OD&D's Dungeon Design - The Primordial Stack

Gus L said:

Basically monster and treasure placement in OD&D is a mess, but at the least I can say the following about how I approach it.
A) It's "Level Based" Balance. I.E. the challenges you face on the first level of the dungeon is for first level PCs etc. This uses the singular dungeon approach of OD&D - quite interesting how this evolved with referees having their unique signature dungeons. It's less useful in published adventures for campaign use.
B) It may focus on random stocking ... but I don't think it really needs to or random stocking works. It wouldn't work well with the traps and tricks really - like you want a treasure reward for figuring out the stupid spinning room etc. Some early dungeons follow the random stocking idea seemingly - Palace of the Vampire Queen feels this way. Again I don't think it works well with the size of the levels though.
OD&D monster design and placement deserves its own post... someday.

Click to expand...

This is a great overview, but I feel like you're underselling a bit the paragraph titled Distribution of Monsters and Treasure on page 6 of The Underworld and Wilderness Adventures, and the dungeon-stocking procedure which follows. I read that as prescribing that the referee is supposed to intentionally place multiple big treasure stashes and monster lairs on each level, and then purely for convenience randomly finishing stocking the remainder of the rooms.

"The determination of just where monsters should be placed, and whether or not they will be guarding treasure, and how much of the latter if they are guarding something, can become burdensome when faced with several levels to do at one time. It is a good idea to thoughtfully place several of the most important treasures, with or without monsterous guardians, and then switch to a random determination for the balance of the level. Naturally, the more important treasures will consist of various magical items and large amounts of wealth in the form of gems and jewelry. Once these have been secreted in out-of-the-way locations, a random distribution using a six-sided die can be made as follows:"

(bold emphasis mine in all of these quotes)

This is more specific about treasures than, but consistent with, the language in Moldvay Basic on B52 that "Special monsters should be first placed in the appropriate rooms along with special treasures. The remaining rooms can be stocked as the DM wishes. If there is no preference as to how certain rooms are stocked, the following system may be used."

Holmes, on page 39, gives much the same instructions- "Each new room or area is given a code number and a record made on a separate page of what it contains, treasure, monsters, hidden items, etc. Place a few special items first, then randomly assign treasure and monsters to the other rooms using the selection provided in the game or appropriate tables."

Dan Collins broke down expected treasure yields if you just use the OD&D random tables some years ago on Delta's D&D Hotspot, and pointed out that just using the random table on page 7 of TU&WA means there will be trivial amounts of treasure, except for the ~1/20 rooms on level 1 which on average have gems or jewelry, or ~1/10 rooms on levels 2-3. 85% of random treasures found on level 1 can be expected to be worth less than 100gp. When you hit the 5% for gems or jewelry it'll usually be worth thousands, but it's entirely possible and reasonably likely to miss those entirely in stocking a level.

When we look at the published OD&D dungeon designs and Gary's Castle Greyhawk notes/key, we see that he definitely included more treasure than that. So what we actually see in his dungeons is very much inconsistent with random stocking.

For me, reading that passage and procedure on page 6 of TU&WA brought it all home. The tables are intentionally scant, because random treasures aren't meant to make up the bulk of the treasure, although they allow a lottery-like chance of a random big score, for excitement. The bulk of the treasure is expected to be "thoughtfully placed" by the referee with intention and forethought. "Several" big stashes of gems, jewelry, and magic items consciously chosen and selected with care, defended by appropriate guardians or traps to ensure that they need to be earned.

Mannahnin said:

The tables are intentionally scant, because random treasures aren't meant to make the bulk of the rewards, although they allow a lottery-like chance of a random big score, for excitement. The bulk of the treasure is expected to be "thoughtfully placed" by the referee with intention and forethought.

Click to expand...

I think this is a lost fact, or at least a very sound interpretation of OD&D - which of course demands interpretation. It's like the other stocking advice you point to ... great and easy to miss, meaning I suspect that a lot of people have missed it over the last fifty years. Of course there is evidence of the best tables using it, but there's also a lot of evidence of the opposite. Certainly very few people were running D&D entirely as randomly generated dungeons (though again, some seemingly were if one looks at A&E), but the treasure tables, monster tables and such offer the idea and offer the possibility, especially for the uninitiated.

There's also a fascinating aspect of how this all feeds into OSR design....

Anyway treasure, yeah I am really not a fan of the treasure tables. I get that they serve a purpose of showing how much treasure one should maybe offer ... but the coin weights don't really work well with the encumbrance rules (especially post OD&D), and well coin hoards are boring. I wrote another long post on that but of course never finished the follow up on alternate tables and treasure design. I really should, no one likes gripes without answers.

