In Shadowdark, the dark dangers of the dungeon are infamously made tangible at the gaming table by linking the duration of light sources to real time: Have you been playing for an hour? Then your torches, lanterns, and light spells burn out and you’ll need to ready new light sources.

This is a mechanic that streamlines bookkeeping, centers expedition-based play, and encourages fast-paced action. (If you sit around dithering, then you’re literally burning the candle at both ends!) It can also be a controversial mechanic due to its lightly dissociated nature, so it’s ultimately up to you whether the visceral immersion of the real-world time pressure is worth the tradeoff.

DARKER DEPTHS

But we can take this concept even further by embracing the ideal of the Mythic Underworld. Within those strange depths, the darkness does not flee the light, but rather turns upon it. Crawling through those tunnels you can feel it pressing in — little eddies of shadow testing the flickering weakness of your torch until finally the stygian murk literally snuffs out any source of light.

And the deeper you dare? The stronger the darkness becomes.

If you’re on the first level of the dungeon, then torches, lanterns, light spells, and other sources of illumination have a duration of 1 hour. But as you descend to lower levels of the dungeon, light source duration decreases as shown on the table below:

If the PCs switch dungeon levels in the middle of a duration, increase or decrease the end time of the light source appropriately. For example, if a torch was going to burn out at 8:50 pm and the PCs descend from Level 1 to Level 2, the torch should now burn out at 8:40 pm. (Going up a level should always relieve pressure; going down should always increase it, and could even cause a torch to immediately go out!)

Sublevels can be treated as a level of equivalent depth (e.g., Level 3A would have the same light duration as Level 3).

DESIGN NOTES

The rules for darker depths are, obviously, designed to increase the stakes and costs of mounting expeditions to lower (and, in classic megadungeon play, more profitable) levels of the dungeon. In more narrative-driven dungeons (e.g., we must follow the dragon into the depths to recover our comrade’s body!) where the logistics of torch management may not be a primary focus, it will nevertheless push a sense of rising dread and danger.

Big, ten-minute chunks are taken out of the time initially so that the players can immediately feel the difference from one dungeon level to the next. Once the timer reaches 30 minutes, this pace is reduced to five minutes per level in order to sustain the effect for larger dungeons.

Similarly, the progression is ultimately capped at 10 minutes per light source because shorter durations (e.g., 5 minutes or 1 minute) simply become too much of a hassle to implement, serving as more of an annoying distraction than a terrifying reality.

You could experiment with the idea that, beyond Level 8, the stygian depths actually prevent any light source from being ignited, but this would obviously represent a fundamental shift in the paradigm of play rather than simply putting pressure on the existing forms of play.

Did you know that the D&D 5^th Edition core rulebooks don’t include an example of a keyed dungeon map?

This is because D&D no longer teaches you how to prep or run a dungeon.

This is a wild thing to think about, and it actually gets weirder the more you think about it. Back in 2020, in Whither the Dungeon?, I pointed out that this had created an entire generation of Dungeon Masters who had, paradoxically, never learned how to master a dungeon. Even the oral traditions which had once passed this knowledge from one DM to another were breaking down, partly because this was actually the end point of a long-term trendline (the dungeon instruction in D&D 4^th Edition, 3^rd Edition, and even 2^nd Edition had grown increasingly anemic) and partly because of the huge influx of new gamers via channels other than playing in someone else’s game over the past decade.

By the end of the 2010s, you could already see the effects of this manifesting in DMs Guild and other third-party adventures: An ever larger number of dungeon adventures were being published without numbered maps; the contents of those dungeons described in haphazard paragraphs which were often little more than a rambling stream of consciousness.

This was bad in its own right, but this degradation of published books was only a reflection of an even deeper rot at casual gaming tables – a malaise invisible to those afflicted, because they didn’t even know what they were missing.

Then, in 2023, Wizards of the Coast published The Shattered Obelisk. As noted in my review, this campaign book included multiple unkeyed dungeons. This, in my opinion, was a red alert: It wasn’t just that Wizards was neglecting to teach basic dungeon design; it appeared that the design team itself was losing the institutional knowledge to design dungeon adventures. (The call was coming from inside the house!)

But maybe this was just a fluke, right?

