How do you feel about roll under systems that use modifiers (eg Difificult Action -2 to the target number) to simulate difficulty?

Quote from: Ghost Whistler;658082How do you feel about roll under systems that use modifiers (eg Difificult Action -2 to the target number) to simulate difficulty?

Absolutely neutral, as I do to any system at this extreme level of abstraction. It's all in the details.

-clash

It has the disadvantge of not being what people are used to, both in terms of "roll high is good" in most games, and "modifiers are usually to die rolls, not target numbers."

Aside from that, mathematically irrelevant.

what do you mean by mathematically irrelevant.

Roll Under == Roll Over mathematically. Modifier to TN == Modifier to Die Roll mathematically. It's like you are saying "Roll dice and modify! How does that grab you?".

-clash

Quote from: Ghost Whistler;658277what do you mean by mathematically irrelevant.

It makes do difference, mathematically, whether you have to roll over 15 or under 5 on a d20. It's exactly the same thing. Change the number of dice you roll, alter the way modifiers work in one direction but not the other, there are things that make a difference mathematically. But roll under and reserving the sign of the modifiers is exactly the same as roll over and normal modifiers.

The only difference, to me, is that if it's different from what we usually use, it tends to slow things down because people have to stop and think about it.

(That said, all three of the games I've played recently, Fantasy Hero, Chivalry & Sorcery and Top Secret, are all roll under mechanics, now that I think about it, with modifiers being to target numbers, as you propose. So for me, it'd be roll over that'd be distracting.)

But only for about five die rolls, then I'd be adapated to the new method, and again, it's mathematically irrelevant.

Yeah, there's no real difference but it feels backwards to me at least.

Quote from: The Traveller;658319Yeah, there's no real difference but it feels backwards to me at least.

Yeah, that's kinda my point. It makes no objective difference, but if it's different from what you're used to, it distracts the hell out of you until you get used to it.

Maybe that's why I don't play D&D any more.

My 2c would be that roll-under systems generally require slightly less math - marginally - though only given the right design in other parts of the game, and probably really not much.

The probability distribution is identical mathematically to rolling over, yes, but with roll-under you're comparing rather than having to add. There is also potential effort-saving if you're rolling lots of dice simultaneously e.g. lots of attack rolls for large combats, although TBH if the target number is known you could just subtract the attack bonus from the final target to find a roll-over threshold and then upend the bucket of dice.You don't get that benefit using roll-under with d100 either unless you have a bucket of actual 100-sided dice, since otherwise the d10s need to be paired and so must be rolled one at a time, or if the rolls have significant levels of success/failure which must be individually calculated.
Both approaches are intermediate in complexity between step-dice (e.g. Savage Worlds) at the simpler end, and dice pools (e.g. Storyteller) at the further end.

In many roll-over additive system there is also extra calculations at the step of generating modifiers - compare how D&D in 2E uses 3-18 stats by just rolling against them, while in 3E the same range of ability scores required an intermediate bonus table to convert the numbers into a bonus/penalty to a check (i.e. [Stat-10]/2). You can very easily make an additive system that doesn't require the extra step by having stats simply equal to the modifier i.e. -4 to +4 or whatever, though there could be other constraints that make a low level of granularity problematic, such as wanting to have detailed ability damage.

Roll-under is also somewhat simpler for working out proportions of success chance (i.e. special success = 1/5th of your base chance of success).

A good chunk of effort saving with roll-under does get lost as soon as you start adding in difficulty modifiers. You can avoid difficulty adjustments to skill checks to some extent with a detailed skill list e.g. how Call of Cthulhu or Runequest have separate base skill percentages for every skill, but I expect this might lengthen skill lists somewhat - essentially front-loading complexity to save calculations during the game.


All that being said, all of this in most cases probably doesn't add up to much total difference in table time, and despite cost-saving I personally don't like roll-under at all, I find the 'lower numbers are better' thing weird, same for blackjack success where higher is better unless you roll over.

I had considered purely roll under with no modifiers but i'm not sure that isn't too simplistic.

Want to hack a computer, make a roll.

Want to climb down the drain in the rain and not be seen? make a roll.

Mathematically, there's no real difference.

Mechanically it helps in accounting, in my experience. Just like BRPs or CoC where you know at a glance what is your % chance to succeed, you know the same from any other roll under (as long as you understand the % increment each pip of the die means).

That may not seem like much, but I personally find it helpful for two things:

a) maintaining bounded accuracy. naturally you can retain it otherwise, but it's just visually easier for me to know where's the extremes with zero as a hard demarcation point.

b) easy to calculate improvisational modifiers. just like roll-over fans acclaim addition over subtraction, I acclaim the same works for the GM, but the GM is the one who really needs it. with a floating TN and unbound accuracy, it makes on the fly modifiers an effort in calculation. with basic percentage expressed in different increments, it's dirt easy to determine on the fly modifiers' change in percentage.

Those two things help keep my games within a certain scope, while assisting with improvisation, as I am very much an environmental GM.

So, outside of the feel good of rolling big numbers, I find no real benefit of roll-over in comparison for my needs. (I mean, how hard is AC + BAB = roll under? It's still addition.)

But would you be satisfied with a roll under system if it didn't have some kind of modifier system for difficulty levels?

Roll under in many ways is perfect: it's simple and intuititve, but it only goes so far before becoming counter intuitive and clumsy it seems to me.

Quote from: Ghost Whistler;659241But would you be satisfied with a roll under system if it didn't have some kind of modifier system for difficulty levels?

Roll under in many ways is perfect: it's simple and intuititve, but it only goes so far before becoming counter intuitive and clumsy it seems to me.

For myself I'd say no, because the idea that all actions are equally difficult is an issue for me. (though it seems that Apocalypse World does just that, in an additive system).

As another option to a straight plus/minus, multiple dice or step-dice are a possibility, where the difficulty adjusts either the sort of dice roll (e.g. larger difficulty ->larger dice) or number of dice a la Summerland/The Fantasy Trip (where a simple task is roll under with 2d6, a normal task 3d6 and a harder  4d6 or 5d6). The latter type of game does have addition steps as well of course, since you're adding the dice together, but nothing too complex. Probabilities a bit weird maybe.

Quote from: Bloody Stupid Johnson;659257For myself I'd say no, because the idea that all actions are equally difficult is an issue for me. (though it seems that Apocalypse World does just that, in an additive system).

As another option to a straight plus/minus, multiple dice or step-dice are a possibility, where the difficulty adjusts either the sort of dice roll (e.g. larger difficulty ->larger dice) or number of dice a la Summerland/The Fantasy Trip (where a simple task is roll under with 2d6, a normal task 3d6 and a harder  4d6 or 5d6). The latter type of game does have addition steps as well of course, since you're adding the dice together, but nothing too complex. Probabilities a bit weird maybe.

I had considered that. Adding dice seems heavy handed. Stepping dice isn't bad but, proabilities aside, seems more counter intuitive.

I am considering treating difficulty levels as opposed rolls. So if you're hacking a secure system, the GM imposes a difficulty level and makes an opposed roll using that as the score. The player rolls against that using blackjack style resolution (highest wins).

Hadn't thought of that - that would work. I guess you'd still need difficulty modifiers for opposed rolls though.