Quote from: Brander;731474Going from 11- to 12- isn't an 11.57% change from the original skill, even though the actual percentage goes up by that amount, it's an 18.5% change from the original skill ((74.07-62.5)/62.5 = 18.5%).  That change is what is important to me in this case, it might not be to you.

If you apply the same logic to your original "sniper in a breeze" example, then dropping from 17- to 16- quadruples the sniper's chance of failure, whereas dropping from 12- to 11- isn't even doubling the average character's chance of failure.

Quote from: Herr Arnulfe;731472Yeah but players wouldn't be as likely to "feel" the effect of miniscule relative modifiers at the gaming table when rolling the bones, whereas they might feel the effect of a +7% difference in absolute modifiers. Personally I'm dubious about the players feeling any of this at the gaming table, but for arguments sake I'm assuming they do.

While I notice such things more or less immediately (mostly due to caring about it) it's likely more of an "over time" feel than an individual roll feel, and, yes, it's a system designer thing more often than a player thing.

Quote from: Herr Arnulfe;731472I mean, if we're talking about relative increase/decrease, a skilled character's chance of failure on a bell curve actually suffers larger relative penalties than an average character's, even though the % is smaller.

True, but when the desired feel is focused on chance of success that matters much less.  Also other system specifics can matter as well.  If failure brings great consequence more often than success, it might be more important to care about the relative chance of failure instead of success.  I'm not saying one or the other of any system is better here, at most I'm saying it can be important that the dice system backs up the desired feel of the game.

Quote from: Herr Arnulfe;731475If you apply the same logic to your original "sniper in a breeze" example, then dropping from 17- to 16- quadruples the sniper's chance of failure, whereas dropping from 12- to 11- isn't even doubling the average character's chance of failure.

It's the change in success I'm caring about in this case.  The miniscule chance of failure going up massively bothers me less as long as it remains small.  None of this, except the actual numbers, is objective as such.  It's about what a system designer or player who will notice will care about.  Some people like the feel of a bell curve, some people like the feel of a dice pool, and some people like the feel of a d% system.  Most probably don't care at all

Quote from: Brander;731476True, but when the desired feel is focused on chance of success that matters much less.

The way you described the "sniper in a breeze" example, it sounded like you were interested in mitigating risk-of-failure margins for skilled PCs compared to average ones as a consequence of negative modifiers.

Quote from: Herr Arnulfe;731478The way you described the "sniper in a breeze" example, it sounded like you were interested in mitigating risk-of-failure margins for skilled PCs compared to average ones as a consequence of negative modifiers.

It was just an example to illustrate a point.  Though it kind of turned into a dissertation.  I still think it's a good example, especially since it highlighted how it can be turned over if you focus on failure over success.  And in my experience it does indeed produce a tangible feel in games like Gurps and Hero for 3d6, Traveller and BESM for 2d6.

I have a preference for multiple dice bell curves.  Despite that, and despite knowing it's statistical quirks, Savage Worlds has somehow become my favorite system, over Gurps even (a previous favorite and a game I still have tremendous respect for).  Turns out the simplicity and speed of rolling one or two mostly flat randomizers per character (among many other things) trumps a lot of what I like about multiple dice bell curves.  The wild die produces a triangular-ish shaped "curve" for success with wild cards, but each die is independent since you (almost?) never apply both and that's before we get into the statistical wonkiness of the dice exploding, which is clearly trumped for me by the sheer fun in rolling again.  And from what I understand they chose what was fun to them over that (quite minor in actual play) statistical wonkiness to get the overall feel they wanted.

Quote from: Brander;731480It was just an example to illustrate a point.  Though it kind of turned into a dissertation.  I still think it's a good example, especially since it highlighted how it can be turned over if you focus on failure over success.

If you're analyzing modifiers as a proportion of the base value instead of as absolute values, you're already well off the path of tangibles and into theory-land anyway. At that point, you'd be remiss IMO to overlook the effect of bell curves on risk-of-failure at different proficiency levels. But, we're talking about subjective realism here, so of course you're entitled to define the parameters of your own personal analysis.