In 2024, the new Dungeon Master’s Guide was released. Like the 2014 Dungeon Master’s Giude, it failed to include even an example of a keyed dungeon map, let alone any sort of actual instruction in keying a dungeon map or running dungeon adventures. Except it was even wore than that: The new DMG included multiple sample adventures for new DMs, including three dungeon adventures… none of which were keyed: One just said “there are some monsters in there.” Another tried to vaguely describe through words where each encounter was located on the map (e.g., “in a sidecave to the southeast” or “at the north end of the stream”). A third tried to use a broken method of random encounter checks.

So where the 2014 DMG simply neglected to teach new DMs how dungeons work, the 2024 DMG escalated to only showing examples of exactly how you should not design a dungeon. This, in my opinion, was now a five-alarm fire.

All of which brings us, in 2025, to Borderlands Quest: Goblin Trouble, one of the first official adventures for 2024 D&D, released as introductory adventure to promote the upcoming Starter Set: Heroes of the Borderlands.

It’s a dungeon adventure.

It’s unkeyed.

Something is rotten in the state of D&D.

WHERE, OH WHERE IS THIS ROOM?

At this point, many reading this might be thinking, “Well, so what? What’s the big deal?”

To understand the problem, let’s take a peek at what happens when you try to actually run Goblin Trouble, starting with the PCs following some bandits and “enter[ing] the cave from the eastern edge of the map.” Checking the map, it’s pretty easy to figure out where this is:

It will later turn out that the actual location being described is several hundred feet to the west, but we can let that slide. So far, so good!

But things quickly get more complicated:

From the cave entrance, a passage continues deeper beneath the hills and slopes downward. You travel for several minutes before the passage turns north and leads up a set of natural stone steps. A group of caverns continues out ahead of you.

The ceiling of these caverns is choked with webs, and the footprints you’ve been following continue through these caves. In the center of the floor in the first cave is a human-sized boot.

I’m fairly certain that this is what the boxed text is describing:

But, as you can see, the PCs have moved past a major intersection without the adventure even mentioning that it exists. This is quite strange, but let’s put a pin in that. We’ll come back to it later.

At this point, the adventure describes an encounter with some spider webs and dead spiders… but is this actually where this takes place? It’s unclear. When the PCs went up the stairs, north, and into/through a “group of caverns,” how far did they go exactly? How many caverns are we talking here?

At first it seems like it could be any or all of these caves, but later, after the encounter with the dead spiders is described, we’re told that

The passage continues onwards, taking the characters north for a while, and then switching back south. The caves here are empty, save for some old cobwebs on the ceiling.

Eventually, the characters find themselves on the southeastern side of the map, near some steps that lead into a cavern with an underground stream flowing in it.

So the dead spiders must have been in the southern cave (so that the PCs could leave it going north). Furthermore, there’s only one cave on the southeastern side of the map, so that must mean that the PCs follow this path:

As we can see, the adventure is once again skipping past several more intersections and the PCs end up… in the cave immediately to the left of where they entered? That’s bizarre.

It turns out, though, that this must be a typo. This is supposed to be “near some stone steps that lead into a cavern with an underground stream flowing in it.” The stream is on the other side of the map, so they must mean one of these two locations:

But which one?

I can guess, but there’s no way to actually know.

At this point we’re told that the bandits the PCs are following are “in the cavern ahead” and “trying to summon the courage to peek around the corner and see what the goblins are up to.” But what does this mean, exactly?

They must be somewhere in this area:

But where’s the corner they’re “peeking around”? Reading ahead, we can figure out that the goblins are hiding inside the secret lair, but the bandits don’t seem to know that. Nevertheless, maybe we can play this as “they look around the corner and are confused to discover a blank wall of stone.”

Okay, so “southeast” is probably a typo for “southwest.” And the “corner” must be that outcropping of rock, so the bandits must be standing at the X:

We also know the path the PCs must have taken (which is still bonkers, but let’s stick another pin in that!). We’ve figured it out!

Phew!

…

Just kidding. The bandits can’t be standing there, because several pages later we discover that the goblins have set up a trap in the mushroom patch. This is not, of course, keyed to the map. (Why would it be?) But if the bandits were there, they would have triggered it.

So where ARE the bandits supposed to be?

I have no idea! Good luck, first-time DMs!

CONSEQUENCES

It’s important to understand that this isn’t some fluke. This is what always happens when you try to describe the contents of a dungeon through rambling paragraphs instead of clearly indicating where things are located on the map.

