Derakon's combat revamp

[Mikko Lehtinen]

STR is already important for carrying capacity. It's the most overloaded stat in the game right now (to-dam and blows and carrying capacity and minimum weapon weight). Making it a bit less important is a good thing.

That still leaves carrying capacity and maximum weapon weight in the new system. Maximum weapon weight will probably be less important in the new system, would you think, with low-STR characters preferring light weapons anyway?

Clearly STR is too important in Angband when compared to WIS, INT, or CHR. But warrior characters tend to minimize all those stats anyway and choose between STR and DEX. So I'd pay attention to the balance of those two stats.

DEX will have number of blows in melee, number of shots with missile weapons, to-hit in both melee and ranged combat, and miscellaneous other effects. But maybe it's ok for DEX to be twice as important in combat than STR, because carrying capacity really is important. You're probably a better judge here than I, since I've messed up my perceptions by playing too much FayAngband.

And don't think for a minute that there's no skill behind knowing how to apply a lot of force behind your blows. Having Power be a class skill makes sense for realism reasons in addition to for game balance.

I tend to not think very much about realism when I'm designing games, I just want everything to be as simple, elegant, and tactically interesting as possible. So we just have a very different point of view. From the realism point of view, I agree completely.

Game balance can really only be judged when we see the whole puzzle, archery and all, and what's the role of STR and DEX there.

A single skill for Power is elegant design. But having a different Power skill for ranged weapons (if that is necessary) starts to get messy.

[Estie]

Hmm I kindof liked the approach to let weapon weight handle speed, and str/dex attack speed, at least if it was done intuitively and not "more str gives more attacks with the dagger". However, its not that important, good bye.

The longsword sample:

(to hit, to finesse, to dam), %finesse, %power ...do we really need the option to have plain adders to both to hit and finesse ? What would one give that the other cant ?

Also, the sum of finesse and power is 1; only the ratio is relevant. While changing that for certain weapons might look intersting, is it different enough from changing the base damage dice to warrant another number ?

Coming from playing v4, i am worried about increasing the length of item descriptions. The presented sample obviously has entries wherever possible to show the extremes. However, I very much dislike that in v4 almost all weapons have a ton of entries about things that have little or no impact on gameplay.

When I come upon a new weapon, I first and foremost need to decide wether it is an upgrade or not. Given a longsword 2d5 (+5,+8) thats easily done at a glance. Given a Dwarven longsword of Agility 2d6 (+5,+8) <+1,+2> [+4] with 3 lines of additional information it becomes annoying.

With the new system we get even more numbers. That by itself is, I think, not a problem, as long as there is some restraint when applying all the new options.

What I mean is this: when deciding about a new ego or property, first ask "does it have enough impact to matter". A [+3] to AC does not, and hence shouldnt be on every other weapon found. In the current vanilla, it comes with the defender ego which itself is so rare that the (useless) AC adder doesnt matter. If a sizable percentage of weapons is "dwarven" (or whatever), it shouldnt spam the information line with [+3] or [+4].
If there is to be an ego with bonus to AC, make it [+30] so its noticable and rare to keep balance.

In all the v4 threads, I havent seen this topic surface, so I am wondering if its just me and everyone else is happy.

[Derakon]

STR is already important for carrying capacity. It's the most overloaded stat in the game right now (to-dam and blows and carrying capacity and minimum weapon weight). Making it a bit less important is a good thing.

And don't think for a minute that there's no skill behind knowing how to apply a lot of force behind your blows. Having Power be a class skill makes sense for realism reasons in addition to for game balance.

[Mikko Lehtinen]

"Weapon speed" under the proposed system is functionally identical to the weapon's finesse multiplier. So yes, a whip would be a very light weapon that nonetheless isn't especially fast, because it would have a low finesse multiplier. And probably a lower power multiplier. Whips just aren't really meant for combat.

In Rolemaster, whip is a very good weapon choice for unskilled characters (against lightly armored opponents), but skill increase doesn't really help much. Sounds exactly like your whip, except the base damage should be higher! I think every weapon should have at least a niche use.

