Oy,

it's difficult trying to be cryptic about one's system while discussing it with a community. Okay, I'm going to be a lot more honest in this post, though the two buddies I'm working with will probably kill me. (However, they've gotten hideously busy and taken a backseat, so I'm mostly working on the system alone anyway). Serendipity struck and we collectively came up with an idea for a physical game tool, which will serve as the main conflict resolution engine for our new RPG. We're all a bit nervous of having the idea stolen but I've read several times here that all newbie developers feel that way, even though typically no one gives a rats patootie in reality anyway. I'm convinced the idea is gold tho. One of the reasons I'm working so hard to make a hardy, excellent RPG is because the tool itself deserves nothing less. (Sorry for the rampant ego.)

The Precursor to the Idea
Remember those old slide rules from WAY back when, before calculators and computers? Useful little guys for making fast calculations because they're logarithmic, which means that if you want to multiply or divide two numbers, just line them up on the slide rule and it produces the result. For example, if I want to perform 5 x 7, I take the number 5, advance the other slide by 7, and the result I get out of the slide rule is 35. Similarly, if I take the number 10 and move the other slide back by 5, I get out the number 2. Sweet! I don't even need to know how to multiply or divide!

Multiplication and division is all well and good, but what's it got to do with roleplaying? Well, let's call the outside of the slide rule the "attacker", and the middle part of the ruler the "defender". Furthermore, let's associate with each result the probability that the attack will succeed. You can define pretty much all conflicts as having an "attacker" and a "defender", though it might be a strongman versus a stuck door, an arrow versus an elf's chest, or a catapulted boulder versus a castle wall. In any case, you have two sides of a conflict, and at any moment, one is attempting to "win" the conflict.

The Spark
Now, we have a slide rule of attacker versus defender, and each point on the rule is associated with a chance of success. There are two reasons this is interesting:

• a) the same percentage chance applies to a wide range of attacker/defender combinations
• b) the percentage can be arbitrarily defined

With (b), an arbitrary percentile means nothing yet, since each place you may slide the rule have a number of attacker/defender combinations. So, let's look at (a) first: the percentile applies to a wide range of a/d combinations, so what? I'll tell you: remember that, as well as multiplication, the slide rule does division? Division leads us to a fraction; let's call that the attacker/defender (A/D) ratio. Here's some examples of A/D ratios:
      100/50
      10/5
      30/15
      246/123

These ratios have in common the fact that the attacker is precisely twice as strong as the defender. Wait a minute! It is also true that, due to the fact that the slide rule uses a logarithmic scale, they would also line up at the same time. In fact, all the A/D ratios are exactly equal to each other, for any given point where you slide the rule, and equal A/D ratios means they all the have the same probability of success.

This "solves" (b). By assigning percentiles to fixed points on the slide rule, you get a device which, when you input an attack score and a defense score, it outputs the probability of success. Now that we have this, it can be treated as a black-box, screw the math!!

Serendipity
Oops, a possible flaw: the slide rule would describe one particular set of probabilities. It might work in a world where everyone is human and no one fights on horseback, but that's hideously boring, isn't it? To make the slide rule work, we'd have to adapt it to each and every case. Meh, lots of work, pain in the ass. I know I certainly wouldn't want to play a game that came with a bunch of slide rules.

You see, the problem I'm talking about is that two cases of conflict resolution may not be described by the same probability function, and furthermore, even if they are, most probability functions take in several parameters to derive their shape which most likely differ between the two conflicts.

All is lost?

We're not done for yet: here's the kernel of our game, the nut. The chewy centre. The Normal Function is a very special, quasi-magical function. If you modify the parameters, you can describe a vast, almost universal variety of naturally occurring probabilities. What about the function parameters, wouldn't they vary greatly between all the conflicts? Yes, but as I mentioned, the Normal Function is magic. Any normal function can be converted to any other function with a few simple operations.

Hrm, "a few simple operations". Even a few operations are irritating during an RPG. Bah, just a stumbling block! If you can't adapt the curve to the rest of the world, adapt the rest of the world to the curve! We're designing our game such that all the parameter conversions are embedded into the statistics already. To facilitate this, we are using another kind of slide rule: a circular slide rule. Rather than just having a limited scale, we could expand this new circular-slide-rule-with-probabilities to encompass a wide scale. By having a wide scale, you can fit a lot of numbers on to it.

