Tabletop Review: Shadowdark RPG by Kelsey Dionne - 6/10 - ******
Enter the Shadow-verse.
Thank you to my proofreaders Bobloblah and Teleros.
18/01/2024 - Some changes have been noted to adhere to the current version of the Shadowdark rules.
Headline Info
System: d20-based
Type: Rule System
Genre: General Fantasy / Mudcore
Read or Played: Read
Page Count: ~330 pages
Cost: $29 pdf / $59 pdf + hardback
Purchased or Provided: Purchased
Introduction
Shadowdark RPG (2023), referred to in marketing as Shadowdark,1 is a d20-based system which had a $1.3mil Kickstarter in March 2023.2 There was a large outpouring of support for the Kickstarter, including some of the biggest members of the old-school renaissance (OSR) community coming out to give glowing descriptions of the gameplay, mechanics and design.
The Kickstarter marketing gave an estimated April 2023 delivery date for digital files, and an estimated July 2023 date of delivery for physical books. This would have given a lead time of under a month for the digital files, and three to four months for physical books.
This was an obviously ambitious timeline but likely would have been achievable if moderate success had been achieved. Given that this was The Arcane Library’s first Kickstarter and an unprecedented success, it should not be surprising that the timeline has been shifted slightly.3
As of writing, digital files have been delivered but the books are still waiting on delivery. The last update on the campaign was a few weeks ago stating that the Arcane Library had received the physical books and was now performing distribution.4
The Kickstarter marketing claimed that the system would unite old-school gaming and new-school mechanics. Functionally, Shadowdark’s goal was to give an old-school renaissance (OSR) feel to a system closer to 5th edition Dungeons & Dragons.
And it did, to some extent, achieve that goal.
Review
The first thing that pops out from the rulebook is the cleanliness of the presentation.
The trade dress, as done by Brandish Gilhelm,5 is pleasant and readable.6
The main artist appears to be Lucas Korte aka shoggoth_kinetics,7 and his art is used extensively in the marketing and to set the tone of Shadowdark. The most common artist in the book is Yuri Pewkowski Domingos8 who has a slightly more playful and early Warhammer style.
I found very few things to fault in the presentation. The only problem that stood out to me is the fact that the text is left-aligned,9 but that is likely for the purposes of assisting readability for some people with disabilities.10
As far as I can tell, there is no editor listed in the attributions section, which is unusual.11
The rulebook is separated into seven sections:
Introduction (~5 pages) - Describes Shadowdark and the basics of the ruleset
Characters (~30 pages) - How to create a character, how to level up and a list of equipment
Magic (~30 pages) - How casting spells works and a full list of spells for wizards and clerics
Gameplay (~25 pages) - The player-facing rules section
Game Master (~80 pages) - The gamemaster-facing rules section
Monsters (~90 pages) - The Shadowdark bestiary
Treasure (~50 pages) - The treasure generator and magic item section
This review will mostly focus on each section in turn. However, there are some design decisions that I will need to check across multiple sections to shed light on them.
#1 - Introduction
The introduction section contains a preamble which sets the game tone, a section that attempts to define Shadowdark against other tabletop games, and a basics section that goes over a small amount of key rules.
This section states that, “Shadowdark RPG is rules-light and intuitive”, but the rules are not quite rules-light. The core rules (characters, magic and gameplay) are 85 pages long.
I imagine that to the target audience of 5th edition Dungeons & Dragons players this is likely unbelievably stripped down, but there are many systems that manage similar in ten pages or less.
The system definitely required me to have some intuition. I had times where something was not clearly defined or the definition was linked to another page which had subtle differences in a rule.
The basics section has descriptions of the core rules, as follows:
Standard d20 plus modifier roll over difficulty check in cases when needed
Each character has a turn per round - uses the 5th edition naming conventions12
Characters can move 30ft per turn
Advantage system ported directly from 5e - if you are in a good position you roll two and pick the best, if you are in a bad position you roll two and pick the worst
Standard stats from 3-18, modifier bounds are the same as 5th edition with every 2 points away from 10 adding an extra +1 modifier
Hit points are rolled per level, at zero HP13 you will die after 1d4 + Constitution modifier rounds
Two points of frustration rear their heads in the Basics section:
Lack of clarity leading to conflicts in when a roll should occur
Mixing natural language and keywording leading to confusing paragraphs
These problems are consistent throughout the book, and I will highlight some examples as they occur. To me, both of these problems should have been solved through an editing pass.
Lack of Clarity
In the Basics section, there is no clear description on when a roll should occur. The below image is taken from various parts of the basics section.
Functionally, there appear to be three different methods here:
“Characters”: The GM determines if actions are successful with no mention of a roll.
“Describe an Action”: The GM uses GM fiat to determine whether a player needs to roll.
