d% roll under system and combat.

[hogarth]

It solves absolutely nothing compared to just counting up instead of fucking around with counting down. There is no advantage to fiddling with roll under. None. It is in no way better than having a fixed TN and counting up.

I agree. It isn't any better and it isn't any worse other than being a bit counterintuitive in terms of the "roll high, but not too high" opposed check thing.

[echoVanguard]

Its not really much different to having an additive system capping bonuses at +100%, and having the total be the maximum possible quality achievable on a linear scale out of 200. However, with this method, you can no longer easily compare 'fail' results and 'success' results on the same linear scale, since the 'fail' results are the scores that are higher.

This is a pretty good argument, CCarter. However, I'm not sure that comparing "fail" and "success" results on the same scale is a valid goal in any system - can you give a couple of examples in which you might want to do such a thing?

echo

[Username17]

However, I'm not sure that comparing "fail" and "success" results on the same scale is a valid goal in any system - can you give a couple of examples in which you might want to do such a thing?

• Player announces their roll result.
• MC begins describing the effects.
• One of the other players realizes that some modifier applies.
• Original Player now recalculates a new result.

In an instance where one player or another notices that some buff or debuff or circumstance should be modifying the action and was not taken into account, which happens like all the fucking time, it is rally important that the number the player reported to the table be the number that is being modified. Because that way you just add or subtract the modifier from the result and move on. If you're doing some kind of roll-under bullshit, you have to recalculate the whole TN and recompare.

-Username17

[echoVanguard]

However, I'm not sure that comparing "fail" and "success" results on the same scale is a valid goal in any system - can you give a couple of examples in which you might want to do such a thing?

• Player announces their roll result.
• MC begins describing the effects.
• One of the other players realizes that some modifier applies.
• Original Player now recalculates a new result.

In an instance where one player or another notices that some buff or debuff or circumstance should be modifying the action and was not taken into account, which happens like all the fucking time, it is rally important that the number the player reported to the table be the number that is being modified. Because that way you just add or subtract the modifier from the result and move on. If you're doing some kind of roll-under bullshit, you have to recalculate the whole TN and recompare.

-Username17

Maybe I'm just being thick, but this seems like it wouldn't be much of a problem either way. Let's look at 3 scenarios:

Scenario 1: Player rolls a 55 with a skill of 66. He realizes that he forgot a +10 modifier, modifying his skill to 76. The roll is still a 55, so there's no difference to the resolved action.

Scenario 2: Player rolls a 68 with a skill of 66, which is a failure. He realizes that he forgot a +10 modifier, modifying his skill to 76. His failure is now a success at quality 68, and the MC updates the action resolution accordingly.

Scenario 3: Player rolls an 84 with a skill of 66, which is a failure. He realizes he forgot a +10 modifier, modifying his skill to 76. His failure is still a failure, so there's no difference to the resolved action.

As far as I can tell, these are the only three scenarios when applying a modifier after the roll would make a difference, and they're quantifiably no different than the normal sequence of events: "I got a 23." "That fails." "No, wait, I forgot a +2, it's a 25." "Ok, that succeeds." In both cases, the mathematical operations are identical, as shown:

Fixed TN: Initial calculation result recalculated, compared to TN.
Roll-Under: TN recalculated, compared to initial result.

It seems to me that the only situation in which the system might cause problems is if you wanted to compare between degrees of failure - that is, if you wanted to see who failed least - and I remain unsure that it's a worthwhile goal. However, if you wanted to do such a thing, you could just say that the highest roll is the best, even if it was greater than your skill.

echo

[DSMatticus]

akin to playing D&D with 3d6 instead of a d20. without THAC0.

Get with the times, grandpa, we made the math better. You'll love it.

[Orion]

Echo,

The trouble with your example is that you only did forgotten bonuses, and not forgotten penalties. If you roll "very well" on a count-up system, like an 18 on a d20, then usually nobody has to interrupt your celebration if they suddenly realize the dude has cover or a range penalty or something. You can see that because you got a very good roll, a -2 isn't going to make a difference.

If you do "very well" on the roll under and compare die rolls sytem, then you need to watch out for modifiers. If you rolled 65 on a 70% stealth skill, then remember that you forgot a 10% lighting penalty, you're screwed.