You could have:

• man versus man: say, 250 versus 524
• tank versus tank: say, 1505 versus 4912
• god versus god: say, 50345 versus 20944
• ant versus ant: say, 1 versus 2
• even, a herd of cats fighting a couple of guys: say, sum=150 versus 250+524

Final Thoughts
Another brilliant thing about the slide rule, which we've called the Normal Engine, or Combat Wheel (though it's application far exceeds just combat), is that is allows us to ignore that which truly doesn't matter and focus upon that which does. For example, in the case of god versus god, does it matter if the attacking god has 50345 or 50346?? NO, on this scale, a +1 difference has no noticeable effect on the final probability. The players who don't care about minutia are free to rough it, and the min-maxers who try to squeeze every +1 out of a system are limited by the granularity of the wheel.

Also, because the Combat Wheel is based upon a circular slide rule, you have access to easy multiplication which opens up new avenues for the game that are undesirable in other games. Thanks to this ability, we can battle on the level of vast armies as easily as we would individuals, greatly increasing the scope and flexibility of the core system. Further, we can even have inter-scale conflicts, something that is very, very difficult to pull off with a linear scale probability system.

We've made one more critical addition to the combat wheel, which I won't reveal, partially to hold something back. :D On the other hand, think of it as a teaser for when this game finally makes it to market, which probably won't be for a few years at least.

Dan

Dan,
  OK, I am not trying to discourage you, or shut you down. But, I do want to frankly state that something like this might be more suited for a CRPG.
  Also, you shouldn't use a logarithmic scale if you want the math to work at any scale of conflict. Because if the value behind a point increases logarithmically, then even if the ratio is 2:1, being 10 point higher than someone and being 100 points higher than someone else makes a BIG difference. So, for instance:
20 vs 10 means that you have the same odds as 200 vs 100
But, if it increases logarithmically and say, it doubles in value every 10 points than that means that 10 is twice as good as 20, but that 200 is 1024 x better than 100.
  Maybe I am arguing semantics and you did not mean logarithmically or you meant it differently than I am interpreting, if that is the case, I apologize.

  I think maybe it might do you well to play a game with looser ability definitions. Can I suggest Issaries HeroQuest or Dogs in the Vineyard? I have played these games and they allow the players to define their own abilities, interpret them loosely and then back up those player definitions with real mechanics. I am not saying you should make your game like that, but maybe if you play one of them, you will be inspired to try different designs,

  Also, I still don't see what the scale is here? If I graduate high school with all C's including phys ed, what are my ability scores? Are they all 100? 50? 200? Why so high? Because I have to be able to mathematically prove I am superior to an ant? Couldn't you just say, you don't have to roll against a single ant and then give an ant swarm stats?

  Maybe that is the piece you are missing from your design goals, the conditions when the mechanics are not used...

  Don't get me wrong, I think you are on the brink of a good universal system, I just think that something in your design goals are flawed and forcing you to chase an objective that will not necessarily yield a fun game.

  I feel like maybe you took my last post to say, "you are doing it wrong" or "this will never work." And if I left you that impression, I apologize. I like that idea of APTS, and I think it is well executed from the perspective of structure. I just think that some of the math needs to be toned down. Like maybe "A" should be rated from 1-10, and so should "P", "T" and "S". then set your target numbers as if everything is rated from 2-20 (A+P) and then when a player can bring T and/or S to bear, it will have a huge impact, right? Then no math is required. And as far as scaling, maybe make a scaling modifier to the TNs. Like Ant vs Human gets a +20 to their TN for the Ant. And for Ants they get a -1 to the TN for ever 20 Ants (or whatever number makes sense to you). This way, there is a little more work to calculate the TN for odd scenarios, but the standard scenario (man vs man) is simple. You don't have to use this exact idea or scale of numbers, but maybe its a good frame work to build on...
[code]Durkon Thundershield  Professions          Techniques
Strength      8        Cleric of Thor*  4    Shield-use            4
Agility      2        Field Medic      6    Basic Weapon Combat    4
Dexterity    1                              Martial Weapon Combat  6
Intuition    5                              Healing Spells        4
Intelligence  3        Specializations
                        Heavy Shield    2
                        Dwarven Hammer  6[/code]

