High-Level Thief Abilities

On the odds for danger follow-up post, we got into a discussion about thief abilities. The Greyhawk 14th level Master Thief maxes out at 100% for every ability. My own thief starts with a 5+ on 1d6 chance of success (2 in 6, in other words,) adding the thief's hit dice to the roll; so, at 7th level, a thief has a 6 in 6 chance of success. The question Pearce asked: what happens when thieves hit that threshold?

I do treat it as automatic success for typical thief tasks, but not for atypical tasks. Some tasks may have penalties, which means there's still a chance of failure. Which I guess raises the question: what are the typical thief tasks? I'm a firm believer in thief abilities being slightly unnatural, not just mundane skills that anyone can learn. Here are the Greyhawk descriptions, and my expansions in emphasized text.

"open locks by picking or foiling magical closures"

Emphasis added. Original ability was not mundane, no need to expand.

"remove small trap devices (such as poisoned needles)"

Anyone can learn to remove small traps (5+ on 1d6 chance,) but failure triggers the trap. Thieves get a bonus on the roll.

"listen for noise behind closed doors"

Anyone can do this, but thieves have better odds and improve at it.

"move with great stealth"

This is total silence; it is impossible to hear the thief if the roll succeeds. Otherwise, the thief has normal surprise chances. Lasts until the thief does something that might change this (pick up something that could make noise, or deliberately make noise.)

"filtch items and pick pockets" [sic]

Learnable by anyone, but thieves get really good at this. Subtract 1 point if the victim is higher level.

"hide in shadows"

Anyone can hide behind an object, but only thieves can hide "in plain sight" with nothing but shadows as cover. Remains hidden until thief decides to step out of shadows or a different kind of vision makes the thief obvious; if the light changes but there are still shadows, thief gets a new roll to remain hidden.

"strike silently from behind"

Expand to any surprise attack.

"climb nearly sheer surfaces, upwards or downwards"

No equipment needed, must have at least tiny cracks. Non-thieves can normally only climb with equipment. 5+ on 1d6 roll is for failure, not success; subtract hit dice from roll. Add 1 point each for slippery or crumbling surface.

Which leads to another idea: after 7th level, when thief abilities succeed automatically for typical situations, I think they should have an extra "impossible" level for some abilities. The odds for the new ability is a flat 5+ on 1d6 for all thieves of level 8+, with no adjustment for higher levels. The additional abilities are:

• Remove small magical trap (not area traps)
• Hear a whisper in the same building
• Disappear into shadows while being observed
• Climb perfectly smooth surface (glass)

Intelligence and Spells

In case it wasn't clear from my spellbooks house rules, I don't use the Intelligence Table from Greyhawk. That's the one that sets a percent chance to know any given spell, and a maximum number of spells knowable per spell level. I see spells as equipment, something for Magic-Users to purchase, acquire, or make themselves. Spells are certainly more powerful than mundane equipment like weapons, but there aren't any D&D Strength-based limits on which weapons can be used, although there are in The Fantasy Trip... and there aren't any limits on how many weapons a Fighter can know how to use, before AD&D and its weapon proficiencies. So why limit the number of spells a Magic-User can learn? There's already a limit on the number of spells usable per day, and there's a limit based on cost, although many groups drop those limits entirely.

It's not that I think Intelligence shouldn't have an effect on magic at all. It already affects the speed of advancement to the next level. I'd probably use Int instead of Dex when figuring out who goes first in a spell duel. Beyond that, I think the spell research and spell prep times might deserve modification. I've tinkered with spell prep times before, but the simplest approach is: a Magic-User spends downtime prepping spells before going on an expedition. No need to track exact prep times that way. Re-memorizing spells in the dungeon counts as "emergency memorization":

• Base time: 1d6 turns per spell, regardless of level.
• If Int is higher than 3 x spell level, halve the time required.

Under these rules, the miserable Int 3 Magic-User will always have a chance of taking one hour per spell, depending on the luck of the roll. Higher Int casters can cut the time for some or most spells.