If you think about it, this principle isn’t just limited to dungeon design. It’s a fundamental property of maps. Imagine, for example, if the maps app on your phone just displayed a bunch of unlabeled streets and said, “Turn left somewhere up ahead. I’m sure you’ll figure it out!”

And the problem here isn’t just that you’re being forced to solve a Where Is This Room? puzzle before you can run the adventure. The deeper problem is that, without a proper scenario structure, you’ll end up prepping the wrong stuff. And even if you prep the right stuff, it’ll be organized in ways that make it difficult or even impossible to use at the gaming table.

In truth, what usually happens is that the designer or GM will end up defaulting back to the only structure they have left: A railroad.

And that’s exactly what happens in Goblin Trouble.

The opening advice for the new DM (which is quite good, actually; check out Nerd Immersion’s video that takes a closer look at this) spends a good deal of time talking about the importance of player choice and how players should be driving the action of the adventure. But the very first thing the adventure says is that (a) there’s only one thing the players can do; (b) if they don’t, the DM should tell them what to do; and (c) if that also doesn’t work, the DM should then use their “power to change the world around them to spur them to action.”

Then, as we’ve seen, the rest of the adventure is forced onto a purely linear track: The players can only choose to go forward or backward. Despite a xandered dungeon map, the predetermined narrative simply shuffles them along a track that ignores all opportunities for choice or navigation.

THE KICKER

Goblin Trouble continues in this vein for a while, but then, right near the end, we find this:

The Forge. The makeshift forge in area 4a occupies the spot at the back of the second room.

Area 4a?

This means that, at some point, somebody actually keyed a map. And then somebody else decided to remove the map key and rewrite Goblin Trouble as a nearly incoherent linear railroad.

There’s some nice stuff in Goblin Trouble: The opening GM advice. The texture of the encounters. The roleplaying vs. combat opportunities set up with the bandits.

But it’s fatally sabotaged by a fundamental failure of basic scenario structure and adventure design. I find it particularly alarming that this appears to have been done to an adventure which initially didn’t have these problems, suggesting that there’s something systemically wrong with the development process for adventures at Wizards of the Coast.

It’s a “the call is coming from inside the house” moment and, in my opinion, bodes ill for future Wizards of the Coast adventures.

And, frankly, for the hobby.

Gygax must be chortling in his non-existent grave: Diablo II has come to your gaming table, and it’s brought random dungeons with it!

Review Originally Published May 21st, 2001

When I first cracked open Diablo II: To Hell and Back my first reaction was one of excitement: It looked like Carl, Eckelberry, Quick, and Redman were going to deliver an epic campaign for D&D.

As I read, though, my excitement quickly turned to disappointment, and disappointment turned to boredom. As I look back over what I’ve read, I’m still of a mixed mind over this book.

Diablo II: To Hell and Back is designed to bring the Diablo II computer game to your gaming table, adapting it for use with the D&D rules. To a certain extent, it succeeds. And to a certain extent, it fails. It succeeds in the sense that every location, every dungeon, every NPC, every quest, and every monster from the computer game is to be found between its covers. It fails, however, in the sense that it is so busy attempting to emulate the computer game experience that it never gets around to establishing itself as a D&D supplement.

Or, to put it another way: Playing Diablo II: To Hell and Back is just like playing the computer game… except without the graphics. (And you also have to do all the number-crunching and random generation yourself.)

But why would you want to do that? I mean, what’s the point? If I want to play something just like the Diablo II computer game, why wouldn’t I just play the Diablo II computer game?

At the end of the day, Diablo II: To Hell and Back fails in being a good D&D supplement – which is what it should have been, and needs to be, above all else. In fact, the only thing it truly succeeds at is emphasizing the vast gulf which exists between tabletop RPGs and Diablo-style CRPGs.

What’s frustrating here is that, with just a little more effort, the authors could have used the same basic structure of this product as the foundation for a truly epic D&D supplement. Something which builds upon the basic plot, characters, and villains of Diablo II — but also brings with it the unique strengths of table-top gaming (more realistic character interaction, the ability for the DM to handle more complicated plot dynamics, etc.). The opportunity, however, is wasted… just as your money will be if you make the mistake of dropping it on this book.

