randSpellbooks

[Mark]

I was thinking about spells today, and how I tend to use a small subset of all the spells available to my class. (at least for 95% of the time)

Maybe it would be a fun challenge to survive on a random subset - giving the player different/less options, and so having to develop new strategies.

Like Randarts, there could be an option that randomly assigned spells to books, (respecting a certain range and spread of spell level).

• Which spells were found in dungeon / town would vary.
• By choosing a subset of, say 70% of all spells (i.e. a random 30% didn't exist in your game), then even more variety would occur. (Possibly mages/priest wouldn't like this, but could be fairer for rogue/ranger/paladin perhaps.)
• The 'missing' 30% could still be found in scrolls, devices, etc. thus increasing the value of such items.
• If any spells were considered almost-required to win, or class-defining, they could be marked as 'always include' (like the One ring in Randarts).

Having to live without your standard detection/attack/escape routine would surely be a fun and challenging option?

[Nomad]

I think fixed town books and randart dungeon books might be a potential compromise. The basic spells that you use over and over involve a lot of macros and muscle memory that it would be a pain to relearn every game, but it might make the dungeon books more interesting if they were randomized (and potentially solve issues like certain books being all but useless to some classes). You could make sure all the 'higher' spells available to the player's class are distributed across the five dungeon books, and then fill out the extra slots with random picks from the town books.

Or maybe you could have just the first one or two books of each type fixed and available in town, and randomize all of the others. (I've been playing a lot of rogues lately, and I find I basically do 99% of my spellcasting from MB1, and only occasionally reach for a different book when certain specific situations come up.)

[fizzix]

I think fixed town books and randart dungeon books might be a potential compromise. The basic spells that you use over and over involve a lot of macros and muscle memory that it would be a pain to relearn every game

There's no need for this. Limit it to 52 spells, all of which are assigned a slot a-x, A-X. All spells are all immediately mapped to the letter. To cast magic missile you use "ma" and if you don't have a book with magic missile equipped, it says, "you don't have a book with that spell in it" Easy-peasy.

If you want to be fancy, do the DCSS thing that allows the player to remap spells to letters of their choice, and then store that in pref files.

And of course, while we're doing this, add the feature of "confirm on casting this spell"

[Nick]

There's no need for this. Limit it to 52 spells, all of which are assigned a slot a-x, A-X. All spells are all immediately mapped to the letter. To cast magic missile you use "ma" and if you don't have a book with magic missile equipped, it says, "you don't have a book with that spell in it" Easy-peasy.

If you want to be fancy, do the DCSS thing that allows the player to remap spells to letters of their choice, and then store that in pref files.

And of course, while we're doing this, add the feature of "confirm on casting this spell"

This sounds really good (after the restructure, of course) - I think getting spells down to 52 without randomness would not be too hard.

If we were going to throw in randomness, I think the best way would be to introduce a new class.

[Bogatyr]

I was thinking about spells today, and how I tend to use a small subset of all the spells available to my class. (at least for 95% of the time)

Maybe it would be a fun challenge to survive on a random subset - giving the player different/less options, and so having to develop new strategies.

Like Randarts, there could be an option that randomly assigned spells to books, (respecting a certain range and spread of spell level).

• Which spells were found in dungeon / town would vary.
• By choosing a subset of, say 70% of all spells (i.e. a random 30% didn't exist in your game), then even more variety would occur. (Possibly mages/priest wouldn't like this, but could be fairer for rogue/ranger/paladin perhaps.)
• The 'missing' 30% could still be found in scrolls, devices, etc. thus increasing the value of such items.
• If any spells were considered almost-required to win, or class-defining, they could be marked as 'always include' (like the One ring in Randarts).

Having to live without your standard detection/attack/escape routine would surely be a fun and challenging option?

I think a more profitable tack to take is asking why do you only use a subset? I also tend to use a subset and for one thing it's because so many of the existing spells are almost pointless: confuse and sleep work so seldomly that I barely even bother. Does anyone actually cast polymorph except once or twice for amusement value? Or wonder? If confuse/sleep spells would work with a much higher percentage against more monsters then I would certainly use them more.

Surviving/winning as a mage is already tough enough IMO. Maybe warrior should be made harder by introducing "fatigue" to make fighting more limited, like spell-casting (fight points as an analogy to mana points)?

I like the direction of the recent mana reduction for attack spells, but I think more needs to be done to allow mages to make it all the way through the game including the final fight without ever needing to swing a weapon.

[Derakon]

Surviving/winning as a mage is already tough enough IMO. Maybe warrior should be made harder by introducing "fatigue" to make fighting more limited, like spell-casting (fight points as an analogy to mana points)?

Mages and warriors are hard at different parts of the game. Warriors are hardest in the midgame where they sorely lack detection; mages are hardest in the early and midgame where they have somewhat poor offense and lack survivability.[/quote]

I like the direction of the recent mana reduction for attack spells, but I think more needs to be done to allow mages to make it all the way through the game including the final fight without ever needing to swing a weapon.

