Well, Mike Mearls got promoted. Any hope for 5e?

[tzor]

Yeah, the problem comes in when it's not a batter facing a pitcher but a rock climber facing a wall.

But it really is the same thing. Sometimes an action is just an action. They don't need to scale and they don't need to be considered. Walking across a room, climbing a wall. It's an action, not per se a challenge of your skill. Those professional baseball players throwing the ball back and forth before the game begins. It's an action they do and succeed with.

But when that challenge becomes the obstacle to overcome then it has to be challenging. That walk suddenly is over a rocking board with fire pushing searing heat from below. The cliff is now tiling backwards. Because it is supposed to be a challenge it is supposed to be an opponent. Anything that is not a challenge becomes a SOP as easy as ... well walking.

(NB: Climbing and walking start to look silly when everyone has flight, not all skills are usefull ad infinitum.)

It's not that everything gets tougher, it's that the things that matter get tougher, and the rest just gets glossed over.

If in the middle of an ordinary combat, a high level character wants to climb a wall for some minor advantage, then he does it, as easy as a lower level character would have had walking across the room. On the other hand if a high level challenge requires high level climbing then it better be worth the effort to set it up and describe the damn thing.

[hogarth]

But it really is the same thing. Sometimes an action is just an action. They don't need to scale and they don't need to be considered. Walking across a room, climbing a wall. It's an action, not per se a challenge of your skill.

So when you were playing AD&D, and a heavily armored fighter or a wimpy magic-user wanted to climb a wall or jump over a pit, you would just say "Sure, you succeed" every time? I wish you had been my DM.

[icyshadowlord]

But it really is the same thing. Sometimes an action is just an action. They don't need to scale and they don't need to be considered. Walking across a room, climbing a wall. It's an action, not per se a challenge of your skill.

So when you were playing AD&D, and a heavily armored fighter or a wimpy magic-user wanted to climb a wall or jump over a pit, you would just say "Sure, you succeed" every time? I wish you had been my DM.

Ever heard of Free Actions? Like, for example, talking?

Okay, that was a VERY lame throw on my part, but I see what you're both getting at. Unfortunately, I have no witty comeback to either of those.

[CapnTthePirateG]

New Mearls article!

http://www.wizards.com/DnD/Article.aspx ... l/20110823

Ok, looks like skill ranks come back, but in that form from last week and determine whether you can make the check. Then you just use your ability modifer. So it's the 3.5 skill system, but with the skill ranks rounded.

This could actually not suck. I'm surprised.

The opposed checks seems to be that the person with the lower rank checks makes a check against a DC pulled out of someone's ass. One rank level difference (i.e you beat a dude one level higher than you) is a DC of 20, which, to me, seems to be kind of a "fuck you" level as you still have a 7/10 chance of failing with a 20 in your attribute. We still don't know how ability rates will progress in Mearlsland yet.

[Username17]

I'm going to disagree with you. That's definitely going to suck. There is a lot of hand waving, but ultimately what he's talking about with the stealth vs. perception checks is getting things really complex for no apparent benefit.

Obviously what he is talking about isn't remotely fleshed out, but let's consider the checklist for doing an opposed test:

• Compare skill ranks, set DCs.
• Rogue makes stealth check against his DC.
• Cleric makes perception check against his different DC.
• Subtract DCs from results and compare modified results.

Now there are basically two ways this could shake out. Either:

• The DC push from differing rank totals is linear.
  or
• The DC push from differing rank totals is not linear.

I think it is trivially obvious that these two possibilities consist of the entirety of the potential play space. And yet: both options are atrocious, albeit in very slightly different ways.

In the first possibility, that is exactly the same as 3.5 from a mathematical standpoint. The only difference is that instead of adding ranks to stat bonuses and rolling and then comparing, you are separating out the ranks and comparing, and then taking that comparison and using it to generate THAC0s for both participants and then having them roll their stat modifiers and doing separate comparisons and then comparing the comparisons. By repeatedly separating steps and inverting signs, Mearls will have transformed a simple comparison operation into a nightmarish four comparison staged scenario that produces no extra or different outcomes. Holy shit.

