This post relates back to the ‘Forget polyhedrals – Cubism is in !!‘ post I made recently. The original idea of that blog post was how to make a (linear) die using 2 dice – so a D12 or D18 using 2 x D6.

In that post I had throw away remark saying you could use this general idea to make D&D like stats – anyway I couldn’t quite help myself, and so I thought I’d try this out to see how well it could be used to generate D&D type stats – seems it works quite well (but I can’t really see anyone really doing this) :

Typical D&D stat rolling methods (orange line = 3D6 method and the blue line = 4D6 drop the lowest method) shown along side using this method where one method is D6 + D12 (where the 1 and 2 are replaced by 6 and 7 – the green line) and D6 + 12 (where the 1 to 4 are replaced by 8 to 11 -the brown line)

The two dice give a result that looks quite like the result from using 3D6 – the main differences being that the 3 result is excluded (hardcore players can frown now), the 10 and 11 scores are slightly lower and the 15+ scores are slightly higher. Other than that, surprisingly close!

The two dice give a result that looks quite like the result from using 4D6 drop the lowest – the main difference is that you don’t get any 3 to 5 scores (players using this method would probably be happy with that!). Other than that difference, surprisingly close!

Hopefully, I’ve got the maths correct!

:O)

This post relates back to a ‘composite dice‘ post I made a little while ago.

Recently, I got thinking there is another way to generate composite dice in a similar way. I figured, let’s do this with D6 first (to explain the thing), and then back it out to all dice.

“Composite die” I hear you say?

As you will know percentile dice (D100s) are made of two dice: a tens die (with faces: 00, 10, 20, 30, 40, 50, 60, 70, 80, 90) and a units die (numbered 1 to 10). But, this idea can be done with any two kinds of dice, not just with 2 x D10s.

So, for example, a D36 can be made of 2 x D6s, with one die labelled one to six (1 to 6) and the other labelled 0, 6, 12, 18, 24, and 30 (i.e. going up in 6s, not 10s like a precentile) – roll both and add them together.

Recently, I realised that you don’t need to go up in even intervals of 6s for each face on the equivalent of the 10s die. For example:

Option 1 to generate a D12 using two D6s: One D6 is a standard D6 (1to 6), and the other D6 is labelled 0, 0, 0, 6, 6, 6. Roll both and sum the result.

D12

I faintly recall that the DMG might talk about making a D20 in a similar way from a D10 …?

Option 2 to generate a D18 using two D6s: One D6 is a standard D6 (1to 6), and the other is labelled 0, 0, 6, 6, 12, 12. Roll both and sum the result.

D18

Option 3 to generate a D36 using two D6s: … again, you are back to the ‘standard composite die’ I originally blogged about – One D6 is a standard D6 (1to 6), and the other is a D6 labelled 0, 6, 12, 18, 24, 30. Roll both and sum the result.

D36

The trick of these ‘new composite dice’, of course, is to make sure that the ranges do not overlap (and each number has the same probability of coming out); otherwise, you start getting into bell-curve-like situations or uneven results (so no 0, 6, 6, 6, 12, 12 etc. – unless you are trying to generate something like a pseudo-bell curve for something resembling a D&D type character stat roll)

So, just using the humble D6, you can generate a D6, D12, D18, and D36.

If you make the D6 cube a D3, you can also generate (I believe) a D3, D9, and D33 (in addition to the above options).

And if you make the D6 cube a D2, you can also generate a D2, D4, and D14, I believe (in addition to the above options).

So, using the simple cube, you can make (unless I’ve made a mistake) D2, D3, D4, D6, D9, D12, D14, D18, D33, and D36 (who needs a percentile anyway?).

Of course, you are not restricted to D6s. You can use D8s, D10s, D12s (just noting that D12s are like 2 D6s), etc., and also mixtures of those dice, so D6 and D8, to generate lots of further different composite dice.
:O|

That’s it.  I’ve thrown it out there in case anyone finds use for it.

