress.

    First ~5 minutes of gameplay solo, next ~15 in a group of 3, next ~30 in a group of 5, next 60 in a group of 8, etc.

    I like that it does double duty filtering out idiots, scaling difficulty/flexibility and providing an additional challenge vector.

  11. The 235M is probably from Zynga and rightfully doubted but the 72M in June is from a site that tracks app usage on Facebook. That is the number of distinct users who actually loaded the app in the previous month. (To be fair, that counts people like me who accidentally clicked on a Farmville button and logged in for about 10 seconds...)

    Zynga reports 3-5% of users actually pay money to buy things. You can doubt that as well, but even the low end of their estimate brings in over 2M paying customers in June. Actual money stats are hard to find, but it looks like Zynga brought in over $150M in 2009, which would be about $12M per month. WoW brings in about $175M per month, so they're making a lot more, but they also have substantially higher costs and a much more established presence.

    I'm not sure even a 10 minute chunk is entirely reasonable and I think expecting your users to be able to spend that much time straight is going to cost you potential users. That's not necessarily a problem, and if the people you do get in are more likely to stay as a result it may be a benefit. But you are intentionally limiting your user base with that choice.

    Consider something like a Scrabble app. You can play all day if your opponent can, but if one of you needs to go do something you can drop it dead and no one is hurt by doing so. I think that mechanic is important for a mobile app.

  12. The ability to leave on a moment's notice with no repercussions for anyone allows people to ditch any time and start games they have no intention of finishing. That appeals to some people, but if I was playing a team game that people dropped from constantly because there was no reason to complete the task and no incentive to play a full round then I wouldn't bother with it.

    I want to play with people who are willing to make some minimal commitment, so making a game with zero commitment is going to lose me. Regardless of the commitment level you ask for you can't please everyone, so you just need to pick something that reasonably fits the format and go with it I think.

  13. Also, the model here is a "buy the app" model, with some kind of restricted/limited free trial, not an ad-view driven model.

    Ad-driven games are much messier for a lot of reasons business-wise, and make it pretty hard (IMHO) to build a good game, since so much must be optimized for the business model.

  14. It allows people to ditch at any time, but it doesn't have to allow people to start games they have no intention of finishing. Continue the Scrabble scenario. I get called away from my phone, all my current games get paused. But when I go to play again, they can all get resumed. Allowing me to ditch at any time doesn't hurt the game world at all. My opponent at the time can go play Scrabble with someone else while I'm sleeping or working or underground. Our game can essentially be paused to no detriment to either side.

    To prevent me from just stopping to play and starting new games with other people because I'm bitter that you drew both blanks and all 4 S's you can put in a rule preventing me from starting or joining new games when I have X outstanding games waiting on me. Games waiting on other players don't hurt me here, just ones where I am the cause of the delay.

  15. Holy crap I just got screwed by the comment system!

    Summary of what I wrote:
    A game that plays like puzzle quest (it's a puzzle game like bejeweled or whatever where making matches powers up your attacks)

    You can get gear to add to your damage as well as level up to learn new abilities (which would add more tiles to your board and probably make the game harder)

    When other nearby people are playing you are cooperating against the same opponent

    Each day (maybe week) there is a set goal you are all trying to do and all successes playing the game by anyone contribute to that goal

    When you log in it tells you whether the last goal was completed or not and rewards you for your efforts towards it

    Enemies scale in difficulty to the number of opponents and to the power and skill of opponents (basically to your dps) so that there is always a chance of failure and having someone terrible join your team doesn't screw you

    Overarching objectives are put on a scoreboard for regions so everyone feels like they are playing for their home team. A success/failure scoreboard and a total spoils scoreboard are both maintained so that small regions can compete in the former, but big regions are recognized just for being big (we do want *more* people to play)

    This has: cooperative play with little reason for recrimination of those who aren't as good as you; an internalized social pressure to play because you are representing your region; a built-in reason to recruit friends to play; accessible gameplay; and, if it is to be successful, cute graphics.

