Monthly Archives: April 2013

Simulacrum strengths

IMG_4416 OSR games

On the OSR Google Plus community, Mike M. asked: what is your favorite OSR game?

Personally, I think a number of the simulacrum games have strengths, and while I don’t exactly have a favorite, all of those mentioned below have influenced and educated me. Here then, in alphabetical order, are what I consider the selling points of the various retro and neo clones. I think this is a good answer to those that see the clones as mere cut and paste jobs, adding nothing to the games that inspired them.

ACKS (neo-clone, B/X with a focus on domain play) has my favorite take on simple balanced old school classes, working within the B/X level range. Demi-humans have some choice as to class, but still feel distinct. There are good rules for ritual magic, constructs, armies, and other high level exploits. There is also a proficiency system (encompasses skills and feats) which I find unnecessary. There is no free version.

Crypts & Things (neo-clone, a Swords & Wizardry variant for swords & sorcery) has a “colors of magic” system which combines all of the cleric and magic-user spells into one class, the magician. Using black magic can cause loss of sanity (based on wisdom) or corruption. Many nice flavor touches (casting invisibility involves temporarily existing on a dimension called “the shroud” and can potentially summon otherworldly horrors, for example). Most (but not all) of the good parts can also be found in Akrasia’s free house rules document, if you don’t care about professional layout or having an all-in-one game.

DCC (neo-clone, B/X crossed with 3E and a heavy dose of Appendix N) probably has my favorite magic system (unpredictable, lots of weird effects, still recognizable compared to other classic and simulacrum games). Illustrated by the likes of Russ Nicholson, Peter Mullen, and Erol Otus (though I love much of the art, I could do without the 70s fashion). The beta version is available for free.

Labyrinth Lord (retro-clone, B/X with a hint of AD&D via the AEC) is the lingua Franca of FLAILSNAILS, but it incorporates many third edition SRD-isms which I find distracting (too many armor types, 20 levels, first level clerics get spells, etc). The Advanced Edition Companion is an excellent halfway point between B/X and AD&D. It is illustrated by Steve Zieser and has the most consistent style throughout (both positives, to me). Free unillustrated versions of the core book and the AEC are available, but then you don’t get to see the Zieser art.

Lamentations of the Flame Princess (neo-clone, B/X through a “weird” lens) has excellent referee advice, a decent encumbrance system, and weird-themed spells (including the first level summon spell). A free unillustrated version of the Rules & Magic book is available, though that doesn’t contain the referee advice, which is the best part. LotFP also makes the nicest physical books, if you care about that.

Swords & Wizardry (retro-clone, OD&D) has numerous very useful meta-discussions about the rules, how they work, and how you might want to change them. Very much in the spirit of the original “do your own imagining.” Free, illustrated versions of WhiteBox and Core are available. If I was going to start another game right now, there would be a very good chance that I would base it on the S&W WhiteBox chassis. WhiteBox approximates the 3 LBBs, Core approximates the 3 LBBs + Supplement I: Greyhawk, and Complete approximates the 3 LBBs + all the supplements.


  • B/X: the Basic and Expert Dungeons & Dragons rules, by Moldvay and Cook/Marsh. Characterized by the 4 classic human classes (cleric, fighter, magic-user, thief) and 3 demi-human classes (dwarf, elf, halfling). Covers levels 1-14 (though demi-humans have level limits). If you were to only acquire one D&D or simulacrum game, this is it.
  • OD&D: original D&D from 1974, 3 little brown books with some supplements. The orignal boxed set (the 3 LBBs) only supported 3 classes: cleric, fighting man, and magic-user. Noted for opaque, ambiguous, and suggestive language. If you use all the supplements, the complexity approaches AD&D levels (so don’t do that).
  • Retro-clone: game designed primarily to mimic the play experience of an earlier game.
  • Neo-clone: game that uses an older game as a stepping off point for to express a (generally) more focused vision.

Note, this list is intentionally meant to not be comprehensive. It is an editorial selection of available games based on my taste and what I am familiar with. All of these games are available in hard copy and/or full digital versions, but for the sake of parsimony I only linked to the free PDFs.