2024 was a reasonably quiet year for me (reasons). In 2024 I posted 6, arguably 7 posts (that were more than just update posts), and one of those was on 31 December. 

In 2025 by mid-February I’ve put out 8 posts, all of which (I think) have something to offer a niche within a niche (which is my niche): 

• Word Play – Old monster New monster

•   Simple Monster Hack | Generic creature + PC spell

•   D&D type Stats – all 1s are 2s

•   Usage Dice – there can be only one (left)

•  Random Encounter Tables – When the Boss Monster comes out swinging

•  Faction Action – Burning the Candle towards both ends | An approach to Random Encounter Table design

•   Dynamic Random Tables | Controlling Access with Rule “Windows”

•  Can you Push it? | … are pushed rolls a thing in D&D? 

oh… and one shit post: Permission to eat the PC’s horses?

I think the Bloggies are an interesting celebration of Blogging, and thinking and reading blogs has spurred me on to post more (which of course is good; and perhaps a niche within a niche has had a bit more to think about – so presumably a tad happier?!) … but will it last? Perhaps that doesn’t matter.

How about you? 

Edit: No animals were harmed in the writing of this post 

Very important context

I recently got included in this Meme (What the dice doin??? †):

†= felt the Meme was in need of a modest but absolutely necessary codifying tweak ^ 

… and within a 24hr window another Redditor stated the below about my post that included the tables used in the above Meme†:

“I was reading it and thinking “This sounds like a blog from that hexflower person” before realizing that, indeed, it was a Goblin Henchmen blog“

So, I thought I best do something to mix things up a bit, be a bit less predictable, do something inside the box, do something that is all RPG content and no RPG design.

– End of very important context – 

So, I’m going with:

Permission to eat the PC’s horses? 

Postulate: when the PCs are in the dungeon have someone or something steal, and preferably eat, their horses. Just do it once.

Would I be breaking some sort of social contract with the PCs, like burning down their safehouse, or taking their fav. magic items (soz) etc … is this HEE  or HAA ?

To me, these grass-eating, mobile storage enablers feel like the weakest link in the otherwise impenetrable PCs’ defenses.  I can’t help thinking a watching monster/villain/Big-Bad would notice this weakness and act accordingly.

Forget Mouse Guard, introducing Horse Guards …

Thoughts?
Are there any sacred cows that DMs shalt not touch?

I’ve been listening to a bunch of Call of Cthulhu (CoC) actual plays recently (Andy Goodman’s Grizzly Peak Radio) and was interested in the dynamic of a “Pushed Roll“.  

Skorpadillo – not on my watch (… wonder if this chap thought this critter was a Rust Monster [same vibe] and stripped to his loin cloth; also love that fluid snake) – Credit: Erol Otis – Skorpadillo from Arduin Dungeon #2 The Howling Tower by Dave Hargrave, Grimoire Games 1979.

From my understanding (forgive me if I’m preaching to the choir here), a ‘Pushed Roll’ is an option when a PC fails a skill type check in CoC. The PC often has the option to roll again, but if they  fail a second time it has dire consequences. 

Sounds like a fun option for D&D combat (or even other skill type checks). I’d probably avoid using this for saving throws (CoC excludes this for Luck Rolls I believe), because it would be a no-brainer to reroll a failed Save vs Death, unless there is something worse than death in your campaign, … OK, scrap that).

Here are some ideas:

Pushed roll on –  missed a combat hit?

Missed a hit and want to push the roll? Obviously pushing it and missing again should be bad – how bad? Well that’s up to the GM – critical fail, lost grip on weapon, shattered weapon, injure self etc.

