Things I don't like about current V (long-ish)

[Derakon]

Yeah noselling is so unbalanced. It actually makes sense to buy a pick and dig out a level on your first dive. enough $ for the game pretty much :P

Multiplier should really be 1.5x (1.0 if its a challenge option, not a birth option)

We keep having this discussion. You say there's too much money with no-selling. I say my characters with no-selling have barely enough money to afford basic consumables (especially, staves and stacks of Cure Critical Wounds) through 1500', let alone to buy nice equipment. Clearly anecdotal data is insufficient for deciding this issue.

[Grotug]

Provides immunity to Acid! baahahhahahahahhwhhhh@@ hahhhh11hh1h@1! not to mention a couple other fine resistances. (picked up Shield of Thorin)

Yes, rAnything should reduce the amount of damage and frequency of damage done to equipment by acid, cold, fire and others. This only seems obvious to me that it would.

True, my melee weapon only took one or two to-dam hits off that I replaced later. Bought a mace of disruption from the black market. 837 damage to dragons. Still, two multihued ancient took me to the 100s hp twice, one with fire and the other with lightning. 500 something damage on dl52. crazy.

Max wisdom and regeneration sure do make one helluva powerful spell-caster!

I've just gotten used to getting armor damaged by acid and having potions get blown up by cold attacks, so it stopped bothering me ages ago...but it's clearly a big deal for other players. Do we want to rethink acid damage somehow? Currently every acid attack will damage something (assuming you have acid-vulnerable armor in every slot). And this is in addition to destroying a wide range of inventory items. We could nerf this:

* Only have a chance of damaging armor at all (say, 50%, or 75% without resistance, 25% with, or whatever)
* Only ever destroy inventory items or armor, but not both

Or something else?

As for disenchantment, I'm afraid this is Working As Intended. Monsters that disenchant gear are rare, and you're usually best-served by avoiding them. You said you couldn't run away; why not? Because he's faster than you? That's why you should be carrying a Staff of Teleportation (or casting Portal, but as a paladin you're probably not very good at that). Angband is always going to have monsters that are not worth fighting at your current power level, and Mim is definitely one of those. Come back later when you have one of resistance to disenchantment, a speed advantage, or strong ranged attacks.

And look at it this way: at least he (presumably) didn't badly drain your melee weapon!

As for holy spells, there's rather a lot of redundancy in them and the paladin spell difficulties and costs haven't been seriously examined in awhile. That said, holy spells are far and away better than arcane spells; they're only missing Haste Self, Resistance, and a high-DPS attack spell (Orb is great but a bit slow). If you're going to make more holy spells viable, then you'll need to take away something else to compensate. There've been plenty of discussions on ways to better-differentiate the different spell types and the different caster classes.

[emulord]

Yeah noselling is so unbalanced. It actually makes sense to buy a pick and dig out a level on your first dive. enough $ for the game pretty much :P

Multiplier should really be 1.5x (1.0 if its a challenge option, not a birth option)


Stacking resists vs not is a big enough change that balancing both is hard. I think the [1/3 1/9 imm ---] display would make it clear enough for not stacking.

If we do stacking, I like [1/2, 1/4, 1/6, 1/8...] for low resists
[2/3, 2/5, 2/7...] for high resists. Poison will have to have something specific. Maybe [1/3, 1/4, 1/5...]?
Its obvious what the progression is, while still being close to the unstacked behavior.

[Nick]

You must be joking, right? Of all the things to balance around an option being on or off, stacking resists must be about the most difficult suggested so far!

That sounds like a challenge

Seriously, I think this might well be possible. As an alternative to the current model, instead make each resistance let, say 60% of damage through. Then one source gives 40% resistance, two sources 64% (roughly what one gives now), three sources about 78% and so on. Now you need more than one source...

The 60% was only a guess; it would need tweaking. Also it might be hard to balance for all elements - some might need a different percentage.

[buzzkill]

You must be joking, right? Of all the things to balance around an option being on or off, stacking resists must be about the most difficult suggested so far!

Well, that's my point. We don't have to fine tune the balance against the option being on. We fine tune against the default setting, which in this case would be the status quo. Doesn't mean that we can't take a broad brush and paint in an option, even if it is more or less unbalanced. It's not like Angband is running with the precision of a Swiss timepiece to begin with.

Isn't this how no selling came to be. Jack it in there, slap a 3x multiplier on GP and see if it flies??? Balanced? I don't think so, but a great option none the less.

[Magnate]

I don't necessarily disagree with any of your substance, but the standard game is the default, and if Angband is to try to achieve any degree of balance then that balance must be struck against the default setting. Optional percentile stacking resists, Hellllll yeah!!! Maybe it's time to revert that whole "No, I not adding it as an option".

