Dice methods

http://www.tcp-ip.or.jp/~panurge/imaz.htm

It's a good review for someone who hasn't played the game -- I've noted that actual experience with the system creates total converts in a way that simply reading the rules won't do.

Between Imazine and Serendipity's Circle, Sorc is starting to take on a presence in the RPGs-for-grownups small press. **********

RON EDWARDS: One thing that surprises me in the several reviews of Sorcerer out there is that no one seems to pick up on the unique dice system. Or "unique" in the RPG sense, that is, a unique combination of effects.

It's not the only dice-pool system (i.e., the changing variable is the number of dice): that would include among others Over the Edge, Morpheus, Extreme Vengeance, the White Wolf system, and even all the way back in time to Tunnels & Trolls. Like these games (except WW), the value rolled is only relevant to the other roll.

Nor is it the only system that uses highest-die-only as the value: the Dream Pod 9 games and Deadlands also do this. I don't count the WW games or Shadowrun in this category because they concern themselves with the NUMBER of dice above a certain value.

But it's the only dice-pool system with no target numbers (unlike WW, DP9, Deadlands) AND highest-die reading method (unlike OtE, Morpheus, EV). For those of you who didn't know this, Sorcerer is simply the Prince Valiant system using dice instead of coins.

My claim is that the Sorc dice method has ALL the advantages of these other systems and none of the disadvantages: extra time to compare target values, extra numbers to add and subtract once the dice are rolled, extra time to total up the dice to determine the value. (Dice are faster than coins, too.) Like many Champions veterans, I can total up a handful (or double handful) of rolled d6's in seconds -- but after playing Sorc, I have no desire to do this ever again.

So why is this being ignored in reviews? Could it be that reviewers are just not interested in dice/system mechanics? *************

ED: In 1997, I was developing my own game system, when I came across Sorcerer. That is the day I shelved my ideas on dice and conflict resolution... they were 95% identicle to Ron's. Besides, he was in the final stages and I was just beginning. So, I totally agree that the dice system is the best there is.... for such a game.

That said, on to the Q: at hand.

>So why is this being ignored in reviews? Could it be that reviewers are >just not interested in dice/system mechanics?

One reason is that half the 'reviewers' never even play the game... they just read it. So, they have no playtest experience with it, and so tend to gloss over the system and focus on the setting.... Which is one reason Sorcerer may get short shift when it comes to reviews. ***********

RON EDWARDS: This ties into a larger issue which has been irking me for a number of years now. My observations of most so-called roleplayers is that they rarely actually PLAY. In fact, the more GURPS supplements, the more Shadowrun supplements, the more Vampire supplements they buy and own, the less they play!

So when I think of the market for Sorcerer (or Zero, or Castle Falkenstein, or Extreme Vengeance, or any of the other really innovative games), how big is it, really? I see it as a fairly small subset of the gaming customer -- or more accurately, I see that bulk of gaming customers as representing mostly BOOK BUYERS and not really roleplayers at all. So of course, little old Sorc and similar games aren't going to be their thing. They WANT all those coffee table books.

So the RPG author has a choice. (1) Go ahead and work with that market, much to the delight of retailers and distributors, and design one's product to be as glitzy and expensive as possible. Rely on advertising to create a demand ("this game is the cool, in thing"), then hope that expenses won't outweigh profits, then hang on until you can get out and laugh all the way to the bank. Funny ... everyone who's tried this has gone bankrupt!!

(2) Grit one's teeth and target that minority roleplayer group as overtly as possible -- state exactly how this game is worth their time, and (opposite from tactic #1) make it as efficient and usable as possible. This is what Zero and Extreme Vengeance tried to do, and a few others -- and interestingly, they failed IN THE STORES. I am convinced this is because this tactic runs directly counter to the economic interests of retailers and distributors, despite its potential attractiveness to that small sub-set of customers.

(3) A hybrid of #1 and #2, which is what Feng Shui, Over the Edge, and Castle Falkenstein tried to do. Look and cost like a big expensive many-supplement game, present an innovative and interesting system. They failed miserably, financially. I think this shows that the two types of customer are rarely compatible.