So it sounds like the proposal to have finesse and power have secondary effects on to-hit / criticals isn't liked. Remember that my original intent was to have to-hit be replaced by finesse, and likewise to-dam replaced by power. Does that still make sense?

Yes, definitely get rid of plusses to-hit and to-damage.

I favour having two Finesse skills, one for melee weapons and one for missile weapons. Both skills would also help you hit things. I think we agree here, too.

I would prefer having pure Strength handle the Melee Power part of the equation. That way we wouldn't need to invent any secondary effects for the Power skill or add any new skills to the character sheet. Besides, STR needs to be really important here for balance reasons, because DEX already affects both melee and ranged skills.

[Mikko Lehtinen]

Has someone thought about how to balance the races with the new combat system?

The problem I see is that most races that would thematically be good at Finesse are currently good in archery and bad in melee, and most good warrior races clearly excel in Power. So we'll either end up having more points in Power than in Finesse overall, or we have to alter the game balance.

One other option, which I wrote about in an earlier post, is to make Finesse somewhat more important in ranged combat, and Power more important in melee.

I think we should have at least a vague idea about the place of archery in the new system before actually coding it.

[Derakon]

"Weapon speed" under the proposed system is functionally identical to the weapon's finesse multiplier. So yes, a whip would be a very light weapon that nonetheless isn't especially fast, because it would have a low finesse multiplier. And probably a lower power multiplier. Whips just aren't really meant for combat.

So it sounds like the proposal to have finesse and power have secondary effects on to-hit / criticals isn't liked. Remember that my original intent was to have to-hit be replaced by finesse, and likewise to-dam replaced by power. Does that still make sense? If so, how do we calculate the player's chance to hit something without having a to-hit value? If not, are we going to have three separate plus values on each weapon? (to-hit, to-finesse, to-power) Or are we going to do away with on-weapon bonuses of that type? I mean, imagine this:

A Giant Longsword of Quickness (3d5) (-2,+4,+10) [+5] <+2>
40% finesse, 60% power
It is unusually large, favoring brute force over skill. 
It is skillfully crafted and easy to maneuver.
It is designed with an eye towards defense. 

It increases your searching by 2, so I can stick a pval on this thing.

Here I'm imagining the "Giant's" affix modifying the base dice, tweaking the finesse/power ratio by 10%, and adding to Power while reducing to-hit and Finesse. Meanwhile, "Quickness" gives a straight Finesse bonus. So that's -2 to-hit, +4 to finesse, +10 to power. Is that too many numbers to keep track of for the weapon's primary display?

As for stat caps, that should really be in its own thread IMO.

[Magnate]

I've probably suggested this before (or something vaguely similar). Late game stat and race/class stat differentiation can be accomplished by adding a finer granularity to stats and having stat potions and stat boosts affect different races and classes differently (without necessarily the need for finesse and power). A base hobbits might only get 1/3 of a point of STR form a potion, or a full 1 point boost for a ring of STR+3. A hobbit warrior maybe twice as much. For a hobbit mage it would be insignificant. As with the system being currently discussed, further balancing would be required, probably an expansion of the min/max stat range and forced to conform to a curve where the extremes are not only unattainable, but practically probably not worth attaining (else scumming will rule). Custom character birthing would be a problem too.

I am half with you here. I completely agree that revisiting ranges and caps and so on, with a view to more race/class distinction at endgame, is a good thing. I'm not convinced that having gains/boosts affect different races/classes differently is worth the trouble. I think there are plenty of levers to distinguish races and classes without making stats that complicated. But happy to hear more.

[buzzkill]

- changing the stats system is on the to-do list for v4, but shouldn't impact on Derakon's proposal. When STR/DEX are recalibrated, we can adjust their impact on Finesse and Power accordingly. (I disagree that stats should be the sole determinants of Finesse and Power. Part of the point is to make races and classes more distinct.)

