Megadungeons as publishing projects

I'm still working on my "state of the OSR" post, which will come out in the next couple of days. But I wanted to comment on something that is becoming more and more important in the world of independent old school publishing: the megadungeon. The whole debate over Dwimmermount resulted in this thread at theRPGSite, which in turn really got me thinking. (My comment on that thread is as Cadriel, the name I use on most RPG boards.)

For me, the great appeal of megadungeons is not a long series of levels, even if they are relatively well jaquayed. It's the wondrous levels; Castle Greyhawk is not the center of so many imaginations because it's so many levels deep but because it's got the truly strange beneath it, like Bottle City, the Machine Level, the Black Reservoir, the Garden of the Plantmaster, the portals to weird places; it's not just a big dungeon, it's a complex where the payoff is quite big. And it's full of things like the Great Stone Face that just capture the imagination.

The attempts to publish megadungeons so far have not been overwhelming. Stonehell was a neat dungeon, and there are some really good bits in its five levels, but it doesn't really transcend the "standard" dungeon levels. Similarly Dwimmermount, which is only in draft, is far from awful for a vanilla fantasy dungeon, but at the same time nothing so far hits that "amazing" button that really needed to be pushed to reward players for adventuring so long in one ruined pile. (This review is useful for non-backers; I think it's too harsh but has some good points.) Anomalous Subsurface Environment goes too far for me in the other direction; so much of the science fantasy that it becomes passé. I do think that some balance is required.

When I came to the OSR, one of the things I thought a lot about was what we needed to be putting out as a community. The answer has been "dungeon modules." That makes a certain kind of sense, and I'll admit that I like buying and reading dungeon modules; I have an extensive collection of TSR, Judges Guild and OSR modules that I enjoy perusing. For most of them, though, I don't see myself running the things. One exception is Robert J. Kuntz's Original Bottle City, which I've played in. It's a shame that the book is hard to find, as it's IMO the best thing to come out of the recent wave of publishing, specifically because it's something that is meant to be worked into your own dungeon and not provide you with a "site and plot" as many dungeons do.

As I put it in the RPGSite thread, I think what we need is less people trying to publish their megadungeon. It's an arduous task to try and run somebody else's massive campaign dungeon for a prolonged period of time; the only one I could see gaining a lot of traction is Gary Gygax's Castle Greyhawk. Instead, what I would really like is a "megadungeon kit": some of the really cool sub-levels and concepts put out into a product that highlights just the "good stuff." This would be better for DMs slotting into their own campaigns, and the kind of thing that would work really well as one-shot or convention material as well.

There are 3 things that I think would be useful as part of this. One is really good dungeon entrances - opportunities to do caverns, abandoned mines, old ruins etc that serve as quality introductions, and can be used to lead to a larger dungeon. This is good for getting PCs up to 2nd or 3rd level before they hit the dungeon proper, so that they can have enough HPs that exploration is really viable. Second is the "special" levels - what's Dwimmermount's Bottle City or even its Machine Level? These are the kind of thing that deserve a room-by-room full detail, not every single chamber in a 500+ room dungeon. As impressive of a feat as that is, it's going to get kind of repetitive.

And third and last, is a "greatest hits" of the other levels. If level 2 could be done by sticking entries from Monster & Treasure Assortment into Dungeon Geomorphs maps except for 3 rooms, give us what's in those 3 rooms. Why add 50+ variations on empty rooms, giant rats and 2000 cp when you really want to highlight a few really great tricks, traps and monsters? Heck, I wouldn't even object to an Undermountain-style product where the whole map is provided but most of it is "fill it in yourself" except for the special areas. Which after all hearkens back to the suggested method of filling a dungeon in OD&D.

That's my take on publishing megadungeons as we are in January 2013. I most certainly welcome somebody proving me wrong by publishing a 500+ room megadungeon where I feel it wouldn't have worked if every single room description wasn't in the book, but I'd also love to see a project like the one I lay out above.