“Where’s Carlos?”

“Who?”

“Your hireling.”

“Oh. Crap. Right… uhh…. He’s been here the whole time.”

“Really? Even when that dragon attacked us?”

“I guess so?”

Never let this happen to you again.

Subscribe Now!

Go to Table of Contents

It’s time to explore Hell itself on the war-torn plains of Avernus!

… except there’s a problem.

Descent Into Avernus frequently talks about how the PCs are going to be “exploring” Avernus. Unfortunately, it goes out of its way to stop them from doing that in almost every way possible.

For example, it’s impossible to make a map. And the reason it’s impossible to make a map is because navigation is meaningless. Descent Into Avernus revamps the metaphysics of the Lawful Evil plane of Avernus into a chaotic, ever-shifting place where direction and distance are unfathomable and any attempt to make a map drives the cartographer insane. If the PCs attempt to go somewhere, you just roll a completely random die to determine if they get there or not.

Why have they done this?

I honestly have no idea

Despite maps being both impossible and meaningless, the adventure nevertheless gives the players a poster map (pictured above).

It’s unlabeled and, again, the spatial relationships it depicts don’t actually exist, so it’s utterly useless for literally anything you might actually use a map for. The map also doesn’t include Elturel (where the PCs would begin their exploration of Avernus) or Fort Knucklebones (the first place they’re likely to be going), but the DM is instructed to put these locations anywhere on the map they want. (Although, once again, this is totally meaningless.)

The module actually says:

Your players should never feel like they understand Avernus geographically, providing an unsettled and disorienting feeling as they roam the hellscape.

And, to be clear, this is not how things used to work in Avernus. The designers of Descent Into Avernus did this deliberately. They repeatedly sell the experience – in both marketing and the book itself – as “exploring” Avernus, but then go out of their way to discard the existing lore and the established 5^th Edition rules for wilderness travel in order to create a custom experience designed to make actual exploration completely impossible.

I should mention that the one thing the map DOES do is magically talk to the PCs: Every time they go somewhere, the map tells them exactly what the location is before they have a chance to explore and find out for themselves.

Without actually seeing it in the book, it’s difficult to believe the lengths Descent Into Avernus goes to in order to make sure that the players absolutely cannot explore Avernus in any possible way.

THE AVERNIAN HEXCRAWL

For the Remix, we’re basically going to just toss all of that out:

• The geography of Avernus will be comprehensible.
• Navigation (and therefore exploration) will be possible.
• The map the players receive will be meaningful.

To achieve this, we’ll be redesigning Avernus to use a hexcrawl scenario structure for wilderness exploration.

Now, one thing I’ve heard from a number of people while working on the Remix is that they’re really looking forward to seeing how I redesign Avernus as a sandbox. So I think it’s important to make something clear:

I am NOT going to be redesigning Avernus as a sandbox.

An RPG sandbox exists when the players can either choose or define what the next scenario of the campaign will be. For example, the opening chapters of Icewind Dale: Rime of the Frostmaiden are a sandbox.

Descent Into Avernus, on the other hand, isn’t a sandbox. It isn’t designed for the players to choose or define the scenario. The scenario is saving Elturel.

By using a hexcrawl structure to create a massively non-linear exploration experience while simultaneously blowing open the potential solution space for saving Elturel (as described in Part 6B: The Avernian Quest), we’re creating a huge playground in which the players will have enormous freedom to chart their own course through the campaign.

But, ultimately, they are not choosing their destination. Their destination is saving Elturel.

And this is significant because our design of the Avernian hexcrawl is going to be driven by that knowledge; it is going to be designed and structured for the PCs to save Elturel.

Now, with that being said, there are almost certainly going to be a lot of sandbox-like elements in our Avernian hexcrawl, and the hexcrawl itself is a very robust and flexible structure. So if you (or your PCs) were to ditch the idea of saving Elturel, it would probably be just a hop, skip, and a jump to use the Avernian hexcrawl as the basis for a pretty cool sandbox campaign. But there are meaningful structural changes and accompanying expansions of material that you’d want (or need) to make to truly shift the focus from saving Elturel and realize a fully functional and effective sandbox.

Note: There’s a fairly popular supplement called Avernus as a Sandbox. It’s also not a sandbox. They are misusing the term to just mean “non-linear scenario.” It’s actually a Choose Your Own Adventure structure. This is a really effective way of repackaging the Choose Your Railroad structure of the published campaign, but it’s not a sandbox.

MAKING THE HEX MAP

The first thing we’ll need for an Avernian hexcrawl is a hex map.

