Gloamhold: In Megadungeon Design, How Much is Too Much?

Literally the bedrock of the hobby, megadungeons have a special place in my heart, and in the hearts of gamers everywhere. But can you have too much of a good thing? Can you have too much megadungeon?




I’ve been bravely buying and reading as many commercially available megadungeons as I can get my hands on in preparation for getting stuck into the nitty gritty of designing Gloamhold. What’s struck me as interesting is the vast differences in styles. For example, Undermountain provides a tremendous amount of material and detail. It’s staggering how much information Ed Greenwood manages to cram into that original boxed set. In contrast, Barrowmaze takes a much lighter approach to describing individual areas. Both are excellent delves, but they are clearly the result of differing design philosophies.

In considering the design of Gloamhold’s various levels and sub-levels two niggling question keep popping into my mind:

  • How much detail is too much?
  • How much detail is not enough?

With the twin tyrannies of page count and print costs looming over the project (as well as the small matter of GM usability) it doesn’t seem possible to provide as much detail for every encounter area as I would normally. That is unless I want to print a book you could use to beat a zombie elephant to death. While there is a a certain attraction to writing such a book —you never know when a zombie apocalypse might strike — I’m not sure I want to catalogue every nook and cranny of Gloamhold. It doesn’t seem practical, and more importantly it feels wrong— like I’m missing the point of a megadungeon.

Beyond the flavour and the cool adventures, one of the reasons I love Greyhawk so much as a campaign setting is that Gary gave just enough detail to get started. He provided the GM with ample design space to make the world his own by stamping his individuality and personality on the setting and the adventures. It was easy to make Greyhawk your own and no two campaign were identical.

I think a similar approach to Gloamhold is the way to go. Providing GMs with a robust framework and thematic tools to bring the dungeon to life seems much more appropriate than providing reams of notes on endless encounter areas. Slavishly detailing the whole place is great for the first time the party explore a certain area, but provides much less replay value. Off the top of my head, I can think of several advantages to this strategy:

  • It enables the GM to stock the dungeon as he sees fit without throwing out or extensively modifying existing material.
  • It enables the GM to restock the dungeon in response to campaign events. After all, once cleared a room, sub-level or whatever isn’t going to stay empty forever. This kind of ebb and flow among Gloamhold’s denizens will be a vital component of breathing life into the dungeon.
  • It enables the GM to easily maintain the illusion of detail I wrote about several months ago.
  • It enables the GM to stamp his own personality on the complex.

Providing tools to enable a GM to restock the dungeon quickly and thematically (or to reveal new secrets and mysteries to the PCs) are worth more in the long run than presenting voluminous notes on every chamber, passageway and cave in Gloamhold that are useful only once.

Of course, I’m planning to write adventures set in Gloamhold — it would be madness not to — and also I really want to detail certain places in greater depth. (I’m particularly keen to shed more light on the Twilight City, which I think is going to be a terrific adventure locale). To achieve this designing — along with Raging Swan Press’s heroic band of freelancers — exciting, flavoursome adventures the GM can modify and weave into his own campaign is the way to go!

And let’s face it a campaign set in Gloamhold should hopefully go on for a very long time. 

What Do You Think?

What do you think? Do you want everything lavishly detailed or would you prefer getting the tools to build your own Gloamhold? Let me know if the comments below and help me make a better Gloamhold!


Published by


Creighton is the publisher at Raging Swan Press and the designer of the award winning adventure Madness at Gardmore Abbey. He has designed many critically acclaimed modules such as Retribution and Shadowed Keep on the Borderlands and worked with Wizards of the Coast, Paizo, Expeditious Retreat Press, Rite Publishing and Kobold Press.

15 thoughts on “Gloamhold: In Megadungeon Design, How Much is Too Much?”

  1. The biggest issue I see with lots of detail is that it locks the megadungeon (or any adventure, really) into a specific location and world. For example, B2 The Keep on the Borderlands is generic enough that I can place it in my world pretty much anywhere. Frog God Games Slumbering Tsar, not so much. There’s world history to interweave, there’s powerful beings to integrate, there’s all that detail that I need to change (or change my world). I don’t necessarily mind this, as the set designer in me gets clues on how to integrate epic themes into my world, but placing such a location gets more difficult with more detail.

