Lots of ways of doing it, although dice pools are fairly granular (not much fine distinction possible) so you might want to limit the distinctions to say two-handed/one-handed/light or similar categories. Maybe consider the perks possible when using a shield (or using two pointy things simultaneously) when setting the damage for big weapons.
 
One option might be to have a large weapon set aside damage from the to-hit roll as automatic successes - so you miss more often but likely do more damage if you do hit ? You could also have say 10s roll over for damage on big weapons, with smaller weapons instead getting the reroll on to-hit only?. Or give larger weapons x2 damage and small weapons get two attacks a turn instead. You could also add a soak roll to resist damage, with larger weapons having a higher soak difficulty. Lots of possibilities.

Quote from: Artifacts of Amber;801146Interesting thing about dice pools is you have to ways to modify, Number of dice and Difficulty or how you determine successes.

I can imagine a lot of fiddly things you could also do:

• Replace some dice with larger or smaller dice (e.g., d12 or d8 in the present case);
• Allow special rules for at most one die (e.g., if any die rolls 10, then it represents two successes rather than one, but only applies once to the entire pool - so two 10's is three successes, not four; or allow one die to succeed on one lower than the target, so if the target is 7 then three sixes would be one success plus successes from rolls above six);
• Roll a distinguished die that gets a bonus (lower target number or two successes if it reaches the same target number);
• Allow limited rerolls (e.g., no more than one die if it shows 1).

The difficulty I have with dice pools is that the success and damage (or degree of success) are all tied up together and it's hard to see the result of various ideas; it's easier with typical d20 to separate success factors from damage roll factors because they can be analyzed separately.

The thing with weapons is its hard to know how to fix it, unless we know how you handle damage. I go back to Basic D&D where weapons didn't make a HUGE difference, just a mild one. (Especially at higher levels.) In my own FRPG, High Valor, a weapons value is not separate from the skill or trait that drives, unless it is very high quality/folkloric-important/magical--at that point it is its OWN trait (and you can always use two.)

But High Valor has a very fiction driven damage system--either your hit and killed, or hit and wounded.  The quality of the wound depends entirely on the aims of the attacker, the quality of your defense (players only roll in the system. If you fail defense entirely--your PC died. If you manage some reasonable level of defense you suffer a wound--all wounds impair mechanics based on their fiction. There is none of this dramatics not equaling the result, for the dramatics are the result, as arbitrated by the roll.
You can't just whittle someone down like with HP, instead you have to make significant, notable, fiction driven wounds that impair them enough so you can basically out fight them and land a killing blow.  (Lopping of a dragons limb for example, might make it easier to fight and kill, depending on the limb. But take out a leg and its a flying creature? Its just going to rain some blood and be a worse foe.)


Because at the heart of any game, "realistic" or "tactical" or whatever, is going to be an abstraction because wounds and living creatures are funny things, they don't work by reliable persistent, always on rules. (They've got rules that are "mostly" true, with exceptions.)


So why not aim for the most believable for the table abstraction?


What is believable with your system?

My first suggestion is to check out The Pool, a free rpg that uses dice pools. It has an elegant and exceptionally clever method of "refreshing" that adds an element of strategy and gambling to using the dice pools that is incredibly fun and satisfying in play.

My second sggestion is to maybe look at the skill system in the old TSR conan game (retrocloned as ZEFRs I believe, free online). The way it works is you have a small number of skills, like your list, and then any number of specialist skills. Characters can spend XP to improve specialist skills, while the main skill goes up based on the total.  So for example if you have Athletics as the main skill, you'd have 1 dice in that. If you then took the specialties Sports (2), climbing (1), and swimming (2), then in normal cases you'd roll 1 dice, but when employing one of the specialties you'd roll 1 + the specialty (ex: when playing football, the aforementioned player would roll 3 dice). Once the specialties add up to, say, 10 dice, the main skill goes up by one die. (ex: a player with the specialties Sports (2), Climb (3), Swim (2), and Ride (3), would have an overall Athletics of 2. If they were to test for, say , horsemanship, they would roll 5 dice). This means that to improve overall in a main skill, a player would have to keep developing more and more specialties that fall under that skill's umbrella.

Putting a cap of 3 on a single specialty (I use the terms apprentice for 1 dice, journeyman for 2, and master for 3 dice), this will encourage the pc to become more well-rounded as time goes on and keeps them from advancing at crazy speeds.

That's incredibly useful; thanks everyone!  

I'll write a more detailed response tomorrow (Friday night here), but wanted to say how useful this has already been for me.