You must be joking, right? Of all the things to balance around an option being on or off, stacking resists must be about the most difficult suggested so far!

Nomad raises an interesting point though - I was indeed stuck in the mindset that the standart game is the "normal" one, and I think I agree that the optimisation puzzle is already pretty interesting and complicated with randarts.

I wouldn't be averse to making randarts the default - or, as has been oft suggested, having a mixture of the two.

[Patashu]

I could be wrong (haven't played a ton of DCSS), but I don't think any of those games have anything as bad as disenchantment (re: item destruction) or time attacks (re: stat drain).

Closest thing I can think of that is nearly as rage-inducing in DCSS is sticky flame?

DCSS has rot, which permanently reduces your maximum HP, stacks indefinitely and usually requires expending resources to cure points at a time. (There might be other things but it's been forever since I played)
And Nethack has a squillion ways to damage your equipment, items and stats/exp.

[debo]

I don't think item damage is a problem that needs fixing.
Nethack has item damage and destruction.
DCSS has item damage and destruction.

I could be wrong (haven't played a ton of DCSS), but I don't think any of those games have anything as bad as disenchantment (re: item destruction) or time attacks (re: stat drain).

Closest thing I can think of that is nearly as rage-inducing in DCSS is sticky flame?

[Patashu]

I don't think item damage is a problem that needs fixing.
Nethack has item damage and destruction.
DCSS has item damage and destruction.

[Raxmei]

As for holy spells, there's rather a lot of redundancy in them and the paladin spell difficulties and costs haven't been seriously examined in awhile. That said, holy spells are far and away better than arcane spells; they're only missing Haste Self, Resistance, and a high-DPS attack spell (Orb is great but a bit slow). If you're going to make more holy spells viable, then you'll need to take away something else to compensate. There've been plenty of discussions on ways to better-differentiate the different spell types and the different caster classes.

Yeah, conspicuously the Purifications and Healing versions of Cure Serious and Mortal Wounds are actually inferior (more expenseive) to the town book versions if you're playing a paladin. The only reason to keep a copy of that book at all as a paladin is restoration. Not that there's much point to casting anything between CLW and Heal anyway as a priest or paladin. CLW blows away anything short of a late game Heal in terms of SP efficiency and the other cure spells don't offer enough action efficiency to make up the difference. Cure Mortal heals twice as much as Cure Light, but costs five times as much SP for priests, ten times as much SP for paladins. That line of spells basically offers small incremental improvements in effect at rapidly escalating cost. And then Heal comes along, which is more expensive but actually worth it.

And then there are all the issues with their spell list other than healing. This would be an undertaking.

[buzzkill]

Basically, as you say, you're solving a problem that's only an issue in the standart game, and I think your fix for it would unbalance the difficulty of the randart equipment puzzle to the point where it would stop being fun.

I don't necessarily disagree with any of your substance, but the standard game is the default, and if Angband is to try to achieve any degree of balance then that balance must be struck against the default setting. Optional percentile stacking resists, Hellllll yeah!!! Maybe it's time to revert that whole "No, I not adding it as an option".

[Nomad]

I think they would make a difference to the everyone-has-the-same-endgame-kit problem. (We are of course talking about standarts here - I need no convincing that this isn't a problem with randarts!) This is what I thought Derakon meant by "solved" - I didn't think he was referring to mid-game kit.

I guess that's why we're coming from different angles - I stopped playing standart games years ago because I find having similar kit every game intolerably boring, even/especially in the very early stages. Adding stacking resistances to make the standart endgame kit less rigid would ruin randart games for me, because there the complexity of building a kit out of random elements is already sufficiently difficult that most of the satisfaction comes out finding a new arrangement that closes one more resistance gap than you had before, rather than the one in a thousand chance of assembling a genuinely 'solved' kit that covers everything. Make that equipment puzzle even more complicated, with fuzzier measures of having achieved something, and it stops being satisfyingly difficult and becomes too frustrating to even bother trying to optimise.

Basically, as you say, you're solving a problem that's only an issue in the standart game, and I think your fix for it would unbalance the difficulty of the randart equipment puzzle to the point where it would stop being fun. (While still not making me want to play with standarts, because it's the fact that artefacts stay the same from game to game that I find boring. So my favoured solution would be adding more randomness to standart games, whether by a mix of fixed and random artefacts, random artifact properties like you get with higher resistances on egos, a greater range of high-level egos that could potentially compete with artefacts, etc.)

