Thursday 25 July 2019

Size and the Inert Sandbox



When I was a lad, I spent quite a bit of time being befuddled and disappointed by the Commodore Amiga game Frontier: Elite II. For those either too young or too cool to have been into computer games in the early 1990s, this was a space exploration and trading simulation which was fairly unusual for its era in being entirely open-ended and, well, sandboxy.

The promise of Frontier has rarely been matched (although it's my understanding that nowadays there are games available which have a similar ambition). Although it came on a single floppy disk its scale was truly vast - it produced procedurally generated galaxies thousands of light years across, with realistic distances between its stars, and complete planetary systems around each of them which could be freely visited and explored. It aimed to accurately model Newtonian physics in the movements of objects in space, and everything in its universe - from stars to planets to space ships - was entirely to scale. It was a truly monumental achievement for its era.

I loved thinking about Frontier; I loved talking about Frontier with my friends; and I loved the thought of getting home from school each day so I could play it. Unfortunately, though, the actual "lived" experience of playing the game was dull. Much of the gameplay was simply glorified arbitrage on a grand scale: buy grain in Planet X at such-and-such a price; sell it in Planet Y for such-and-such-a price and then buy slaves; go on to Planet Y and sell them for a huge markup; buy narcotics; return to Planet X and buy a shitload more grain. Repeat ad nauseum. This was punctuated by fiddly, repetitive space combat which was always either too easy or too difficult, and the occasional attempt to find more profitable trading routes elsewhere, or other things to do such as carrying passengers from one place to another, out of sheer boredom.

The problem was scale. For all its astronomical and physical accuracy and for all of its vast scope, there just wasn't very much going on in Frontier. There was a galaxy of 100,000,000,000 stars to interact with, but there was nothing much other than that. You couldn't get out of your space craft. You couldn't talk to people. You couldn't make friends or enemies. You couldn't get involved in intrigue or espionage. And you couldn't really do much with your money except for buying more commodities to trade, a bigger space ship to carry them round in, and better weapons to vanquish the irritating pirates and other enemies who might come your way. If it had had only 100 stars to visit, but real stuff going on - wars, factional rivalries, interesting NPCs with real motivations - it would have been infinitely more interesting and exciting to actually play.

There is a lesson here for DMs. When it comes to a campaign setting, less is more (within reason). A small hexmap - both in actual area and in number and size of hexes - with plenty going on in it is a far better basis for long-term enjoyable play than an entire painstakingly mapped-out continent with too much space to fill. A 10 x 10 hex map of 1-mile hexes you can stuff to the gills with lairs, dungeons, settlements and interesting NPCs who all know, or know of, each other and have networks of rivalries and alliances as a result. A 100 x 100 hex map of 6-mile hexes takes a lifetime to even key, let alone run, and will feel largely empty and featureless to run. The bigger the sandbox, the more inert it becomes. And the smaller the sandbox, the more enjoyably dense (to the point, of course, at which it becomes so small it is quickly exhausted). When it comes to campaign settings, big may be beautiful, but small is fun.

(You don't have to take my word for it as far as Frontier is concerned: you can download a freeware version of it here. However, I strongly recommend you try the smaller and denser and hence more enjoyable Oolite instead.)


20 comments:

  1. That’s a very good analysis and a good analogy.

    I would add that the key is not the size of the hexes or how you key them, but how you fill in the manner in which you are able to fill things in procedurally.

    Ex. Day two of a four day walk over the mountains pursuing slavers up their smuggling route.

    Weather a little rainy but not cold. No serious weather event occurs.

    Players decide to go, reluctantly, risking colds.

    Encounter indicated. between noon and 6. Wandering monster: dragon. Number, 2. Type: gold, 12 HD.

    DM thinks. Finds two old prospectors sitting in homemade rocking chairs at the mouth of a cave with a nice fire.

    Reaction roll: 6. Neutral.

    They welcome company. RP goes fine. Ask party to stay for dinner.

    Prospectors have golden eyes. The Dwarf smells gold nearby and lots of it.

    Party doesn’t detect they’re gold dragons. Has dinner, explains quest.

    Reaction roll: 10. Good.

    One dragon sneaks out to them while they sleep and wakes up party leader, reveals herself, and offers to fly them the rest of The way.

    This was all off of a few die rolls and a little bit of thinking. It turned out to be a well remembered encounter and resulted in a new ally.

    ReplyDelete
  2. I had Elite on the C64 and was mystified. Although I like to think I was a bright kid, my attention waned between disk swaps

    ReplyDelete
  3. Interesting points. I think it depends on how you run the sandbox. If it's a hex by hex hex crawl, then yes, maybe 10x10 is plenty. A larger area could be made just as interesting by making it easy to find the locations of interest, with between 30 and 100 locations of interest. A point crawl would be one way of doing this, but also just having a map you share which allows hex by hex travel so routes can be chosen to the players liking, but is otherwise like a point crawl.

