The Tavern

One of the biggest struggles I have with this game is knowing what weapon is best for me. I'm not talking about which of the weapon type I should train, I'm talking about which specific weapon I should be wielding at this very moment, given my stats. With extremely complicated formulas for base damage, delay, enchantment modifier, brand - it gets really hard to actually know which weapon I should be wielding. What I suggest would be to display a crude measure of your character's damage with a given weapon with a "damage per aut" rating upon inspection of a weapon.

Now, there are already spoilery ways of being able to figure this out, but even though I would say I'm fairly involved with this game, I've never bothered to actually go look it up because it feels somehow wrong, and it takes too much time. I would venture to guess that the vast majority of people who play crawl have a similar attitude. Yet if this were somehow easily visible, it would be a massive quality of life change for me.

In terms of the math, it could just be an average of the unreduced damage roll of your weapon, divided by the average number of auts (0.1 turns) that it takes you to swing it.

For example:


It would also help if it displayed damage per aut from your brand (if applicable) as well. Vorpal weapons could just add it directly to the score because it's all physical damage, whereas other brands (pain, electrocution, flaming, freezing, draining, holy wrath) would display it separately, with the corresponding damage type. For example:

You know it will differ for like any single monster/equipment combination?

Right, it wouldn't count AC of monsters because of course that's impossible, and it would would just recalculate the "dpa" each time you inspect the weapon, based on the current equipment you have on.

The point of this is not to know exactly how much damage you deal, but rather comparing the relative damage of weapons you might want to wield.

Knowing the average damage roll per aut is only good for comparing relative damage of two weapons of the same delay -- a quick blade suffers around twice as much from AC than a demon blade does, for example, when both are at min delay.

I have proposed picking a small set of 'representative' monsters in the past to give dpa figures for, but nobody seemed to like the idea.

Your 'average' melee damage really doesn't mean much.

The real problem is that the damage calculation for melee is much, much swingier than that for spells. You can kill a high-threat black mamba or cyclops in three hits or twelve with the same character, pretty much discounting misses AND variation in the monster's starting HP. If you've destroyed 4 naga warriors without much trouble and are wrecked by the 5th, how are you to know that you should have been more careful?

You're right, Hurkyl, DPA wouldn't give a precise picture of how much damage you can deal based on AC. That is not the point of the feature.

The idea here is to help players choose between, for instance, a broad axe and a battleaxe of different enchantment levels and brands, at a non-optimal aptitude ratings. That's it. It's for relative comparisons of damage, not an absolute measure. Just like no one looks at the "base damage" of a weapon and assumes that means they'll be doing at least that amount every turn. They just use it to compare between two weapons.

For experienced players this amounts to no change at all if they already have a firm grasp of the mechanics. For inexperienced players, it will help them to make better decisions most of the time, and while they might be misled in a few cases, I don't see how the current system of displaying only base damage and delay is any less misleading. It still doesn't explain to me why a quick blade sucks against a high armored opponent, for example.

Swinginess of damage rolls is a different issue, and maybe you can make a different thread about that. I think swinginess is just a way in which the game tries to kill you, exactly as you described above with the 5th naga. I don't see the relevance with the issue of picking which weapon to wield, though.

And to get an accurate comparison between weapons with different attack delays, one must account for AC, since that affects the relative performance of the two weapons. If you ignore it, you will be biasing things to make low delay weapons look better than they are in relation to higher delay weapons -- significantly so in some cases.

Please don't continue commonly known "it is not perfect, so let's not add it" strategy. There are multiple factors which can be accounted for (slaying, aux attacks, antimagic-bite, might, augmentation, EV, AC etc.) but in 80%+ cases you don't care much about those, you have a +4 eveningstar of freezing, +2 greatmace of "vorpal" and +6 morningstar of electro and you want to know which weapon you should use.

I'm not saying "don't add something" -- I'm saying "add the better thing" or at least "add the simpler thing that tells you the same information as the thing you're suggesting in the situations it's actually accurate but is less likely to be misleading about its usefulness" (i.e. "per hit" figures rather than "per aut assuming perfect accuracy" figures).

I am not sure if it is technically possible to calculate without running fight simulations. If it is, then yes, getting more accurate info is always welcome of course.