In designing the hex map, we’re going to want to keep the player’s map of Avernus in mind: If the players use their map to navigate, the navigational choices they make need to product coherent results when translated into the hexcrawl system.

(For example, they might be at the Hellwasp Nests and see, on their map, that the Demon Zapper lies to the “southwest.” If they tell the GM that they’re heading southwest, then when the GM plugs that movement into the hex map, they should generally end up at the Demon Zapper.)

The good news is that the player’s map is (a) illustrative instead of strictly representational and (b) lacks a scale. In other words, the only specific navigational information encoded into the map is a directional relationship (and even this can be arguably distorted by the imprecision of the illustrations). This gives us a great deal of latitude in tweaking the map.

(No pun intended.)

To create the map, I’m simply going to drape a hex grid over the top of the labeled DM’s version of the map:

In doing this, I’ve played around with the precise positioning of the grid to (as much as possible):

• Keep each location in a unique hex; and
• Maintain as many of the clear directional relationships as possible.

I’m fairly satisfied with the results here, so the next step is to start creating our hex key by assigning all the locations on the map to the corresponding hex. For example:

• A4 is the Arches of Ulloch
• B1 is Arkhan’s Tower
• C1 is the Monument to Tiamat.
• G6 is the Wrecked Flying Fortress.

And so forth.

MAP DIMENSIONS

I’ve used a 6 x 10 grid here, but because the player’s map has no scale, we have a pretty free rein in what the size of the hex map can be. (We could, for example, have just as easily made this a 10 x 20 grid or even distorted the map to achieve something like a 30 x 30 grid.)

In deciding the initial size of your hex map in a hexcrawl, there are basically two factors. First, you need to stock all these hexes with content, so you don’t want so many hexes that the task is overwhelming. Second, the hex map needs to be large enough that the PCs won’t immediately run off the edge of it and into uncharted territory.

(It’s quite likely they’ll eventually end up deciding to see what’s beyond the edge of the map, but if it happens in every single session including the first it can be problematic.)

In the case of Descent Into Avernus, we also have to consider the amount of time and narrative space we want to dedicate to exploring Avernus. If we had a 30 x 30 grid, for example, there would be 900 (!) hexes. That would probably mean spending twenty or thirty times longer in Avernus than the rest of the campaign combined (which would make it very difficult to maintain focus) OR it would mean the players only experiencing a tiny, tiny fraction of the hexcrawl (which would mean a ton of wasted prep).

Based on my experience, I generally recommend a 10 x 10 or 12 x 12 hex grid for an initial hexcrawl campaign (assuming the PCs’ homebase is located in the center of the map).

Our 6 x 10 map, therefore, is a little undersized in one dimension. I think this will be fine in practice because the existence of the map will act as a soft constraint on the PCs’ navigation: Having a map in hand focuses a group’s attention on the area depicted by the map, and they will tend to make their decisions within the confines of the map. There are also mountain ranges depicted to both the north and south, which will act as natural navigational constraints as well.

ADDING LOCATIONS

There are roughly two dozen locations described in Descent Into Avernus. So even after placing those locations, we’ll still have three dozen empty hexes that we’ll need to fill with new content. We’ll take a closer look at that in Part 7C: Avernian Hex Key.

However, among the existing locations, as mentioned above, there are two that need to be added to map: The Dock of Fallen Cities (where Elturel is currently located) and Fort Knucklebones.

These two locations need to be placed in relation to each other, as Lulu can hypothetically see Fort Knucklebones from the Dock of Fallen Cities. Furthermore, we know that the Dock of Fallen Cities is located on the Styx.

We could put both of these locations in a single hex, but it probably works just as well to put them into two separate hexes.

The map actually has a number of unlabeled locations, and to simplify my life it would also be great if I could grab one of them and use it for either the Dock or the Fort or both. I’d also prefer a relatively central location.

Scanning the map, this hill looks promising:

You could pretty easily make the case for that depicting Fort Knucklebones. That’s in Hex I3, so we’ll go ahead and slot the Dock of Fallen Cities in Hex H3.

Go to Part 7B: Avernian Hex Map

Feng Shui is the action movie RPG of fast and furious combat inspired by the classic Hong Kong films of Jackie Chan, John Woo, Tsui Hark, Michelle Yeoh, and Jet Li. It uses a really cool initiative system featuring a shot counter:

• Your initiative check determines your initial shot count in the sequence.
• Each action is rated by the number of “shots” it will take to resolve.
• This shot cost is subtracted from your initial shot count to determine the next shot count on which you’ll be able to act.
• When everyone’s shot count hits 0, the current sequence ends and a new sequence begins with fresh initiative checks.