    ReplyDelete
    Replies
    1. That is a good point - and even more true for e.g. sea-based or spaced-based campaigns.

      Delete
  4. Ah, Frontier: Elite II. I can remember being so excited about that game and bouncing off it so hard.

    I've long been dubious about RPG books that approach a subject from too high and remote a level. Shortly after I started playing D&D, I spent my hard-earned spending money on The World of Greyhawk boxed set. I already had a few modules, but how exciting it was going to be to have a world of adventure to share with my players.

    The map was really cool, and it was my first encounter with weather charts, and there was indeed a whole world in the box, but it was all descriptions of entire countries that were no more than a few paragraphs that didn't even have much in the way of hooks to be fleshed out. It was my first major disappointment in a gaming product.

    During the '90s, it felt like most RPG books followed the pattern of being more interested the high and distant overview of their setting rather than giving you ground-level things for the PC to interact with.

    My favorite development of the last fifteen years has been the way that things have swung back to focusing on the encounter level.

    ReplyDelete
    Replies
    1. Good comments - and I completely agree about 1990s RPG books, especially TSR ones.

      Delete
  5. I completely agree with you that the size of the hex matters, because things become implausible when the hexes are too big. It's much more dynamic then it makes sense that a goblin survivor goes to fetch his orgre allies 2 hexes by lying to them about how rich the PCs are if the hexes are 1 mile wide vs 12... by the time the ogre return in the latter case, the PCs will be long gone.

    I think it's one of the reason I like urban campaigns - everyone is close to one another.

    ReplyDelete
  6. Nothing changes: the current version of the game, Elite Dangerous is criticised for being wonderful to explore but with little for players to do.

    ReplyDelete
    Replies
    1. Ha! Oolite is a lot of fun, as I remember.

      Delete
  7. I think what it really comes down to is Quality over Quantity. Less is More only in the sense that the creators of game worlds have limited time to work on it, and the more time you spend creating more stuff, the less time is left to refine existing stuff.

    For RPGs, I think one very important thing is that both durations and distances are all imaginary. You always transition from one scene to the next, no matter how many years pass and parsecs you've been flying. In the end, all that really matters is that there is content that is interesting for players to interact with, and that the players can find it.

    ReplyDelete
    Replies
    1. True, although believability gets stretched too far where the distances between things don't seem realistic.

      Delete
  8. I remember the original Elite, which sounds much the same: an exciting concept, lots of boring trading stuff and then - all too often - blisteringly quick dogfights with the hexagonal craft of some insect species that tended to kill you in short order. The high points of the game occurred when you survived an attack of those particular baddies and managed to limp on to your original destination, where your profits were entirely devoured by the cost of repairs.

    (Actually, there's something in there that mirrors the platonic RPG combat encounter: the party mostly survives - but only just - and is extremely relieved about it.)

    Agree entirely on your main point. I'm preparing a one-shot piratical/island adventure for The Fantasy Trip at the moment, and I'm conscious that even 10 lightly detailed locations is going to be ample for six hours' entertainment. And if the players manage to visit four or five of them, their knowledge of the others (from their starting 'treasure map') is going to make the whole experience seem richer and deeper despite the remaining mysteries - especially as some of the random encounters may riff on those mysteries.

    One point on the map size/scale. It's just as possible to run an effective campaign with (say) 30 encounters over a continent as it is with 30 over an island. In the former case, things should be framed more episodically, but that can have its own magic: "Three weeks after crossing the Great River ...".

    ReplyDelete
    Replies
    1. This comment has been removed by the author.

      Delete
    2. That is true, and I do think it depends on the type of campaign, of course. (SF and fantasy can differ hugely, for example - although they don't have to.)

      Delete
  9. I think is relates to your most recent post (about Vance). The larger world serves as a backdrop to lend flavor to the much smaller locale. The smaller-scale is more manangeable for game-play.

    I like to think is terms "how many days travel" are things apart. One or two days is good---it allows for interconnections. 1-2 weeks is cumbersome. Too close and your get the B2-syndrome: "Why are these orcs living next-door to the gnolls?"

    However, some things need to be remote. Its intrinsic to their existence.

    ReplyDelete
    Replies
    1. Also, I've noticed that there is a multiplicative factor that exists when locations are revised multiple times. Often the party is very cautious and meek when they first encounter a new vista---you might even think they are not paying attention. However, the next time they return, suddenly they are hatching plans about things you assumed they had missed. It's very rewarding.

      Delete
    2. Oh yes, I love that feeling as a referee when I realise my players were actually picking up on things; they were just playing it very safe up until that point.

      Delete