[echoVanguard]

Orion - I see what you're getting at. Your argument is that a forgotten penalty can turn a phenomenal roll into a failure, while the same event only happens in a roll-up system with non-phenomenal rolls.

That being said, it seems like something of an edge case.

echo

[Username17]

Orion - I see what you're getting at. Your argument is that a forgotten penalty can turn a phenomenal roll into a failure, while the same event only happens in a roll-up system with non-phenomenal rolls.

That being said, it seems like something of an edge case.

echo

No, god dammit. The point is that a forgotten modifier can change a success into a failure or a failure into a success in either system. But in the Roll Under system, the thing that is being modified is neither the number you reported to the group nor is it a number actually listed on your character sheet.

You have a skill of 65. You were carrying +18 in modifiers, so you needed an 83. And you rolled a 76 and reported a 76 success. And now someone comes back and tells you that 10 points of those modifiers didn't apply because Recitation ran out last turn and... what? The number you reported to the group was a 76, it doesn't get modified. The number on your character sheet is only one of the numbers that goes into the operation, so it's not much help either. You have to go back and add all the modifiers from first principles. It's fucked.

In the roll high version you retroactively add or subtract a number from whatever the fuck you told the group, meaning that there are six times as many people who can help you remember that shit.

-Username17

[echoVanguard]

A lot of people have made some excellent points in this discussion - this has been a really interesting topic.

Frank - I see what you mean, but I don't see how it's any different than having an environmental factor affect the DC. Moreover, worse things happen in lots of systems - for example, in Shadowrun 4E, getting your dice pool modifier incorrect can often force you to reroll your entire check.

Lastly, roll-under (including modified roll-under such as we've been discussing here) has an additional intrinsic benefit: in any situation with no situational modifiers, it operates entirely based on the numbers on the die, and sets it own TN. If I have a 63 skill and no modifiers, I know I need to roll a 63 or lower to succeed, period. In a TN100 system, I know I need to roll 100-63, or 37, to succeed. I have to take the extra step of computing my situational modifiers in *all* situations, even if my situational modifiers are 0. Now, in a system where there's always at least one situational modifier, this is a non-issue, but if most checks are made with no situational modifiers, it's an enormous increase in the amount of work required to make a check for no benefit.

echo

[CCarter]

This is a pretty good argument, CCarter. However, I'm not sure that comparing "fail" and "success" results on the same scale is a valid goal in any system - can you give a couple of examples in which you might want to do such a thing?

echo

Possibly rolling initiative...in an additive system this is just an additive check without a DC. In roll under....its gonna get messy.

With roll-under you presumably want to order all the people who succeeded, then all the people who failed in some sort of order. [possibly 'counting down' by die roll, then with the losers going on a negative score of [die result -success chance], which entails a messy subtraction step.

I don't know how Eclipse Phase handles this. Pendragon uses d20 blackjack style and doens't have initiative (all opponents roll combat rolls simultaneously, higher roll is resolved first and it doesn't matter when the loser goes, because they missed). Kult also uses d20 blackjack style but uses a different resolution mechanic for initiative (opposed d10 + agility mod).

[DSMatticus]

in any situation with no situational modifiers

A roll-under system with no situational modifiers is weird. That's certainly not a normal case. Roll-unders depend on situational modifiers to create easy and difficult tasks, and in the case where the status quo is that there are no situational modifiers that implies many tasks are no harder or easier than others.

This is also equivalent to a +0 bonus in a roll-over system, so it also doesn't occur uniquely in roll under. We certainly imagine +0 bonuses in roll-overs as weird, but that's because we're used to D&D mechanics where they are weird. There's no reason you couldn't have a system where it's normal, though. You simply gear the proficient, average roll as +0 and anything less than that as a minus and anything more as a plus, so the typical guard swinging a sword is +0. A peasant with a sword is -2, or some such, and a trained swordsman is +2. Scale as appropriate. And this system would harness all the benefits your proposed roll-under has (in the simplest case, it's only a single direct comparison; no mathematical operation), while retaining all the many, many, many, many, many (etcetera, etcetera) strengths roll over has over roll under.