  16. While I was re-writing that I was thinking about kinds of game play. You know that game where you get a bunch of different coloured blocks and you can remove any group of matching colour but you get big score by removing large groups at once?

    What if you had similar gameplay to that but when you removed blocks more blocked filled in so the screen was always full, and matching larger sections did more damage? Also, to make it obvious what's going on, you could get a glowing aura around you on the screen that gets bigger/brighter depending on how big your next attack *could* be. Then when you unleash the attack everyone would know it was you who laid that hurt on the monster. How quickly you made moves and attacked would be throttled so its not just a matter of bashing the screen quickly and touch screen devices don't have an advantage over devices with other inputs. Basically every time you cleared blocks your character would have to animate their attack before you could do it again.

    I'm not sure whether that level of stealing game play is something you can get away with. My observation is that puzzle games is a field where people flagrantly steal game play and furnish it with new graphics to make the difference. Anyway, thinking of a good gameplay mechanic could be important. Just having the different colours be different attacks or spells, however, already differentiates it (so there is a difference between matching yellow tiles vs. red tiles). Anyway, some kind of tile matching.

    Maybe instead of just choosing tiles to remove you can also switch tiles around, but doing so removes tiles if you put them in a group with your biggest current group (which would be glowing on the screen). Double tap to remove, tap one and an adjacent one to switch.

    Maybe even give each class a different ability to manipulate tiles that recharges over time? A warrior might have tile switching as I just described, a wizard can swap colours (so change all red to yellow and yellow to red to cast the spell you want). A rogue can randomize the colour of a group of tiles, etc.

  17. You're basically describing the game Puzzle Quest. It's a Nintendo DS/xBox 360 Arcade game which has spawned two sequels. (Galactrix and Puzzle Quest 2).

    It's certainly possible to have a similar but different game and it is a reasonably well received game so the market is there. You just want to make sure you're not too close to Puzzle Quest itself.

  18. I love the PuzzleQuest angle, the genre is certainly not exhaustively explored. Mixing puzzles w/ a team/MMO RPG seems to hit a lot of hot buttons.

    Of course, it puts a lot of stress on game design, but I like to think that's a good thing.

  19. I know I was describing Puzzle Quest (I mentioned it early on). I don't know how the puzzle quest game play works at all, if it is very similar to what I described then obviously the puzzle part would have be pretty different, but that was kind of a brainstorm/rough sketch thing. Is Puzzle Quest multiplayer at all?

  20. It's not co-operative but you can fight other players. I used to puzzle brawl with Randrew on the bus home. Puzzle Quest 2 has a ladder/tournament feature of some kind built into it for the xBox version, though I don't know much about it since if I get the game it will be on the DS. Games I can play on the TTC are worth more to me than games I can only play at home... If I'm home I can just play WoW or Starcraft!

  21. First, Puzzle Quest +1!

    Second, back in the day before the Internet got big, PBeM style games were pretty popular. VGA Planets was one of the better known ones.

    Third, for a while there has been a resurgence of remakes of old school games or games in the style of old school games. One example of the former would be (a "PBeM" remake of Laser Squad), another would be Aah Little Atlantis--a game written in 48 hours:
    Now I know Sky wouldn't get addicted to the former--I suspect few of his readers would, either, but I can certainly see several getting addicted to the latter.

    I could go on, but not at the moment.

  22. Arg! Rant to BlogSpot: Stripping the contents between <...> is not filtering for invalid html! Still, I should've checked before hitting 'Post Comment'

    To reiterate:
    Third, for a while there has been a resurgence of remakes of old school games or games in the style of old school games. One example of the former would be <> (a "PBeM" remake of Laser Squad), another would be Aah Little Atlantis--a game written in 48 hours: <>
    Now I know Sky wouldn't get addicted to the former--I suspect few of his readers would, either, but I can certainly see several getting addicted to the latter.