  Then when he wanted to heal someone, maybe disinfecting a wound has a TN of 12, meaning that the player has to roll 2d10 and add his APTS and get 12 or higher. While mending a broken leg might have a TN of 20 again vs 2d10+APTS. As you can see from this example, Durkon would probably have an APTS for healing of 15 meaning that either task would be trivial for him, which is appropriate for this character no? And PvP conflicts are simple APTS vs APTS if you like. So you see how if we had a poorly trained healer, they have a marginal chance to help people, but if the character has APT and S to bear, most tasks in that realm become trivial. You can change the TNs or mechanics to suit your play preferences, but this is just an example of how you can keep the APTS structure and scale the math down a bit...

  Anyways, I hope your game is fun and successful, I surely think this is a good start.

I had a similar reaction to the logarithmic scale part. I think the main problem I'm having with the original post is that it seems to describe a solution at great length, without actually stating the problem that it solves.

It seems like what you are saying is something like: "We want a system that does X, but it requires too much math to be practical; however, a slide rule allows us to do it." Assuming that's right, what is X? It apparently uses ratios of some kind?

Hello Dindenver, Wordman,

I'll be responding to you both at the same time. Unfortunately, rather than clarifying my other cryptic posts, I suppose I've muddled it all, which I'm afraid has led to some misinterpretation. X-) Sorry about that.

I'm going to try this from a different tack: the player's perspective. From this point of view, using the Combat Wheel is no more difficult than a table look-up. With a table, you have two numbers, A and D. Look up A on the columns, D on the rows, and you get out the attacker's % chance to hit.

Similarly, with the Combat wheel, you also have two numbers A and D. Find A on one scale, line it up with D on the other scale, and you get out the attacker's % chance to hit. NO hint of tricky math stuff here. (Dindenver, although it involves logarithms inside the machine, they cancel each other out. You put in linear numbers and you get out linear numbers .. they don't explode to crazy-high values and the player never even has to know what a "logarithm" is.)

I want to use a combat wheel for conflict resolution because, as I mentioned, it allows me to remain independent of scale. Why am I putting humans at the scale of 100's if I can choose whatever scale I want? The answer to this is that it allows me to put smaller stuff in, while leaving plenty of room for bigger stuff (for the base game.. expansion games will go even smaller and bigger.) I don't want to prove that a human is mathematically superior to an ant. What I want to do is allow for the possibility that a human gets shrunk to ant-size (say, by magic) and competing with an ant on it's own terms. (Pixies and halflings might be on the scale of 10-100.) This opens the door for campaigns of vastly different scales than those we're used to, but which play out in universes that are the common stuff of movies and TV. (Tom Thumb versus Thumbelina, anyone? Or maybe, Godzilla versus Mothra?)

Dindenver, your retort to this was that I should simply build stats for an ant swarm. Okay, but what if they're not ants? What if they're cockroaches? Pixies? Mutant pixie/cockroach-warrior hybrids? You might scoff at this, but think about it .. do I really want to be building swarms all over the place? It's a hack solution, because I'm making a rough guess at what the collective stats would be. This might work once, but it takes a lot of effort, and if you start involving more groups of different things, the error introduced with each rough guess starts accumulating, until the system collapses. With the Combat Wheel, there is no significant error introduced by building groups and swarms from individual unit statistics, as though they were pieces of lego. (Of course, this assumes uniform creatures, i.e. clones in terms of game mechanics.) Furthermore, scale is relative. Simple 0th level NPCs look like ants to titans, but does that mean I should build stats for a "swarm of 0th level NPCs"? No, that's silly. By the same token, although it's silly, I do want the ability to have a collection of 0th level NPCs for when it becomes necessary (say, the villager mob finally gets too angry at the dragon and attempt to kill it, even though each villager would get creamed on his/her own.)

Does this make things clearer?  (And don't worry about dampening my spirits. I'm waaaaay too convinced about this system. I'm practically a religious convert.)