Back to reviews: one of the most interesting things about reviews of all kinds is that they pre-suppose a shared set of values between reviewer and reader. The reader relies on the idea that the reviewer is "one of his kind" and looks at the book or whatever from an appropriate perspective. Given these two approaches to gaming, though, I think that reviewers can end up pretty confused. What is the appropriate look & feel to a roleplaying game? NO ONE KNOWS. I think this leads to incoherence across many reviews: to some, L5R is a great game, mainly because it's pretty and (to use the jargon) well-supported; to others, L5R is a stinking blight on RPG-dom because it's a gaudy over-produced monster that could have cost a third of its retail price; to still others, it's got a character creation system, a monster list, a spell list, and a weaopns list, so it's fine, right?

I'm starting to spray the screen with indignant spit now, so it's time to stop. But please, comments, anyone? **********

RON EDWARDS: To continue my musings on dice and RPGs ...

Taking a look at two variables: type of curve and determining effect. That is, take a look at how you resolve an action in one of your RPGs. We'll take Champions just 'cause I played it for so long. In Champs, to resolve (say) whether Blastopore zaps Archenteron, one rolls 3d6 against a target value set (ultimately) by the concerned characters' Dexterities. Equal to or lower? Got'im! Then one rolls an entirely different set of dice, completely independently of the first roll, to determine the damage (effect). So Champs gets categorized as BELL curve (3d6, total'em) and SEPARATE effect (new roll to determine damage).

Then there's Zero: 2d6, multiply them together, and compare them to the character's Focus (a number from 2 to 9 on the character's sheet, usually 4-6). Equal to or higher, you got it. The math is a little tricky, but it turns out to be a bell curve of possible values. However, to determine the effect, just compare the product (e.g. you rolled a 3 and a 5, so that's 15) to the Focus. Beating Focus by more points means an increased effect, in units of 10. In other words, the effect is determined straight out of your original roll. So Zero gets categorized as BELL curve and INCORPORATED effect.

Keep goin'. Good old 1st edition AD&D ... roll that d20 and compare it to the weapon/level/armor class matrix. E.g. a 3rd level thief with a shortsword stabs a bugbear with armor class 6 ... damned if I can remember what value that is to hit, but it's probably around 14. So, roll 14 or higher on d20 -- a single die, equal chance for each separate value. Then you roll separately for damage (bugbears are big, right? so that's ... oh, forget it, probably d8 or d6 or something). So that's a FLAT curve and SEPARATE effect.

And then there's Rolemaster, with a d100 roll compared to a target value ... if I remember correctly, the goal was to roll high, with re-rolls and modifiers pumping up final values above 150 sometimes. Anyway, the higher, that is, the more you beat the target value by, the better. Or an example closer to my memory-access units would be Fading Suns, in which one rolls a d20, hoping to get equal to or under the target value -- and the closer to you get to it, the better. So if your skill in zap-pistol is 13 and you roll 12 or 13, you roll more damage dice than if you rolled a 1 or 2. Anyway, these games would be categorized as FLAT curve and INCORPORATED effect.

Quick terms definition: in a flat dice technique, every possible result is different from every other; in a curve technique, one can get the same value in multiple ways. Other points: I'm aware that crits make games like AD&D a wee bit incorporated, on occasional rolls anyway. Also, beware of calling a system "flat" when the result is generated by comparing offensive and defensive rolls -- if defender rolls d20 and offender rolls d20, then the possible outcomes are actually belled.

Examples - Flat & separate: AD&D, Rifts, RuneQuest, Call of Cthulhu, Pendragon, Fifth Cycle, Jorune, Flashing Blades - Flat & incorporated: Rolemaster, Armageddon, Kult, Sun & Storm, Neverworld, Fading Suns - Bell & separate: The Fantasy Trip, Champions, GURPS, Whispering Vault, Over the Edge, Albedo, In Nomine - Bell & incorporated: Alternity (most of the time), Tunnels and Trolls, Feng Shui, Zero, Babylon Project (?), Lace & Steel (dice) - Weirdos: Ars Magica (original), In Nomine, Cyberpunk. Alternity varies from flat to bell depending on the situation, Talislanta has incorporated for noncombat actions and separate for combat actions.