Honorable mention goes to Delving Deeper and Astonishing Swordsmen & Sorcerers of Hyperboria. Delving Deeper (retro-clone of OD&D; free version) for being, in some ways, an even closer clone of the original game than Swords & Wizardry (though without as much useful explanatory text). AS&SH (neo-clone; AD&D influenced swords & sorcery) for having art like this by Ian Baggley (though I haven’t had a chance to look closely at the rules). I don’t have much experience with OSR sci-fi games, but I’ve heard good things about Stars Without Number and Other Dust.

2d6 casting again

Nicholas Roerich - Snakes facing

Nicholas Roerich – Snakes facing (source)

On monday, I got to test out the petition system for cleric magic that uses a 2d6 casting roll. Here are my findings. I think they apply to other 2d6 casting systems that I have considered in the past as well.

Overall, I was pleased, but I would like to simplify the presentation somehow (the fourfold categorization did not seem immediately obvious to the players). I think there also needs to be some sort of exhaustion mechanic built in. There were no problems during the session exactly, but magic did feel a bit to accessible, especially compared to the party magic-user. Thus, I think I’m going to modify the 2 and 3-4-5 ranges to apply a cumulative penalty. So:

Spurned (further attempting this petition is at -1)

Becomes something like:

Spurned (failure, cumulative -1)

This is similar to a previous idea I had for accumulating arcane stress. It’s also related to this other recent post about another cleric magic system, which allocated “disfavor” points for successfully casting spells. I think the arcane stress post had the right of it by only causing cumulative penalties on lower rolls, as disfavor arising from success seems slightly strange.

Here is an adjusted cleric magic roll:

Petition Roll
2d6 Result
2 or less Abandoned (specific petition unavailable, any abjuration ends, cumulative -1)
3, 4, 5 Spurned (failure, cumulative -1)
6, 7, 8 Ignored (failure, may try again next turn)
9, 10, 11 Answered (standard success)
12 or more Rewarded (double effect, demons or undead destroyed, etc)

And a magic-user version:

Sorcery Roll
2d6 Result
2 or less Catastrophe (chaos surge/mutation/backfire, spell lost, cumulative -1)
3, 4, 5 Miscast (failure, chaos leak, cumulative -1)
6, 7, 8 Delayed (goes off at the end of all actions, may be interrupted)
9, 10, 11 Success
12 or more Puissant success (extended duration, full damage, or something similar)

Cumulative penalties go away and spells may be re-prepared after characters return to civilization and rest for a night, along with appropriate prayer or study.

The differences between cleric petitions and sorcery are as follows. Cleric magic need not be prepared, but is limited to the powers granted by a particular order or patron. It also is more ritualistic, and with the exception of a few limited combat effects (such as turn undead or hold person) requires at least an exploration turn (and often a full day) to attempt. Sorcery, on the other hand, requires preparation, but the set of effects to choose from is limited only by spells known. Sorcery is also more directly potent and more dangerous (potentially causing chaos leaks, mutations, backfires, and all kinds of nastiness). Both kinds of magic become harder to use as failures accumulate, which is important for the resource management aspect of game play.

Solipsistic hexes

Landscape by Nicholas Roerich

Landscape by Nicholas Roerich (source)

Starting from the idea of distance may not be the most productive way to approach either running or mapping the wilderness. This is counterintuitive, because measurement and mapping are so tightly linked conceptually. However, a graph of locations with adjacency (sometimes called a point-crawl recently) seems like too much abstraction. As an attempt to navigate between these two extremes, consider the following system, which I have been using in my Vaults of Pahvelorn game.

Hexes don’t have any determinate size at all. They are abstractions built around three things: sites, travel time, and landmarks. Sites are the locations (towns, dungeons, towers, ancient battlefields, etc) within a hex that can be visited. Sites are either obvious or hidden. One day of travel allows visiting any obvious site within the current hex or within an adjacent hex (this is “moving through hexes” mode). A day may also be spent to search the current hex (which provides a chance of finding a hidden site). Or, to rephrase it in more game-oriented terms, players get one “move” per wilderness turn, which can either be moving to an obvious location within one step or searching the current hex for hidden locations.