[Magnate]

I'm with MattB and Derakon on this. Covering as many resistances, abilities and immunities as possible with the highest possible stat boosts is a fun and satisfying optimisation puzzle, especially with randarts. But endless optimisation of stacking gear for diminishing returns lacks that "Ah-ha!" moment of triumph of having achieved something concrete with your new kit. If juggling gear after my latest find allows me to gain Nether resistance without losing anything else, I'm smugly triumphant. If it just gives me 20% more Nether resistance at the cost of losing 10% of my Chaos and 5% of my Sound resistance, well, I don't feel like I've achieved much and it's hard to even judge if my new kit is any better than the old one.

Your thoughts on spellcasting captured the subsequent discussion, but I thought it worth offering up why I differ on this. I don't disagree that the satisfaction of ticking a new resist after a kit swap (without giving up anything else) is greater than the compromise you describe, but I don't think you're comparing like with like. If I got 20% more nether resistance without giving up anything else (consistent with your first scenario), that would feel good too. So I don't think that stacking resists make any difference to individual kit swaps.

I think they would make a difference to the everyone-has-the-same-endgame-kit problem. (We are of course talking about standarts here - I need no convincing that this isn't a problem with randarts!) This is what I thought Derakon meant by "solved" - I didn't think he was referring to mid-game kit.

Once upon a time I nerfed Thorin during one of my iterations over V's standarts - I took away one of the high resists (chaos or sound, I forget which). Thorin is one of - if not the single - most common piece(s) of endgame kit, and I wanted to make those decisions more difficult. A significant minority of old school players howled in protest precisely for this reason - they didn't want those decisions made harder! They wanted to be able to find Thorin easily and tick off one of the endgame kit boxes.

I accept that this problem is addressed more directly by randarts, but I think that stacking resists would also add subtlety, even in standarts, to what the optimal endgame kit is.

[Magnate]

Regarding monster mana: I've played NPP a decent bit and I found uniques to be intensely frustrating in that game, because you almost literally cannot make progress in fighting them until they run out of mana, and there's no realistic way to hasten that process. In the meantime, you just have to eat their most powerful spells over and over again (necessitating a gigantic amount of healing) while dealing as much damage as possible (in the hopes that they spend their mana on self-heal spells instead of attack spells).

Fights in Vanilla aren't exactly tactical puzzles, but they are unpredictable, and that has value. Fights in NPP almost universally followed a two-phase pattern: in the first phase, the monster blows all their mana on hitting the player, who has to tank the attacks as best they can; in the second phase, the monster is out of mana and thus has no effective offense, so the player can beat on them with impunity.

I'm oversimplifying, but not by a whole lot.

It seems like our problem is that monster spells are very powerful, with the unintended consequence that intelligent spellcasters are horrifically dangerous. To me, that says we need to do one of three things:

1) Not have intelligent spellcasters (current Vanilla)
2) Not have very powerful monster spells (casting a spell is roughly as dangerous as any other action they could take)
3) Put hard limits on how frequently monster spells can be cast (i.e. cooldowns)

Monster mana does none of these things.

It does do the third if the costs are high enough (buzzkill's suggestion). I think the key part of your post is "there's no realistic way to hasten that process". If we do bring in monster mana, we also need to bring in ways to drain it (for all classes).

But I also like your take that monster spells are unnecessarily powerful. Changing this would be a very long job with huge amounts of rebalancing, but ultimately for the better.

I like Nick's suggestion for intelligence with retained randomness.

[Nick]

How about randomness but with intelligent discarding of unsuitable options? e.g. If a caster has full health, they will discard healing or escape spells as options and choose randomly from among their other possible actions

This already does happen for intelligent spellcasters in Vanilla. What the 4GAI variants then do is rate the desirability of each of the spells and choose the most desirable.

It seems like our problem is that monster spells are very powerful, with the unintended consequence that intelligent spellcasters are horrifically dangerous. To me, that says we need to do one of three things:

1) Not have intelligent spellcasters (current Vanilla)
2) Not have very powerful monster spells (casting a spell is roughly as dangerous as any other action they could take)
3) Put hard limits on how frequently monster spells can be cast (i.e. cooldowns)

This is a good summary, but I think this may not be an exhaustive list, and there's also the possibility for having some combination.

After a lot of thought ("What would I do if I was a monster?", etc) I find myself somewhat surprisingly leaning towards randomness. So here's a possible model:

• Monsters assess the value of their spell choices; how well they do this depends on how smart they are
• Each spell then gets a score
• Chance of spell is then (spell score)/(total score)

In fact, the current model is this, with only intelligent spellcasters assessing at all, and the scores they assign then being other 0 (reject this spell) or 1. One of the virtues of this model is that granularity can be increased easily, and in a number of different ways (introduce variable smartness, introduce greater range of scores, allow for different scoring methods in different monsters, and so on).

Resistances: There will be a thread. You have been warned.