In the second possibility it is less atrocious, since at least it does something that 3.5 does not already do. Of course, it's still fucking awful, since you could get the same result with only two comparisons by handing out a flat bonus for being more skilled during an opposed test (with that bonus scaling in whatever non-linear fashion you desired).

The fact that someone would go to print with something that is so obviously retarded would be disturbing if it wasn't for the fact that we're talking about Mike Mearls and he lost the ability to surprise me with his hubris and stupidity long ago.

-Username17

[Psychic Robot]

mearls's idea is not terrible. however, without seeing an actual system, all I'm left with thinking is, "couldn't we just fix the RNG instead?" as he describes it the system feels very clunky and requires metagame knowledge

Under this system, a skill or an ability contains its own DC. Let’s say I have a 16 Dexterity and expert training in that ability. I need to sneak up on a cleric who has a 14 Wisdom and journeyman training. There’s no opposed check. Instead, the active party, the sneaking guy, makes a journeyman Dexterity check. My expert training wins, meaning that I can sneak up on the cleric. so wait the cleric doesn't even get a spot check now?

However, let’s now say that the cleric is suspicious of being followed. He takes a moment to scan the area and make a Wisdom check. That check’s DC is matched against my expert Dexterity. He is one rank below me, so he needs to make a DC 20 Wisdom check to notice me. so you have to cross-reference the difference between the characters to know the actual DC rather than simply making an opposed check. rolling a +12 for the elite assassin vs. the +4 of the novice cleric seems a lot easier to me imo

Let’s assume that I’m playing a rogue and I’m tracking the cleric in a shadowy ruin. The cleric, suspecting that he’s being followed, casts a light spell down the hallway where I’m lurking before he makes his check. The DM can then drop the DC down one level, to journeyman. The cleric’s target number is now 10. Just as in the system described last week, players (and DMs) are encouraged to engage with the game setting and come up with ways to tilt the odds in their favor. so we're supposed to tell the players that they're going to autofail so that they engage with the system?

okay so how does this even work. why do we have varying DCs and cross-referencing stats and it sounds like you're going to have to say "well the rogue has +1 skill levels on you so he automatically wins, you had better do something to make it easier to spot him" and what happens if the rogue fails his first stealth check then? does the cleric automatically spot him or what?

yet another problem: what about tracking NPCs. the 4e skill system is quick and easy, and I would love it if the skill list weren't neutered and the RNG was a little more solid. since I'm the kind of guy who like statting up NPCs using PC classes (I'm aware this is a no-no in 4e) it makes dealing with them much easier than using the 3e skill system. with the skill tag system, I can easily calculate a skill number on the fly and not have to worry about anything--just roll an opposed check and see if the monster spots you. with this system however, we're looking at a bunch of extra, behind-the-scenes math which works well when you have everything planned out exactly in advance. but what happens if the PCs are trying to follow an NPC that isn't statted up? using the 4e example, if they were trying to follow a ranger, I'd say "okay he's around level four with a 16 Wis and trained in perception, so he's going to have a 5 + 2 + 4 = +11 perception modifier, make your roll" (I'm aware they have a system for passive perception but I dislike that). with this system I have to decide his wisdom modifier and his specific level of training and then recalculate the DCs based on that

like all mearls ideas this one has promise but lacks execution. also I'm sensing a bit of FUDGE/SotC in this system where you're rolling your "extraordinary" stealth or whatever vs. the "average" spot check of the cleric.

imo if mearls wanted to do this right he would do something like the following (using the rogue vs. cleric example):

1. the rogue makes a stealth check. does he succeed? IF YES: he follows the cleric because his skill level is so high. IF NO: his skill level drops by one for the encounter; proceed to step two.

2. is the rogue's skill training now equal to the cleric's? IF YES: make an opposed roll using the normal rules and proceed to step three. IF NO: return to step one.