To be fair, not everything here is lackluster. As I mentioned above, the structure the authors employ to emulate the computer game is not without merit: They succeed in creating randomly generated adventures which, at the same time, have a structure and purpose. They don’t entirely succeed (I would have actually preferred to seen less left in the hands of the DM – after all, I can always ignore randomness if I want to), but they do explore some interesting ideas. The monsters to be found in this volume are also nice, particularly the demonic creatures with lower CRs. If my understanding is correct, however, these can also be found in Diablo II: Diablerie — so you might be better advised to look there, instead of here. (Without the monsters, I would most likely have given this book a Substance rating of 1 instead of 2.)

While the random structure holds some promise, in one area they drop the ball big time: “Rather than list a specific number of monsters for such Fixed Encounters, we list an Encounter Level. For instance, Corpsefire (a special Zombie boss) has CR 3 Zombies with him. You can use the Dungeon Master’s Guide (Table 4-1) to calculate just how many Zombies that means for the party level when the PCs encounter Corpsefire. That way the game adjusts the challenges the PCs face as they go up in level.”

When I first read that I was incredibly excited – not only as a game player, but also as a freelance writer. What a clever way of building scalability right into the adventure without having any significant amount of hassle for the DM!

And, indeed, it would have been clever… that is, if it actually worked. It doesn’t: First, it’s a misuse of the CR/EL system (because groups of creatures have EL; individual creatures have CRs – they were using terms interchangeably that aren’t interchangeable). Second, because the CR/EL system doesn’t work that way no matter how you look at it. Zombies which have a CR or EL of 3 will always have a CR or EL of 3 – no matter what level the adventuring party facing them is. The XP awarded for defeating a CR 3 creature scales as the party’s level changes… not the CR or EL of those creatures.

When I figured out that it didn’t work (about twenty seconds after reading it), I proceeded to get pissed off: Not only had someone designed a system that didn’t work, they had then – obviously – not even bothered to playtest it.

In short, let me say this again: Don’t waste your money on this book. It had potential, but it was horrendously squandered.

Style: 4
Substance: 2

Authors: Jason Carl, David Eckelberry, Jeff Quick, and Rich Redman
Company: Wizards of the Coast
Line: Dungeons & Dragons
Price: $29.95
ISBN: 0-7869-1831-4
Production Code: WTC11831
Pages: 192

I really don’t envy the design team for this one. Wizards of the Coast had licensed Diablo II and started releasing licensed D&D tie-in books in 1999. The idea of adapting what was, at the time, arguably the most popular fantasy CRPG on the planet to D&D was actually a pretty great way to try to expand a D&D fanbase that had been contracting for more than a decade. The only problem? Wizards was simultaneously getting ready to release a new edition of D&D in the summer of 2000.

So the Diablo II D&D sourcebooks were awkwardly split across editions, and the 3rd Edition books — like this one — were being developed before the core rulebooks had actually been finalized (which is always a difficult position for a designer to be in). Add in all the headaches that come from licensed games (e.g., getting approval from the licensor for everything you do), and, as a designer, you’re being put into an almost impossible situation.

Unfortunately, the result here was an almost perfect example of exactly what NOT to do when adapting a CRPG to the tabletop.

Glass Cannon Unplugged was supposed to be releasing a new Diablo TTRPG in 2024 that was going to be somehow compatible with a Diablo board game. But this currently seems to be vaporware.

For an explanation of where these reviews came from and why you can no longer find them at RPGNet, click here.

Frontier Justice is an incredibly well-researched supplement, providing a valuable investment not only for Blue Planet GMs, but GMs of any game system.

Review Originally Published May 21st, 2001

I’m of a mixed mind when it comes to Frontier Justice. It’s meant to be the sourcebook for “Crime and Law Enforcement” for Blue Planet — and it does that. In fact, it does that with a copious amount of detail and information. Multiple criminal and law enforcement organizations are given a history, culture, scope of affairs, and individual character; the process of committing a wide variety of crimes (and then solving them) is gone over with a figurative pair of tweezers; and, finally, an entire adventure is presented.

So, without a doubt, Frontier Justice accomplishes what it sets out to do.