The strange thing with D&D's treasure placement to me is that as much as there may be some sources for decent amounts of treasure stocked (as you mention) a lot of flagship TSR stuff ended up ending tournament modules - which of course don't really need much treasure, being one shots. Later editions of course lean into combat XP and then I think a lot of OSR design followed these as TSR adventures as models without a lot of examination of baked in assumptions about treasure amounts. You end up with very paltry treasure in a lot of published stuff - look at Stonehell for a gruesome example. Then there's also that famous Gygax piece in Strat Review about pacing and leveling - which I think is pretty polemical. No one I know gets to (or wants to?) play 3 days a week for 10-12 hours a session... let alone do that for 5 years to get to level 15.

Personally I like to think of dungeons like "who many sessions will this take a group of 6 v. how often do I want them to level?". I was shocked when I ran treasure calcs on the 50 room dungeon I'm finishing up and if a party of 6 recovers all the treasure in it they will go from 1-6th level. Now they won't and it will take I suspect 10 - 20 sessions to really deal with this thing, so I figure that's about right.

Gus L said:

I think this is a lost fact, or at least a very sound interpretation of OD&D - which of course demands interpretation. It's like the other stocking advice you point to ... great and easy to miss, meaning I suspect that a lot of people have missed it over the last fifty years. Of course there is evidence of the best tables using it, but there's also a lot of evidence of the opposite. Certainly very few people were running D&D entirely as randomly generated dungeons (though again, some seemingly were if one looks at A&E), but the treasure tables, monster tables and such offer the idea and offer the possibility, especially for the uninitiated.

There's also a fascinating aspect of how this all feeds into OSR design....

Anyway treasure, yeah I am really not a fan of the treasure tables. I get that they serve a purpose of showing how much treasure one should maybe offer ... but the coin weights don't really work well with the encumbrance rules (especially post OD&D), and well coin hoards are boring. I wrote another long post on that but of course never finished the follow up on alternate tables and treasure design. I really should, no one likes gripes without answers.

Click to expand...

Absolutely. In practice the advice is easy to miss, especially because AD&D gives almost no guidance on treasure placement in dungeons! The DMG has sections titled Placement of Monetary Treasure and Placement of Magic Items (p91-93), but they're mostly general essays cautioning you not to overdo it and recommending that you start out with weak items and small treasures, and giving some advice on making monetarily-valuable treasures interesting, plausible, and challenging to identify and transport. But no guidelines at all on exactly how much treasure to place in a dungeon!

I'm not sure if they expected you to have read OD&D or Holmes Basic first, or expected you to use one of the Monster & Treasure Assortments, but the actual DMG gives no guidelines*, unless the reader decides to infer use of the treasure table (p171) from the random dungeon generation rules (appendix A). But that's still random, even if it's a bit more generous than the OD&D random stocking table.

But Holmes and even the justly-vaunted Moldvay still overlooked the importance of telling new DMs explicitly, as OD&D does "place multiple REAL treasures- magic items, gems & jewelry- per level before you roll random stocking for the remaining rooms"! Moldvay drops the ball by giving us an example of dungeon design and stocking (with The Haunted Keep (B55-56)) which relies entirely on random tables for the room contents.

When I learned to play it was initially with the Mentzer Basic set, quickly followed by Cook Expert and some AD&D books, and the impression I got was that the monster treasure types were the primary source of treasure, rolled per the treasure tables for lairs and cut down a bit if the monster numbers weren't high.

By the time I actually had a regular gaming group it was for 2nd ed AD&D, xp was primarily for monsters and quests/plot goals, and the difficulties of understanding GP for XP and proper treasure placement were left behind under the pretense that GP for XP had never made sense and we were happy to be focusing on "heroics" rather than treasure hunting.

Gus L said:

The strange thing with D&D's treasure placement to me is that as much as there may be some sources for decent amounts of treasure stocked (as you mention) a lot of flagship TSR stuff ended up ending tournament modules - which of course don't really need much treasure, being one shots. Later editions of course lean into combat XP and then I think a lot of OSR design followed these as TSR adventures as models without a lot of examination of baked in assumptions about treasure amounts. You end up with very paltry treasure in a lot of published stuff - look at Stonehell for a gruesome example. Then there's also that famous Gygax piece in Strat Review about pacing and leveling - which I think is pretty polemical. No one I know gets to (or wants to?) play 3 days a week for 10-12 hours a session... let alone do that for 5 years to get to level 15.

Click to expand...

Yup. I bought into those Tough DM Gary editorials when I was learning, but in practice giving paltry treasure means excessively slow advancement and frustrated players. You're dead right about the tournament modules, and of course Stonehell, for all its virtues, does show indications that Mike Curtis made the mistake of relying on the random stocking tables as his measuring stick.