“Checks”: Checks are made when attempting risky actions.14
I understand this is slightly pedantic, but this is the foundational mechanic for the game.
To have even two different descriptions a hundred pages apart would be poor editing, to have three different potential interpretations across two pages is unbelievably sloppy.
There is another, seemingly definitive, method in the Gameplay checks section of the book.
Natural Language and Keywording
Natural language is the art of writing rules in a way that flows as if you were explaining them in conversation. Keywording is where the writer indicates and ties concepts to specific words so that they can be referenced later on.
Texts written in natural language can be more understandable at a basic level, but there are problems with semantics and definition. As an example sentence: “I saw a man on a hill with a telescope.” There are at least three possible interpretations of that sentence.15
Shadowdark is overwhelmingly written in natural language, with the inclusion of keywords in an attempt to shore up ambiguity. With very clear writing direction and a good editor, it can work.
Unfortunately, that is not the case here.
There are situations where the use of natural language runs into a brick wall because of the keywords. See the following paragraph taken from Basics and note that bold is sometimes used to indicate keywords in Shadowdark:
MOVEMENT
During your turn, you can move up to near (roughly 30 feet). You can break this movement up in any way. If you don’t take an action, you can move near again.
The use of the keyword near makes the natural language uncomfortable to read.
I should note that it is not clear why near must be keyworded here instead of simply using 30 feet. I imagine that it is to stop players carefully avoiding enemies by being 31 feet away, but it makes reading the rules uncomfortable and clunky.
This issue with near is repeated later on several times.
With an editing pass, it could easily be have been rephrased as:
MOVEMENT
During your turn, you can move anywhere nearby (roughly 30ft). This movement can be broken up in any way. If you take no other action, you can move nearby again from your new location.
The keyword and the natural language are both maintained.
In addition, there is a frustrating lack of consistency on keywording. For example, the following two paragraphs are from the Basics section of the book:
DESCRIBE AN ACTION
On your turn, describe an action you want to do. For example, you could say you’re going to shoot your bow at a troll.
The GM might ask you to do a check (see Checks) to determine whether your action succeeds.
Is action a keyword or not? It is bolded in the first paragraph but then not in the second.
This reoccurs throughout the book in four different forms:
Some words that are bolded do not have a definition
Some words that are not bolded should be as there is an underlying keyword
Some concepts should have a keyword associated with them but do not
Some words are inconsistently bolded and then used in natural language throughout the text
This made understanding the rules much harder than it really needed to be, as I had to repeatedly check whether a word was a keyword.
#2 - Characters
The “Characters” section is about character creation, equipment and some information on advancement. For the most part, I found it functional.
Characters themselves are mostly standard with the following traits:
Stats referring to Ability Scores
Derived stats which are not clearly listed but do exist
Ancestries16
Classes - A package of abilities that all characters of that class get
Talents - Every odd level, characters gain a talent which is determined by randomly rolling on a table individualised for each class
Background - what your character did before taking their class
Alignment - standard BX alignments
Equipment
Stats
Stats are rolled 3d6 in order, but players can reroll all their stats if they do not get a 14 or higher in at least one.
This means that players will reroll their stats once ~30% of the time, twice ~10% of the time and three times ~3% of the time.17 To put that into perspective, there is a ~3% chance of rolling at least one 18 on 3d6 down the line.
The chance of a player rerolling their stats three times is the same as the chance of a player rolling an 18 on a standard stat line.
In contrast to most other OSR games, high stats are essential to characters. Class abilities use stats heavily and the usage of 5th edition ability score break points means that every point very much does count.
A Magic-User with an Intelligence of 11 might be fine in Moldvay D&D, but in Shadowdark they will likely lose all their spells within a few turns.
Ancestries
The ancestries aka races are relatively basic.
Each ancestry has a singular bonus associated with it.18 For example, goblins have the bonus:
Keen Senses. You can’t be surprised.
While elves have the bonus:
Farsight. You get a +1 bonus to attack rolls with ranged weapons or a +1 bonus to spellcasting checks.
To me, these bonuses appear to be not well balanced. The elf bonus is a weaker version of a class talent19 whereas the goblin bonus cannot be replicated in any way.
Classes
For some reason, the book drops straight into the “Fighter Class” without any description of what a class is or what one does.
However, the start of the “Characters” section has an overview of each component, and this is how it describes classes:
CLASS
Your character’s job.
Fantastic.
The classes are a relatively standard grouping of Fighter, Priest, Thief and Wizard.
Fighters get skills that are based off of Strength or Dexterity and Constitution, Priests get bonuses based off of Wisdom, Thieves are actually completely unlinked from stats according to the rules, and Wizards get bonuses based off of Intelligence.
These are a pretty ordinary fare. Fighters get hit and hit other creatures, Priests cast buff spells, Thieves try to backstab and Wizards blow things up.