3. who won the opposed roll? ROGUE: sneak successful. CLERIC: notices the rogue.

obviously this is far more complicated than simply making an opposed roll but if mearls wants this weird kind of granularity then so be it.

alternatively what I might do is give characters a number of "training bonuses" for their skill training (untrained has zero, basic training has two, expert training has four, etc.). at the start of each encounter, these pools refresh and allow them to reroll their skill checks or perform special abilities, each one taking a certain number of bonuses (for instance a sneak attack might take two points where hide in plain sight might take three or more). obviously this wouldn't really work for the D&D system as-is but it's just a thought

[CapnTthePirateG]

I wasn't saying the opposed checks didn't suck ass, just the non-opposed ones. Sorry if that was unclear. Thanks for the great analysis though, damn I hadn't seen the crazy ass steps.

My first thought of the first part was "dammit, it's just like 3.5" but I wasn't sure if I was on the right track.

[Psychic Robot]

oh okay here's his article of the week before. he goes on for a bit about basic problems with the d20 system (specialists significantly outclassing non-specialists) and his solution is as follows:

In this world, a check DC is no longer a number. Instead, like several rules-light gaming engines ahah I was right, he has been playing too much SotC, it uses a series of descriptors that illustrate the minimum skill level needed to attempt a task. This list gives the ranks from lowest to highest:

* Novice
* Journeyman
* Expert
* Master
* Grandmaster
* Impossible

When the DM asks for a check, you compare the DC to your rating in the appropriate skill. Impossible is essentially a placeholder, a rank that applies to things that the DM or the rules deems are doomed to failure. For example, a character who leaps off a cliff and flaps his arms in order to fly is attempting (and failing) an impossible task.

If your skill rank is greater than the task’s DC rank, you automatically succeed. You are so skilled that you can complete the task without any special effort. Think of a tightrope walker at the circus. She has enough training and experience that performing her act is an automatic success. It would take some outside factor, like a sudden injury, an equipment failure, and so on, to cause her to fall. On the other hand, as a sedentary game designer I wouldn’t even imagine trying to walk across a tightrope myself. I’d fall after a step or two.

If your skill rank equals the task’s DC rank, you need to make a check with a result of 15 or higher to succeed. You’re skilled enough that you might succeed. In this world, skill checks use an ability score modifier (chosen to fit the task by the DM; a skill uses whatever ability is the best match for the actual action) with perhaps a small modifier based on feats or a skill bonus.

Going back to our tightrope walker, perhaps an earthquake strikes in the middle of her act. As the rope sways, the DC shifts one category up. Now she has to make a check, perhaps with her 18 Dexterity for a +4 bonus as well as a +4 bonus from a feat or other benefit she took. That gives her a 65% chance to remain on the rope.

If your skill rank is below the task’s DC rank, you automatically fail. Your training and experience are not enough to complete the task. Going back to the tightrope walker, let’s say that as the earth shakes she also steps on a length of the rope that her rival covered in grease. The difficulty shifts one more category up, causing her to fall to the net below. autofailure is going to piss people off even if it does make sense. for instance there's no way I could walk on a tightrope but players are going to say BUT WHY CAN'T I TRY. imo better to add a hefty penalty to the task instead allowing for extraordinary success

The DC is based on two things. First, the DM determines a task’s basic difficulty. Then, for each element in play that makes it more difficult, he shifts the difficulty down one rank to the harder level. For elements that make it easier, he shifts it up. Typically, the DM informs you of the DC before you make the attempt. A hidden threat might mean that you don’t know the true DC until you make your attempt, but that’s the exception rather than the rule.

Going back to our hapless tightrope walker, perhaps she is carrying a pole to help balance her. The pole gives her an advantage that shifts the DC down one category. When the earthquake hits, she still keeps her balance. When she steps on the greased section of rope, she must then make a check.

Benefits of this Approach
I will readily admit that is a radically different approach to skills in D&D, one that looks nothing like anything in any version of the game. I imagine that 90% of you are ready to lambaste it. However, it’s worth pointing out some of its benefits:

* It dramatically simplifies the math and removes the escalating bonus race. instead we have an escalating skill training race. how is this sufficiently different? if some guy is a grandmaster at sneaking and another guy is a journeyman at spotting, then the grandmaster is going to automatically win. it's basically the same as if you gave him a +30 to his skill...the only difference is that the numbers are smaller

* It speeds up play by eliminating die rolls in some cases.

