Derakon's combat revamp

Collapse
X
 
  • Time
  • Show
Clear All
new posts
  • Derakon
    Prophet
    • Dec 2009
    • 9022

    Derakon's combat revamp

    Since we're posting our ideas for how to rework combat, I guess I may as well give this its own thread.

    I brought up this idea first in this thread. The idea is to replace to-hit with "weapon finesse", and to-dam with "weapon power". The former represents the character's ability to quickly send a weapon exactly where it needs to be; the latter represents the character's ability to put a lot of force behind each blow.

    A character with high finesse (as given by their racial and class skill bonuses, their DEX score, any on- or off-weapon bonuses, and to an extent their level) is more likely to hit his opponents. A character with a high power score (ditto, except STR instead of DEX) is more likely to score critical hits. These two effects are global and don't depend on what weapon you're using.

    However, each individual weapon also has two values: a finesse multiplier and a power multiplier. These determine how many blows you get with the weapon, and how much damage for each blow is multiplied by, respectively. For every 100 converted finesse points, you get 1 extra blow; for every 100 converted power points, your damage gets an extra multiplier. Of course, fractions are allowed.

    Examples
    A 1d4 dagger might have a finesse multiplier of .8 and a power multiplier of .2. If your finesse before multiplying is 200 (that is, your race, class, level, DEX, and equipment modifiers sum to 200), then you would get 200 * .8 / 100 = 1.6 bonus blows per round. Everyone gets at least 1 blow per round so your actual blows per round would be 2.6. If your power was 400, then each blow would deal 400 * .2 / 100 = .8x bonus damage, so 1.8 times normal.

    Alternatively, a 2d5 longsword could have a balanced set of multipliers: .5 finesse and .5 power. Given those same skill levels from before, we get 2 blows per round at 3x normal damage.

    Finally, a 4d4 maul would be massively overdirected towards power in its multipliers: .1 finesse, .9 power. Again, with those same stats, we end up with 1.2 blows per round at 4.6x normal damage.

    Breaking things up by class
    Now, we can bias different classes towards different weapon types by changing their skill growths. Rogues would presumably get very high finesse scores, which would bias them towards weapons which benefit strongly from finesse, which would tend to be lightweight weapons. Paladins could have good power but poor finesse, directing them to the heavy maces and axes; warriors would be good all-rounders.

    Relative balance of finesse and power
    Because both blows and the damage multiplier effectively do the same thing in terms of total damage dealt, we may need to re-think how many dice weapons get. Of course, finesse-biased characters will tend to hit more often, while power-biased characters will tend to get more and better criticals. The net result is that power characters will have more unreliable damage, which I tend to think means they should get better base damage, though perhaps not to the current extent (where a heavy weapon can do 3-5x more damage than a light weapon just from dice alone). Power characters will also be worse at fighting weak groups, because they have to spend most of a round pancaking each enemy; a finesse character might need two blows to precisely kill each foe, but that still leaves plenty of un-used energy to deal with his buddy.

    Implications for item design
    Right now all we care about for items is the base damage, the weight, and whether or not the item is pointy -- and that last bit only if we're priests! There's a lot of functionally identical weapons out there. By giving each one different relative weightings towards finesse and balance, we can inject some variety. For example, longswords would be balanced weapons, while pikes would be power-oriented and scimitars skill-oriented.

    We can also make weapons that are intentionally awkward, or intentionally easy, to use. The infamous whip, for example could get a .6x finesse multiplier and a .2x power multiplier -- summing to less than 1! But take a standard weapon, and apply the "masterwork" affix: now its finesse gets an additional bonus. This would be the new "extra blows" affix, in other words. Or the "lead-filled" affix could reduce finesse by .1 while raising power by .3.

    What about missile weapons?
    There's no reason they couldn't be handled by the same framework. High-finesse bows can be fired faster, high-power ones get a better multiplier. This actually matches reality better than our current system, where it doesn't matter how strong you are, you can still use that Longbow of Extra Might x5 which probably has a draw weight of over 100 pounds.

    What about slays?
    Apply a 1.x damage multiplier to each shot/blow. Done.