Looking at thieves specifically, they have Backstab as a class ability. The text for this is as follows:
Backstab. If you hit a creature who is unaware of your attack, you deal an extra weapon die of damage. Add additional weapon dice of damage equal to half your level (round down).
The problem that I see is that “unaware of your attack” is not defined. It is ambiguous, and will cause many arguments at a game table. It is not clear why Backstab is not linked to the surprise mechanics found later.
Continuing on, each class has a list of talents that they can roll on at each odd level (1st, 3rd etc). It is a 2d6 roll and therefore grouped heavily towards the 7-9 range.
The talents themselves are dull and unimaginative. +2 to some stat, +1 to attacks and damage etc. You can see the Fighter Talents table below.
This is a system that has been used well in a few games, most notably small scale wargames. The best example is the Middle-Earth Strategy Game Battle Companies supplement.
In that game, there is a list of special abilities that the characters in the company can roll on. Your characters are often defined outright by those special abilities.
It would have been nice to see the same here.
Backgrounds
After classes, players can pick a background for their character. Backgrounds have no listed bonuses but players should “work with the GM to determine if your background provides you advantages in a given situation”.
It is not clear if this is intended keyword advantages, mechanical bonuses or just roleplay benefits.
It is strange to have backgrounds after classes. A minor point in this case but similar layout issues occur throughout the book and result in a lot of flicking back and forth.
Alignment, Gods, and Equipment
The alignment, gods and equipment sections are relatively standard. The gods seem to imply some interesting things about the Shadowdark default setting that would be nice to see expanded.20
Then there is another problem in the use of natural language in the “Gear Slots” section:
GEAR SLOTS
You can carry a number of items equal to your Strength stat or 10, whichever is higher.
Unless noted, all gear besides typical clothing fills one gear slot. Gear that is hard to transport might fill more than one slot.
This is due to the continuous mix of natural language and keywording. The number of items is not equal to the amount of gear slots available, and some of the things that take up gear slots are not pieces of gear.
There is no real reason for them to be called “gear slots” except to differentiate them from whatever is used in other games.
In reality, the text should read something along the lines of:
INVENTORY SLOTS
You have a number of inventory slots equal to your Strength stat or 10, whichever is higher.
Unless noted, all items besides clothing fills one inventory slot. Items that are hard to transport may fill more than one slot.
A single editing pass would clear up potential points of confusion like this.
Level Advancement
Next is a description of how XP and level advancement works:
LEVELING UP
To gain a level, you need to earn your current level x 10 XP.
Once you reach a new level, your total XP resets back to zero.
You get any new title, spells, and talent improvements listed for your level. Your maximum HP increases, and you might also gain a talent roll.
This means that the XP requirement to level increases by an extra 10 XP per level. In reality, this will likely not be felt by players due to the way that the treasure tables work but this will be commented on later in the Gamemaster section.
Functionally, this looks like the image below if total XP is included.
There are issues with levelling, but they are tied to XP awards and treasure values rather than the table itself. Just keep this in mind for later.
Finally, there is a section to generate random characters if needed. It is functional and clear enough.
#3 - Magic
The Magic section contains the rules on:
Casting mechanics
Scrolls and wands
Spell lists
Casting Mechanics
Spellcasters have no concrete limitations on how many spells they can cast, except that they lose daily access to any spells that they have failed to cast.
Every time spellcasters attempt to cast the spell, they have to roll a d20 plus their spellcasting stat. This is Wisdom for Priests, and Intelligence for Wizards. The casting DC to beat is 10 plus the tier of the spell. Therefore, a 1st tier spell has a DC of 11, a 2nd tier spell has a DC of 12 etc.
This does mean that a Priest with a Wisdom of 14 and a modifier of +2 only has a 60% chance of success at casting any 1st level spell. There are talents that make casting successfully more likely, but they are few and far between.
If the player rolls a 1, then it is always classed as a failure. The spell is lost and some negative effect happens. Priests have to atone through sacrificing items, and wizards get to roll on the mishap tables.
The spell lists are reasonably well defined, and the spells themselves are varied.
It is obvious that there has been a lot of work that has gone into the Magic system overall. I appreciate the renaming of spell levels to tiers as it reduces the overhead on the word level.
In my opinion, this is actually one of the best written sections in the entire book, except for some issues with scrolls.
Scrolls and Wands
Activating scrolls and wands acts the same as normal spellcasting with the same difficulty level.
Wands themselves are interesting, and function as so:
WANDS
On a failed casting attempt, the wand stops functioning until you complete a rest.
On a critical failure, the wand breaks and permanently ceases to work. The spellcaster incurs a mishap or penance as usual.
I think I will port some form of this rule to other games to avoid having to count wand charges.
On the other hand, scrolls have this description:
SCROLLS
After a successful or failed casting attempt, the magical writing disappears from the scroll and it ceases to work.