Thanks for your input! :)

Dan Blain

OH and, meant to say, ALL of that aside...

hmmm, yeah, the whole APTS thing might be way over-complicated, and I consider your reactions and snapshot of reactions of the general population at large. Even though it's just two people, it's enough to convince me that this isn't the way to go with abilities and skills.

(Technically this should be posted in that thread .. oh well.)

Dan,
  OK, so, there is some kind of mis-communication happening. The trick is this, the combat wheel is OK. I don't have a big issue with it (in fact, I would do something like it if I could figure out a way to make and package them cheaply in a game). The problem I see is two fold:
1) Right now, I can't look at a PC and tell you if their stats are good or not. You never did say what the scale is for abilities. Is 100 an average human? 500? Don't forget, these numbers do more than feed into mechanics, they are a value that players use to judge if their character is good at a task or not. If I gave my character a 200 STR, I don't know if that is average, high or low. Moreover, since each segment of APTS is on a different scale, players have to learn that 100 is average for A, 20 is average for P, 50 is average for T and 200 is average for S, or whatever the scale is, I can't really tell from the examples you gave.
2) Scale. By that I mean scale of the numbers themselves, not the scale of the conflicts (in fact, I was never referring to the scale of the conflicts, to be clear). I talked about scale from the perspective of why you don't need to prove that a PC is around 100 times more powerful than an ant. OK, but there is another component to the scale that abilities are measured on. how characters are measured against each other. I mean, to put it another way, do the players really care if Fred has a 499 STR and John has a 498 STR? Even using percentile dice, that is not enough of a difference to change the die roll. But you still have to track numbers at that scale and buy/advance numbers at that scale. Also, there are things that your wheel doesn't solve. Like lift capacity. There has to be some way to determine if Fred can lift this fallen rock off of John's Leg. Your two choices are: Make a table, but then either the 499 vs 498 is in the same weight category or you have to give the players a formula and at that point either the logarithmic scale is apparent or it is disproven. There are other approaches to scale, maybe you should look at Mekton, most stats are measured 0-12 in that game, but it handles man-sized to death star sized conflicts pretty well.

  Again, I am not saying you are wrong or stupid. but I wanted to be clear, the problem I have is with the numbers themselves, not with your resolution mechanic. And I don't think your resolution mechanic really depends on the numbers being this big.

  Good luck with your game, with your passion, I bet it will be awesome when its done.

dindenver wrote:
.. Is 100 an average human? 500? Don't forget, these numbers do more than feed into mechanics, they are a value that players use to judge if their character is good at a task or not. If I gave my character a 200 STR, I don't know if that is average, high or low.

dindenver wrote:
Moreover, since each segment of APTS is on a different scale, players have to learn that 100 is average for A, 20 is average for P, 50 is average for T and 200 is average for S, or whatever the scale is, I can't really tell from the examples you gave.

dindenver wrote:
2) Scale. By that I mean scale of the numbers themselves, not the scale of the conflicts (in fact, I was never referring to the scale of the conflicts, to be clear). I talked about scale from the perspective of why you don't need to prove that a PC is around 100 times more powerful than an ant. OK, but there is another component to the scale that abilities are measured on. how characters are measured against each other. I mean, to put it another way, do the players really care if Fred has a 499 STR and John has a 498 STR? Even using percentile dice, that is not enough of a difference to change the die roll. But you still have to track numbers at that scale and buy/advance numbers at that scale.

dindenver wrote:
Also, there are things that your wheel doesn't solve. Like lift capacity. There has to be some way to determine if Fred can lift this fallen rock off of John's Leg. Your two choices are: Make a table, but then either the 499 vs 498 is in the same weight category or you have to give the players a formula and at that point either the logarithmic scale is apparent or it is disproven. There are other approaches to scale, maybe you should look at Mekton, most stats are measured 0-12 in that game, but it handles man-sized to death star sized conflicts pretty well.

dindenver wrote:
  Again, I am not saying you are wrong or stupid. but I wanted to be clear, the problem I have is with the numbers themselves, not with your resolution mechanic. And I don't think your resolution mechanic really depends on the numbers being this big.

Dan,
  OK, again, read what I write, not what you think I am implying.