But after finishing Frontier Justice I felt a reservation, and it took me awhile to nail down exactly what it was. For a long time I thought that the book simply lacked – for want of a better word — excitement. But, upon reflection, I think my problem actually stems from the fact that Frontier Justice is so busy presenting the copious research and creative effort undertaken by its creative team, that it forgets to do so form the viewpoint of a game supplement.

Certainly there is more than enough information here for any GM who wants to get his players intimately involved with either side of the law, but the GM is – to a certain extent – left adrift in this sea of information without a compass or rudder.

Then again, I may just be paranoid.

On the other hand, the wealth of research which has gone into Frontier Justice pays huge dividends – making this a valuable supplement not only for Blue Planet GMs, but GMs of any game system (regardless of genre or time period).

Finally, I should make a note of the adventure – which provides the interesting opportunity of playing both or either side of the story (either the criminals committing the crime or the law enforcement agents trying to solve it).

Style: 4
Substance: 4

Writers: Greg Benage, Catten Ely, Jason Werner
Publisher: Fantasy Flight Games
Line: Blue Planet
Price: $23.95
ISBN: 1-887911-44-8
Production Code: BP05
Page Count: 128

As I discussed in my review of John Tynes’ Power Kill, this was a time frame in which I was beginning to wrap my head around the concept of what I would later call “game structures” (and what I would not refer to more specifically as campaign structures and scenario structures). The thing I was trying to put my finger on in this review of Frontier Justice is the same thing: A ton of valuable information, but lacking the structure that would bring all of that material meaningfully to the game table. (If you want to see the difference, check out how Kenneth Hite’s Nights Black Agents does more than just inundate you with espionage facts — it packs the entire spy thriller and vampire horror genres into structures that make it shockingly easy to fully realize them at the gaming table.)

But, as I said all those years ago, don’t let that overly detract you from Frontier Justice. There’s a TON of really cool reference material here for running crime-based campaigns. (Which is something I’ve been thinking about quite a bit recently.)

For an explanation of where these reviews came from and why you can no longer find them at RPGNet, click here.

It’s time for the thrilling finale!

The villain gives a little monologue, the initiative dice roll, and now it’s time for an epic—

Oh. Never mind. The PCs already killed him.

It turns out it’s quite difficult to keep a single target alive when five highly trained killing machines (i.e., the PCs) are highly motivated to simultaneously release all of their death-dealing abilities on them. (Particularly if your stat blocks are simulating reality and/or you want that same stat block to function as Not a Boss™ in other situations.) So this is a systemic problem that you’ll find in a lot of different RPGs.

There are various solutions to this – mechanical, structural, and otherwise – but here’s one that works surprisingly well:

Have the boss show up AFTER the fight starts.

In other words, the PCs get into a fight with a bunch of the bad guy’s minions, and then two or three rounds later the bad guy shows up:

• The door is slammed open dramatically!
• The summoning ritual completes and the demon materializes!
• They teleport in with reinforcements!
• A helicopter swoops down from the sky and they jump into the middle of the melee!
• They were invisible the whole time and suddenly reveal themselves!
• A car with blackout windows drives through the wall of the warehouse and the vampire lord leaps out!

However the bad guy makes their dramatic appearance, this has three effects.

First, it’s a cool and memorable moment. This really shouldn’t be undervalued.

Second, the PCs will already be engaged with other bad guys. Their tactical positions may be far muddier than they were at the beginning of the fight. They are likely to have already blasted some of their most powerful combat options. In other words, it will be much more difficult for the PCs to focus their fire on a target that appears in the middle of the fight than one that’s available when the fight begins.

Third, even if they do wipe out the boss nigh-instantaneously, you’ll have pulled off an important bit of legerdemain: Yes, the boss died in two rounds. But that didn’t happen until the fourth or fifth round of the fight. So it will, no matter how illogically, feel like a big, satisfying fight instead of a curb-stomping.

Another variant here is to have the bad guys retreat to wherever the boss is, drawing the PCs after them in pursuit. This inverts the dynamic while expanding the encounter’s theater of operations and giving it a more epic scope.

It should also be noted that this whole dynamic can often organically arise if you’re using adversary rosters (either because the PCs are pulled into a running fight that takes them to the boss or because the boss is drawn to them).

You shouldn’t do this every time, of course, or it will become predictable and trite. (Although with enough variation in the boss’ dramatic entrance you can cover your tracks quite a bit.) But it’s definitely something to keep in your toolbox.