The problem here is that Wizards can add spells permanently using scrolls:
Learning Spells. You can permanently learn a wizard spell from a spell scroll by studying it for a day and succeeding on a DC 15 Intelligence check.
Whether you succeed or fail, you expend the spell scroll.
Spells you learn in this way don't count toward your known spells.
This is a lot more valuable than a potential scroll use.
This reads to me as a potential pit trap for groups, although it may introduce some interesting potential consequences.
In addition, Wizards can, read as written, learn spells of any tier from scrolls and then cast them as much as they like until they hit failure. I could not find any restrictions on casting spells from higher tiers.
#4 - Gameplay
The Gameplay section has a lot of sub-sections. And, to me, most of them read as optional rules rather than being at the heart of gameplay.
To list them all:
Adventuring - Describes what is expected from the gameplay style
Rolling the Dice - Goes into advantage / disadvantage, critical successes and critical failures
Luck Tokens - Describes giving out reroll tokens to players for “exceptional roleplaying, heroism, or just plain coolness”
Using Stats - Gives a list of what to use each stat for, which was already described in the “Characters” section
Making Checks (*) - When to roll, what the standard difficulty checks are and how to resolve contested checks
Time (*) - How time works in the system
Turn Order - How turn order works in the system
Crawling (*) - How exploration works
Resting - How resting works and how often it gets interrupted
Stealth and Surprise (*) - How stealth works and how characters get surprised
Combat - Explanations on combat mechanics
Overland Travel (*) - How overland travel works
Downtime - What downtime activities are available
Carousing - How to get XP from spending gold
Wizards and Thieves - A gambling game added on to the end for some reason
I will go into deeper detail on sections marked with (*).
These are listed in that above order. So luck tokens are before the core crawling section, and overland travel is near the end despite likely being the first section needed.
For the most part, these are competent and standard descriptions of mechanics, and contain no major surprises. The highlighted sections have positive or negative traits and therefore are going to be dipped into in more detail.
Making Checks
The first section that pops up with issues is “Making Checks”. If you recall, the Basics section listed checks as being done in risky situations.
Here that changes to the following description:
The GM asks for a check when the following is true:
• The action has a negative consequence for failure
• The action requires skill
• There is time pressure
Reading this as written, all of the conditions have to apply in order for the GM to ask for a check. This would mean that, as an example, copying a spell scroll into a spell book does not require a skill check as there is no time pressure.
I think, in reality, the intent was for any one of the conditions to apply. This works as an acceptable definition of when checks need to be rolled.
Another problem a paragraph over:
DIFFICULTY CLASS
The four standard DCs represent how difficult an action is.
Easy, DC 9. Leaping a narrow chasm, sneaking up on an inattentive guard.
Normal, DC 12. Kicking open a stuck door, picking a poor lock.
Hard, DC 15. Swimming against a strong current, giving first aid to stop a character from dying.
Extreme, DC 18. Climbing a slippery cliff one-handed, restraining a frenzied lion.
The difficulty classes do not match their descriptions.
A character with a +4 modifier in a relevant stat will fail at a DC 9 difficulty check ~20% of the time. With a +0 modifier, it is a ~40% chance of failure.
The Advantage system can help with this, but a +4 modifier and advantage still has a ~5% of failure. A +0 modifier has a ~15% chance of failure.
And this is against an Easy check.
Time
I have heard that Shadowdark has “real-time torches”.
And it does, until it does not.
Here is the description of the real-time element:
REAL TIME
Time passes in the game world at the same pace it's passing in the real world. One minute or hour of game time is equal to one minute or hour in real time.
This matters for tracking light sources, because most light sources only last for one hour of real time.
If you can't track real time in your game, assume one hour is equal to 10 rounds.
I think this is good advice. Players need to feel time pressure. It is a negative to have players spend twenty minutes arguing and then try to continue on as if no time has passed in game.21
But then the next section along has:
TIME PASSES
Every moment in the game doesn't have to be accounted for in real time.
For example, if the characters want to spend 10 minutes examining a room from top to bottom, the GM and players can agree that time passes.
When time passes, the GM and players move any timers down by that amount.
So time passes according to real-time until it does not.
This is how the vast majority of people treat time in TTRPGs. There is nothing innovative in this, but it was used extensively in the Kickstarter and general marketing.
Functionally, it just acts the exact same as every other time keeping system in every other dungeon crawling RPG.
Turn Order
This section describes initiative and what players and the GM should be doing. The initiative system is a relatively standard d20 plus Dex modifier.
There is also this:
INITIATIVE
Shadowdark RPG is played in turn order right from the start.
At the beginning of the game, the GM establishes initiative, or the order in which the players act. Everyone rolls a d20 and adds their DEX modifier. The GM adds the highest DEX modifier of any monsters, if relevant.