I describe this mostly to provide context for the cool new toy we’re talking about today: mission critical objectives.

Most fight scenes end when everybody on the opposing team is incapacitated. You stand triumphant with bruised knuckles over their unconscious, dead, or groaning forms.

But sometimes, that’s not the point.

Sometimes you’re fighting the bad guys because they’re trying to launch their hijacked nuclear missile or open the gates to the Mongol invaders. And sometimes they’re trying to stop YOU from uploading the virus to their computer network or destroying the Crystal of All-Knowledge.

One way to stop an opposing force, of course, is the aforementioned carpet of corpses. But often these goals can be logically achieved in the middle of the fight (“Hold off those Nazis while I disable the V-2’s fuel line!”), and ideally we’d like to capture the frenetic thrill of the struggle, creating fight scenes that are more varied than just punching hit point pinatas.

The problem is that these scenarios often become anti-climactic when combined with the turn-based nature of most RPG combat systems: On the hacker’s turn they make a Computer Systems check and… that’s it. The virus is uploaded. It’s over. There are ways to work around this, but they can be extremely situational and difficult to pull off.

Fortunately, the Feng Shui initiative system makes it easy to implement a solution.

MISSION CRITICAL OBJECTIVES

Attempting mission critical objective requires a 5-shot action, often accompanied by a skill check (i.e., a Sabotage check to cut the fuel lines). If successful, then the character can achieve the mission critical objective on their next action.

While a character is attempting a mission critical objective, however, other combatants can attempt to interrupt them as a 3-shot action.

If a hero is attempting to stop a GMC from achieving a mission critical objective, they can attempt either an attack stunt or an appropriate skill check (with the foe’s skill AV as the difficult). If they succeed, the GMC fails to achieve the mission critical objective (although they can try again later). If they fail, the GMC achieves the mission critical objective.

If a GMC is attempting to top a hero from achieving a mission critical objective, the hero must make an appropriate skill check (with the foe’s skill AV or Speed as the difficulty). On a success, the hero achieves the mission critical objective. On a failure, they’ve failed (although they can try again later).

ADDITIONAL OPTIONS

Multi-Step Objectives: It can often be satisfying to have multiple mission critical objectives that have to be achieved. (For example, you might need to shut down all three etheric turbines to prevent the Martian death laser from firing.)

Tug-of-War: As a variant of a multi-step objective, each mission critical objective might exist as a kind of “toggle” that can be swapped back and forth. For example, you might need to cut the three pirate harpoon lines to free your ship and sail away. While you can attempt mission critical objectives to cut a lines, the pirates can attempt one to resecure a line you’ve cut.

Mission Critical Boost: Characters can spend 1 Fortune to grant a +3 bonus or +1d6 (their choice) to a character attempting a mission critical objective.

Non-Critical Mission Objectives: For objectives that are involved but not the ultimate goal of the confrontation, you might choose to attempt them as 3-shot actions (instead of 5-shot actions).

Just One Chance: For some mission critical objectives, you might only get one chance. If the mission critical objective is attempted and fails, it cannot be attempted again.

ENDING THE FIGHT

Achieving the mission critical objective means that victory has been attained… but there may still be a bunch of bad guys. So what happens next?

One option, of course, is to just finish the fight – i.e., keep fighting until everyone on one side has been KOed. But this can be rather unsatisfying: Victory (or defeat), after all, has already been achieved.

Mission done? Time to leave! Feng Shui conveniently includes rules for doing that! (See the Cheesing It mechanics on p. 111 of the rulebook.)

Probably the key thing to keep in mind here is the central lesson of the Art of Pacing: The question we used to frame the fight scene (“Can we achieve the mission critical objective / prevent the bad guys from achieving the mission critical objective?”) has been answered. The scene is, therefore, functionally done. And when a scene is done, it’s okay to just wrap it up, cut away, and start the next scene.

CODA: IN OTHER SYSTEMS

Although well-suited to Feng Shui, this same basic concept can be hacked into other roleplaying games. In D&D, for example, mission critical objectives can be actions that take 1 round to complete, finishing on the character’s next turn and allowing other combatants to potentially interrupt them.

This is a useful cheat sheet I created for understanding what characters can see in the wilderness. In practice, sight lines will vary quite a bit (due to hills, forest canopies, atmospheric haze, and other obstructions), but I’ve found it’s useful to have some reference points and a few rules of thumb.

HORIZON: The horizon is 3 miles away at sea level.