[hyzmarca]

Assuming http://www.tgdmb.com/viewtopic.php?t=50 ... sc&start=0 is the thread in question, I'm reading it now, and thank you for pointing it out to me. However, I'm a little skeptical that "subverting the lower-is-better goal" is a meaningful problem, especially considering how many other problems it seems to solve.

echo

It solves absolutely nothing compared to just counting up instead of fucking around with counting down. There is no advantage to fiddling with roll under. None. It is in no way better than having a fixed TN and counting up.

Let's say for a moment that you rolled really well and you forgot a modifier and wanted to go back and add it in. How is it easier to have your "good rolls" be the ones that are razor's edge from not having succeeded? What the fuck?

-Username17

It does have one advantage. Because the range is inherently limited supplement writers can't totally break the system by adding absurdly large stacking bonuses. That's not much of an endorsement, as good editing and plays-testing does that much better, but it is an advantage.

[DSMatticus]

Because the range is inherently limited supplement writers can't totally break the system

Limited to what, a 100% success rate (effective skill of 100)?

It's a strange argument to make that it's harder to break because it prevents you from succeeding more than 100% of the time.

[Username17]

Frank - I see what you mean, but I don't see how it's any different than having an environmental factor affect the DC.

You realize that Roll Under is equivalent to fixed DC roll-over, right? As in, you already paid the price of being unable to have hidden modifiers or post-roll surprises by fixing the roll-under TN to the character's skill (or the roll-over TN to 100). So anything that makes you jump through the hoops and calculations of varying the DC on top of that is automatically shitty.

If your system "always has a TN of 100" that makes certain things easier and limits certain other things. In roll under, you have all the limitations of fixing the TN to 100, but as you just artfully conceded, on top of that you have all the difficulties of having a variable TN.

Check and Mate.

-Username17

[TheFlatline]

Having run long-term games with "roll under" systems, it's a frickin pain in the ass to do a string of math.

In fact, I'm going to flat say it. Yes, a percentile system is great for granularity, but continuously adding in penalties and bonuses to calculate each fucking roll is akin to pulling teeth.

I never finished, for example, a session of Dark Heresy that was combat heavy without a splitting headache at the end of the night from running the entire game and doing modifier calculations for a couple hundred rolls.

The math itself isn't hard in itself, but mentally adding columns of 4 or 5 numbers together and subtracting 2 or 3 others while keeping track of a game, and doing that maybe 10 times a combat round (4 players and say 6 NPCs) means I end up focusing on the math and *not* on running combat in an interesting and engaging manner.

And it's a problem, since in most percentile systems I've seen modifiers are handed out like fucking candy, and forgetting to take them into account can sometimes result in a 20% swing in the target number, which is significant.

[echoVanguard]

A roll-under system with no situational modifiers is weird. That's certainly not a normal case. Roll-unders depend on situational modifiers to create easy and difficult tasks, and in the case where the status quo is that there are no situational modifiers that implies many tasks are no harder or easier than others.

A good example is the AD&D Skill System for Rogue characters. The vast majority of rogue skill checks were made with no penalty or adjustment, with the exception of halving your Detect/Disarm for magical traps.

I agree that it's not an optimal system for combat, which I probably should have made clear earlier in the thread, but I do think it has merit for the use in a skill-based subsystem.

You realize that Roll Under is equivalent to fixed DC roll-over, right? As in, you already paid the price of being unable to have hidden modifiers or post-roll surprises by fixing the roll-under TN to the character's skill (or the roll-over TN to 100). So anything that makes you jump through the hoops and calculations of varying the DC on top of that is automatically shitty.

If your system "always has a TN of 100" that makes certain things easier and limits certain other things. In roll under, you have all the limitations of fixing the TN to 100, but as you just artfully conceded, on top of that you have all the difficulties of having a variable TN.

I concur that it its statistically equivalent, but disagree that it is equally worthwhile. Fixed DC roll-over (or TN100, as I've referred to it thus far), has several idiosyncracies which I posted previously that can make it less desirable in certain situations.

I don't disagree that TN100 is a perfectly workable system. I do disagree that it is inherently superior to roll-under in every possible circumstance. I think that the most complete assessment is that they are both viable systems with complementary strengths and liabilities, and that TN100 tends to work better in a broader range of circumstances.

Math

This is a really great point. One of the patterns I'm definitely seeing emerge from this discussion is that roll-under suffers exponentially the more modifiers you apply.

echo

[DSMatticus]

One of the patterns I'm definitely seeing emerge from this discussion is that roll-under suffers exponentially the more modifiers you apply.