  I don't care how many numbers you have to add together. You are right, I have played games where the number of mods you apply to a roll are crazy.

  The issue is those numbers are too big, unwieldy and meaningless.

  If the difference between 498 and 499 are meaningless, why do the players have to track it? What is the point of having that granularity if it doesn't mean anything? I mean it doesn't even change the roll. Think about that, statistically, our character with 499 is superior to 498, but he has no mechanical advantage because of it (because we are doing some kind of crazy division and converting this difference to % and the 1 point difference is less than 1%.

  I like APTS, I don't think its too many numbers to keep track of, I just don't like keeping track of numbers that are over 100 (really over about 10 or 20, but 100 just has to be acceptable for d% systems, I understand that).

  As to Mekton, I played a ton of this and it handles scale pretty smoothly. It uses a traditional stat+skill+mods+roll vs TN system. And you use the same mechanic for differing scales, the difference is, the damage is modded after the roll. So, full sized mech vs power armor means the mech does x10 damage vs the power suit and the power suit does 1/10th to the Mech. Its not super elegant, but it works well and lets you play that fight for players that are stupid enough to try it.

  And really, how often is the scale of conflict an issue in your games? If I say to the players, you are a normal person, and there is godzilla standing there, what do you do? They run.
  It seems like that is what this mechanics is designed to "fix" Rambo vs Godzilla. Is this a setting where that will be a common issue? And if I have Durkon and you have a titan, do you really want me to get crushed like a puny ant with literally no chance of survival? If so, why put a number on it? Just say, if you are human-sized and you come up against a titan, roll to escape or you die. Then, you can get back to designing a system that addresses the problems your characters will most likely face. And if you have 20 PCs (overlooking the fact that this would be a nightmare to GM) versus a Titan, do you want them to win, but lose 10 or so characters?

  So, to sum up:
APTS - Its a cool concept, don't drop it because you like big numbers.
Combat wheel - Great idea, how do you make one with out owning a rivet gun? All of the wheels I have seen have that cool rivet for the center of the wheel.
Skill flexibility - Sounds great, I hope you don't lose that design goal.
Number of mods - I don't care, I trust you not to go overboard, or to balance them so the game is still playable
Lots of Ants vs people - I don't care, this doesn't come up much in my games, a "hack" would be fine, or just a roll modifier or sidebar rule would work too
People vs big stuff people shouldn't fight - Not interested, if people shouldn't fight them, I don't need to beat the players over the head with giant numbers to make my point. For instance, in Exalted, the stats for Mask of Winters (a Deathlord that no PC should be fighting) it listed "How ever much you need to win the fight" for most of his stats. This is sufficient, it's a great signal to send to the players and it gives the ST great license to make sure Mask of Winters isn't defeated by a frontal assault (he shouldn't die like that, but be defeated through clever use of magic and/or trickery).
Lots of People vs big stuff people shouldn't fight - I play with a big group (8 players and even then, I don't want to pit all of them against big stuff people shouldn't fight unless they are magical or otherwise have a chance to survive the fight. Because the Titan's scores are so huge that its a guaranteed one hit one kill vs puny mortals. So, even if you get enough puny mortals, to bring down a titan, you will still lose at least one puny mortal every time the titan attacks. Now, besides the obvious "not fun"of one hit one kill situations. It puts the GM in an awkward social situation. Essentially, they have to "pick on" one character. I mean, whoever the titan attacks, they are gonna die. How do you decide who dies? Not only that, but there is the whole death spiral issue. Because as PCs die, the effectiveness of of the group of PCs goes down, but thte Titan remains at full value until he dies.
Big numbers - Don't like them. I will put up with big numbers, if they are meaningful. I mean if a character with a 499 has some sort of advantage over a 498, then its worth the paperwork to track it. Even so, I wonder si the mechanics will work fin if 0th lvl NPCs start with Abilities rated from 1-5, if the game will "survive contact" with the smaller scale?

  Does that make sense?

  Well, even if you ignore my advice (and I expect you to, its your design after all), I hope your game is successful. Good luck man.

TORG and Masterbook both use a similar log-based system, and it's definitely got some nice aspects to it.  Every 5 points of "value" (skill/attribute) is 10x.