I think these categories have extremely concrete effects on how the game is played and what sort of tactics and standards will arise within them. Comments? ***************

RAN: Well? Keep going, dammit! What are they? ;) **************

RON EDWARDS: It's your input I'm looking for. To my certain knowledge, you've played more RPGs than, well, God. What do YOU think? Is any of these options better than the others? Why or why not?

And everyone else, too!

P.S. And I didn't even bring up the cards vs. dice issue yet. **************

RAN: To paraphrase Kid Miracle Man (post lithocranial fusion), "Guh."

Here's a question to start things off: is it screwy to apply linear modifiers to a bell-curve system? If you're rolling, say, 3d6 against a target number... if you need a 17 or better, your +1 is worth "more" than if you needed a 7 or better, isn't it?

How else would you do it, you ask? Hell, I dunno... this is off the top of my head. ****************

ED: I think that each type of conflict resolution / dice system has its relative merits. The determining factor, the way I see it, is that you need to match the system you use to the type of game you are developing. Some games work best with one system but not with another. In much the same way that the choice of using attributes, skills, levels, alignment, etc... is tied to the type of game you are making, so too, the choice of your dice/less system should be just as important a decision. ******************

CHARLES: I've only played a handful of these games mentioned by Ran & Ron, so I'm out of my league when it comes to comparing the different game systems & their specific methods.

However in broader terms I think the merit of a dice system (or any other game mechanism, but dice for now) is this: how well does it serve the parameters of the game?

1) Playability: how often do you have to roll, & how easy is it to calc the effects of the roll? 2) Depth: how much information do the dice results convey? 3) Game sense: do the things you roll for make sense in the context of the game?

Pretty obviously every game puts a different value on each of these. The first 2 especially are kind of opposite ends of the same spectrum: speed & ease of play is always going to be a trade-off against complexity & level of detail within a given set of game mechanics, although a more efficient engine can provide way more of both than a poorly designed one. THe trick is to decide the mix the game is designed/ best suited/ aiming for, and build an engine that reflects & enhances this.

Number 3 is a little more global. No matter if the game is fast & loose or intricately detailed, the things you roll for should always make *sense*. One thing that always used to piss me off no end when playing AD&D was the way AC combined the target's evasiveness (how hard they were to hit) with actual armour (how easy it is to damage them when you do hit them). In a system that uses dice as the action-resolution mechanism, when I roll the die I'm *doing something*, & I like to know what it is I'm doing. Since the AD&D system uses 2 rolls to determine the result of an attack, one to hit & one to damage, it always made far more sense to me to use the second roll--which must measure how hard you hit the target, since it has no bearing on *where* you hit--to determine if the blow penetrated the target's armour or not.

To me this is far more satisying, because it makes sense to me, and the other method doesn't. If what I'm rolling for doesn't make sense, if I can't picture or get into what it is I'm actually doing or rolling for on some level, then I just feel like I'm throwing a bunch of dice around the table.

Another aspect which probably comes under playabililty is consistency. The fewer different dice rolling methods in the game, the easier they are to remember & employ. ******************

RON EDWARDS: >I think that each type of conflict resolution / dice system has its relative >merits.

It might be good to get these articulated. Personally, I have a hard time with flat-curve action resolution, always have. The reason is that if my character has an 12 or less on 3d6, the chance for a SINGLE missing value (e.g. 18, the "aw, crap!" result) is tiny compared to the chance for a SINGLE better value (e.g. a 10). But if I'm rolling a d10 or d20, the chance for a single "aw crap!" value is exactly the same as the chance for one I want.

Ran is correct in pointing out the problem with bell methods though, in regard to modifiers. We Champions veterans well remember the key to success: skill levels! Why spend 3 character points per Dexterity point, 2 or 3 of which (6 or 9 character points) are necessary to boost Combat Value, when those same three points give you a +1 skill level (a whole Combat Value unit!) on an offensive technique?