The person who rolled the highest takes the first turn, and the turn order moves clockwise from that individual.
This is what is usually described as “always on initiative”.
There are benefits to this. It can be used to force shy players to actually engage with the game and prevent some players from getting distracted.
I have found that it has a tendency to grind games to a halt, especially if players do get distracted.
It has the same problems that groups tend to face in combat where people do not pay attention and then, on their turn, have to ask a dozen questions to understand what is going on.
Crawling
Crawling is the way that the game system describes exploration. This section begins with a pleasant description of what “Shadowdark” is in setting.
And then there is a description of out-of-combat rounds:
CRAWLING ROUNDS
Characters are in crawling rounds while not in combat. They are exploring, talking, and engaging with the environment.
I think it is bizarre to call it crawling. I understand that it is a nudge towards “dungeon crawling” but it leads to weird sentences like:
ACTIONS
While crawling, characters can take actions that might include:
• Prying a gem from a statue
• Sneaking up on a slumbering manticore
• Swimming to the bottom of an oily pool of water
This section has the distance keywords noted quite clearly:
Distances. Distances are broken up into close (5 feet), near (up to 30 feet), and far (within sight during an encounter or scene).
And then a bit further down there is another section with clunky use of the near keyword, but without bolding for some reason:
REGROUP
During crawling rounds, the GM can ask the players if they want toregroup. To regroup, any PCs within eyesight or earshot of each other who are able to move freely may immediately move within near of each other. Then, initiative continues as normal.
Changed in the current version of the rules to:
REGROUP
During crawling rounds, the GM can allow the players to regroup. PCs within reasonable reach of each other can come together into a marching order. They can also move as a group, taking a round each time. When needed, initiative shifts back to turns.
This is the first section where light is explained in detail:
LIGHT SOURCES
Most light sources last for up to an hour of real time and illuminate a limited area.
Players have two options when lighting more than one light source at a time:
1. The newer light source "rides along" on the current timer.
2. Extinguish all old light sources. Start a new timer with the fresh light source.
The GM can make rulings counter to the above; the purpose is to make tracking light timers simple, not frustrating.
Up until this point, the only lights mentioned in the text were torches which illuminate near the character. After examination, there are lanterns which demonstrate the ridiculousness of the distance keyword system:
Lantern. Casts light up to a double near distance (see Light, pg. 84). Requires oil. Has a shutter to hide the light.
The light source system is otherwise workable. It does not seem any simpler or easier to track than other methods, but it exists in a functional state.
Stealth and Surprise
This section is mostly acceptable.
My main problem is that “surprised” is not indicated as a keyword, and the word is reused several times. See this description:
SURPRISE
A creature who begins its turn undetected has the advantage of surprise. If combat hasn't started yet, those with surprise take one turn before the combat initiative roll (see Combat Initiative! on pg. 88).
A creature has advantage on attack rolls against surprised targets.
As was mentioned earlier, goblins have the following bonus:
Keen Senses. You can’t be surprised.
Therefore, it is not clear what being surprised means.
I see several potential interpretations of the goblin bonus, from weakest to strongest:
Creatures that attack goblins during a surprise round do not get advantage against them.
Goblins can act during surprise rounds.
Parties with goblins can never be surprised by stealthy creatures.
I think the most likely choice is #1, but I know a lot of players that would argue #3. There are a lot of permissive GMs who would also likely allow #3 to avoid disappointing players.
Overland Travel
THIS IS UPDATED IN THE CURRENT VERSION OF THE RULES - THE NEW SYSTEM UTILISES TIME ON THE WAYS TO TRAVEL PORTION AND IS MUCH BETTER - BELOW INCLUDED FOR POSTERITY
Overland Travel has two rules sections that do not meld together.
A description of how overland travel works:
TURNS AND TIME
The game moves in initiative order as normal while traveling overland, and groups often travel in chunks that comprise several hours. Use the Time Passes rules (pg. 82) as needed.
Check for random encounters using the following cadence during overland travel:
Unsafe. Check every 3 hours.
Risky. Check every 2 hours.
Deadly. Check every hour.
And then a description of how far players can travel:22
WAYS TO TRAVEL
PCs can cross a number of 6-mile hexes per day based on travel method. Pass a CON check for each extra hex attempted.
Therefore, the random encounter chance is measured in hours and the distance travelled is measured in hexes per day.
These are two completely separate measures. The ridiculous part is that there was time travelled in an early version of the rules, but then this was changed to this unhelpful system.
I could calculate what I think it is, but it should have been done in the rules.
What I find especially strange is that encounter chance is measured in days and hexes in older versions of Dungeons & Dragons. This is a solved issue.
Secondly, arduous and difficult terrain are not defined anywhere. There is a section later on called “Overland Hex Maps”, but that does not contain links between terrain type and difficulty to traverse.