The pattern you should be seeing is that they are immediately inferior in every way the second you have to apply any modifier that isn't +/- 0. As in, rollover has tons of problems, and it's only 'advantage' is that you can roll against your skill, allowing you to construct a skill check for people with varying levels of skill while only doing one comparison (roll vs skill) and no math whatsoever. If the normal situation isn't an unmodified roll vs skill check, the system is stupid for using roll-under. If you're going to do math, you use the system where the math is easier to do and modifications are easier, and that's roll-over. If the system uses opposed rolls and is a roll-under, someone needs taken out back and shot.

The fact that you can make roll-under work in those situations is not impressive, because I can do the same thing in a roll-over with people doing half as much work and mental arithmetic, and keeping track of half as many numbers.

[CCarter]

The pattern you should be seeing is that they are immediately inferior in every way the second you have to apply any modifier that isn't +/- 0. As in, rollover has tons of problems, and it's only 'advantage' is that you can roll against your skill, allowing you to construct a skill check for people with varying levels of skill while only doing one comparison (roll vs skill) and no math whatsoever. If the normal situation isn't an unmodified roll vs skill check, the system is stupid for using roll-under.

Add to this that basically the point of rolling d100 is to have an exact percentage. If your system goes down to a 1% detail level, its more likely that you're going to have +2% or +3% modifiers being applied that wouldn't even be possible in a d20 system. If you're not going to add -2% to the archery roll for the wind blowing the wrong way, you may as well stick with d20.

[echoVanguard]

basically the point of rolling d100 is to have an exact percentage. If your system goes down to a 1% detail level, its more likely that you're going to have +2% or +3% modifiers being applied that wouldn't even be possible in a d20 system. If you're not going to add -2% to the archery roll for the wind blowing the wrong way, you may as well stick with d20.

There are other reasons to use d100 scales - most commonly, more granular advancement.

echo

[Psychic Robot]

The main reason why I am shying away from the d20 system in favor of the d100 system is that the d20 system has so much baggage attached to it. Best to avoid that entirely.

[CCarter]

There are other reasons to use d100 scales - most commonly, more granular advancement.
echo

Well, maybe. I guess maybe having -1% to surprise per level worked for Synnibarr since the designer wanted the game to go to 50th level...


On-topic (i.e. fixing Warhammer) - and I don't know if this is within the PRs parameters so far?.. running out of ideas here but HarnMaster is d100 roll under (somewhat like Runequest) and it uses a success system where the tens place of the d100 is a critical success (or failure, if you failed) if the ones place is an 0 or 5.
You cross-reference Critical Success, Success, Failure and Critical Failure results or both attacker and defender to find an attack's success (in number of damage dice) i.e. attacker critical success vs. defender critical fail gives the most damage.
IMHO its ass, but still marginally better than having a dodge roll automatically dodge. There would of course be complications for Warhammer since you'd have to roll hit location separately instead of just reversing the d100 roll.

Or..idea 2...keep with the usual system have a bidding system for attacks/defenses. The attacker can choose to roll to hit at a penalty, and the same penalty then applies to the defender's dodge roll. You could potential name and flavour combat moves at particular penalty levels ('I stab him with Accurate Attack/Inescapable Blow/Feint), and/or add other rider effects.

[Draco_Argentum]

Couldn't you just use a dice pool system? The wargame is all about rolling handfuls of d6, I'm sure any real Warhammer fan would be cool with using that.

[Psychic Robot]

Couldn't you just use a dice pool system? The wargame is all about rolling handfuls of d6, I'm sure any real Warhammer fan would be cool with using that.

Well, yes, technically.

Anyway, something I've been thinking of is applying a penalty to WS/BS checks based on the target's Agility without any sort of complex calculations.

• �0-15%: +20%.
• 16-30%: +10%.
• 31-45%: +0%.
• 46-60%: -10%.
• 61-75%: -20%.
• 76% - 100%: -30%.

I'd probably cap the bonuses to WS/BS at a maximum of +30%, all things factored in (on top of aiming and whatnot), just to make the math a little easier and to avoid total RNG fuckery. Shields would probably worsen the penalty by 10% (so the really agile characters wouldn't even need shields).