Whatever it is, it should be a deterrent to just doing it every combat round. Someone should make a table … like this one (but better): 

1. 1. Stunned self – opponent gets double attacks next round 
   2. Broken weapon (magic weapons get a save)
   3. Injure self on own weapon  – weapon damage/2
   4. Injure friend – weapon damage/2
   5. Wicked vibration – numbing pain in arm; further attacks at -4
   6. Fall flat on the floor fully prone – lost next attack, opponent gets +4 to hit 
   7. Winded – lose attacks until save made (save vs Breath Weapon?!?)
   8. Armour confounded – AC downgraded by 4 units until fixed
   9. Spun around – back is facing opponent – their hits do double damage 
   10. Weapon stuck in object/monster – Bend Bars Lift Gates roll (or alike, e.g. roll % under STR) to free weapon

Pushed roll – damage roll too low?  

Maybe this is a non-option if you roll a damage of 1, or perhaps better you need to roll higher than last time otherwise it is a massive failure.

But, what sanction for a re-fail? Maybe use the same table as above with 0 damage done to the opponent?

Why do this?

Well maybe the PC and the monster have 3 HPs left, the Monster has 3 attacks and goes next, and the PC misses or rolls 2 damage. The player might figure it is a do or die move and so takes this chance to win it. 

Gaming the system

Should this be a declaration by a player after they roll (use it or lose it), or a declaration by any player before the monsters turn in initiative based combat? Probably the former for simplicity. 

Apply this to Magic? 

Why not, but again the consequence of failing a pushed roll should be disproportionally  bad  … which way does that stinking cloud go now …? 

Feels like this must be a thing already, if so, I haven’t seen too much chat about it …
:O)  

I’ve only recently realised that I’ve had a Random Table Itch that I’ve needed to scratch for some time.

In this post, I’m going to delve into the general idea of making a single random table which uses different access “windows” (i.e. rules) to situationally access the content of the table.

Setting Up Random Tables

Random tables are great for generating everything from encounters to treasure:

A common approach used by people is to create multiple specialized tables (e.g., one for plains encounters, another for forest, hills and mountain encounters etc.). However, arguably this approach can sometimes become unwieldy:

A potentially more streamlined approach is to use a single more comprehensive/encompassing table with rules to determine which parts of the table are accessed. These rules act as “windows” sliding across or highlighting different portions of the table. This dynamically filters results without generally altering the table itself.

Here are some ways this could be done.

Structuring Your Table

Table structure is important when using this ‘windowing’ approach. If your table is completely homogeneous, the “window” concept is not much relevant (except for perhaps introducing a level of unpredictability/chaos). Just to cover what I mean, here are some common random table structures:

• Rarity: Entries arranged from common to rare (or vice versa), or with the rarest entries at both extremes of the table etc.:

• Power: Order encounters from weak to powerful (or vice versa etc.):

• Factions/Alignment/Theme: Organize/group the table entries by faction type, good-evil axis, or even treasure chest contents

Applying the “Windowing” approach 

Here are some ways you could use to implement these rule-driven access “windows”:

• Sliding Window: A fixed-size access ‘window’ moves across the table, only selecting entries within the window range (this is usually done by adding a situational modifier, e.g. 2D6 + a situation modifier). The window’s position is situationally-dependent e.g. a monster type is modified by terrain:

The window can be spaced apart (e.g. left above), or can overlap (e.g. right above). An example of this can be found in my In the Heart of the Unknown where encounters are dictated by a terrain dependent modifier.

• Probability Curve: A linear, peak or bell curve, centred on the table, could expand or contract (as dictated by the situation), making entries under the curve selectable by the situation. These curves can also be made to gradually extend across the table:

I posted something about this idea here and here, and here’s a linear example from Ktrey of the D4 Caltrops blog fame, covering day encounters and extending the table to cover night encounters.

** PS – these are not real probability curves, I’ve drawn them in Word! **

• Probability Curves: Multiple probability curves overlay the whole table, but where each curve favours a different section of the table:

An example of this can be found in my “Carapace”adventure (basic version here and the art version here) and blog post, where it becomes progressively more probable to encounter the Queen Ant as you progress deeper into The Hive.

