Saturday, 28th June 2014, 18:40 by tedric
My strategy depends on the character, but the rule of thumb is that you want to do the most damage as quickly as possible. That means weighing the relative value of each specific weapon's combination of delay, brand, base damage, and enchantment (roughly in that order).
For melee, I pick a starting weapon class and stick to it pretty much exclusively -- I don't even try out weapons of other classes. The one possible exception is if I find an exceptionally good weapon of a type that cross-trains with my starting skill rather early in the game, then I might switch. I upgrade base types as soon as possible (unless I'm lacking the skill to make the delay acceptable) and sometimes downgrade base types for a useful brand.
For casters, I don't train any weapon skill until I've got plenty of magic and defensive skills. Along the way I look for low-base-delay weapons with brands like elec, pois, draining (or pain if I'm training necro). Whip or short blade of elec is usually my top choice. If I run across any higher-tier weapons that could have potential if I choose to train their skill later (a glowing great mace if I've already got a whip, a good artefact of any type I'd be willing to train, etc), I collect them. I'm usually ready to make a decision and start weapon training around Lair time, but sometimes even later.
In general I'd say delay is the first concern: ideally your delay is always 1.0 or less so as not to give enemies extra attacks (but that doesn't stop me from using a 1.1 base delay whip of elec with 0 skill). Brand also matters quite a bit: additive brands like elec are better on fast weapons, which tend to have lower base damage; multiplicative brands like flaming/freezing are better on high-base damage weapons, which tend to be slower. As Thalfon said, base damage matters more than enchantment -- so all else being equal, a base type upgrade is usually worth losing a few points of enchantment.
Of course, all else is usually not equal, so to resolve this issue you need to choose whether you want to treat weapon selection as an art form or as a science. If you pick science, there are formulae and spreadsheets and wizmode and fsim to help you get a mathematically precise answer, at the cost of time and potentially fun. If you choose art, you'll listen to various players' thought processes, experiment in-game, and develop your own intuitive method. The cost here is that intuition can lead you astray -- it may seem like a good idea to use a slow weapon that is likely to one-shot most enemies if/when the blow lands, even though you might allow a couple free hits as you're swinging it; this may work for a while, but is likely to get you killed the very first time you underestimate an enemy even slightly. An extreme example is the artefact Dark Maul, which delivers tremendous damage but can never be brought down to a "reasonable" delay even at max skill: its damage output is tempting, but pretty much nobody takes it seriously as a primary weapon (it is fun to play with, though).
My strategies above are based on a combination of art and science approaches: knowing some of the formulae and mathematical principles improves my intuition, though I tend not to crunch numbers in the midst of play.
Wins (Does not include my GrEE^Veh 15-runer...stupid experimental branch)
- For this message the author tedric has received thanks:
- linear