Thanks again for the helpful comments, questions and recommendations here (both the Pool and ZEFRs were well worth a look -- thank you TristramEvans).

Skills
I haven't really resolved what I want to do with skill progression. As there is no default statistic for tasks and my players want to do a wide range of different tasks, I'm not overly worried about hyper-specialisation at this stage (though you're quite right to advise watching it, Rawma!).

I'm more interested in creating meaningful routes of progression that diversify character specialisation without significantly increasing mechanical complexity. I'm cautious about building in too much granularity to the skill system, because I'd rather the players have a chance to attempt any tasks that seem interesting or exciting to them -- and I worry that adding lists will stifle their creativity.

Also, I'm quite taken with the idea of replacing skills with more advanced versions (telling myself that "In writing, you must kill all your darlings"). However, it's early days with the group, so this is basically theory-crafting at this stage.

Combat
Great stuff here, guys. Thank you! My primary intention is for the combat system to be fun, logical (insofar as outcomes should be believable and reasonably dangerous), and resolved quickly (the Pathfinder game I tried here ran at nearly an hour per turn -- I presume this is atypical?).

Inspired by the ideas here, I've created a very broad weapon system, which includes weapon ranges:

• Range (bows, thrown weapons, etc.)
• Reach (spears, polearms, etc.)
• Fencing (swords, axes, maces, etc.)
• Close (daggers, unarmed, etc.)

and some basic descriptors:

• Balanced   -1 defence target number
• Cleaving   -1 attack target number
• Crushing   -1 enemy armour
• Thrown

As a caveat, I'm not actually sure whether I'll do anything with weapon ranges in practice -- them might be fiddly. I've also linked the ranges into armour types, which might make things even more fiddly but hopefully not: e.g. bucklers are good vs close and fencing ranges, while kite shields are good at all ranges apart from close, etc.

I've also decided to allow the players to manipulate their pools a bit. Currently, they roll their Fighting to attack and their Fighting to defend (well there are other skill options, but for simplicity...). Having thought about the gambling / strategy element mentioned above, I'm going to allow the players to move some of their dice from attack or defence on a round-by-round basis. The idea being to allow them to manage their risk / reward, as well as to manipulate the battlefield a bit more.

As you can see, though, nothing very scientific going on here!    I'm mainly posting the decisions to show your comments were useful, however any further critique or comments are welcome. I do understand it only needs to work for my particular group and campaign, and the testing of that will begin tomorrow!

Quote from: Majus;801843I've also decided to allow the players to manipulate their pools a bit. Currently, they roll their Fighting to attack and their Fighting to defend (well there are other skill options, but for simplicity...). Having thought about the gambling / strategy element mentioned above, I'm going to allow the players to move some of their dice from attack or defence on a round-by-round basis. The idea being to allow them to manage their risk / reward, as well as to manipulate the battlefield a bit more.

Moving combat pools between attack and defence is a mechanic that's got a proven history, but it can make being outnumbered incredibly dangerous... Of course, you might want that to be the case.

The issue I've found with free form skills, is players who create skills that are universally applicable - rubbish like "savvy", frex. It requires strong GM input to make sure players don't try and take the piss.

Note that neither of these are bad things, just handy to bear in mind. They may not be fun for everyone.

Quote from: Ladybird;801863Moving combat pools between attack and defence is a mechanic that's got a proven history, but it can make being outnumbered incredibly dangerous... Of course, you might want that to be the case.

I do want it to be dangerous (character health is low and combat has already been brutal), but I'm not sure whether this will push things over the edge. So this is very useful to know. Will try it and see!

Quote from: Ladybird;801863The issue I've found with free form skills, is players who create skills that are universally applicable - rubbish like "savvy", frex. It requires strong GM input to make sure players don't try and take the piss.

I can see how that would happen. When all you've got is a hammer...

That could definitely happen in my system, too, but I think I'll have different skills accomplish different tasks in different ways and at different speeds. So while you might quietly, efficiently pick a lock with Thieving (I bet that example surprised you all!), you could also spike the hinges with Engineering or bash it down with Athletics -- but neither option will be as "good".

So, I guess my design approach is that I don't mind what the players try to do, but I don't want them to be unable to act. Also, a non-optimal skill use, will create other kinds of outcomes that they then get to play with, so that's interesting, too.

By the way, these questions are helpful in making me think about and articulate the why's of my design. Thanks Ladybird (and others)!

Quote from: Ladybird;801863Note that neither of these are bad things, just handy to bear in mind. They may not be fun for everyone.

Cool. I need a thumbs up icon (because that's just how cool I am), but I don't have one. So a "cheers mate" will have to suffice.