v4 - a few characters

[Jungle_Boy]

Is there a limit to the number of affixes an item can have? Also is there a list of current names? I might look into it and see if there is some convention I can come up with.

Magnate - your granularity example was not quite correct +10 and +20 could just as easily be +1 and +2 with no loss of detail. The loss would be between +10 and +15, half the amount you mentioned though still perhaps significant.

As far as dragonscale as a material how difficult would it be to have it add a constant as well as a percent boost to AC? That way it still affects smaller items without boosting the percent too high for armors. Alternatively you could tie the percent boost to armor weight so lighter items get a higher percent boost than heavier ones.

[Magnate]

Aren't the per-level increments in p_class.txt actually given as "bonus per 10 levels"? So they already have that much granularity.

Good point.

Percentage isn't a terrible idea either. The main issue is that we have the heft multiplier to take into account; +100 prowess is nowhere near as good on a dagger as it is on a warhammer.

Indeed - how about dispensing with the plusses altogether - after all, the same applies to +finesse. How about we just show A Dagger (1d4) (2.47, 1.08x) where the first number is the blows you get if you wield it, and the second is your damage mult.

[Magnate]

Is there a limit to the number of affixes an item can have? Also is there a list of current names? I might look into it and see if there is some convention I can come up with.

lib/edit/ego_item.txt in v4. MAX_AFFIXES is currently 8. @ekolis: I don't like the idea of multiple affixes providing the same ability - unnecessary complexity. And no, we don't currently restrict what affixes can go with each other (but see ticket #1562).

As far as dragonscale as a material how difficult would it be to have it add a constant as well as a percent boost to AC? That way it still affects smaller items without boosting the percent too high for armors. Alternatively you could tie the percent boost to armor weight so lighter items get a higher percent boost than heavier ones.

Neither of those are possible in the current affix system, without creating multiple separate affixes for the same dragon scale (i.e. a percent boost for heavy armours and a straight AC boost for light items). We could limit weak DSMs (blue, white, etc.) to light items and more powerful ones to heavier items and do it that way, but that then plays havoc with the multi-DSM themes.

Personally I'd prefer recalibrating AC so that a percent boost is always meaningful. In fact I'd prefer not to worry about it, but it seems to be an emotive issue.

[Magnate]

My opinion would not decide anything aside from a Variant I would decide to create. There simply has to be some kind of consensus, it never is going to be uniform, people will always fall on the boundaries of both sides. It is just a discussion and then some kind of majority opinion which the devteam then respects, just like the user base respects the devteam isn't a bottle of genies that can do anything regardless of complexity / effort or personal desire/enjoyment.

Ok, so if I understand your previous post correctly, what you meant was "it's all very well the devteam agreeing with each other, but if the rest of the user base doesn't agree with them then it isn't Angband proper". That's fair enough - I'm pretty confident that at least half the user base didn't want to encourage store scumming, and if the user base is split, we can only ever please half.

[Magnate]

Seems like this logic may be at least part of the problem-- adding a die isn't necessarily more powerful than adding a side (consider say a 6d2 weapon becoming 7d2 (10.5avg) vs 6d3 (12avg)). What about having a single ego that boosts base damage by some amount or some fraction of base damage and then using some logic to figure out how that increase should be mapped to dice based on base item balance/heft (I'd be happy to share the same logic I used for redistributing weapon dice).

Well, as Derakon observes in the Pyrel thread, the real solution to this is to use object_power to check the power of the item - that way we can tell whether an affix that adds a side is trivial (1d10 to 1d11) or awesome (10d1 to 10d2).

I'm not averse to that, but it's fairly major surgery. First I'd like to try adding more weapon affixes, using balance and heft, and making the dice/sides affixes less common. If we still have overpowered weapons after that, we can revisit.

[Derakon]

Indeed - how about dispensing with the plusses altogether - after all, the same applies to +finesse. How about we just show A Dagger (1d4) (2.47, 1.08x) where the first number is the blows you get if you wield it, and the second is your damage mult.

Then how do you handle off-weapon combat bonuses? Say you're wielding a Cesti of Combat that, under the current system, would be (+80, +110). Under the new system, if you're wielding a dagger it would be "Cesti of Combat (.64, .22x)", but if you switch to wielding a warhammer it'd be "Cesti of Combat (.2, .82x)".

(numbers probably slightly off, but you get the point)

[Magnate]

Then how do you handle off-weapon combat bonuses? Say you're wielding a Cesti of Combat that, under the current system, would be (+80, +110). Under the new system, if you're wielding a dagger it would be "Cesti of Combat (.64, .22x)", but if you switch to wielding a warhammer it'd be "Cesti of Combat (.2, .82x)".

(numbers probably slightly off, but you get the point)

I do. If we move to this system, I don't see any problem with the numbers changing as you change your kit, since people are often asking for the game to do more of the calculations for them. Again, it would take a bit of getting used to.

[CliffStamp]

I can understand the ideas here, but it sounds a little munchkin to have so much revealed. I never even for example look at how damage is calculate from slays, etc. . as I prefer to actually use the weapons and see what does better. It is not like you are tripping over them anyway and this actually adds a bit to the game to see on average which is better for raw damage.

Is this going to start a trend where armor is going to show the effects on damage reduction/blows landed, etc. ?

