This post first appeared on Patreon. To get access to all posts at least a week before they go live on Loot The Room, plus more exclusive content, sign up here.

As we’ve established, I’m trying to write magic items for a dungeon game that all do something unique while also providing generators for GMs to make their own stuff for the game. What I’m realising as I work on this is that I can’t really have both – generators are functionally incapable of creating wholly unique results. It feels unsatisfying, but I’m starting to suspect that my best option is to write small minor abilities (this weapon knows a random Phrase, this weapon sheds light, etc.), and for major abilities to direct GMs to invent something. This then goes together with a bunch of example/sample magic weapons that I write, to show GMs what sort of things these items should do.

As I think about this, I’m wondering if the book that I release these in (or blog post, or web page, or whatever I choose to do) should focus more on the philosophy behind how magic weapons are treated in ADG rather than on simply providing the weapon stats and the generators. And with that in mind, I’ve been thinking about what magic weapons actually are in a dungeon game.

Something I’ve had in mind for a long time is that magic weapons (as opposed to magic items in general) aren’t something that’s easily crafted. Instead, they’re forged through acts of unimaginable violence. The intelligence inside a magic weapon isn’t an embodiment of the weapon itself and isn’t something that was created out of the ether – it’s a remnant of the person who was slain by this weapon when it was simply a mundane item, and the magical abilities the weapon possesses are reflections of the power that person held in life. The result of this is that magic weapons are the rarest form of magic item, and they often have antagonistic relationships with the people who wield them – especially if they’re newly-forged.

This adds a couple of layers to my generator. We need to know who the intelligence was in the past, who killed them, and how long ago that occurred. We probably also need to know why they were killed. We’ll also need to decide what type of weapon they are (right now ADG only has Small, Medium, and Large weapons but maybe for the magic weapons we can get more specific within those categories). It might also be fun to decide how complete or fragmented their memories of their past life are, and what they want to accomplish – do they seek revenge, or redemption, or peace, etc. Perhaps we should also decide whether their killer is still alive, and whether the weapon is aware of this or not. I’m also thinking about my game Go Alone, which asks you to roleplay a sentient weapon and reflect on what it means to be used as a tool for violence, and I’m wondering how much of that I can bring into this. I like the idea of determining how many previous bearers the weapon has had, and how hey feel about being used for violence. Finally, we should also decide whether the weapon is able to grow or diminish in power over time.

With that in mind, I’ve taken a run at some more tables for this generator. Here’s the one we ended up with last time, for reference:

And here are the new ones:

I feel like there’s enough here to generate a fairly complex, unique intelligence, and for the GM to know what drives it and how to roleplay it. The only thing I’m missing now is “what do the powers actually look like?”

My instinct is that minor abilities should be drawn from a random table – things like “knows a random Ritual”, “functions as an additional Scar”, etc. Small, rules-based tweaks that are essentially the a dungeon game equivalent of a +1 style bonus. Major abilities should be left up to the GM to create, with a few (maybe 10?) pre-written example weapons to consult for ideas.

So my next step is to go away and write those minor abilities and to create some example weapons. In the mean time, I’d love to hear any thoughts you have about this!

This post first appeared on Patreon. To get access to all posts at least a week before they go live on Loot The Room, plus more exclusive content, sign up here.

Last time I talked about my aims for magic weapons in a dungeon game. To quickly recap:

1. All (or most) magic weapons should be intelligent.
2. All magic weapons should have at least one unique ability that “breaks” the game rules in some way.
3. The GM should have tools to easily create new magic weapons for their game.

It’s no secret that I’ve drawn on 2nd Edition AD&D a lot in writing a dungeon game. This has largely been in the way I balance the game – my experience thresholds and treasure quantities all use the numbers in 2nd Edition as their base, and some my monster stats are at least partially derived from it. None of it is an exact copy, but why reinvent the wheel?

With that in mind, I thought it was worth looking at AD&D’s magic sword generators for inspiration. Both the first and second editions of AD&D have fairly extensive tables for generating intelligent weapons. For the rest of this post I’ll be referring to the original version of the 2nd edition Dungeon Master Guide, not the Revised edition, so if you want to follow along your page numbers may not match, but generally this stuff is found in the appendices (I believe it’s Appendix 3 in the Revised edition, as in the edition I’m using).

This procedure starts on page 186 with table 113: Weapon Intelligence and Capabilities, which lets us determine the weapon’s intelligence, method of communication, and its capabilities. It looks like this:

• ± The weapon can also read languages/maps of any nonmagical type.
• ±± The weapon can read languages as well as magical writings.
• ±±± The weapon can use either communication mode at will, with language use as any speaking weapon.

Before I go any further in this process I can already see that this can be fairly easily adapted to a dungeon game. I will need to adjust the percentages to account for the fact that the game only uses d6 and d20, but otherwise I can more or less copy this.

When I’m writing for a dungeon game I often run into tension between my desire for the game to be fast and “rules light”, and my love of crunchy trad games with lots of cool bits to play around with. This is something that pops up most often when it comes to monsters. When I write for Pathfinder 2 I love making monsters and coming up with cool things that they can do, and I find a lot of satisfaction in putting together a complex stat block that has a set of abilities that work well together and make the monster feel unique. With a dungeon game I’m always tempted to go too far with monsters when really I want the majority of them to do just one cool thing, and I’m feeling that same urge when it comes to these weapons. For that reason – and because I’m still struggling to see how I’m going to “tool-ify” this process – I don’t know if I want quite so many results that grant 3 primary abilities here.

