Tuesday 29 July 2014

Random Tables as Tools to Compress and Communicate Information.

Let's think about random generators. What makes a good one? On A Gaming Podcast About Nothing a few episodes back I made the rather banal observation that while the contents of a table are important, what you really want is interaction between columns. This is obvious to anyone with half a brain.

But let's demonstrate anyway. As said, what follows is probably obvious, but it is also interesting.


The first table gives a bog-standard result. A random encounter with one of 6 possible monsters. The second table is more interesting, because it gives a combination of results through the interaction between columns: 36 possibilities. It's richer. And the third table is richer still: 216 possibilities.

Now, at the most superficial level that provides more variety, which is probably a good thing, all else being equal. Variety is nice. It avoids repetition. But let's drill down a little.

  • The second table communicates information vastly more efficiently than the first, and the third vastly more efficiently than the second. There is simply more stuff concentrated in the table as columns increase. Writing out all 216 possibilities for the third table (orcs near a crevasse in a thick blizzard, orcs near a crevasse in a hail storm, orcs near a crevasse in high wing, orcs near a crevasse in a fog...) would take a long time but also a lot of space. A random table with a number of columns is like a mechanism for compressing data ready to be unpacked through the use of the dice - there is no better tool available to the writer of an RPG product for doing this. 
  • The second and third tables require much less thought on the part of the DM in order to make them interesting. "Orc" requires spur-of the moment creativity. "Orc, abandoned yak herder village" is easier to deal with. "Orc, abandoned yak herder village, in high wind" practically runs itself. The third table lifts the burden of having to come up with interesting things on the spur of the moment. 
  • Paradoxically, while requiring less thought on the part of the DM than the first table, the third also cannot help but make him more creative. "Snow nymph, abandoned yak herder village, third party involvement" can't help but make him come up with a creative solution - why is the snow nymph there and what's the third party? "Frost giant, frozen lake, fog". What's going on there? A frost giant engaging in impromptu ice fishing, invisible to the players because of the mist - with the added danger of possibly falling through the ice. So the table both requires less creative thought, and focuses it. 
Random tables, then, are little packages of concentrated usefulness. They hold compressed information and creative power and release it with great efficiency when called upon to do so by the rolling of dice. 

8 comments:

  1. Great post! Thanks for the inspiration for building tables... This will be very helpful!

    ReplyDelete
  2. Yes, I concur heartily with this post. I like the metaphor of random tables being kind of like dehydrated concentrate of creativity. Instead of having to long form write out a story or scenario, just store the cogent bits in a table and re-constitute with a few dice rolls.

    Creativity is tough in a wide open space, having some things to bounce off of gives you a lot more to work with.

    ReplyDelete
  3. I agree and wrote something similar last year. Databases and Game Design

    ReplyDelete
  4. I'm not quite sure if the second and third tables are not too specific to get repetitive. An encounter with an orc is good - an encounter with an orc somewhere with a hail storm is... strange... if it happens a second time. Or even if you have any two encounters after another within a hail storm. On the other hand, you could still improvise it away, what you have to do anyway with the simple table...
    Hmm. I use the d12+d8 mechanic of AD&D for my encounter tables... so I have to do a little work, maybe... ;)

    ReplyDelete
    Replies
    1. Yeah, the tables are not very inspiring. I just threw them together to illustrate the point. Much more interesting contents are available.

      Delete
    2. I like 'em, I could imagine a sub table for 3rd party involvement with same species (rival/mentor/dependent/captive etc) and other species (fighting, parlaying etc) .

      Delete
  5. Where is the interaction between the columns? The example you provide works just as well if the three tables are separated. It works even if the tables are several pages apart and the DM has to flip through her notes to find them all. In other words, the information contained in each column is independent of the other columns. I'm not sure that I see your point here...

    ReplyDelete
    Replies
    1. Er...the interaction between the columns comes from the fact you can roll 3d6 and just look across and come up with an interaction. The example doesn't "work just as well" if the three tables are separated, because then you have to look at three separate tables, especially if they are several pages apart. I think I can be forgiven for thinking that was obvious.

      Delete