#5 - Game Master
The Game Master portion of the book has a ridiculously high amount of sub-sections:
Your Role - What the game master should be doing.
The Players - Advice on how to hook players in.
Core Ethos - Advice on how to treat different aspects of the game.
On Balance - When to act against the normal rules, how to treat danger and choices.
Not the Enemy - Advice on the GM versus player balance should be.
Set the Tone - Advice on starting to build a world
Let There Be Darkness (*) - Advice on threatening light sources.
Modes of Play - Optional rules.
Random Encounters (*) - Rules on random encounters.
Traps (*) - Building traps.
Hazards - Environmental dangers in the world.
The Gauntlet - Funnels for 0th level characters in Shadowdark.
Awarding XP (#) - How much XP for any given item.
11 more sections and 20 pages dedicated to a very random selection of tables.
Random Encounter Tables (*) - Random roll tables for encounters.
I will go into deeper detail on sections marked with (*). I will discuss Awarding XP (#) in Treasure.
The vast majority of advice in this section is very, very good. There is clear distillation of knowledge from an experienced and naturally skilled GM in The Players, Core Ethos, On Balance and other sections.
I think all of these sections show that Kelsey understands gaming and what makes players tick.
The Hazards section, as an example, is a reasonable source of things that could block or make more difficult player progress, and a good explanation about why this should be used.
The only real issue with concepts in this section is at the very start:
THE ONLY RULE
The Only Rule is that you make the rules.
What's written in this book is a guide, not a constraint, and none of it takes precedence over your judgment.
If something doesn't work at your table, change it or throw it out and don't look back.
This is one of those pieces of platitude often expressed under the guise of Rule 0. This makes me question why I paid for the book in the first place, if I am expected to act as a game designer.
Ultimately, though, this is an expected piece of advice in rulebooks.
Let There Be Darkness
This section is a distillation of some excellent advice.
To quote:
ATTACK THE LIGHT
A main source of time pressure in Shadowdark RPG is the limited duration of light sources.
Don't let light sources be "set it and forget it." The characters must protect their light sources and keep them from going out.
Alongside this are recommendations on what should be done by enemies to eliminate player light sources, and a table of “Light Mishaps” that could occur during play.
Random Encounters
The random encounters section is acceptable.
Except for the way encounter check frequency is defined:
HOW OFTEN?
The GM checks for random encounters during crawling rounds (not combat) based on the environment's danger level.
Unsafe. Check every 3 rounds.
Risky. Check every 2 rounds.
Deadly. Check every round.
To check for a random encounter, the GM rolls 1d6. An encounter occurs on a result of 1.
The danger level of the environment is listed as “Unsafe”, “Risky” or “Deadly”. The danger level of an area is generated through random tables later on in the “Shadowdark Maps” and “Overland Hex Maps” sections.
These danger levels are not defined in any formal manner.
I feel this is an issue for two reasons:
The check rate does not actually correspond to how dangerous an area is. It corresponds to how busy an area is. A lot of encounters are neutral or positive.
There is no clear way to determine how to implement designed dungeons, or translate dungeons from previous editions into Shadowdark.
If I tried to bring a classic Dungeons & Dragons module like S4 - The Lost Caverns of Tsojcanth (1982) into Shadowdark then I have to guess what danger level should be applied to the areas of the module.
Similarly, if I design a dungeon, instead of randomly rolling one, then I have to guess what Shadowdark means by danger level.
These are negatives that could easily have been avoided either by aligning further with older Dungeons & Dragons or by keywording these clearly.
Traps
Traps are detected by searching a “specific area or object”. Players have to describe how their characters disable them via a “reasonable method”.
I hate this method. It leads to what is often described as “pixel bitching”, as neither specific nor reasonable are defined.
Some groups enjoy it, so it is a matter of personal taste. There are no alternative methods offered.
Random Encounter Tables
The random encounter tables are strange:
To me, this reads as an encounter table for a specific module or dungeon, not a generic table to be used to generate adventures or worlds.
There are no level ranges for generating monsters and there is no clear indicator of how dangerous any of these tables are.
#6 - Monsters
The Monsters section has descriptions of what monsters have as attributes, a monster generator table series and some advice on how to design monsters.
It has a good list of monsters. There are over two hundred monsters listed. That is a very nice bestiary.
Monster Creation
The Monster creation system is reasonable and I appreciate it.
But it really highlights is that at no point is it clear whether the roll tables are linked across or not:
If you are wondering why arachnids would be weak to vanity, I did the same. I believe a d20 is meant to be rolled for each of these sections, but this is not mentioned anywhere in the book.
This is followed with a recommendation to be aware of the role that monsters take in the game:
COMBAT ROLE
Monsters typically fall into one of the below combat roles.
Mook. Low damage, HP, AC, and attack bonus. Swarmy.
