Monster AC -> absorption + evasion

[Atarlost]

There needs to be a bonus for light weapons to hit too, because if there isn't then heavy and light weapons are similarly powerful for evaders, but light sucks for absorbers.

Light weapons are on the winning end of the blows table. They have a damage advantage whenever they aren't facing absorption. If you can't hit with weapons you probably should be using spells or the currently pointless damage wands and rods.

[will_asher]

I'm also thinking about this idea. If I did this, I would have absorbtion be flat, and I'd keep the automatic 5% to hit/miss. Also, I think if this was done, then it would also be best to implement some kind of partial element resists for monsters (which is already done for 1.3.0 in DAJ), so that most (or all) spells could also have damage reduction applied, but it would be another type of damage reduction.

EDIT: DAJ sprites will be really hard to hit...

[pampl]

Light weapons are on the winning end of the blows table. They have a damage advantage whenever they aren't facing absorption. If you can't hit with weapons you probably should be using spells or the currently pointless damage wands and rods.

Light weapons are also mostly cheaper and lower level than heavy ones, so shouldn't really be equal in power anyway.

[Timo Pietilä]

In the heavy blows thread, Eddie suggested splitting monster AC into an absorption parameter and an evasion parameter, with the notable caveat that this would create a lot of work in rebalancing (he's right).

I have a bit difficult time figuring out what has changed in attitude recently. I suggested this about year ago and then everybody seem to be against changing AC abstraction to absorbion and evasion. Now everybody seem to like the idea. Odd. Nice to see it happening though.

[fizzix]

Another outcome of this is that we can make to-hit bonuses be less opaque. As in each + to-hit increases some flat % to hit.

Current code to hit calculation (for melee) looks like:

Chance = (intrinsic player melee skill) + (bonuses to hit * 3)
if (Chance > (AC * 3/4)) then player hits.

The intrinsic player melee skill is class based for warriors it starts out at 70 and gets +45 every 10 levels. A mage starts out at 34 and gets +15 every 10 levels. Critical hits are calculated later and are a complete mess.

So for the highest AC monster (cat lord, ac = 200) a 40+ level warrior with +50 to hit (fairly common) will have a 150/400 chance of missing or a 5/8 chance of hitting.

Overall, this system is not bad, however it does lose quite a bit with opaqueness. Even if you know the AC of the monster it takes some time to figure out your chance of hitting.

I'm trying to brainstorm new models. Preferably simple ones. Here's what I came up with, after a little thinking.

player rolls 1d100 to hit. If the roll is less than "hit value" then the player hits.

Hit value = intrinsic skill + to-hit bonus - monster evasion score.

Intrinsic scores may need to be modified some (specifically the level dependent part.) I don't think we'll need to mess with to_hit values, though. Priests and Mages may have a lot of trouble hitting in the early game and we may need to bump up their starting skill so that they can fight a little in the beginning.

One side effect of the old system is that buffs like bless and heroism made a much bigger impact on the early game than they will under this new system. If this is an aspect that is worth keeping, we can probably find a way to incorporate it.

[Magnate]

Another outcome of this is that we can make to-hit bonuses be less opaque. As in each + to-hit increases some flat % to hit.

Current code to hit calculation (for melee) looks like:

Chance = (intrinsic player melee skill) + (bonuses to hit * 3)
if (Chance > (AC * 3/4)) then player hits.

The intrinsic player melee skill is class based for warriors it starts out at 70 and gets +45 every 10 levels. A mage starts out at 34 and gets +15 every 10 levels. Critical hits are calculated later and are a complete mess.

So for the highest AC monster (cat lord, ac = 200) a 40+ level warrior with +50 to hit (fairly common) will have a 150/400 chance of missing or a 5/8 chance of hitting.

Overall, this system is not bad, however it does lose quite a bit with opaqueness. Even if you know the AC of the monster it takes some time to figure out your chance of hitting.

Once you know a monster's AC, recall now tells you your % chance of hitting it with your current melee weapon.

(But I agree that a more transparent to-hit formula would be an excellent result of the new combat model.)

[Magnate]

I have a bit difficult time figuring out what has changed in attitude recently. I suggested this about year ago and then everybody seem to be against changing AC abstraction to absorbion and evasion. Now everybody seem to like the idea. Odd. Nice to see it happening though.

I think a year ago the devs active at the time all had other things they were working on for 3.2 - now is the time to start ambitious new branches ...

[fizzix]

I think a year ago the devs active at the time all had other things they were working on for 3.2 - now is the time to start ambitious new branches ...

I have an AC branch now up on github. It has my edited monster list and the changes to the parser, monster race structs, and all calls to r_ptr->ac so the game compiles and runs. I haven't done any tinkering with the to_hit or to_dam calcs yet.

I've called the new AC variables eva (evasion) and sorb (absorption) but that's only because I am uncreative with names... Perhaps sorb should just be the old ac name?

[Derakon]

Is there some Angband style rule that says that variables should have names of 5 characters or less? I'd just call them "evasion" and "absorption" myself.

A quick survey of code I've written recently includes delayBeforeImaging, shouldAutoTrack, numDatapoints, numericalDisplays, and other such verbose variable names.

[fizzix]

Is there some Angband style rule that says that variables should have names of 5 characters or less? I'd just call them "evasion" and "absorption" myself.

I think it's more that I'm not sure absorption is a good term. Evasion means exactly what it's supposed to. Absorption seems a bit odd. Just plain armor may be better.

[takkaria]

I have an AC branch now up on github. It has my edited monster list and the changes to the parser, monster race structs, and all calls to r_ptr->ac so the game compiles and runs. I haven't done any tinkering with the to_hit or to_dam calcs yet.

I've called the new AC variables eva (evasion) and sorb (absorption) but that's only because I am uncreative with names... Perhaps sorb should just be the old ac name?

Why not call them evasion and absorption? One of the last things the game needs is more obscure abbreviations...

[ThunderToads]

Hydra should lean absorption IMHO because one of their mythical attributes was heads regrowing after they were cut off. I do get the idea that the heads are all snakey and writhing which makes them evasive, but they have a kinda squat, fat reptile body too.

[Atarlost]

Hydra should lean absorption IMHO because one of their mythical attributes was heads regrowing after they were cut off. I do get the idea that the heads are all snakey and writhing which makes them evasive, but they have a kinda squat, fat reptile body too.

I'd make them neither evade nor absorb, but regenerate very quickly. At least the Lernian.

[fizzix]

I'd make them neither evade nor absorb, but regenerate very quickly. At least the Lernian.

Making increased regen something beyond an on/off switch goes well beyond the scope of this change. (not to say that someone else shouldn't take it up)

To some extent absorption works as instant regen though.

[ewert]

Light weapons are on the winning end of the blows table. They have a damage advantage whenever they aren't facing absorption. If you can't hit with weapons you probably should be using spells or the currently pointless damage wands and rods.

Except this system should work at all points of game. In the end, you have max blows with or without the weapon being light. In the end only heavy weapons would make sense then. Or atleast my view was that the light vs evaders heavy vs armor would be an universal game design, not just prior to maxing your blows per round.