NEIGHBORING HEXES: Passing through the center of a 12-mile hex, neighboring hexes cannot be seen. If the path is biased, the nearest hexes can usually be discerned (depending on the terrain).

Design Note: This is one of the reasons I prefer a 12-mile hex. Unless the PCs take special action to see farther, you’ll generally be able to focus exclusively on the hex they’re currently traveling through.

MOUNTAINS: Mountains can be seen from 6 hexes (72 miles) away.

Design Note: My research indicates that most mountain ranges have an average height of 3,000 feet. If you do the math, you can see an object 3,000 feet high from about 68 miles away, which I then rounded up to 6 hexes. Or, if you reverse the math, I’m saying that at 72 miles you can see the occasional peak that’s up to 3,500 feet high in that range or thereabouts. A very tall peak of 10,000 feet could theoretically be seen from 10 hexes away on a clear day.

ELEVATION: Distance to the horizon in miles is the square root of (feet above sea level x 1.5 feet). Add the height of tall objects to the viewer’s. Atmospheric haze will eliminate the ability to see even the largest objects more than 3-5 hexes away.

SIGHTING

Characters may seek out a good location for seeing long distances by taking the Sighting watch action. Height is obviously a factor here, but finding the right sight lines can be equally important. (Climbing a tree is all well and good, but if it’s at the bottom of a valley the effect will be mitigated.) This is also why simply being “on a mountain” doesn’t automatically translate to great sighting: you’re usually surrounded by other mountains.

As a general rule of thumb, assume that characters can find a location granting them 50 feet of height (or the equivalent thereof). This will allow them to see into neighboring hexes, and possibly even see notable locations within those hexes or their current hex (which may or may not require a Wisdom (Perception) check).

If you call for a skill check to find a sighting location, consider using a fail forward technique: On a success, the PCs get the normal benefits of sighting. On a failure, they might only be able to make out the terrain type of two or three of the nearby hexes (and no details thereof).

If circumstances suggest that the PCs would have great sight lines without needing to take special effort, that’s great. It might still be appropriate to allow them to take the Sighting action to improve their line of sight even more, extending their vision by another hex.

ENCOUNTER DISTANCE

When an encounter is generated, the distance at which the encounter may be detected will depend on the terrain in which it is occurring. (If the encounter is with a group of creatures and both sides are surprised – i.e., they do not detect each other – it’s technically possible they will pass each other without ever realizing it.)

The figures here represent typical circumstances on the ground. If the PCs are keeping watch from the top of a stone tower, for example, it’s quite possible for them to spot potential threats at much greater distances.

Back to 5E Hexcrawls

Go to Part 1

Characters can take watch actions to contribute to the expedition’s success or achieve other tasks while traveling. Some watch actions are limited to specific types of watch or travel pace.

Playtest Tip: It’s usually a good idea to get an expedition’s “standing orders” instead of asking everyone to declare their watch action during every single watch. An easy example is that if the ranger has been doing the navigating for the last eight days, he’s probably going to continue navigating for the next four hours and you don’t need to confirm that.

FORAGER

Characters can forage during an active watch or while traveling at a slow pace. Foragers make a Wisdom (Survival) check against the Forage DC of the terrain. On a success, the forager either gains 1 ration of food or finds a source of fresh water (allowing the expedition to drink their daily ration of water and for waterskins to be refilled). An additional ration of food or source of fresh water can be found for every 2 points by which the check result exceeds the DC.

Advanced Rule – Sparse Biome: At the GM’s discretion, a biome may be deemed sparse. In a sparse biome, each source of fresh water discovered only yields one gallon of water.

Advanced Rule – Grazing: Some animals (like horses) can simply graze for sustenance. In an appropriate biome (as determined by the GM), they will be fully fed as long as they are allowed to graze for one watch per day. In a sparse biome, they must graze for two watches per day and it may be necessary to also provide fresh water for them.

FOOD & WATER

Food: Small or Medium creatures require 1 ration of food per day. They can go without food for a number of days equal to 3 + their Constitution modifier (minimum 1) before suffering 1 exhaustion level per day thereafter. A normal day of eating resets the count of days without food to zero.

A creature on half rations counts as going a ½ day without food (and these half days accumulate until they can eat full rations).

Water: Small or Medium creatures require 1 gallon of water per day, or twice that in hot weather. A creature on a half ration of water must succeed on a DC 15 Constitution check at the end of each day or suffer a level of exhaustion. If they drink less water than that, they suffer a level of exhaustion automatically. If the character already has one or more levels of exhaustion, the character takes two levels instead of one level in either case.