* It makes DCs the same across all levels. An expert task is always expert level. We don’t need to shift difficulties assuming that your bonus continually increases because some characters can remain untrained. The system works by removing the link between difficulty class and level. Instead, we just use a simple, descriptive system as applied to reality. uh but couldn't you just make an expert task a DC 25 and leave it at that

* It allows trained experts to repeatedly achieve impressive results through practice and training. Just as in real life, a highly trained character can do amazing feats without any real risk of failure (barring any complications or unexpected hazards).

* It more closely models the real world (in my opinion, at least) by shaping how we approach tasks. We have an innate sense of things we can do without thought, things that we know to not try, and things that can be challenging. I know that I can walk a mile in 20 minutes without any real effort. I can jog a mile in 15 minutes with some (OK, a lot of) exertion. My sedentary butt would collapse long before I hit a mile if I ran at a 10-minute pace. I don’t need to try these tasks to determine this. I have 36 years of experience to establish what I can do.

* It makes skill training even more valuable because it grants automatic success in easier situations rather than a better chance of success. A sure thing is more valuable than improved odds (a bird in the hand versus two in the bush). A rogue highly trained in Acrobatics or Balance simply scurries across a tightrope while the fighter looks for another route. This approach actually encourages players to use their skills more often in dangerous situations by removing random, chance-driven failure as the norm.

* It encourages smart play and engagement. A player with a clever idea can shift the DC one level and turn a check into an automatic success, or an impossible challenge into one with a chance of success. I personally like this because it gives the DM a lot of leeway to use the system to shape his or her game. yes but couldn't that smart player also work to get situational bonuses? also I thought 4e was about saying "yes" to players not "no you automatically fail at this"

* It bakes “impossible” directly into the game. No amount of Diplomacy can sway the raging, bloodthirsty barbarian. It’s off the chart. This may sound like a minor thing, but I think it’s important to set the expectation that the DM can simply invoke common sense or logic to rule that a check will fail regardless of the die roll. Relying on the rules to set what are meant to be impossible DCs is simply asking for trouble. this already existed good lord

[hogarth]

New Mearls article!

http://www.wizards.com/DnD/Article.aspx ... l/20110823

Ok, looks like skill ranks come back, but in that form from last week and determine whether you can make the check. Then you just use your ability modifer. So it's the 3.5 skill system, but with the skill ranks rounded.

It seems dumb to graft the 3.5 skill system on to the system from the previous week, except only for opposed checks.

autofailure is going to piss people off even if it does make sense. for instance there's no way I could walk on a tightrope but players are going to say BUT WHY CAN'T I TRY.

This happens all the time in video game RPGs (e.g. in Fallout 3, I can't open a Hard lock unless I have 75 ranks in Lockpicking) and it's not particularly bothersome.

[CapnTthePirateG]

Of course, in videogame RPGs, you can't jump the fence either.

No one is suggesting that "because videogames get away with it" is an acceptable excuse for a tabletop rpg.

[hogarth]

No one is suggesting that "because videogames get away with it" is an acceptable excuse for a tabletop rpg.

I am. Really. (At least in this case.)

[Josh_Kablack]

This happens all the time in video game RPGs (e.g. in Fallout 3, I can't open a Hard lock unless I have 75 ranks in Lockpicking) and it's not particularly bothersome.

But in videogames I can

• Read a walkthough first to know that I'll need X ranks in ability Y when I get to area Z if I want to get through the door
• Go grind mobs somewhere else to level up then come back to the lock, which will still be there no matter how long I take (so long as I don't progress the plot to the next stage)
• Revert to a save file from before I leveled and distribute my advancement ranks differently this time.
• Start the game over (or new game+) to experiment with different ability setups in the exact same world

None of those are at all applicable to a TTRPG.

[Psychic Robot]

I don't care if Fallout 3 does it it's still going to piss people off

[DSMatticus]

"You must be this tall to ride" is fine. What it really helps you do is break up sections of the RNG without hyper-inflating it. If you want a level 1 character to not be able to do X, and a level 10 character to be able to do X 25% of the time, the difference between those two characters has to be +15 or so. And that's pretty much where the average lock should be: it should be a challenge for a level 1, and it should be somewhere in the neighborhood of 'walking and chewing gum at the same time' for level 10's.