Travel time is probably the most controversial aspect of this scheme. One day per hex, irrespective of anything else. Travelling to an adjacent hex allows characters to interact with any of the obvious features of that hex (such as stopping in a town). Exploring a hex provides a 1 in 6 chance to find one of the hidden sites. Any site found is determined randomly, unless the characters are looking for something specific, in which case chance might be weighted in that direction based on if the characters know something about the location they are looking for (“the tomb is by a stream”). A hidden site is treated as obvious if a knowledgeable guide or accurate map is available. The relative locations of sites within a hex are usually not important, and are determined randomly or arbitrarily as needed.

Landmarks, in addition to large obvious features of the current hex, include large obvious features of adjacent hexes. This provides players with information so that they can make meaningful choices about where to go. Most of the time, characters should be able to tell the basic terrain type of all adjacent hexes, though occasionally local terrain will prevent this (such as wandering at the bottom of canyons, or journeying through a very dense forest). This should be clear by context, and limitations are often easy to overcome (such as by climbing a tree).

Each day spent in the wilderness necessitates a random encounter check, as does each night. This is a 1 in 6 chance, but can be adjusted per-hex (based on terrain type or general danger level). It is perfectly functional to stick with the 1 in 6 chance in general, for simplicity’s sake. If an encounter is indicated, I sometimes roll another d6 to see if more than one encounter might occur (a 6 on the second die) or if the encounter will involve more than one NPC group (a 1 on the second die). In the second case, “more than one NPC group” means that the PCs encounter two other groups that are already engaged in an encounter themselves. The exact probabilities for the rarer occurrences are not important as long as they are impartially determined and remain uncommon.

Exploring off the beaten path carries with it the risk of getting lost. There is no chance of getting lost when following a known route, such as a road, but in other cases the chance is 1 in 6 (or greater, of course, depending on the situation and terrain). In game terms, getting lost means wandering accidentally into a hex adjacent to the one intended (determine which randomly). This can happen either during movement toward a known site (if a path is not followed) or during searching for hidden sites.

By implication, the “real” size (whatever that might mean) of a predominantly mountain hex is smaller that the size of a plains hex (because you can travel much farther on plains than on mountains). What this does is pull the wilderness into a loose mesh similar to a point-crawl, but with more enforced structure (as there will always be six adjacent nodes at any given location).

Just like in the dungeon, the default rate of travel assumes caution, resting occasionally, and so forth. Journeys are purposeful but not forced marches. This mode engages all the standard rules (mounts not dying on you, standard getting lost chances, standard encounter chances, standard surprise chances, standard encounter distances). If a group wants to throw caution to the wind and make like a bat out of hell to their destination, more than one hex may be traversed in a single day of travel, but chances of mishap are be greater. Roll or pick any number of possibilities from the following list:

  1. One encounter check (with increased chance) per hex traversed
  2. Mounts must save or die when destination is reached
  3. Increased chance of getting lost (if appropriate)
  4. Guaranteed attention from bandits (haste implies value)
  5. Guaranteed pursuit from origin location (haste attracts attention)
  6. Force retainer morale check (“I didn’t sign up for this garbage”)
  7. Otherwise obvious landmarks or sites go unnoticed
  8. Increased chance of being surprised during encounters
  9. Decreased encounter distance
  10. Penalty to encounter reaction rolls

Alex S. uses a similar one day per hex method, and his post helped lead me to my current method, though I do not expose hexes directly to players.

Current hexcrawl procedure:

  1. Roll for weather (2d6 reaction roll with cosmos)
  2. Move or search?
  3. Day encounter check
  4. Lost?
  5. Describe travel, note obvious sites and landmarks
  6. Resolve any day encounters
  7. Camp procedures? (establishing “default” procedures is reasonable)
  8. Mark off rations (I always forget to do this — bad referee!)
  9. Night encounter check
  10. Resolve nocturnal encounters


Perseus fleeing the gorgons

Perseus fleeing the gorgons (source)

How do you handle pursuit? Most of the official answers are either tedious or uninteresting (such as comparing movement rates directly). It seems like situations that are both uncertain and dangerous (like combat, or running away from a giant tentacle horror) deserve to incorporate some randomness. Knowing that your “movement 6” plate armored fighter will never be able to outrun anything is boring, and also causes perverse decision-making (any kind of certainty is rarely good for maintaining tension).

