Quote from: DestroyYouAlot;621698To the OP:  One of the things I've been getting into lately is the idea of the thief skill roll as a backup.  Think about it this way:  Before the thief was added to the game, any character could disarm a trap.  You got a description, you interacted with it Q&A-style (I have begun to call this the "text adventure" method), and - failing that - you had your basic 2-in-6 chance to disarm a simple trap.  So the thief's 15% (or whatever) chance to disarm is ON TOP of that (if both those methods fail).  It was a real revelation when I started to look at it this way.  Suddenly the thief gets a lot better at his job.

Yeah, that's a good idea, I'm going to try this out next time. When I ran 2e last year the thief did seem like the weakest PC, although this was also because after he found a +1 sword, he tried to play like a fighter.

So in effect the thief abilities become

Move Silently (as opposed to the automatic ability Move Quietly, that every PC gets by describing themselves taking relevant precautions) allows automatic surprise maybe? And can be achieved at any speed?.  

Hide in Shadows (as opposed to Hide Behind Stuff, that every PC gets by describing themselves as hiding behind something sufficient) lets thieves try to hide in implausible places such as the corner of the room or 'in the half-light'.

Climb Walls needs no rope etc and can act as a save against falling whenever the thief tries to climb something like a normal PC (ie using a rope and making a Dex check).

Find Remove Traps is effectively a saving throw for when the player's described precautions wouldn't be enough to automatically find the trap through exploration. Ditto for Open Locks and Detect Noise.  

Pick Pockets is something that only thieves can even try maybe?

Should thieves also get some kind of Spot Secret Doors and Dodge abilities? Maybe combine both with Detect Noise for an Extraordinary Senses % ability?

Quote from: soviet;622128Move Silently (as opposed to the automatic ability Move Quietly, that every PC gets by describing themselves taking relevant precautions) allows automatic surprise maybe?

That's how I've been handling it.  (Normal surprise rolls should assume the PCs are at least attempting to move quietly.)

QuoteHide in Shadows (as opposed to Hide Behind Stuff, that every PC gets by describing themselves as hiding behind something sufficient) lets thieves try to hide in implausible places such as the corner of the room or 'in the half-light'.

Ditto.

QuoteClimb Walls needs no rope etc and can act as a save against falling whenever the thief tries to climb something like a normal PC (ie using a rope and making a Dex check).

Worth noting that B/X calls it "Climb Sheer Surfaces".

Aside from http://web.fisher.cx/robert/infogami/On_thief_skills_in_classic_D&D what I've thought of doing is saying that for stealth activities, a non-thief will generally leave a chance for the "opponent" to detect by some means, whether it be a perception roll against an ability (or saving throw, etc.) or a straight die roll (2 in 6 or whatever). The DM could even decide arbitrarily based on situation. Whereas if a thief makes the attempt and succeeds, it's no longer an issue of whether the opponent hears/sees; rather there is nothing to hear/see. Even here the DM could overrule (e.g. if there simply are no shadows large enough, or if someone is looking straight at the thief before the hide attempt is made) but should generally give the player the benefit of the doubt.

The reason this stands out in my mind is that the thief skills are pretty awful at low levels in some versions of D&D. Interestingly I don't think that's true for all versions.

Quote from: DestroyYouAlot;622130Worth noting that B/X calls it "Climb Sheer Surfaces".

Random footnote:  I only noticed 'cause I just received a LOVELY copy of "Shadows over Bögenhafen" and I was scanning the handouts, but Warhammer FRP 1e has the "Scale Sheer Surface" skill, which specifically exempts the character from risk tests when climbing (except for especially difficult climbs).  Just another perspective on how this might be handled (especially relevant if, like me, you see WFRP as basically "old D&D smashed into Call of Cthulhu").

The following is assuming you want to run a RPG with a focus on roleplaying in an IC manner.  If you just want to play "old school" by going Full KoDT, then just do whatever is funny and fun and blow the doors off the place.

As has been said, Roleplaying not "narration".