I've probably suggested this before (or something vaguely similar). Late game stat and race/class stat differentiation can be accomplished by adding a finer granularity to stats and having stat potions and stat boosts affect different races and classes differently (without necessarily the need for finesse and power). A base hobbits might only get 1/3 of a point of STR form a potion, or a full 1 point boost for a ring of STR+3. A hobbit warrior maybe twice as much. For a hobbit mage it would be insignificant. As with the system being currently discussed, further balancing would be required, probably an expansion of the min/max stat range and forced to conform to a curve where the extremes are not only unattainable, but practically probably not worth attaining (else scumming will rule). Custom character birthing would be a problem too.

EDIT: I forgot one. I think we need to distinguish between a weapon's weight and its speed.

This has been a long time coming and will work well, but probably not before the split of movement speed and attack speed. Start thinking of extra attacks not as extra attacks but as extra attack speed. Fractional and further modified by the weapon and finesse/DEX.

[Magnate]

This new combat system will make finding good weapons much more dependant on luck than before. In the current system, most any good weapon will do; after the change, you will have to find a good weapon that is suitable for your combat style.

Yes and no. As Derakon pointed out earlier, many weapons will have only minor preferences for power or finesse (the dagger and maul being extreme examples) - finding a really good longsword should be a good thing whatever your combat style, really. But yes, *optimising* will be more difficult than it is now - I wonder if that will upset people ...

So, I've missed a ton of posts, but I've read them all. Some brief thoughts:

- we should absolutely break this up into chunks, which means not touching missiles at all in the first instance. (Personally I think the mechanics ought to be very different for the three launcher types, and I know d_m wanted to make throwing weapons competitive in their own right.)

- changing the stats system is on the to-do list for v4, but shouldn't impact on Derakon's proposal. When STR/DEX are recalibrated, we can adjust their impact on Finesse and Power accordingly. (I disagree that stats should be the sole determinants of Finesse and Power. Part of the point is to make races and classes more distinct.)

- one thing we *should* bear in mind is that AC will be split into Evasion and Absorption, as discussed earlier this year (this is exactly the kind of change v4 was created for). This will actually work very well with Derakon's proposals, since Finesse will make it easier to hit monsters with high evasion, and power will make it easier to damage monsters with high absorption

- similarly, we should be open-minded about changes to the way slays, brands and extra blows work. Slays/brands can be re-worked around evasion and absorption, and the OF_BLOWS flag can be an adjustment to the blows calculation (e.g. increasing effective weapon speed, or finesse, or whatever) rather than adding to the result

- I agree with the point about criticals being rare, and not the primary mechanism for power increasing damage. Often a 'critical' hit is an exceptionally well-placed one, not a hard one (i.e. Finesse over Power).

- I agree that these changes will mean more differentiation between endgame characters (especially combined with changes to stats). I think this can only be a good thing, and the answer to Estie's question is a resounding Yes. Ringil will continue to be an awesome weapon for just about anybody, but hobbit rogues may end up better off with Aglarang than Deathwreaker, for example.

EDIT: I forgot one. I think we need to distinguish between a weapon's weight and its speed. A whip is light but very slow. The blows calculation needs to use a weapon's speed, not its weight. (And in fact this solves the question of what to do about OF_BLOWS: it simply increases a weapon's speed rating.)

[Estie]

After a night of sleep, my thoughts on the matter have congealed....

The critical point where every new suggestion for a combat system breaks is the transition from early to late game. In the former, any amount of changes is welcome, but the latter is constrained by the request to let stats (that is str and dex) not differ for varying classes/races (in a matter that would affect combat).

The question I asked myself is _why_ this is the case. What is the advantage or effect of everyone having the same maxed stats ?
Well, it means that everyone, be he mage or warrior, halfling or troll, wants the same gear. The best weapon is always Deathwreaker, followed by Ringil etc.

Obviously this makes it easy to balance the endgame, gives everyone the same bragging rights if he finds Ringil and so on.