Waterskins hold a half-ration of water.

Recovery: Any exhaustion suffered from lack of food or water cannot be removed until after a full day of normal consumption.

Large & Tiny Creatures: Tiny creatures require one-quarter ration of food and water per day. Large creatures (like horses) require four rations  of food and water per day.

NAVIGATOR

The expedition’s navigator is responsible for making navigation checks. A second navigator can assist, granting advantage to the navigation checks.

PACK-PULLER

A pack-puller is responsible for managing an expeditions pack animals. A pack-puller can lead a number of animals equal to their passive Wisdom (Animal Handling) score. (This number includes the pack-puller’s mount, if any.)

RESTING

A character must take the Resting watch action for two rest watches in a row in order to gain the benefits of a Long Rest. (See the rules for Long Rests regarding which types of interruptions are possible without disrupting the Resting action.)

Advanced Rule – Lack of Sleep: If a character does not spend at least one full watch per day resting, they must succeed at a Constitution saving throw (DC 16 – the number of hours they slept, if any) or suffer a level of exhaustion.

SCOUT

A scout can journey out from an expedition in an effort to chart a course or learn the lay of the land. When scouting, they can choose one of two actions:

• Reporting: If the navigator receives a scout’s report, they gain advantage on their navigation checks for the next travel watch.
• Pathfinding: The scout attempts a Wisdom (Survival) score using the area’s Navigation DC. On a success, the expedition can treat trackless terrain as if it had a trail for one watch.

A character can scout during a watch in which an expedition is traveling only if their speed is faster than the expedition’s. Alternatively, scouts can be sent out during watches in which the rest of the expedition are active or resting (but not traveling).

An additional encounter check is made for each scouting group. (They are effectively a separate expedition while engaged in scouting.)

SENTINEL

A member of an expedition acting as a sentinel can make Wisdom (Perception) checks to detect threats or notice anything else out of the ordinary.

Design Note: If you use passive Wisdom (Perception) scores, their use on journeys is also limited to sentinels. I do not.

Advanced Rule – Outrider: During a travel watch, a character can journey out in an effort to protect the expedition from detection or threats. Whenever an encounter is generated for the expedition, each outrider group can attempt a Wisdom (Perception) check (DC 15 or opposed by the encounter’s Stealth check) to detect the encounter before it’s encountered by the expedition.

However, an additional encounter check is also made for each outrider group at one-half the normal chance of an encounter. (It’s possible for outriders to discover locations or encounter creatures which would have otherwise been missed by the main expedition.)

SIGHTING

During an active watch, a character can take a watch action to find an outlook – a good sighting place which will allow the character to significantly extend the distance to which they can see the surrounding wilderness. (See Hexcrawl Tools: Spot Distances.)

At the GM’s discretion, a skill check may be required to identify or reach the outlook. (For example, a Strength (Athletics) check to reach the top of an appropriately positioned tree.)

TRACKER

Finding Tracks: Searching a significant wilderness area for tracks is an active watch action. The tracker makes a Wisdom (Perception) check against the appropriate Track DC.

Following Tracks: Once tracks have been found, a tracker can follow the trail during a travel watch by making a Wisdom (Survival) check against the appropriate Track DC. A new check must be made each time the trail enter a new hex.

If a trail is lost, it may be possible to reacquire it using the Finding Tracks action.

Design Note: Generally speaking, it’s appropriate to use let it ride techniques when tracking a quarry. In this case, however, navigation (and possibly getting lost) is a significant component of the hexcrawl structure and these additional checks are meaningful.

However, you could easily decide to go the other direction: A single successful Wisdom (Survival) check could follow the trail all the way to its end. Alternatively, you might only call for new checks when the terrain (the tracks enter the desert) or conditions (it starts raining) change, creating points of uncertainty.

TRACK DCs

Advanced Rule – Cover Your Tracks: As a watch action, a character can attempt to cover the expedition’s tracks. This is a stealth action, requiring the expedition to be moving at a slow pace. The character makes a Wisdom (Stealth) or Wisdom (Survival) check to set the Track DC for any pursuer attempting to find or follow their tracks. The check suffers disadvantage in very soft ground, but gains advantage on hard ground. Condition modifiers apply normally to the pursuer’s tracking check.

TRAILBLAZER

When an expedition is trailblazing (see Advanced Rule: Trailblazing), one member of the expedition must use their watch action to blaze the trail. If hidden signs are being employed, an additional character can assist the trailblazer on their Wisdom (Stealth) check.

Go to Part 4: Navigation