Die rolls for running down stairs when chased by zombies if you're looking at things from an "old school" roleplaying perspective (as I think you are meaning the term) aren't made because they are "interesting" or because the consequence for failure is dramatically important, but because running down stairs when chased by zombies has a pretty good chance of failure, hence you should check.

If you are familiar with an NPC enough to know how they would react in a given situation to a PC, then have the NPC react appropriately.  If you're not sure, then roll using the Reaction Tables.

As an Old School GM, you Roleplay the inhabitants of the setting from animals to gods as honestly and as impartially as possible based on how those beings would actually act, randomly determining (with probability aided by rules sometimes by not always) when the outcome is not clear.

When the outcome of the player's actions is not clear, you use the game system to resolve it.

Quote from: soviet;620510So after re-engaging with AD&D 2e last year I'm interested in exploring old school play again. As I understand it, one of the key tenets of old school play is the use of narration as a resolution method wherever possible. So instead of making a check traps roll, describe to the GM how you open the chest.

Proper Old School would be to roll for everything.

New Old School would be to roleplay and then roll for things that need rolling.

New School would be to narrate everything.

A sensible compromise would be to narrate those things that need narrating, roleplay those things that need roleplaying and roll those things that need rolling.

The flip-side of using less rolling to decide what happens in conflict is using more rolling to decide what the conflict will be in the first place (wandering monsters, random encounters, reaction tables and a bunch of other random tables).

Quote from: soltakss;622351Proper Old School would be to roll for everything.

New Old School would be to roleplay and then roll for things that need rolling.

New School would be to narrate everything.

Amusingly, I don't think you're right about any of these.

RPGPundit

Quote from: RPGPundit;622619Amusingly, I don't think you're right about any of these.

RPGPundit

Yup. All three are wrong.

Quote from: soltakss;622351A sensible compromise would be to narrate those things that need narrating, roleplay those things that need roleplaying and roll those things that need rolling.

But he's right about this bit, with the understanding that what activity falls into what bucket will vary by DM and group.

Quote from: VectorSigma;622645But he's right about this bit, with the understanding that what activity falls into what bucket will vary by DM and group.

More or less.

RPGPundit

I reckon it depends on which "old school" you want to take marching orders from.

Obviously, the old school that says (a la Arneson, Gygax, St Andre) "Do whatever is fun for you!" won't be any help.

We dare not defy the gaming ideological purity of either Ron Edwards or Pundit.  Glory before fun, after all. Hail the gaming popes!

Quote from: Benoist;622628Yup. All three are wrong.

Nope, speaking from my gaming experiences circa 1984 playing AD&D 1, that's pretty spot on. The whole 'ruling before rules' thing is pretty much a reaction against 'rules matters' as purported by the Storygames movement and as a reaction to the 3.x optimization craze (which was a modern manifestation of what was called min-maxing.)

Nearly every GM I knew back in the 80s, especially those in D&D, made it a big deal that they were interpreting rules objectively, especially for square movement, checking for traps and secret doors, and use of thief abilities. If you made alternative rulings, you were looked down upon as some kind of loosey goosey hippy. I remember more than one conversation at Origins in Baltimore in 1984, for instance.

Would it be fun for a PC to slip in the bath and break his neck, or for a planet-slamming asteroid to wipe out all the characters in the campaign? If not, then don't roll for it!

Even if a low-probability occurrence is in itself fun, a lot of rolling for it may be tedious.

If a question is of no great significance, then coming up with an answer probably does not warrant a great investment of time. Tastes differ, of course; if a group gets a kick out of devoting a quarter hour to the carrying capacity of a mule, fine for them -- but that might be the wrong group for some other players.

Rolling dice can provide surprises for the GM. This can be a way of "keeping one honest" in a sense.

Is it fun to pile on a lot of modifiers? I think this usually depends on how much the players' choices influence those. Routinely invoking penalties to cancel out experience bonuses strikes me as adding labor for no purpose, but YMMV.