There's nothing 'bad' about it that isn't bad in the same way as setting a DC higher than you want a character of level X to be able to beat.

[virgil]

Having there be tiers in the skill system should only really be used when they're present in the rest of the game (see SotC). You might as well just use SAGA/4E version of flat training bonuses and keep bonuses on the RNG for your desired DC chart. Same thing, fewer steps. 4E screwed up by never having a DC chart that took into account anything but the PC's level.

[Josh_Kablack]

If your skill rank equals the task’s DC rank, you need to make a check with a result of 15 or higher to succeed. You’re skilled enough that you might succeed. In this world, skill checks use an ability score modifier (chosen to fit the task by the DM; a skill uses whatever ability is the best match for the actual action) with perhaps a small modifier based on feats or a skill bonus.

Wait, whatthefuck?

I am a journeyman blacksmith - I attempt to forge a journeyman sword

I need to roll a 15 on a d20 to succeed? So you usually fail at tasks equal to your skill level!?

Oh wait, I get a bonus based on ability scores
And Why are those still numeric if skills are adjectives?

More importantly which ability is best for sword-forging?

Oh, right the ability of BS-the-MC-to-use-my-high-stat-anyway is key here. Didn't we go over why this was a bad idea already?

But here goes for this task:

Obviously Strength is the most important attribute because of the need to work the bellows and hammer the metal into shape. Obviously Con is the most important attribute because of the need for extended work in blazing hot conditions. Obviously Dex is the most important attribute because the blade must be hammered and ground to a fine point and the sword must have proper balance. Obviously Int is the most important attribute because I need to know about metallurgy, martinsitic grain structures, annealing, carbon content and japanse metal-folding. Obviously Wisdom is the most important attribute because there is no single best sword - it's about making the correct sword for the wielder and fitting it to their height, strength, fighting style, and matching it best to overcome the armor of their enemies. Obviously Charisma is the most important attribute, because that's the ability for Influencing spirits and Magic Devices and I am forging it with a +1 forge salvaged from the ruins of the Dwarf kings and calling for the spirits of my ancestors to aid me and aid whomsoever wields this blade.

So you'd have a fairer, more transparent system with the exact same success chances, but which involved one fewer additions per task if you just set the base task DC to 15 minus highest expected stat and didn't let people add their stats.

What do you have against fairness, transparency and streamlining resolution Mearls? What did they do to you when you were little to inspirie your irrational hatred of them?

Going back to our tightrope walker, perhaps an earthquake strikes in the middle of her act. As the rope sways, the DC shifts one category up. Now she has to make a check, perhaps with her 18 Dexterity for a +4 bonus as well as a +4 bonus from a feat or other benefit she took. That gives her a 65% chance to remain on the rope.

So +4 is a small bonus in your world?

So, unless they have a decent stat and some sort of feat bonus, a character is going to fail at rank-appropriate tasks more often than they succeed?

Dude, that results in the OPPOSITE of engaging with your system - players will either max out stat and feat bonuses on top of super-expert training to always auto-succeed at anything or they will do everything they can to avoid rolling the die and then piss and moan any time the MC says to roll anyways.

But the true dealbreaker in your latest fractal of fail you are somehow still getting paid to write is that you somehow think rolling 1d20+8 against a DC of 15 is a 65% success chance in your world?. It's like you are actively trying to prove your incompetence here.

[hogarth]

This happens all the time in video game RPGs (e.g. in Fallout 3, I can't open a Hard lock unless I have 75 ranks in Lockpicking) and it's not particularly bothersome.

But in videogames I can

• Read a walkthough first to know that I'll need X ranks in ability Y when I get to area Z if I want to get through the door
• Go grind mobs somewhere else to level up then come back to the lock, which will still be there no matter how long I take (so long as I don't progress the plot to the next stage)
• Revert to a save file from before I leveled and distribute my advancement ranks differently this time.
• Start the game over (or new game+) to experiment with different ability setups in the exact same world

None of those are at all applicable to a TTRPG.