[Derakon]

Is this going to start a trend where armor is going to show the effects on damage reduction/blows landed, etc. ?

I could certainly see the game telling the player somewhere what their damage reduction is, though, figuring out exactly where to put it would be tricky since it comes from so many different slots. Maybe put it on the character screen?

The main issue with damage reduction, though, is that it only applies to physical hits, which is a poorly-defined term. IIRC it's something like hit, bash, trample, and kick are all physical hits so long as they have the HURT attribute, but anything that does something else on a successful hit (e.g. drain life, disenchant) or uses a different verb (e.g. touch) is unaffected by damage reduction.

[Timo Pietilä]

The main issue with damage reduction, though, is that it only applies to physical hits, which is a poorly-defined term. IIRC it's something like hit, bash, trample, and kick are all physical hits so long as they have the HURT attribute, but anything that does something else on a successful hit (e.g. drain life, disenchant) or uses a different verb (e.g. touch) is unaffected by damage reduction.

That's the main reason why Titans hit so hard. Hit to confuse does go thru at full force.

OTOH if it is elemental attack and you have immunity then reverse is true, you don't get damage at all (problem with monsters like Pazuzu).

IMO both cases should be fixed, armor should reduce all physical damages from blows, and immunity should not remove all damage from melee-attacks. Maybe divide damages to physical and elemental and act accordingly with armors and resistances/immunities.

Get base dice for hit, bite, kick etc. which gets reduced by armor + same base damage which gets multiplied by element - reduced by resistances. Something like that.

Because confusion is effect and not element it doesn't affect base dice, but armor should.

[Magnate]

That's the main reason why Titans hit so hard. Hit to confuse does go thru at full force.

OTOH if it is elemental attack and you have immunity then reverse is true, you don't get damage at all (problem with monsters like Pazuzu).

IMO both cases should be fixed, armor should reduce all physical damages from blows, and immunity should not remove all damage from melee-attacks. Maybe divide damages to physical and elemental and act accordingly with armors and resistances/immunities.

Get base dice for hit, bite, kick etc. which gets reduced by armor + same base damage which gets multiplied by element - reduced by resistances. Something like that.

Because confusion is effect and not element it doesn't affect base dice, but armor should.

Agreed. This is ticket #1580 - I've updated it to reflect that it needs to work both ways. This is scheduled for 3.5.0 because it's been uncontroversial so far - but I'll probably implement it in v4 first to check that there are no problems arising.

[CliffStamp]

Ideally the attack is partially physical and partially elementary. If you get hit by a fire demon it would make sense that the heat is going to do some of the damage and if you are immune to fire then some of that should be prevented.

Rather than direct release of all internals, would it be possible to implement a fuzzy system where the information comes from use? You put on armor, you get the base class, when you get hit you get the +, as you get hit more you get information about how much absorption, etc.?

Similar for weapons and damage, it just seems odd to me the player has so much information which there is no way they should have just by picking up a weapon. Even if skilled, there is no way you can know these things without actually trying them out.

[Derakon]

Ideally the attack is partially physical and partially elementary. If you get hit by a fire demon it would make sense that the heat is going to do some of the damage and if you are immune to fire then some of that should be prevented.

Rather than direct release of all internals, would it be possible to implement a fuzzy system where the information comes from use? You put on armor, you get the base class, when you get hit you get the +, as you get hit more you get information about how much absorption, etc.?

Similar for weapons and damage, it just seems odd to me the player has so much information which there is no way they should have just by picking up a weapon. Even if skilled, there is no way you can know these things without actually trying them out.

I'm all for more encouraging of ID by use. But to me, the numbers are the efficacy -- if you know the base AC and the AC bonus, then you know what the absorption is.

What we might want to consider for armor is splitting out ability to deflect blows altogether from the ability to absorb some of the damage. Basically the evasion/absorption thing that monsters now have, applied to the character as well. Shields would increase your evasion (by blocking attacks) while body armor would increase absorption. Functionally I suppose this means that armor stats would be given as [A + B, C + D], and you'd need to learn both A and B to know how evasive the armor makes you, C and D to learn how absorptive.

[CliffStamp]

Yes, know in the sense it is determined, but now know in the sense you would be able to produce it - the raw number. It just seems odd to me that the characters would be able to know the results of some fairly complicated equations.

When things started to be split up, I didn't think I would like it (separate stun and sound, etc.) however after playing it my opinion changed and now it is just another list of attributes to take into account and it balances out gear selection as instead of just trying to maximize con/speed you also can get various status protections which is very nice.

Similar I would suppose for armour, the more options the more that gear choices are not so unbalanced/trivial in general.

[Derakon]

Yes, know in the sense it is determined, but now know in the sense you would be able to produce it - the raw number. It just seems odd to me that the characters would be able to know the results of some fairly complicated equations.

So e.g. instead of saying "Your chance of being hit by this monster is 18%" you'd rather say "You have a superb chance of dodging this monster's attacks"? That is, using broad categorization adjectives in place of specific numbers.

We aren't providing these numbers for the character, but for the player -- of late, Angband's philosophy is that the equations that govern the game should be transparent, and pretty much any math you could figure out by source-diving should be done for you. This isn't necessarily the right way to do things; plenty of roguelikes prefer to keep their players guessing. Personally I like having a more "pure" game of tactics, but I do recognize that we lose something of the flavor and uncertainty of gameplay by doing this.