Apologies to non-Champions sorts ... what I mean is that +1 unit modifiers boosting an average level of skill for a 3d6 system are very, very effective -- far more so than buying the skill itself up. In other words, modifiers are meaningful at a low level of skill but less so at higher ones.

The fellow who wrote Fading Suns told me that this is exactly why he and the other designers went with a flat d20 resolution system.

>The determining factor, the way I see it, is that you need to match >the system you use to the type of game you are developing.

That's the main reason I decided to stay from target numbers altogether in Sorcerer. I wanted it to be almost entirely mysterious to the player just what he or she is rolling against when binding a demon. Yet I also wanted meaningful modifiers based on roleplaying and narrative detail.

>In much the same way that the >choice of using attributes, skills, levels, alignment, etc... is tied to the >type of game you are making,

OH yes. But that's another barrel o'annelids ...

Charles wrote some good stuff too.

>1) Playability: how often do you have to roll, & how easy is it to calc the >effects of the roll? >2) Depth: how much information do the dice results convey? >3) Game sense: do the things you roll for make sense in the context of the >game?

>... The >first 2 especially are kind of opposite ends of the same spectrum: speed & >ease of play is always going to be a trade-off against complexity & level >of detail within a given set of game mechanics, although a more efficient >engine can provide way more of both than a poorly designed one.

I agree so strongly with the final statement that I think it outweighs nearly any consideration of "realism" or "depth" at the one end VERSUS "speed & ease" at the other. In other words, it IS possible to have the best of both worlds.

>Number 3 ... the things you roll for should always make *sense*. >... In a >system that uses dice as the action-resolution mechanism, when I roll the die I'm *doing something*, & I like to know what it is I'm doing.

I agree with this in general, although just what needs resolving per unit "event" in game terms. Let's take combat. Do we really have to have, GURPS-like, one roll to hit, one roll to defend, one roll for hit location, one roll for damage, one roll for shock, etc? Or do we just say, hey, good roll, which directly determines that the damage is enough to knock the guy out? The absolutely bizarre system in Maelstrom has led me to question some long-cherished RPG standards pretty seriously.

>Another aspect which probably comes under playabililty is consistency. The >fewer different dice rolling methods in the game, the easier they are to >remember & employ.

Yet for some reason we see little trend in "the industry" concerning this. For a brief while in the early 90s, RPG systems went through some great streamlining (OtE, The Whispering Vault, Zero, lots more). Now we seem to have regressed, with Deadlands in particular bringing a completely different resolution system to gunplay, three types of magic, and reactions.

I'm still up for more comments on this! How about some new folks' thoughts ... which of The Four is best? - flat, separate effect - flat, incorporated effect - bell, separate effect - bell, incorporated effect ********************

RAN: > 3) Game sense: do the things you roll for make sense in the context of the > game? > I think this is incidental to the actual mathematical *system* of dice used in a game, but it brings up an interesting (and also incidental) point...

> Number 3 is a little more global. No matter if the game is fast & loose or > intricately detailed, the things you roll for should always make *sense*. > One thing that always used to piss me off no end when playing AD&D was the > way AC combined the target's evasiveness (how hard they were to hit) with > actual armour (how easy it is to damage them when you do hit them). In a > system that uses dice as the action-resolution mechanism, when I roll the > die I'm *doing something*, & I like to know what it is I'm doing. Since > the > AD&D system uses 2 rolls to determine the result of an attack, one to hit > & > one to damage, it always made far more sense to me to use the second > roll--which must measure how hard you hit the target, since it has no > bearing on *where* you hit--to determine if the blow penetrated the > target's armour or not. > > To me this is far more satisying, because it makes sense to me, and the > other method doesn't. If what I'm rolling for doesn't make sense, if I > can't picture or get into what it is I'm actually doing or rolling for on > some level, then I just feel like I'm throwing a bunch of dice around the > table. > Back in the old AD&D days, combat in RPGs was generally was abstracted in similar fashion: there was a period of time (generally a "round"), and you rolled once (or engaged in one related series of rolls) to find out what happened during that time. If I'm not mistaken, isn't the AD&D round a whole *minute*? Charles' comments strike home as to a potential problem with this kind of system: since you don't have a clear idea as to what your ZE=-1>character is doing, you tend to find the least complex picture: in AD&D's case, it is that one attack roll = one swing of your sword/mace/etc., which is not the case in the rules. Likewise, hit points, which are supposedly a combination of luck, experience, near misses, etc., become just a damage tally, i.e. a physical representation instead of a metaphysical one.