I don't know what your point is. Are you saying that TTRPG skill systems should always allow a PC a chance to succeed at any given task, no matter how feeble the PC or how difficult the task?

There's nothing 'bad' about it that isn't bad in the same way as setting a DC higher than you want a character of level X to be able to beat.

What he said.

[Stubbazubba]

I don't know what your point is. Are you saying that TTRPG skill systems should always allow a PC a chance to succeed at any given task, no matter how feeble the PC or how difficult the task?

I do believe that's what nat 20 and/or nat 1 are for.

[hogarth]

I don't know what your point is. Are you saying that TTRPG skill systems should always allow a PC a chance to succeed at any given task, no matter how feeble the PC or how difficult the task?

I do believe that's what nat 20 and/or nat 1 are for.

A natural 20 is not a success on a skill check in 3E, and likewise for a natural 1.

[Stubbazubba]

Not a skill check, but for attacks, saves, and casting they are. Apparently the idea is OK for you in the primary mini-game of the system, so why are you so against having extraordinary successes in the skill system?

[echoVanguard]

But the true dealbreaker in your latest fractal of fail you are somehow still getting paid to write is that you somehow think rolling 1d20+8 against a DC of 15 is a 65% success chance in your world?. It's like you are actively trying to prove your incompetence here.

Yikes.

echo

[Lago PARANOIA]

Because it leads to things like Steve the Crap-Covered Peasant being able to forge Excalibur or tell you how to integrate a nested trig function if they're persistent enough at it.

It doesn't offend us as much when Steve has a chance of getting through a Solid Iron Demon's guard and clocking him a good one with a staff, because even though he has a 5% chance of hitting there's still no way he can kill said Solid Iron Demon due to other factors. But when an extraordinary task is resolved in one or only a few rolls then it starts to create results which we find silly.

[Stubbazubba]

Because it leads to things like Steve the Crap-Covered Peasant being able to forge Excalibur or tell you how to integrate a nested trig function if they're persistent enough at it.

It doesn't offend us as much when Steve has a chance of getting through a Solid Iron Demon's guard and clocking him a good one with a staff, because even though he has a 5% chance of hitting there's still no way he can kill said Solid Iron Demon due to other factors. But when an extraordinary task is resolved in one or only a few rolls then it starts to create results which we find silly.

OK, agreed, and I thought of this just after I posted that. If one skill check determines the outcome of an encounter, then yeah, auto success or auto fail on nat 20 or nat 1 is unacceptable. This could be helped by having a more interesting out-of-combat encounter resolution scheme, of course. It could just be as simple as your Stealth has 3 levels; Undetected, detected, and spotted. So that even if you were 'detected,' you would at least have a chance to roll your stealth again and see if you could get back to undetected again. Then, if you fail that, and the detector overcomes your Stealth again, you are spotted. That way there are 2-3 poor rolls before you fail. On top of that, add that the detector can drop you all the way to spotted on the first roll if and only if he beats your Stealth by 10 or something, likely in a way that is handled like critical hits, in that you have to roll again to confirm it. That way you still have the possibility of being spotted, the same way you have the possibility of being killed in combat in one shot, but it's a long, long shot.

[DSMatticus]

Stubbazubba, there's an alternative mechanic in Unearthed Arcana where, instead of 20's/1's being autosuccess autofail, a 20 is a "add 20, roll again" repeated, and a 1 is a "subtract 20, roll again," repeated. This does nearly exactly what you're looking for, and it mostly maintains the impossibility of Steve the Farmer knocking down an adamantine wall. It has its problems, but it's quick and easy.

You really don't really want to create a skill minigame where the only action is "repeat until you succeed or fail," unless it's quickly executable.

[Username17]

Exploding d20s is really not very different from 20s automatically succeeding. A 10th level task is what, DC 30? If you thought it was ridiculous for Steve the Crap Covered Farmer to succeed 5% of the time on this task (whatever it was) that was so difficult that it was a normal task for a character so high level that he wouldn't even get XP for besting an unlimited number of Steves... do you really think that your objections are going to magically go away when Steve has a 3% chance to complete that task?

-Username17