Gus L said:

Personally I like to think of dungeons like "who many sessions will this take a group of 6 v. how often do I want them to level?". I was shocked when I ran treasure calcs on the 50 room dungeon I'm finishing up and if a party of 6 recovers all the treasure in it they will go from 1-6th level. Now they won't and it will take I suspect 10 - 20 sessions to really deal with this thing, so I figure that's about right.

Click to expand...

100%. And this is someplace Moldvay helped so many people, despite his rare fumble on the stocking example, with his advice on page B61 that "If no one has reached 2nd level of experience in three or four adventures, the DM should consider giving more treasure."

In running my last (three year) old-school GP for XP campaign, I found that I hit a solid average rate of 1 level of advancement per 3-4 sessions for most characters, at least through the first 3 levels or so, after which rate it slowed a good bit for the next 2-3 levels, and then WAY down after that, though some of that was probably a shift in adventure pacing and how we were playing. That pace was excellent. Moldvay was dead-on about the early levels, and having that guideline was really helpful. Although a huge key for me was that "thoughtfully place" language on p6 of TU&WA.

I stumbled on/learned from the blogs a similar rubric that you talk about here- I normally want to place enough treasure on a given level of the dungeon that, at least if it's all recovered, the entire party can expect to gain a level. And really, I generally want to pad that a bit, to account for missed treasure and deaths causing xp to be lost. Although I use Carousing for XP and give more generous monster XP (though not as general as in 1974), to help out as well.


*(EDIT: I think this is another example of the issue that the AD&D DMG was in significant part a cobbled together set of essays and expanded advice on DMing that was written with the assumption that the reader had already read OD&D. Which manifests and ways large and small. )

Gus L said:

Personally I like to think of dungeons like "who many sessions will this take a group of 6 v. how often do I want them to level?". I was shocked when I ran treasure calcs on the 50 room dungeon I'm finishing up and if a party of 6 recovers all the treasure in it they will go from 1-6th level. Now they won't and it will take I suspect 10 - 20 sessions to really deal with this thing, so I figure that's about right.

Click to expand...

Mannahnin said:

In running my last (three year) old-school GP for XP campaign, I found that I hit a solid average rate of 1 level of advancement per 3-4 sessions for most characters, at least through the first 3 levels or so, after which rate it slowed a good bit for the next 2-3 levels, and then WAY down after that, though some of that was probably a shift in adventure pacing and how we were playing. That pace was excellent. Moldvay was dead-on about the early levels, and having that guideline was really helpful. Although a huge key for me was that "thoughtfully place" language on p6 of TU&WA.

I stumbled on/learned from the blogs a similar rubric that you talk about here- I normally want to place enough treasure on a given level of the dungeon that, at least if it's all recovered, the entire party can expect to gain a level. And really, I generally want to pad that a bit, to account for missed treasure and deaths causing xp to be lost. Although I use Carousing for XP and give more generous monster XP (though not as general as in 1974), to help out as well.

Click to expand...

I was looking at Philotomy's Musings last night for another reason, and remembered that Jason Cone was already recommending doing this kind of calculation Back In The Day.

His rubric was something like, multiply the number of PCs by the amount needed for a fighter to level up, then roughly double it to account for treasure which goes unfound and XP lost for hirelings and due to deaths. EDIT: And level drain!

Just gotta say, I really enjoyed the article. Very thoughtful and though-provoking.
I think over the past 50 years the whole concept of "level design" has matured a lot, and some design principals that were almost unheard of in the 80's have now become common practice in video games.
For example the Tunic video game has a very nice sequence of gatekeeping and unlocked shortcuts that condenses a lot of variety in a relatively small physical space.

Mannahnin said:

I was looking at Philotomy's Musings last night for another reason, and remembered that Jason Cone was already recommending doing this kind of calculation Back In The Day.

Click to expand...

Philotomy's is well worth a read. I keep wanting to write a follow up on how I see his "Mythic Underworld", its sources and ultimate evolution. In some senses the design sensibilities of a lot of the mid-OSR come out of Philotomy, certainly the OSR megadungoen feels deeply connected to his "Musings".

What hits me about it though is that it's entirely a personal understanding of OD&D, it's not a manifesto or anything like that, despite how much it has been used for that - but one thing I enjoy about it is that it really encourages a close reading of OD&D (or any system really) with an eye for "what sort of game does this make for?"

Mannahnin said:

His rubric was something like, multiply the number of PCs by the amount needed for a fighter to level up, then roughly double it to account for treasure which goes unfound and XP lost for hirelings and due to deaths.

Click to expand...

Patrick Wetmore of ASE had something similar I think - personally it's more a guideline and goal, not a real rigorous thing, but I tend not to place treasure that's worth less then 50GP - except as a sort of greedy bad idea tax - like you can haul 4,000 CN worth of iron ingots out of the dungeon and make your 10 GP ... but maybe it's not the best plan. Adventurers aren't miners.