Most proposals I have read on the web to address this are a bit too complicated. I like this one by Roger, and also have thought about converting the movement tiers to d6 rolls (so movement 6 would end up being 2d6, which has an expected value of 7). But then his system gets into obstacle penalties and lines of sight, and I end up feeling lost. Or this one from Talysman, which breaks the chase down into segments, and modifies the standard movement rates by situation rolls in order to introduce some variation. I like both of these systems, actually, but perhaps a compromise between multi-turn chases and just comparing static movement rates would work best at the table.

Some form of direct roll-off would probably work best. Those fleeing roll some dice based on their movement, those pursuing roll some dice based on their movement, and the higher total wins. Easy to remember, obstacles and hindrances can be handled ad-hoc. Okay, so which dice then? The spread between 1d6 for movement 3 and 4d6 for movement 12 is just too much. I like there to be some effect of armor or encumbrance on chases, but those numbers just don’t feel natural, either from the perspective of what will make for interesting tension in the game or from reasoning about how armor “should” affect movement.

Perhaps we can borrow the d6 “dice chain” from OD&D? That is, 1d6, 1d6+1, 2d6, 2d6+1, and so forth. Then reorganize the chain slightly, so that the three different armor types don’t differ by that much, but still do affect the outcome meaningfully.

Encumbrance Traditional Dice Chain
unarmored 12 2d6 +2
leather armored 9 2d6 +1
metal armored 6 2d6
metal armored
and carrying treasure
3 1d6

Traditional encumbrance and movement numbers are taken from the Moldvay Basic rules, page B20 (though Moldvay includes equivalent coin values as well, for bean counters). When you reorganize the chain though, it starts to feel less intuitive, so while I think this would work, I’m not totally satisfied. Also, I would probably modify Moldvay’s categories so that the progression was unarmored/light, medium, heavy (rather than unarmored, light, medium/heavy).

Here is another proposal. Unencumbered movement is 3d6 (expected value: 10.5). Subtract armor type (for example, heavy armor is -3) and any other encumbrance penalties. This makes the expected values of the movement by different armor types 10.5 (unencumbered), 9.5 (light/leather), 8.5 (medium/chain), 7.5 (heavy/plate). This seems to be a more reasonable spread than 12, 9, 6. Fast monsters would roll 4d6 for pursuit, slow monsters would roll 2d6. Very easy to remember, and supports the goal of adding tension to chases without complexity.

Using an entirely new system for resolving chases does give me pause, but none of the existing systems seems to work well. 3d6 gives a good distribution, and also plugs into numbers from other parts of the game (armor levels as 1 through 3, encumbrance penalties) in a way that makes sense, so I think the new die roll is justified. And since when has D&D ever been afraid of multiple resolution systems, anyways?

This movement scheme could also be applied to dungeon exploration mode. Traditionally, exploration rates are static, and based upon encumbrance. Instead, what if players rolled to move during the exploration context as well? The random encounter check could even be incorporated into this roll. One of the movement dice could be a different color. I don’t think this would work well in a videoconference game, but it might be good for an in-person session.

Postscript: Potential scheme for the 2d6 fantasy game: 2d6 +strength -armor (strength is probably a somewhat reasonable proxy for speed).

Viole Falushe on refereeing

The Palace of Love

The Palace of Love (source)

Viole Falushe, the Demon Prince, speaks:

Artists before me have conveyed their assertions by abstract symbology; the spectators or audience has always been passive. I use a more poignant symbology, essentially abstract but palpable, visible and audible–in short a symbology of events and environments. There are no spectators, no audience, no passivity. There are only participants.

— Jack Vance, The Palace of Love (1967)

Chaos Engines

When using a magic item, roll 2d6, add one half relevant class level (round up) and any item level (use enchantment bonus if it has one, or make a ruling). Then consult the table below.