Soldier. Mid-range damage, HP, AC, and attack bonus.
Striker. High damage and attack bonus. Low HP and AC. Often stealthy and skillful.
Tank. Mid-range damage and attack bonus. High HP and AC.
Controller. Environment-altering talents. Mid-range damage and attack bonus. Low HP and AC.
Legendary. High HP, AC, damage, and/or attack bonus.
This is good advice, if a strange fit for the book. These classifications are lifted wholesale from 4th edition Dungeons & Dragons which is a very mechanised and gameplay-oriented system.23
There is a system a bit further on that has a method for “designing combats of average difficulty”.
Some people would say this is not an OSR method of design, as it conflicts with the idea of combat as war.24
I disagree. It is important to know how much something is likely to hurt players so that you can keep them to the grindstone and make them fight dirty.
Monster Stat Blocks
The below is a standard Shadowdark monster stat block.
It is very pleasantly terse, tight enough to fit on a few lines in an adventure but covers off nearly all situations.
However, there is something missing. I had to go back and check to make sure I had not missed it.
There is no encounter amount listed.
Compare that with the same monster in Moldvay’s Basic Dungeons & Dragons (1983) rules:
The No. Appearing in the top right refers to the number of monsters that make up an encounter. The numbers outside the brackets is how many can be encountered in random encounters in a dungeon, whereas the numbers inside brackets is how many can be encountered in lairs or while wandering in the wilderness.
I think it is important to have this in the stat block so a GM can roll up an encounter with a specific monster during play. A GM could reverse engineer it from the monster design advice, but that would take up unnecessary time during the session.
#7 - Treasure
I think the treasure and treasure to XP systems are the worst part of the book. They are just both overly complicated and still, somehow, lack clear definition.
Generating Treasure
The treasure generation system is just strange:
Monsters give treasure according to their level. A monster has a treasure level equal to its monster level.
“Wandering monsters”25 have a 50% chance of having treasure.
Treasure troves give treasure according to who finds them. A treasure chest has a treasure level equal to the level of the player who finds it.
The treasure level determines which d100 table gets rolled on.
Only one item appears to be generated each time. Some specific monsters, such as dragons, have more than one item.
Every item has a monetary value assigned to it.
In a simple flowchart, it looks like this:
The problem with this system is that it is completely disconnected from how OSR games are played. Players no longer feel clever for avoiding monsters and getting to a treasure trove, because it will give level equivalent treasure.
I think this is an issue with the Random Encounters system, as the lack of dungeon levels means that there is nothing to tie difficulty of encounters to it.
This paragraph also indicates a problem with the system:
DEALING WITH LOOT
Treasure typically occupies one gear slot (coins occupy one slot per 100), but the GM might determine otherwise.
The items really should have had their encumbrance listed against them, because leaving it to GMs is the height of laziness.
Some minor examples of items that would definitely take more than one slot:
Matched trio of warhammers (10 gp each)
Two intact griffon eggs (40 gp each)
Taxidermied smilodon (30 gp)
I think the treasure tables themselves are actually nicely evocative, but they seem likely to get old very fast.
They also do not include any of the items that monsters will be using, and it is not clear to me if monsters should be dropping the items that they are wearing. If they do, it is likely to lead into problems with the XP system.
Awarding XP
Characters can gain XP from treasure.
There are four categories of treasure, with corresponding XP values:
• Poor (0 XP): Mundane, low value, ordinary, unexciting.
• Normal (1 XP): Good value, worth protecting, useful.
• Fabulous (3 XP): Incredible, prized, well-guarded.
• Legendary (10 XP): Mythic, unique, quest-worthy.
This is the table to allow the GM to understand what sort of bounds these fall into:
This table is probably going to be the biggest source of arguments at any Shadowdark gaming table. This section as a whole suggests a complete lack of playtesting with any sort of adversarial player.
Questions that immediately pop into my mind:
“How many coins are in a bag of gold?”
“Is fifty bags of silver still a poor quality treasure?”
“If I pick up two hundred silver worth of items across an entire dungeon, does it get upgraded to normal quality treasure?”
“What happens if me and a player disagree on treasure quality? Does the resentment just fester?”
The worst part is that we know items have a monetary value. Why is that monetary value not actually tied to their quality? Only one thing on this list does not have a monetary value and that is a wish spell.
Since we know the values on this list, we can reverse engineer monetary values for treasure quality:
Poor (0 XP): 0 - 10 gp
Normal (1 XP): 11 - 199 gp
Fabulous (3 XP): 200 - 999 gp
Legendary (10 XP): 1,000+ gp
This gives clear and consistent bounds for players and GMs to understand the rules of the world. You can add on more rules. You can say that GMs can award XP for particularly interesting stunts, but these basic rules are crucial.
