Special Fail Rate for Utility Spells

[Mark]

Hi guys,

A lot of improvements to Angband over the years seem to be 'we removed/changed this feature, because it was tiresome and didn't really offer meaningful choices to the player' (like casting Detect Traps every 30 turns). I think this is good.

I'm wondering if we should consider allowing some spells like Perception, Identify, Clairvoyance, Satisfy Hunger, to have zero or low fail rates if the player has no monsters in LOS (akin to whether the player can run). I find it pretty tiring when I first learn Identify to actually identify something via the spell, because of the amount of fails, mana cost, need to rest, etc.

If a monster came along while I was low on mana from trying to cast Identify, then there would be significance to it all - but still, not exactly fun. I'd rather the game effectively said 'you've got to level X, you've got spellbook Y, so you can identify things easily now'. Currently, it feels the game says: '...you can identify things in a painful way, but save a inventory slot and some gold, or keep doing it the easy way, and using the slot and spending the gold'. At least for the first X level of getting the spell (way more if Paladin/Ranger).

Does anyone else have gripes on the frequency of failing utility spells when you first get them? Do you feel the fails add to the game being fun?

[PowerWyrm]

My principal concern is for spells like Enchant Weapon or Enchant Armor. These will fail depending on the weapon/armor bonus, but will eventually work up to +15. So why waste the player's time by repeatedly fail...

What would be nice is to implement a repeat count for these spells, just like for the digging command. Then you would just use up game turns instead of real player time.

[buzzkill]

Does anyone else have gripes on the frequency of failing utility spells when you first get them? Do you feel the fails add to the game being fun?

I feel that removing the mana cost of spell failure would be disastrous.

[Philip]

I would say it is a major consideration - in O, mages have miniRecharge early, and Identify in the 4th book, expensive and high failrate, so I just keep a staff of Identify around. It leads to one of the nicest tough choices. Do I waste turns, time, and risk getting caught with low mana or do I sacrifice a slot. Later, the balance turns a bit, making the cost lower and fewer wasted turns. Just because you are capable of casting a spell doesn't mean it should be efficient. I wouldn't say repeated failures add fun, but I would say the presence of them does.

[takkaria]

I would say it is a major consideration - in O, mages have miniRecharge early, and Identify in the 4th book, expensive and high failrate, so I just keep a staff of Identify around. It leads to one of the nicest tough choices. Do I waste turns, time, and risk getting caught with low mana or do I sacrifice a slot. Later, the balance turns a bit, making the cost lower and fewer wasted turns. Just because you are capable of casting a spell doesn't mean it should be efficient. I wouldn't say repeated failures add fun, but I would say the presence of them does.

One way to work around this would be to add variable mana costs in return for a lower fail rate. Not sure exactly how this would work, I could imagine two scenarios:

1. Spells that function like this have a 'mastery level', e.g. the character level you get a guaranteed, low, mana cost for. Then the actual cost of the level varies: it's mana required plus 0% to (mastery level - current level) * x % extra, where x is determined by playtesting, and includes an element of randomness.

2. Truly random costs for all characters, e.g. 3 + 2d5 for Identify instead of 7.

[Philip]

Number one would be a good way to keep the mana cost. I like it. You avoid the "I just cast mcg or whatever 7 times in a row, rested up, cast it again, rested up..." while still forcing characters to either waste turns, or have low mana, and even risk being caught while weak.

[Mark]

Currently:

• the player doesn't know how much mana it will take to successfully identify an item (can't predict how many fails)
• the player can bail attempting to identify and move on/rest, after each failed attempt
• at lower levels, the mana pool will only suffice for 1 or 2 attempts without a full rest, so a high fail rate is not only mana-costly but requires a lot of key strokes from the player, and thus potentially a significant amount of game turns

I can see how preserving a high mana cost affects gameplay (although the original intention behind a finite mana pool was probably to reduce casters' ability to endlessly heal or zap things, I doubt there was a real concern you'd 'identify in a OP manner').

In order to not be caught with your mana pants down, presumably the safest way to attempt a utility spell is to attempt it once, and then rest to recover that mana, rather than draining your full supply, and only then resting.

In that case perhaps we should auto-repeat 'cast-once-then-rest-to-recover mana-cost', until the spell succeeds, the player is disturbed, or the repeating reaches a predefined cap. I.e. it works just like digging.

That would preserve the mana cost and fails rates as they are, and automate what a player would like to do, but doesn't because of the patience and keystrokes involved.

[buzzkill]

In that case perhaps we should auto-repeat 'cast-once-then-rest-to-recover mana-cost', until the spell succeeds, the player is disturbed, or the repeating reaches a predefined cap.

This sounds suspiciously similar to a macro, ahem, keymap, ahem, theoretical enhanced keymap.