    Certainly as I’ve become more experienced, I’m happier to have tools to create my own thing instead of being satisfied with what I’m given.

    1. That’s an excellent point, that I hadn’t yet considered for Gloamhold. Adding it into an existing campaign world will probably be a key feature. I’ll have to think about that carefully as I progress. At the moment, I think the only major stuff is in the very deep history of the place – who built it, when and why. I’ve also made the gods, like in the Lonely Coast, pretty generic so they can be easily swapped out if desired.

  2. We’ve discussed my node-based megadungeon before. I stopped at a fairly high level of design, but I’ve discussed with several people how it might be completed.

    My first thought was that I were to prepare it for publication I might go full detail. Take the current graph and drill down into each as needed, ultimately ending up with maps and room descriptions and monster states.

    I had someone tell me he’d probably outline some random tables (wandering monsters, dressing, and weird stuff) appropriate to each region and area and run it from there.

    When I actually ran it myself, I went in with even less than that. I had the entity description for each region (call it a page or two each describing the nature of each area, its major elements, and how it connected to others) and just improvised. Not suitable for publication except as a starter kit, I daresay, but it was enough for me to work with…

    1. At the moment, I’m thinking level-specific tables for helping the GM generate the contents of various areas along with fully fleshed out portions of the dungeon – adventures the GM can drop in as he pleases.

      Less than that makes me a little uncomfortable.

      1. As it happens, I concluded that I really should have done more preparation. I had the shape of the dungeon as a whole, but region-specific elements (even random tables) would have been quite helpful. The stuff I came up with on the fly was okay, but degraded quickly when I was tired… and since we were playing mid-week after work, that happened very quickly.

        1. Part of the thing that attracts me toward megadungeon design and play is that it is pretty easy to prepare for or run on the fly. As a busy gamer anything that makes my prep time easier is to be welcomed. Tons of tables definitely seem to be the way to go for that.

  3. I think it’s the right approach. It also will help it the description of the dungeon is not page up and page down with monster stats and such. Keep that in the back. Makes it fill less pages and helps those of us that don’t think Pathfinder is the optimal choice for ORS dungeon crawl 🙂

      1. On that note.. could you perhaps do a blog post sometime on the topic of choosing which game system to design for?

        This is something I need to consider when preparing (parts of) my own campaign for publication. My gaming group is playing a homeruled version of AD&D 2nd edition, so most of my material is written for this, but there are probably very few players still playing this version.

        I have the feeling most people are now playing Pathfinder, D&D 5th edition, or one of the many systems that imitate 1st or 2nd edition D&D. I have no idea which of these is currently most popular. Even when designing things to be system neutral you still need to balance monster stats and spell selections somehow. In short, your sage advice on this topic would be most welcome.

        1. I’ll start my ponderings!

          A lot will depend on the OGL or GSL WoTC announce for 5e. One of the reasons I didn’t publish any 4e material was the insanely restrictive text of the agreement. You’ll also need to consider how 5e is doing in (say) a year’s time. I think the system is in a honeymoon period at the moment – no one’s really had a chance to get under the hood and see what’s what.

          I think Pathfinder is here to stay and is a good choice. Going down the retro clone route I think could also be a winner. That said, I’d love to know how many people are actually actively playing those games first. On the plus side, at least they are all relatively similar in terms of adventure design. If it’s compatible with one, it’s bound to work with the others!

  4. It seems to me I remember you fitting some Raging Swan material into Greyhawk (Ratik and Bone March perhaps?). Did you unofficially have an area of Greyhawk in mind when you fleshed out the area of Gloamhold?

    1. Good memory, Jason! Yes–essentially, I had an idea to place it either in an isolated part of Northern Ratik just beyond the Timberway or in some other out of the way place. One of my ideas was to simply swap out Dullstrand for Ashlar.

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.