This is made worse by a few paragraphs way back in the Characters section:
EXPERIENCE POINTS
…
The GM can award XP right away or at the end of each session.
LEVELING UP
…
Once you reach a new level, your total XP resets back to zero.
You get any new title, spells, and talent improvements listed for your level. Your maximum HP increases, and you might also gain a talent roll.
Offering the option to award XP right away or at the end of each session is a problem because it means that the writer cannot know which one a group is going to choose.
So now a GM could have a situation where a player can run into a room, grab the legendary Rod of Mount Doom and level up every member of his group. Or, maybe it cannot because of the option offered here.
Again, frustratingly, this is a solved issue. Old school games use XP based off the gold value of items. Old school games give XP when you return to civilisation.26
The gameplay loop is already resolved, there is no need for the confusion.
Final Word
I wanted to like Shadowdark. Kelsey Dionne seems like a genuinely good person, and deserves success.
In reality, my biggest point of frustration is that you can tell some things about Kelsey from reading the book:
She is excellent at descriptive writing and setting tone.
She is certainly a good GM, and the advice given in the Game Master section shows that.
She definitely understands the OSR ethos and gives good advice on adhering to it.
However, the rules seem inconsistent, and the natural language in which they are written is somehow made worse by randomly inserted keywords.
This is exacerbated by the fact that this is not a rules-light release. Shadowdark is over 300 pages long and around 180 pages of that is just rules. Moldvay Basic/Expert Dungeons & Dragons is 130 pages including rules and monster statistics.
I am giving Shadowdark a 6/10.
Shadowdark has definitely had a proof-reader because there are no obvious errors in grammar or spelling, but the lack of a listed editor suggests that the book has not had the editing pass it desperately needs.
There are good ideas throughout, but they never come to a cohesive whole. Experienced GMs can probably carve out components to be used for their own houserules.
In my opinion, this is a very solid first ruleset release from Kelsey. She has had experience in writing adventures for 5th edition, but they require completely different styles of writing.
Partway through writing this, I realised that I would have quite liked a book of Kelsey’s DMing advice and managing game feel. I think she would have done an absolutely brilliant job of that.
On the other hand, she made ten times my yearly salary in a month with Shadowdark, so maybe she should not take my advice.
The trade name and marketing name split is likely due to the use of the term Shadowdark in the Forgotten Realms.
The campaign goal was $10,000. If the backer distributions held down to that level, that would have worked out to around 100 books needing to be printed. Since the campaign achieved over a million, there was a need to print 10,000 books instead.
Brandish is the head of Runehammer Games and has assisted in numerous other titles with a similar clean trade dress.
I feel a physical copy might make the text too large, but that might be balanced by the open book format.
I know this blog is left-aligned. I think I would have to change the css to get it to be justified.
In contrast, I have a disability that makes it difficult to read left-aligned text. Thus is life.
This is called foreshadowing.
BX has rounds per turn instead.
Notably not at or below zero HP.
Risky is not defined for checks, but is used as a keyword for something unrelated later on.
Here is a video demonstrating Hugh Dennis using the semantic ambiguity in a similar sentence to win on a game show.
In an early version of the rules, they were explicitly listed as Races.
People will read these percentages and think this is not likely. In a group of four players, this means an ~80% chance that at least one person will reroll once, a ~35% chance that at least one person will reroll twice, and around a ~12% chance that at least one person will have to reroll three times.
Sometimes these bonuses are referenced as ancestral talents, but that is deep in the book.
Fighters and Thieves both have “+1 to melee and ranged attacks” as one of their most common class talents.
I think Kelsey Dionne has a good skill at evocation of tone and world building and that does come up throughout the book.
Time can be a bit flexible in play. I tend to tell my players that ten minutes spent arguing at “campaign level” (sitting in a tavern in a city) is a day in-game. It makes them move faster.
Some people will already see the issue here.
This is a random footnote to check if you are reading this. If you are planning to send me hate mail, please include the phrase “Klaatu barada nikto” so I know you have actually read this post.
From the OSR Principia Apocryphya.
Another undefined term but it seems to mean any monster found on the encounter tables.
Civilisation can be frustratingly ambiguously defined.
Incredibly detailed review, thank you for going through the system!
Frankly to me, the game seems closer to a 3/10. I guess stuff like "nice layout" and "great art" and "good DMing advice" are wonderful, but the rules seem absolutely terrible and the game feels, based on your review of it, like it's borderline unplayable.
Also, players can wander around a lvl 1 dungeon and still get amazing loot? What were they thinking! It seems like the incentive is for the high level player to always be the one searching for treasure, which is pretty bizarre.
Great review. Great read. I was quite disappointed after reading Shadowdark, too. All the hype, all the money ended up in a good but far from perfect or special product. It shows that marketing is the most important aspect of kickstarter nowadays. Even more so than in the passed years.