Why did this happen in those 1970's games? because although their hearts were in the right place, there was no attempt at giving any sort of a story-related structure to combat. Basically, they said "a minute passes, and in that time you had some effect on the creature," but didn't give any inkling as to how an individual character's actions might affect that outcome, other than "I swing at it."

In the 80's, this system was rejected in favor of shorter rounds, and more specific actions -- look at Dragon Quest, which originally had a 10-second round, but revised it to 5, and Champions, where each "round" was *one* second! Of course the problem here was that 1) combat took *FOREVER* to play out, and 2) it just got too damned ponderous and complex (such as Stalking the Night Fantastic, where you rolled not only to hit something, but also to look on a chart that told you *precisely* where a bullet struck and *how far it penetrated*).

It's interesting to me that the cutting edge of game design in the 90's has gone back, essentially, to the old AD&D-type system of "in this amount of time, thus-and-so happens, and its eventual effects are x..." Only this time, there has been attention paid to how an individual character's actions (and personality) affect what happens. Instead of a literal "one round, one swing" philosophy, we've gone full circle to a period of maneuvering and reacting, as summed up by rolling once" kind of deal.

Anyway, kudos to Ron, Jonathan Tweet, and the others who've done much towards correcting the flaws of the more holistic resolution system while preserving its efficiency and relative simplicity. *****************

RAN: Well, geez, Ron... given the above (deleted) discussion, it sort of depends on the game, now doesn't it? ;)

To be simplistic, I lean towards bell, incorporated, if only because I generally feel it's less likely a given individual will "hit something" really well, as opposed to hitting it in an average way. Likewise, if you hit well, you hit well... why roll again to see how much you hit it?

Actually, you can answer that by saying "it depends," which leads into something Charles pointed out about what each dice roll is for, in a nit-picky but necessary way. Let's take the example of sword combat again: when your man "rolls to hit," what exactly is being determined? The possibilities are endless, e.g.:

-- Whether your swing hits the other body; -- Whether your swings on its way to hitting the other body, pending some defensive motion by that body; -- Whether your swing hits the body and penetrates whatever protection it may have; -- Whether your "swing" is really the culmination of several seconds' worth of maneuvering and attacking/parrying/riposting; -- Whether your swing is representative of a longer period of dueling; --Whether your swing *should* hit the other body, but is possibly only a "near miss" (a la AD&D); -- etc., etc.

Once that is determined beyond a shadow of a doubt, if a second roll has been deemed necessary, it must be noted precisely what *that* roll accomplishes/represents, with the same slew of possibilities. I think a potential flaw in many separate effect systems is that the "areas of determination" in the two (or more) rolls overlap.

It almost seems like one could lay out some general guidelines for the different styles of systems as laid out by Ron regarding what could be included in the area of effect for a particular roll of the dice, and what probably should NOT be included, based on the style of system.

And no, I'm not going to tackle that one :) ******************

GEORGE: Okay I've been following this dice discussion and finally figured out what I wanted to say.

Which one of these do I prefer? It seems to me that the less rolling the better because more into I am more into the story end of RPG. Though I must admit that most players I have had like to roll something. It makes them feel like there doing something. No offense to players, of course. Any dice roll can be superceded or added to. So that would be bell incorporated with seperate rolls to deal with any element not covered by the roll or the GM. And sometimes just to satisfy or concern a player's need to feel randomness. Dice serve a psychological purpose as well as any mechanic need