Here’s what my first run at this table looks like for a dungeon game.

Rather than rolling percentile dice, we simply generate the weapon’s Cunning and derive the rest of its attributes from that. Player Character Cunning is in the 3-18 range, and this only gives us magic weapon intelligence up to 12, which sort of works but also sort of doesn’t. Highly intelligent weapons are fun and introduce a new dynamic into play, and I’m robbing myself of that here. Let’s expand this out to 3d6 and see how it looks.

(You’ll notice I’ve also played around with the communication methods here. That’s just adjusting things to my own personal tastes, and there’s no real reasoning behind these choices other than that it felt right.)

On this second run at it I’ve introduced the concept of Major and Minor abilities. I still don’t know exactly how I’m going to generate these, but I feel like we get a good variety of weapons from this table.

Going back to AD&D, the next set of tables generate:

• Weapon alignment
• Weapon primary abilities
• Weapon extraordinary powers
• Special purpose & special purpose power
• Languages spoken by the weapon
• Weapon ego

Weapon ego (and from that, its derived personality score) makes for some interesting outcomes but it’s also a bit of a pain to work out and keep track of, and I’m not sure it achieves much that isn’t achieved by the weapon having a Cunning score that’s used as the basis for saves. That said, the constant shifting of player character personality scores (which you have to work out once a character begins wielding an intelligent weapon) makes for some interesting outcomes as the weapon’s intelligence takes advantage of moments of weakness. I’ll have to think about whether this is something I want to codify into the rules of a dungeon game or whether this is something that can be dealt with simply through GM advice.

What I’m really interested in here are the powers that AD&D grants to magic weapons. Here’s table 115: Weapon Primary Abilities

And Table 116: Weapon Extraordinary Powers:

The immediate thing that I notice about this abilities is that they’re really boring. If I wanted to replicate this I could simply say that every magic weapon knows X amount of Rituals and have GMs roll for them. That’s not a terrible idea for a minor ability, but I want my weapons to be actually interesting.

For now I don’t have an answer, and so I’m just going to think about it a little more. I like the base table, though, and I think it’s a good starting point. All that remains to be done is to figure out how to draw the rest of the owl.

This post originally appeared on Patreon. Patrons get access to blog posts at least a week before they go live on Loot The Room.

I feel like I’ve been trying to “solve” magic items in a dungeon game for years – because I have. I’ve been thinking about how they’re going to work for literal years and still haven’t quite got there.

Actually writing the weapons isn’t the hard part. I know exactly what I want them to do. They’re all intelligent, and they all have at least one cool thing that they do. They sort of function in the same way as monsters in that regard, where each of them has a way in which they break the rules of the game.

Designing them is fairly straightforward, but where things get tricky is in giving GMs the means to design their own. A core design principle I try to upload when I work on stuff for a dungeon game is that I would rather provide tools to allow you to expand the game and build your own things that work with it, rather than just slapping down a book containing a bunch of magic weapons that you may or may not ever actually utilise.

Of course, Little Miss Scope Creep is always present in my mind, saying “but you could write a book of magic weapons and write the tools to allow GMs to make their own”. And this is why we still don’t have a magic weapons book or creation system for the game, because I want to do both and can’t figure out how to.

So what’s the difficulty? The difficulty comes from figuring out what the generation tools for these things should look like and how they should work. If I’m going to write a tool to create things for the game, I want it to be the same tool that I use to create things for the game. I very firmly don’t want magic weapons that confer the equivalent of a +1 bonus, because that’s boring. They should all do something unique. But it’s very hard to write tables that generate unique effects that are also re-usable. You either end up exhausting the utility of the tables – i.e. you see all of the “unique” things, and thus no longer get unique results – or else there’s a point on the table that instructs you to now draw the rest of the fucking owl.

There are a couple of already-extant magic weapons in a dungeon game, hidden inside The Moss Mother’s Maze. The game has evolved slightly since that adventure was written and I’m taking a more structured approach to stocking dungeons with treasure, but those items still reflect my main philosophy that magic weapons should be cool and unique. Here’s what they look like.

Hadratha

Magic longsword. (Medium melee weapon).

When you roll damage you can Exert 1 point of Cunning to add an additional damage die. You can do this as many times as you like on each attack, provided you have the Cunning to pay.

Sacrificing Cunning strengthens Hadratha, with the following effects.

Hadratha Given Flesh
HD 6 AC 7
Deals damage directly to Cunning.
Can choose for her attacks to hit without making an attack roll. She can do this up to 12 times, after which point she loses this ability forever.

Cursed Spear

Large melee weapon.

While you wield it you can’t be killed by violence.

Curse: The Moss Mother always know the location of the spear. When it is removed from this chamber the GM rolls 1d20 secretly. This is the current location of the Moss Mother. Do not roll for encounters again. Each Turn she moves one room closer to the spear, taking the most direct route. Her approach can be heard throughout the maze.

These really occupy two extremes. Hadratha has intelligence and a fairly complex set of abilities. The cursed spear has a simple ability, and a curse that only really has meaning within the dungeon in which it’s discovered. The question now is, how do I construct tools that yield results that fall somewhere within this spectrum of magic weapons?

I’m not sure what the answer to that question is, and I don’t think I’m going to solve that problem today. This post is really just me coming back to touch the issue again, with a view to finally cracking this nut. I’d love to hear thoughts about how to approach this, though, should you have any.