    Implications for code work
    We would need to rework the areas where the following things are calculated:
    * Weapon definition (loading from object.txt) and in-memory storage
    * Number of blows per round
    * Damage per blow
    * Chance of hitting
    * All critical hit logic
    * How slays apply
    * Race and class growth information (growth in finesse and power skills)
    * Display of finesse and power bonuses (rename to-hit / to-dam)
    * Item 'I'nspect display (to show impact of finesse and power)
    * (Possibly) rework the energy scale to allow finer gradations (e.g. 1.15 blows/round vs. 1.25 blows/round). Just multiplying everything in the current system by 10 should do.

    Implications for non-code changes
    Big. Every weapon, race, and class will need new stats. All of this will take significant balancing work.

    Ultimately, what does this buy us?
    * More sensible weapon decisions -- trading between fast light weak weapons, and slow heavy powerful ones. The logic for the decision is the same at level 1 as it is at level 50, though the factors on the decision will change. Characters at level 1 won't always opt for the biggest weapon they can max blows with.
    * More distinguishable (and thus interesting) weapon types. Daggers, whips, main gauches, and rapiers will all have more to differentiate them than a single die side. Longswords and broadswords will differ by more than just 2 pounds. Angband is ultimately largely about finding new and better equipment; creating finer gradations here will make that part of the game more detailed, ergo more interesting.
    * Unifying melee and ranged combat.
    * Smoother power curves without the sharp breakpoints we get currently (especially with regards to the "Extra Blows" and "Extra Shots" affixes).
    * The impact of STR and DEX is more intuitive. STR directly affects power, which directly affects your multiplier. DEX directly affects finesse, which directly affects your blows/round. They don't combine in some weird interdependent fashion where gaining 5 STR is like gaining 2 STR and 3 DEX is like gaining 7 DEX.

    Why not just O combat?
    OAngband-style combat has a similar approach: to-dam is replaced by a "deadliness" damage multiplier which applies to the base dice, multiplying damage by up to 255% (if I recall correctly). My main beefs with it:
    * There's diminishing returns on deadliness, so it's not necessarily clear exactly how equipment decisions affect your damage.
    * Your number of blows is still calculated based on your STR and DEX and the weapon weight, which I'd like to get away from as being overly opaque.
    Last edited by Derakon; December 13, 2011, 20:28.
  • fizzix
    Prophet
    • Aug 2009
    • 3025

    #2
    I really like the idea of splitting combat up into finesse and power. Although I'm blanching a bit at all that needs changing...

    Comment

    • Antoine
      Ironband/Quickband Maintainer
      • Nov 2007
      • 1010

      #3
      This will all be easier if you can keep the average DPR (by class and level) about the same as what it is now - then you won't have to rebalance monsters?

      A.
      Ironband - http://angband.oook.cz/ironband/

      Comment

      • Derakon
        Prophet
        • Dec 2009
        • 9022

        #4
        Originally posted by Antoine
        This will all be easier if you can keep the average DPR (by class and level) about the same as what it is now - then you won't have to rebalance monsters?

        A.
        That would be ideal, yes, and it's made easier by the fact that in v4 at least we aren't too worried about balance changes at the momentr. In other words, this could make the game too easy (or too hard), and that change could then be corrected for in other ways. It doesn't have to have no net effect on balance in itself.

        Though if this were to end up in v4, would it stand any chance of being "promoted" to Vanilla standard? We've had the same basic rules for melee for the better part of two decades now, as best I can recall.

        Comment

        • PowerDiver
          Prophet
          • Mar 2008
          • 2820

          #5
          Before you go crazy with this, take my word that 3.0 combat was fine except for the availability of enchant scrolls. I've been playing 3.0 for months, with no enchants and blows calculated by min {f(str, weight), g(dex)}, and I am happy with it. You don't have to change the blows calculations for balance purposes, but I think it is better.

          More recent versions are more unbalanced because Takkaria and Magnate kept boosting the damage per blow of early characters, and then sped up stat gain in addition. That's not a flaw with the combat system per se.

          Comment

          • LostTemplar
            Knight
            • Aug 2009
            • 670

            #6
            Combat system by itself is fine, btw if you cannot balance this relatively simple system for 20 years, how can you balance more complicated one ?

            Comment

            • Nick
              Vanilla maintainer
              • Apr 2007
              • 9637

              #7
              Originally posted by Derakon
              Why not just O combat?
              OAngband-style combat has a similar approach: to-dam is replaced by a "deadliness" damage multiplier which applies to the base dice, multiplying damage by up to 255% (if I recall correctly). My main beefs with it:
              * There's diminishing returns on deadliness, so it's not necessarily clear exactly how equipment decisions affect your damage.
              * Your number of blows is still calculated based on your STR and DEX and the weapon weight, which I'd like to get away from as being overly opaque.
              So, if O-combat were your starting point, how would you improve it?
              One for the Dark Lord on his dark throne
              In the Land of Mordor where the Shadows lie.

              Comment

              • fizzix
                Prophet
                • Aug 2009
                • 3025

                #8
                Originally posted by LostTemplar
                Combat system by itself is fine, btw if you cannot balance this relatively simple system for 20 years, how can you balance more complicated one ?
                Sometimes it's harder to balance a simple system than a more complicated one.

                Comment

                • LostTemplar
                  Knight
                  • Aug 2009
                  • 670

                  #9
                  Sometimes it's harder to balance a simple system than a more complicated one.
                  Have you ever try ? In some old versions, maybe 3.0.9, melee combat had no problems, or at least not so much as now. Archery was overpowered, spells were too weak and cost too much, devices were pitifull, but melee was OK. But it was broken at some point and now you try to change core rules, in hope, that it will be easyer to repair after that.

                  Comment

                  • Magnate
                    Angband Devteam member
                    • May 2007
                    • 5110

                    #10
                    Originally posted by Derakon
                    That would be ideal, yes, and it's made easier by the fact that in v4 at least we aren't too worried about balance changes at the momentr. In other words, this could make the game too easy (or too hard), and that change could then be corrected for in other ways. It doesn't have to have no net effect on balance in itself.
                    Very nicely done - I really like this proposal. It's exactly the kind of detail I hoped someone would come up with. I have views on how slays/brands ought to work in this system, but I'm happy to leave that until the code gets to a point where they need changing. How much of the coding are you planning on doing yourself, and how much would you like to outsource? I'm happy to help as soon as I've finished effects.
                    Though if this were to end up in v4, would it stand any chance of being "promoted" to Vanilla standard? We've had the same basic rules for melee for the better part of two decades now, as best I can recall.
                    I think, once it is done and balanced, that it will be near impossible to port it to V without also porting the different item generation and different monsters (since all of slays/brands/blows/shots will work differently). The consensus may be that V should get all of those, but it may not. Given the number of people who currently think that combat in 3.0.x isn't broken, I wouldn't waste too much energy arguing about it at this stage. If we do it and people like it, then's the time to start the debate.
                    "Been away so long I hardly knew the place, gee it's good to be back home" - The Beatles

                    Comment

                    • Derakon
                      Prophet
                      • Dec 2009
                      • 9022

                      #11
                      Originally posted by Magnate
                      Very nicely done - I really like this proposal. It's exactly the kind of detail I hoped someone would come up with. I have views on how slays/brands ought to work in this system, but I'm happy to leave that until the code gets to a point where they need changing. How much of the coding are you planning on doing yourself, and how much would you like to outsource? I'm happy to help as soon as I've finished effects.
                      I don't currently have anywhere near enough experience with the V codebase to realistically tackle this on my own. However, I'll have some spare time soon what with the holidays. Sounds like it may be time for me to hop onto IRC...

                      Nick: my main beefs with O-combat are as I listed them. I'd really like to change how blows are calculated to a more intuitive system, since the current system pretty much necessitates a "try changing your stats and see if you get more blows" approach. And diminishing returns also introduce opacity since you can't easily tell how a given change to your equipment will affect your damage output. I'm sure that Leon put the diminishing returns in for a good reason, but there must be a way to be able to do without them.

                      Regarding 3.0.x combat being balanced, while I certainly agree that the game is better-balanced when pluses are harder to come by (I was singing the praises of v4 with my dwarf warrior, up until I made it to about 2500' and still hadn't found anything approaching a good weapon...), that's not really what this is addressing. This is for the common complaint that, all else being equal, a dagger is preferable to a longsword because you get more blows with it. And that's been a problem for over a decade now. I certainly remember the common advice for new players being "roll up a Half-Troll Warrior, tell the autoroller to give you an X STR and a Y DEX, go to the temple, buy, a whip, and have fun."

                      Comment

                      • Mikko Lehtinen
                        Veteran
                        • Sep 2010
                        • 1246

                        #12
                        Having fractional blows or damage multipliers here is not necessarily a good thing.

                        Bigger jumps in power often feel better to the player, giving a sense of accomplishment.

                        Equipment choices may also be more interesting with bigger jumps. (This is where EyAngband combat system shines.)

                        Consider these cases: "I need only one point of STR or some experience levels to get double damage with this weapon... Was there a ring of STR in the magic shop?" Or: "If I lose even one point of STR, I'm not able to get the double damage. I'd better carry a lighter swap weapon."

                        Comment

                        • Estie
                          Veteran
                          • Apr 2008
                          • 2347

                          #13
                          One thing:

                          I wouldnt handle the "power" branch with criticals. Critical hits are, in my perspective, something that happens rarely. For the purpose though you need something amounting to a damage increase of a factor of 6 or so, if we assume the "finesse" branch ranging from 1 to 6 blows. Just call it a damage multiplier and leave criticals free for possibly special effect (like monsters have them in stunning). Maybe this is only terminology, but there you are.

                          The problem that still remains is the change from early to lategame, where, in the current system, we have characters with varying str and dex scores wheras lategame everyone has the same.

                          You want to give characters an inherent score of finesse and power, increasing with level and depending on class, but still list str and dex as impacting the final number.
                          How exactly is this going to look for a lategame toon ?

                          Comment

                          • Mikko Lehtinen
                            Veteran
                            • Sep 2010
                            • 1246

                            #14
                            Originally posted by Estie
                            The problem that still remains is the change from early to lategame, where, in the current system, we have characters with varying str and dex scores wheras lategame everyone has the same.

                            You want to give characters an inherent score of finesse and power, increasing with level and depending on class, but still list str and dex as impacting the final number.
                            How exactly is this going to look for a lategame toon ?
                            You would have to take care that at all points of the game, there are good weapon base types and artifact weapons for all character types, at least for these three: fast, balanced, and powerful.

                            Ey (and Fay) tries to do exactly this. It takes a lot of time and playtesting, but it's doable.

                            Comment

                            • Derakon
                              Prophet
                              • Dec 2009
                              • 9022

                              #15
                              While it's true that everyone will have maxed their STR and DEX by the endgame, that doesn't mean that those maxima will be the same for each character. Assuming maximize mode is on, there's a big difference between, say, a dwarf warrior (18/150 STR, 18/100 DEX) and a hobbit paladin (18/90 STR, 18/120 DEX). More importantly, we can strongly differentiate bonuses from equipment. Right now everyone only bothers to optimize damage on their gear -- but the equivalent, optimizing weapon power, is only really relevant if your weapon gets a good boost from weapon power. If you have a finesse-oriented weapon then you're better-served with a Ring of Finesse rather than a Ring of ...er, Power.

                              And of course there's the question of skill growth. While I generally don't like mechanics that heavily take level into account, generally you do gain at least 40 levels without really noticing, as you dive through the dungeon. Depending on class those levels can be biased more towards finesse or power, which will strongly differentiate different classes and influence which weapons they want to use.

                              Example weapons, off the top of my head:
                              Finesse: dagger, rapier, scimitar, spear
                              Power: club, mace, hammer, halberd, scythe
                              Balanced: shortsword, longsword, executioner's sword, glaive

                              As for having power not impact criticals, I would like each stat to have a use beyond just influencing your weapon, to give you some utility from boosts to that stat even when you're using an "opposed" weapon type. It seems clear that weapon finesse would have an impact on your ability to hit enemies, and weapon power would have an impact on your ability to hurt enemies; hit chance and critical quality seemed like the natural mechanics to insert, therefore. Do you have a better suggestion?

                              Comment

                              Working...
                              😀
                              😂
                              🥰
                              😘
                              🤢
                              😎
                              😞
                              😡
                              👍
                              👎