In this light, the combat system suggested by Derakon tries to break the uniformity of endgame combat by introducing new stats in finesse and power, which presumably make the wants of a hobbit and a troll different for the lategame, without changing the stat cap.

But....if that is going to happen anyway, and we need to provide for two different wants (finesse and power endgame weapons), what is the point of keeping current stat cap ? Wouldnt it be easier to just let str, dex and weapon weight handle everything (albeit changed in functionality, so the big troll wants a hammer early on and not a dagger) while keeping the stat diversity throughout the game ?

The question I have been asking over and over: what are lategame numbers going to look like ? - can thus be reduced to: will trolls and hobbits, mages and warriors, want _different_ weapons for lategame ?

If the answer is "yes", then I think its best to just modify the str/dex effect of weapons to give the troll his hammer. If the answer is "no", then Derakons finesse and power can solve the early game combat, but _they must not have any effect for lategame_ .

This, I think, is the crux.

[Mikko Lehtinen]

Prismatic: so, to get your applied melee finesse rating, you'd do ((player finesse + DEX mod + other mods + etc.) * (player melee bias)) * (weapon modifier)? That would work, though I worry about making the calculations too obtuse.

It would be simpler if Melee and Ranged skills only affected to-hit chance and perhaps critical hit probability.

If we'd change to-hit to really mean something (even in the end game), that would be enough. This doesn't necessarily mean hitting less often; you could also have very high to-hit chances result in critical hits far more often (like I did in Fay).

The best ranged weapons should be designed so that ranger is the best class at using them. Using a longbow would require both Power and Finesse (ideal for rangers), and a shortbow would require more Finesse than Power (ideal for rogues).

Longbows would also have greater range than shortbows, and shooting far would require far more Ranged skill than shooting near. Only classes with really good Ranged skill could hit an enemy at extreme range.

[Prismatic]

Prismatic: so, to get your applied melee finesse rating, you'd do ((player finesse + DEX mod + other mods + etc.) * (player melee bias)) * (weapon modifier)? That would work, though I worry about making the calculations too obtuse.

The thing about bows is that realistically, you need Power to draw them, and Finesse to aim them. The best way I could see to implement this would be to institute a minimum Power level to effectively use a given bow, after which point further Power is moot. That minimum level would scale with the multiplier on the bow, so you'd really need godlike Power to use an x6 longbow.

I'd still rather push that back to "version 2" if possible, though.

I'm not familiar enough with Angband's internals to give a proper method, but something like that. I don't think it's too obtuse, since while it is a little complicated internally, what the player mostly need to worry about is 'is this melee or ranged?', and 'how much does it favour finesse vs power?'.

Realistically, bows would need power to draw, but I didn't pay attention to that because it /does/ make it a little more complicated and they do use power at least a little bit in their formulas as is.

[Mikko Lehtinen]

For flavour and simplicity, I'd like this option the best:

Pure Strenght determines your Power for both melee and ranged weapons.

Melee skill determines your Finesse with melee weapons, and Bows skill your Finesse with missile weapons. Your dexterity, class, race, and level affect these skills.

Unfortunately we hit the Strength ceiling in the end game again, but that might be possible to fix.

[sethos]

yeah, without going into some ad-hoc way of handling archery, I thing that you'd need at least one more skill for archery(shooting), if not two (Range power / Range Finesse) or more (Throwing skill / Bow Skill / Xbow skill / Sling Skill) The latter would be most useful for fine tuning (hobbits use slings well, warriors and trolls throw things well, rangers and elves are good with bows, etc.) bot the most complicated to implement (not to mention understand!)

[bio_hazard]

Ranged finesse could affect "to hit".

Ranged power could affect damage, and maybe also shooting speed? Bigger bows and especially bigger xbows might take > 1 turn to draw if you are not very good at it, while the extra shots of rangers could come from their higher ranged power skill.

To be honest, I'd be OK with only a single ranged skill. I think I'd want more subtlety added to ranged attacks in general to make it worth to differentiate between slings, bows, and xbows. And certainly beefed up options for thrown objects.