• Or, a bit of both: For example you could use different dice combinations (e.g., d6, 2d6, 3d6) to create varying probability distributions and which cover only some parts of the table:

For examples see Bloggie Silver Winner entry here and my blog post here

• Opposing Dice: Two dice “push” a window in opposite directions, creating a moving window (a window with a sort of “memory” no less):

It’s a bit like the ‘Sliding Window’ approach above, but this window has a random roaming element. See MINDSTORM’s Dice Ladders blog post here

• “Magazine” Table: Table entries are deleted when encountered, these entries are crossed out, and new content feeds into the table from say the top or bottom. The changing situation triggers the use of the table e.g. see an example here

• 2D tables: A d66-style table can become a d44 or d88 table (or even a D48 table), expanding or contracting the available options (e.g., 16 to 36 to 64 entries). For example, as you get lower in the dungeon, the types of random monsters types reduce from 64 to 16, perhaps to the strongest 16

I’ve not seen these kinds of table in the wild, but sure it must have been done.

Or perhaps (also) including a probability gradient to a ‘normal’ D66 type table:

like this and in the wild here.

• Unique Rules: The above are not intended to be a comprehensive list, just some examples.  I’m sure lots of options are out there and still to be done.

Real-World Examples

Above I have linked to some real-world examples. I have favoured my own work just because I know it best … and can remember where it is! But I have included others that I have spotted out in the wild. I’m sure there are more examples out there.

Summary

The core idea is to have one table with comprehensive/diverse content and use rules (windows) to control which parts of the table are accessed or how likely they are to be encountered. The access rules are dictated by the situation. Hopefully, at least this approach offers interesting options and perhaps can reduce the need for numerous separate tables.

With that ridiculous itch scratched, I can move onto something else … !!

Some extra reading

For more on random table design, you could look at my “A Field Guide to Random Table Design“. It covers the basics, but sometimes it’s good to go back and look at these.

CC image – click to see source (in theory the above is the opposite of my idea :O|)

In a similar vein to my “Random Encounter Tables – When the Boss Monster comes out swinging” blog post, I like the idea that only certain parts of the random encounter table become progressively unlocked after some action has already taken place.

Here’s one way of doing it:

Illustrative Random Encounter Table* 

(* = of course, you can mix this up so it’s less of a power progression)

Rules

1. When a random encounter is triggered, roll 3D6 and cross reference the random encounter table above.
2. Remove (cross out) that random encounter (sign or event or trap etc.) from the table
3. If you roll another random encounter and that encounter has already been used (and so deleted from the table), move towards the closest (distil) end of the table, until you encounter an unused random encounter.

For example, if your first random encounters were 5, 15 (e.g. Faction A Standard level Monster, and Faction B Standard Level Monster) you remove these from the table. If next you then roll 13 , then move towards the 20 (distil) end of the table until you get an unused random encounter, in this case 16-18 (e.g. Faction B Brute Level Monster):

If you next roll a 15 as your next random encounter, then again move towards the 20 (closest distil) end of the table until you get an unused random encounter, in this case 19 (e.g. Faction B Boss’s Right-Hand Man).

Of course if you rolled 5, 6, 7 or 8 you’d move towards the other distil end, i.e. towards 1, so this would give 3 to 4, Faction A Brute Level Monster:

So the idea is that as you use up random encounters, the end results unlock and start to be more likely.

Some options?

• Use a smaller table e.g. 2D6 (but you will only unlock one entry at each end, unless you include a 0 and 8) or 3D4 etc
• This table doesn’t have to be about factions – for example one end of the table could unlock something bad (like a trap or a dungeon-wide disease), the other end could unlock something good (like treasure or magic item).
•  You don’t have to use ranges, for example if you went:

13 = Faction B Standard Level Monster
14 = Faction B Standard Level Monster
15 = Faction B Standard Level Monster

instead of

13 to 15 = Faction B Standard Level Monster

you would only deplete all the ‘Faction B Standard Level Monsters’ after 3 occurrences rather than one occurrence if using a range.

Name

Nothing seems to get  very far without a nifty name … Burn Tables (Distil Tables?)
Sounds too cool, and anyway someone must be using this term already for something way more METAL