Enchanted Device (2d6)
2d6 Result
2 or less Destroyed (item ruined, triggers chaos surge)
3, 4, 5 Warped (does not function, chaotic energies twist device)
6, 7, 8 Undermined (functions, but not quite as expected, device left unchanged)
9, 10, 11 Consistent (device functions without surprises)
12 or more Augmented (functions, chaos embues the device with a new permanent ability)

For example, a magic-user deploying a staff would get the class bonus, but a fighter would not. natural 2 and natural 12 should always override any modifiers, just like how nat 1 and nat 20 are often interpreted. If you wanted it to be more swingy, you could only apply item level (and ignore character level). That would also decrease the load on referee creativity (by not requiring the invention of new permanent abilities quite as frequently).

Perhaps especially useful for handling magic items in FLAILSNAILS games.

A great question from Eric B. on Google Plus: Would there be ITEMS OF CHAOS with negative item levels?

To which I responded: yeah, that sounds like a good way to operationalize curses. “Most of the time this wand will just make problems for you, but once in a blue moon it will level the evil overlord’s fortress.”

Monster XP as treasure

Most versions of D&D provide XP values for monsters, usually awarded for defeating the monsters in combat. The Third Edition (and Pathfinder) manuals also give explicit prices in gold pieces for every enchanted item. Here is an idea for modifying how advancement is handled using those numbers. XP value specified per monster is converted into GP. This is the value of the monster’s treasure (and equipment). XP is given for GP spent, as I have been doing in my Vaults of Pahvelorn game. Additionally, magic items can be purchased as is commonly expected (I think) in many Third Edition games.

In a game like this, I still wouldn’t make magic item shops common. Instead, I would put together some sort of system to rate sellers of enchanted items (cunning folk, wizards, specialist merchants, demonic patrons, and so forth). The rating could either be a flat GP threshold (this NPC can sell magic items of up to 1000 GP value) or it could be something like an inventory saving throw (this wizard has a stock number of 11 — roll 1d20 against that target number, modified by desired magic item level). This is somewhat similar to an idea I had for rating sages, which I should probably also write a post about. This creates a motivation to quest for NPCs that are better able to provide powerful items. Particular famous items of power might be the purview of specific individuals.

The entire advancement system could even be replaced by item acquisition, for a slightly lower power game. So, for example, fighters wouldn’t ever get attack bonuses, they would just get more powerful weapons. Wizards wouldn’t prepare spells, they would buy (or create) scrolls and other enchanted devices.

The interesting aspect of this proposal is that it maintains most of the expectations of the 3E/Pathfinder advancement system (assuming that standard class progression is also maintained) while only modifying the specific action incentive. The standard benefits of XP for GP are preserved (rewarding clever planning as opposed to straightforward combat). Obviously, this system would result in a setting with more common magic, but that’s not necessarily a bad thing.


If you are laying out text with narrow columns, using full justification without hyphenation is unlikely to result in an attractive document. Very narrow columns are most commonly the result of wrapping text around images. Consider the following unhyphenated and fully justified example:

Unhyphenated Fully Justified Text

Unhyphenated Fully Justified Text

Pretty ugly, right?

Compare that to a hyphenated version:

Hyphenated Fully Justified Text

Hyphenated Fully Justified Text

The hyphenated example above is still not perfect, but I suspect most readers will agree with me that it looks better than the unhyphenated example. These cases were created with a standard word processor (Mac Pages) by a person who is not a layout expert (me). A professional using real typesetting or layout software could undoubtedly do much better.

Many people prefer fully justified text, because it provides a veneer of profesionalism (since most books use fully justified text). However, fully justified is not the only option. Ragged text is a totally legitimate alternative. Ragged vs. hyphenated-and-fully-justified is a taste thing, but full justification without hyphenation (especially in a context of narrow columns, as shown here) should probably just be avoided.

Given that I have found myself giving this feedback several times now, I figured it merited a page that I could just reference. (And now you can too, if you come across egregious layouts and agree with my document aesthetics.)

Ragged right example

Ragged Right Example

Example images and text taken from the Wikipedia page on Necromancy: