Down With The 6 Mile Hex! A Modest Proposal

Creators often live in the shadow of their own previous work, the big thing they’re known for. Danial Radcliffe isn’t called the guy from Guns Akimbo or Swiss Army Man– he’s Harry Potter to the average person on the street. Steamtunnel at the Hydra’s Grotto can be counted among these shaded creators by their well known 2009 blog post “In Praise of the Six Mile Hex”. In it they laud the 6 mile hex scale as the most convenient for computing distances at the table.

This post has become one of several classic and seminal OSR posts with the blog getting linked in blogs and social media threads everytime someone is discussing how to build and scale a hex crawl. Open any RPG hex crawl product that’s come out since and it seems like there’s at least a 75% chance the module uses a 6 mile hex. (This blog post isn’t the only reason for the scale’s popularity as a 6 miles is the suggested small hex size for detailed wilderness areas in B/X D&D, though older D&D adventure modules use a large variety of hex sizes- still it’s definitely helped popularize the scale.)

There’s just one problem- its conclusions are complete bunk for many styles of hex crawl and the author agrees.

Their second post that revisited the topic of hex scale nearly a decade later has been far less widely shared and spread. In The Ergonomic 3 Mile Hex, Steamtunnel noted that the three stated benefits of their first post are tied to a certain view which views hexes as rulers to measure specific distances instead of as discrete units themselves. Different styles of hexcrawl run hexes as discrete boxes while some use hexes as rulers of specific distance which allows for fractional hex travel. My preferences find the specific measurement method as simultaneously more work and less immersive. To quote the post:

“Human beings generally don’t think about travel in a continuous Cartesian way. We think of it in a discrete linear way. That is as lines between origins and destinations. The rivers, roads, passes, trails, and other linear routes that we use to navigate about our day are really just lines connecting origins and destinations with sequences of landmarks we are familiar with. This is why the point crawl is such a powerful and familiar idea. The most important thing we ask ourselves about these lines is not the question “How far?” but rather “How long?” And this question should drive how we determine hex size.”

One discrete hex can be the same as one wilderness turn or watch. The size of the hex determines the length of a turn, the frequency of random encounters, and the density of wilderness points of interest. I agree with Steamtunnel’s take on the alternative merits of the 3 mile hex for tables that prefer using hexes as discrete units of time and specific game content as opposed to markings to aid exact distance measurements.

“These factors indicate a 3-mile hex is the superior measurement and here is why:

Travel from the middle of one hex to that of the next takes 1 hour over open terrain. This makes counting time easier.

Time to cross can be adjusted to allow for various terrain features.
The center of the next hex can be viewed from the current hex rather than the edge as is the case in the 6 mile hex.
This allows for all movement to be discrete and informed – we no longer need measurements.
A smaller size (approximate to 1/4 of the 6 mile hex) allows for a better focus for what is in the hex and thus a better discrete location.”

Being able to see 3 miles to the horizon allows the PC’s to always see the general terrain contents of all surrounding hexes allowing for informed decision making instead of randomly selecting a direction to be locked into for six miles even if there’s no access to a tree or hill to climb to see to the next hex. Terrain can be more detailed in chunks that measure allowing a “small” region of even 21 miles across to be revealed as a content rich, diverse and adventurous area that a whole campaign could be based in. This makes the world feel bigger and more discrete.

This 3 mile hex scale also helps provide a faux medieval feel if you care about that as it matches the 2-3 miles average minimum distance between villages in much of Medieval Europe if that era is an inspiration for your setting.

3 mile, one hour hexes can also be a great diegetic hex measurement as they are the same distance as one league and it sounds cool and old timey fantastic to have NPCs say “The bandits are camped 4 leagues up the road” instead of miles or hexes. I’m a proponent of a gradually filled PC hex map being a player facing tool and in game artifact. Burn some holes in it with a lighter if the party map carrier gets fireballed. Difficult terrain? Just increase the number of hours it takes to cross the 3 mile hex. A party can travel for a total of 8 hours.

If you are stuck being used to silly imperial units like me, the 3 mile hex also converts well to close enough to 5 km if someone in the metric majority of the world wants to use your hex crawl.

Coins and Scrolls blog has noted that all of medieval and modern Siena, Italy including the surrounding countryside, hills, farms, and over 17 castles fit in a single 6 mile hex. I think it’s a sparse fantasy setting that only has 1 or 2 points of interest in 6 miles. A 3 mile hex is still a large space but at least it’s ¼ of the area of a 6 mile hex resulting in a denser and more discrete region, defined in less abstracted terrain and content detail. 

Outdoor Survival, the wilderness board game that OD&D used for overland exploration also used 3 mile hexes to break down travel across terrain into discrete chunks. This experience tracks better with my experience hiking in the wilderness than chunking the day into huge 6 mile blocks.

Okay Joel, so when do you think larger hexes are good?
If instead of a location designed for exploration like a dungeon is, the wilderness is just connective tissue or purely an obstacle to be traversed between sites of interest where the real meat of your sandbox content is then larger hexes work well.  Also if you’re moving rapidly over great distances, such as in a sailboat or airship, then a hex scale of 18 or 24 miles could be better as the world whips by you and you don’t need to be concerned with more granular terrain for navigation at all. A point crawl between these detailed hexcrawl regions of deeper interest could fill this region space as well and simplify your prep.

(On the topic of point crawls I’m currently mulling over new stuff I’ve been reading including Ava Islam’s response to my post on the topic.)

Example Comparison Region

Below on the left is the 6 mile hex map from Kevin Crawford’s excellent Wolves of God historical fiction setting. It shows all the cities and regions of that era’s Britain after the fall of the Roman Empire and invasion by the English and is a land split into petty sparsely populated kingdoms where even the lords are poor and most of the population lives in tiny hamlets. This is a setting where societies are small and diffuse and dangerous wilderness covers much of the terrain. But I think that the larger scale and smaller density of this version of the map makes this area feel smaller and less imposing and wild than the setting implies due to it’s sparseness. In a setting like this, two days journey should provide a large serving of adventure and danger.

The larger the hex scale the less keyed stuff there is to encounter in the world through exploration gameplay and the smaller and less wild the world is. On the right is a fan made map of the same region done in 3 mile hexes.

Move the slider around to check out what I mean. I don’t think I matched up the scale exactly but you get the jist.

I think the 3 mile version provides a more interesting wilderness navigation experience due to the increased specificity of the terrain and contents. It makes a geographically smaller area feel imposing and dense.
It also might be too large of a region now due to the huge increase in the total amount of wilderness hexes over the same area.
For a more manageable scope a campaign could zoom into just the kingdom of Dyfed and environs and have plenty of content.

This smaller area is still 138 miles wide.
If you’re using hexes as discrete boxes of content then the grander your unit of hex measurement is, the smaller the world feels.
While the characters may travel further distances but the players experience of wilderness adventure is more diffuse because the content engaging with it is more spread out and sparse.

The advice on wilderness journey’s in Dungeon Crawl Classics RPG follows this principle as well. “In a game that accurately attempts to capture the medieval adventuring experience—or, phrased differently, in a game that retains the spirit of Appendix N—you do not need a vast space for adventuring. An area of land only 100 miles square should provide years of adventure.”

I think many GM’s including myself tend to inflate the size of our sandboxes using larger scale hexes due to our modern understanding of distance. This is influenced by our access to cars and trains and accurate maps of the entire globe to the nearest mile in our pockets. Enormous tracts of land feel instinctually more epic and large, but I think by making a smaller setting dense with content and wild we better inspire a feeling of scope and fantastical awe.

P.S If you’re researching how to build a great hexcrawl sandbox this recent post by Prismatic Wasteland is one of the better summaries and synthesis of good advice I’ve seen.


Why I Use Point Crawls More Than Hex Crawls

Point Crawl over Hex Crawl Cinderstrom

Hex crawl sandbox settings can suffer from an overdose of non meaningful choice without enough information to make informed decisions. I think using a point crawl makes choices more interesting by constraining them to detailed specific choices.

You wouldn’t run a dungeon of rooms with six entrances connected to each other in a large grid because the complete openness without sufficient information doesn’t lend itself to meaningful and fun decisions.

I love hex crawls but the more times I’ve run them the more I think that they’re best saved for certain use cases. Specifically they’re best for campaign that covers huge distances and is concerned with surveying the unknown expanding outwards from a point as opposed to making decisions about which route to take. They also are great for low prep pick up games that rely on procedural content over planned description and pre-connected areas pe of game where you’re surveying an expanse of land to eventually catalogue what’s inside each sector of a region and for a feeling of completion as your explore each hex. Because the way things are connected and the nature of the paths in between nodes of a point crawl matters it’s preferable to prepare the connections between locations ahead of time. A hex crawl is path neutral and abstracted so it does have the benefit of being quick to randomly generate the connecting hex, random encounters, and other content at the table and then improvise connections. If I need to play a quick pick up game. But if I’m buying a prepared module for my use or making one for someone else, it seems nice to have the detailed paths and connections done ahead of time.

I’ve grown to be a big fan of point crawls and prefer them in most of my sandbox games for increasing interconnection.

Hex crawls give the illusion of complete exploration and offer a level of top down abstraction that makes navigation choices less compelling then they could be.

When people talk about the value of a good dungeon as a quintessential play structure of old school inspired play they’re usually talk about how it can be designed in a way that maximizes and increases player navigation choices. Strange interactive objects you can mess around with. Verticality. Loops. Multiple paths. Secret entrances. Shortcuts to much deeper levels.

However the other thing about a dungeon that makes it such a strong structure is the constraint it places on choices. Because there’s only so many routes and rooms a GM can create a lot of interesting specific content in the area descriptions that makes these choices fun by spending some prep time providing meaningful information for each of the choices.

There’s a reason that dungeons aren’t designed in one huge cavern with no walls and sightlines for miles, it’s too much to parse and run effectively. Like a computer’s memory a GM only has so much brain power to render encounters and location details. Breaking things into rooms allows the rooms to be described in more detail and for specific connections between rooms to be meaningful. This means that you can improve your sandbox setting by using the same principles used in good dungeon design. I’ve attempted to follow these principles in designing the sandbox region surrounding my Cinderstrom megadungeon.

Some psychological studies have shown people with more options are more likely to be unsatisfied with their eventual decision then those with a limited amount as they suffer from analysis paralysis . When I was a kid growing up in Nevada I didn’t get the concept of the casino buffet right away. I discovered to both mine and the nearby buffet patrons horror that there is an upper limit to how many plates of chicken alfredo a 9 year old can devour. You can spare your players this unpleasant experience by giving them a cultivated set of path options instead of an endless void.

A dungeon of 50 rooms laid out in a grid with all with doors leading to six other rooms wouldn’t be the most satisfactory for exploration. The overload of choices makes navigation a series of random choices with minor information.
Why are so many hex crawl sandboxes formatted this way then? The same joys of making travel choices based on relevant information and limitation shouldn’t be left at the dungeon exit.

Now generally in a hex crawl you’ll be able to tell the party what terrain is in each cardinal direction but if you’re in a great desert that extends all around their location, what meaningful choice do the players have in their choice of direction between South and Southwest? If they know the direction of their objective they can just always head in that direction, cutting through the “walls” and forgoing having to consider the terrain around them. You can have a “getting lost” rule but frankly I’ve never talked to players who enjoyed this, nor do they tend to reflect reality as someone that’s done a good deal of off trail wilderness hiking (I’ll cover this in a future post). Now a good GM can provide more meaningful landmarks and cues to make navigating more of a meaningful and interesting gameplay, a lightly used game path to the east, a towering iron spire that stretches to the heavens to the south, a mirror like glimmering to the south east. But they’ve just effectively created a point crawl on top of their hex crawl off the top of their head. The other three cardinal directions are vestigial organs if not given some kind of more interesting clue to what the direction holds. I say hack them off!

I had originally planned on creating a hex crawl for my megadungeon region but realized that if there was 49 hexes in the region and I wanted to provide details for each of the I would have to create 294 different path descriptions. This didn’t seem worth the effort to me, along with the delay in play of describing each of the 6 exits to the hex to the PC’s each time without being an overcomplicated bore. Some WOTC designers visited Gen Con undercover in 2005 and found that DM’s get about 2-3 sentences of describing a location in a monologue without player responses before eyes glaze over and people stopped paying close attention. Initial node descriptions in a point crawl need to be just as pithy as a dungeon room description.

Switching to a point crawl based sandbox most human navigation in fiction or real life adventure is a point crawl. People follow paths of some kind to get to destinations. Look at Lord of the Rings, the Dying Earth, or Conan and the very specific and fascinating paths they had to decide during in their adventures. The Fellowship of the Ring were on a point crawl to reach their destination to the East when they had to choose between climbing the snowy Redhorn Pass and braving the path through the feared Mines of Moria. It was only when the snows proved too thick on the Redhorn Pass that they felt they had to chance the darkness of Moria (which could be run as a point crawl connecting different dungeon areas itself to give the sense of megadungeon scale but that’s for another post).

On a hex crawl all of these atmospheric and interesting environmental decision points could be summed up within one six mile “Steep Mountains” terrain hex.

Pictured: Not a hex crawl

Even in an open desert or rolling plains trekkers follow ridges, saddles, gulley, and other preexisting paths of less resistance. In my experience the very specific path is the interesting play space between the points of interest not the abstracted general terrain type surrounding around them.
“Should we take the skull and ruby strewn arroyo through or the crumbling remnant of a black marble highway around?”

Again, all this said I think the freedom of the hex crawl structure is still a swell and serviceable sandbox campaign structure and has been used in many of favorite sandbox books like Hot Springs Island and Neverland. I’ve just become a bigger fan of point crawls lately and think all the sweet hex resources out there like these sweet wilderness hexes can also be used for populating nodes and path connections to make a sweet point crawl.

P.S You can always layer a point crawl on top of a hex crawl as I’ve done in the header image and use your hex crawl procedure to adjudicate what happens if players decide to go through the natural “walls” and just pick a direction where there’s no terrain to make it easier. It should take a lot longer then following an occurring path like a road, or stream, or dune ridge though.

Copyright © 2024 Silverarm Press. All rights reserved.