I'm toying with a third rule: halve the time if the caster's level is 3 x spell level. This would stack with the Int rule, so 3rd level M-Us with Int 4+ would be able to re-prep a 1st level spell in 1-2 turns, max.

Odds for Danger, or Ability Check?

Fitz-Badger made a comment on my post yesterday about setting odds for danger:

I guess the 5+ on 1D6 is best for fairly arbitrary things or at least things where character's skills, attributes and abilities don't play a strong role (or if they do you bump the roll up or down to suit)

Not to single him out, but I think it's best for everything, which is kind of the point of the post.
We tend to emphasize ability scores too much, in this game where originally ability scores had little effect. If you change the roll to cross a bridge to a d20 roll less than/equal to Dex, instead of a 5+ on 1d6 roll, what have you accomplished? Average Dex characters will have an increased chance of falling, Low Dex characters may have double the normal chances of falling, or more. And Dex 17 characters wind up with almost a 1 in 6 chance, with Dex 18 only marginally better. We haven't really made things much better for most characters by using a Dex check.

We could allow a +/- 1 on the "5+ on 1d6" roll for significant ability scores, which I do in some cases. But more and more, I prefer using ability scores as a trigger for when to make a roll at all. Always include an option where the players can avoid rolls, only require a roll when operating beyond that scope, for characters with an ability below a certain threshold.

It's part of my view that rolls should not be skill rolls, but situation rolls: chances that something may go wrong, or go right.

The Way Spellbooks Work

After writing up my new take on Read Magic house rules, it occurred to me that it affects my view of spellbooks as well. I've already mentioned that I see "spellbooks" as mostly being the charts and names and procedures for producing spell effects, rather than individual spell descriptions. In other words, an ephemeris, a table of correspondences, a list of sigils for the spirits and intelligences of the planets and constellations, a table of hours... that kind of thing. Magic-Users who lose their spellbooks don't lose any of their known spells: they just need to locate new copies of those charts and tables. Spellbooks might, however, include research notes for a new spell that is not common knowledge, so capturing a spellcaster's spellbooks might be useful. I do not follow the "casting spells from spellbooks" rule seen in some people's house rules or in Delving Deeper/The Fantasy Trip.

My new rules doesn't alter that basic take, but it does alter unique spells and acquiring spells a little.

Spell Research isn't as important

I had a rule before that PC spellcasters get the default list of 1st level spells and must research all other spells. Now, all they have to do is find a scroll or casting notes for that spell and use Read Magic. Spell research only applies to high- level spells (7th+) or creating unique spells.

Scrolls and casting notes become a way to trade spells

Casting notes cost half as much as a scroll, since they can only be used to learn a spell, not to actually cast it without memorizing it. Spellcasters can charge whatever they think they can get on the open market for a spell.

Spellbooks for known spell levels don't need to be identified

It's like Nethack: once you identify something, you will always recognize it. Any Magic-User can cast 1st level spells, so they will automatically recognize a 1st level spellbook. 4th level M-Us can't cast 3rd level spells, so they need to cast Read Magic to identify a 3rd level spellbook.

Unique spells can be learned from a spellbook mid-adventure

They can't be learned all at once, because of the 1d6 spell levels limitation I added. But if an M-U can cast 6th level spells and finds an enemy caster's spellbook with Unleash Kraken (unique 6th level spell) in it, that M-U could prep Unleash Kraken right there, right now. And if that M-U finds a scroll of Unleash Kraken later, that scroll is immediately recognized, without casting Read Magic.

Spellbooks will contain unique spells, if a caster has any

When researching a completely novel spell, the final casting notes are added to the caster's spellbook. Cost is included in as part of the research costs. If the researcher specifies beforehand, the casting notes can be in a separate manuscript and not the main spellbook. So an enemy caster who casts Unleash Kraken must have notes on that spell somewhere.

I can see some possible changes to amnesia and Feeblemind effects. I might have to cover that in a future post.