They don't have a wheel or anything, just some charts with value->number, and various conversions (all addition or subtraction.)  So your base movement rate is in meters per round (say value 9, which is 60.)  If you want Miles per hour, that's -3 to your value, or 6 (=15MPH.)

It does make it a lot easier to easily scale up to supers/vehicles/space scales with the same system and dice.  It's also go some fairly simple rules for combining, say, 10 lower skilled attackers into one roll.  It's also drastically easier to figure out rough estimates for various things like travel times, bonuses for extra time/helpers, etc.

In practice, the value numbers are always fairly close to each other, with 8 being average unskilled human and 20 being stupidly super-powered.

Hi Dindenver,

Honestly, I apologize if it comes across that I'm not listening .. I really understand where you're coming from. You're saying that although the Wheel solves all this, the players still have to track the minutia of 498 vs 499 even though, in this particular case, it makes no difference on the final result. And I don't argue that; in this particular case, it is indeed minutia and makes no difference.

However, think about this; have you ever played a game where you rolled one of the following two cases?

• Big number of dice, small futile modifier: for example, 10d6 + 2, 12d8 + 10, or 14d4 + 4
• Small number of dice, but huge weighty modifier that makes the rolls futile: e.g. 1d10 + 143, 2d4 + 534

In other words, more minutia that you have to keep track of. This plagues games that climb into the high levels (at least, it plagues D&D when you approach epic, in my experience). This is the exact same problem that you are (correctly!) pointing out in my system. It comes into play even when the combatants are of equal challenge, e.g. titan versus titan or ant vs ant.

LONG Story short, the Combat Wheel solves this problem. The dice and modifiers you roll for titan versus titan and ant vs ant is precisely the same as that for 0th-level NPC versus 0th-level NPC. So, although you have to keep track of minutia over the course of the game, when it comes to keeping track of such minutia during the game, when it can make or break the play experience, players can get used to the idea of dropping the futile little modifiers that make no difference.

Anyway, this is getting circular so I'm prepared to let it stop here, or allow you a last word, or even continue if you don't yet have murderous thoughts towards me. Thanks again for the debate X-)

Dan

Dan,
  Alright, I'll stop trying to change your mind and instead give you some examples of how other designers have dealt with vastly differing scales.

  So there are some other games that deal with scale pretty smoothly:
  The old Marvel Super Heroes game was one. Stats ranged from 1 to over 5,000.
  The way it makes the numbers more manageable is twofold:
1) All numbers are on the same scale, Abilities, Skills, Powers, etc.
2) The numbers are divided into brackets that have real meaning.
  Then the resolution system was a universal table, you used the same table for fighting, talking or basket weaving.

  DC Heroes (now in print as Blood of Heroes) had a great system for scaling. Every ability was double the the next lowest, these were called APs. So, Jimmy Olson would have a Strength of 2 APs and Superman has a Strength of 25 APs (to put that in perspective, 25 STR can lift 1.3 million tons). The math for their AP system had to be applied differently, but it actually made life easier. So, if you add two APs of weight you take the highest AP of the to and add one. If you wanted to know how far you could throw something you took the APs of weight and subtracted it from the Strength of the thrower, that is the distance in AP.
  Again this had a universal mechanic, fighting, talking, basket weaving were all resolved on the table. Interestingly, they provided the table on a wheel as well.

  One game that handles scale very poorly is GURPS (especially GURPS supers). Because the stats are measured in a linear scale, it does not handle large numbers very well. so, in GURPS Jimmy Olson might have a 10 STR, and Superman would have to have over 300 STR. These can't resolve or play on the same system. At least this is the complaint made by most people that tried GURPS Supers.

  Exalted has an interesting approach, its not linear. Nor logarithmic. Instead they just add more awesome to each subsequent level. so an ant might have a STR of 0, normal human 2, Titan 6 or Super man might have a 9 or 10.

  These are some games you might want to try, or at least check out to see how other people have handled the issue of scaling.

  Also, it seems like you want to have combat and mass combat in the same rules. Maybe you should look at other mass combat systems and see how they handle this.

  Again, good luck with your game man, APTS seems like a good approach.