Heck, to make modifiers tangible just play EotE and WFRP3e so players can actually feel the modifiers in their fingers. From a purely mathematical POV I don't think players are likely to "feel" anything less than a 5% shift when rolling the bones, and probably won't feel the difference between a 5% and a 10% shift either. What matters is the imagery that the GM attaches to the modifier, and the character's in-world options to negate or enhance them.

I like roll-over dice (ala Rolemaster), but some of my favorite all-time games (Rifts, Palladium Fantasy) and my current system (RQ6) use roll-under. I really don't have an issue with them, though I've converted Rifts to roll-over before, to generally good effect.

Quote from: Zachary The First;731495I like roll-over dice (ala Rolemaster), but some of my favorite all-time games (Rifts, Palladium Fantasy) and my current system (RQ6) use roll-under. I really don't have an issue with them, though I've converted Rifts to roll-over before, to generally good effect.

IIRC the only d% roll-over system I'm familiar with is the Streamline system, used in FJ Gaming's Gunslingers & Gamblers: Streamline Edition and Privateers and Pirates. This system is pretty neat and I don't really have a preference one way or the other.

Quote from: Herr Arnulfe;731481Heck, to make modifiers tangible just play EotE and WFRP3e so players can actually feel the modifiers in their fingers.

Loathing is too bland a word to describe my feelings of EotE's dice and dice system.  I'm in a mostly weekly game, so no one needs to tell me to try them, I'm stuck in their hell now with an otherwise great GM and group, but that is literally another thread.

I only know what the WFRP3e dice look like, the game struck me as too many bits for too little gain.  I do however have a great deal of nostalgia for WHFRP1e which is d%.

Quote from: Brander;731507Loathing is too bland a word to describe my feelings of EotE's dice and dice system.  I'm in a mostly weekly game, so no one needs to tell me to try them, I'm stuck in their hell now with an otherwise great GM and group, but that is literally another thread.

LOL I was being facetious there. Actually I wouldn't recommend those game to you anyway since it sounds like you prefer for modifiers to disappear at certain proficiency levels.

Quote from: Herr Arnulfe;731511LOL I was being facetious there. Actually I wouldn't recommend those game to you anyway since it sounds like you prefer for modifiers to disappear at certain proficiency levels.

Sorry I missed it at first, kneejerks (me) and all :-)

Two of my favourite games are Cyberspace and Call of Cthulhu; one is roll-over and the other roll-under. What shall I say...I like them both!

Personally, I do think that roll-under is a (little) bit more intuitive for players to understand than roll-over...

I always prefer roll under %, actually.  I use it in 90% of the games I run and design.

I don't mind roll under, or even linear progression of success, but for some reason I'm just not fond of the d% dice. Its probably just my own personal neuroses. It won't stop me from playing a game, but for some reason I just don't like them.

Quote from: Shauncat;731182have a number of issues that would make me avoid them, personally, if designing a system from the ground up:

1) Pass/Fail - With no gradient in between, pass/fail mechanics tend to produce fairly boring results.

This has absolutely nothing at all to do with whether we're tossing d100 or d6, low or high! The absurd notion that there is some relevance is equally musterable against whatever arbitrary method of tossing dice anyone cares to single out for disdain.

Also, anything "other than" a binary choice is just a set of binary choices. Different people choose to define different dichotomous domains. YMMV as to convenience of looking up a set table of finite possibilities, or using progressively branching (and perhaps improvised) considerations.

Quote2) Variable Target Numbers - In some systems, you end up with situations where a professional has a 50% chance of doing his 40+ hour a week trade correctly. It could be said that he's under pressure, but pressure could be part of his daily routine! A fireman doesn't have a 50% chance of missing a fire with a torrent of water just because fire is scary.

Again, this has sweet fuck all relevance to ANY way of tossing dice. This is all a matter of the context in which a roll is called for in the first place.

QuoteThe obvious solution to #1 and #2 is only roll if the stakes are interesting, right? Well sure, if it works for your group. Often however, the percentile roll-under games are used at simulationist tables.

Every simulationist table that has ever been, has excluded consideration of an astronomically vast domain of stuff because (A) it wasn't sufficiently interesting; and (B) otherwise, all the time in the universe wouldn't be enough to complete Move One.

I'm sorry, but similar physical realities dictate that writers must assume some common sense in readers. Once upon a time, simulation gamers were on average notably intelligent, too.