I am strongly opposed to the idea that an Angband mage should be restricting the activities that they allow themselves to do. Tolkeinian mages were not afraid to get their hands dirty in combat if that was how they judged it best to resolve a conflict. It's not necessarily going to be their default choice, but neither should they be able to rely on their spells for everything they want to do.

Restricting class capabilities is where the interesting gameplay comes from. Warriors have trouble with detection, which creates some difficult choices. But they have fantastic durability and combat strength, which they can lean on to make up for it. Mages have trouble with mana supply and, until Raals comes around, with having attack spells that work against all monsters. But they have fantastic utility spells, which help make up the balance.

If you want a class that can kill everything with just spells, then it should be a different class. Call it a "blast mage" or something, and give it only attack spells -- no detection spells, no mobility spells, just arcane destruction.

[Bogatyr]

I am strongly opposed to the idea that an Angband mage should be restricting the activities that they allow themselves to do.

Exactly! As it is now, mages are basically forced to use weapons because attack spells are so insufficient, still. That is a restriction! No "blast mages" needed for balance, the mage being so incredibly subject to insta-death throughout most of the game means they basically have to remain out of line-of-sight of most monsters most of the time. This disadvantage more than makes up for the swiss army knife of utilities available, several times over.

Giving mages a better mana per damage so that they *could* do the final fight without a weapon (requiring restore mana, to be sure, they could be hoarded throughout the game just like how a warrior must hoard mass banishment, etc.) creates choice in game play, the opposite of restriction.

[Derakon]

Exactly! As it is now, mages are basically forced to use weapons because attack spells are so insufficient, still.

Okay, I so poorly-stated my thesis that you were able to misread it as saying the opposite of what I intended. My bad.

Classes must have weaknesses. A class that is not threatened is boring to play. You argue that the mage's weakness is their frailty, and that this frailty is such a big weakness that you could, and should, improve their offense to compensate. But a skilled player can compensate for that frailty, and then if there are no other limitations on their abilities, effectively slaughter everything in their path without feeling seriously threatened. The limited magical offense is there to ensure that you are forced to compromise your physical safety at times, because you can't kill everything from a position of safety.

When I said that mages should not restrict their options, what I meant was that a mage that has sufficient spell-based offense to deal with threats has no incentive to not use spells for every scenario. You end up with a class that just mashes "cast level-appropriate attack spell" until all the mans fall down. The mana/damage limitations in the game are there to force you to look to other options for, if nothing else, stretching your mana supply. And those options are there! You have:

* Potions of Restore Mana. Awfully plentiful these days, honestly.
* Attack wands and rods. Fantastic in the hands of a mage.
* Bows and other missile weapons. Lets you keep your squishy neck out of chopping range.
* Melee. Unlimited damage, but you do admittedly kind of suck at it.
* Running the hell away. This advice applies to a level-30 mage just as surely as it applies to a level-1 mage. You don't have to kill everything you meet.

[Bogatyr]

It's sort of confusing arguing that a *mage* must look to his non-magic attacks as a basic matter of course. There are plenty of ways the game has to compensate: crowds for one. Sure, a mage shouldn't be able to deal with arbitrarily large and tough crowds with his spell books alone. But shouldn't a mage be able to deal with *one* opponent just with spells? I think so, even beefy opponents.

And there's nothing wrong with level-appropriate damage spell use, that's the bread and butter of the mage. With inflated damage ability comes beefier opponents, opponents who summon, crowds of uresistible breathing opponents, etc., where one cannot simply stand and cast.

I suppose it's true about attack wands, I usually don't use them much and maybe I should look more to them. But the mystique of the mage to me is the massive power that a high level mage should posses.

[quarague]

But the mystique of the mage to me is the massive power that a high level mage should posses.

I think current high level mages have massive power due to their incredible range of utility spells. I would say, a level 40+ warrior can die due to a combination of unfortunate circumstances, a level 40+ mage only ever dies through stupidity of the human playing him. But this power does not come through the mages damage dealing abilities, but rather his utility spells.

One can imagine a class that has massive power from spell based damage abilities, and that would be interesting as well. To make such a class well balanced it should have a much more limited range of utility spells. As I think current mages are an interesting class that should be kept, this should be a new distinct class. Which one of the two will be called mage is rather irrelevant to me.

[Bogatyr]

I think current high level mages have massive power due to their incredible range of utility spells. I would say, a level 40+ warrior can die due to a combination of unfortunate circumstances, a level 40+ mage only ever dies through stupidity of the human playing him.

That's simply not true. There are unresistable breaths, power curses/mortal wounds, etc., and bad luck or just sheer numbers of opponents can land you in a place where the mage's puny hit points are quickly overwhelmed, even if you're casting detect/reveal every few steps. As Derakon noted, in order to make progress, it is unavoidable that you must put yourself at risk. With a mage, you're rolling a particularly dangerous die standing in line of sight of anything with a heavy distance attack. It takes a mighty long time getting to level 50 via magic missiing molds. At some point you must take on the big boys.

But point taken, the mage does have great utilities, which is why I enjoy playing them.

[quarague]

I'd say a level 40+ mage has enough hitpoints to survive a single attack from any monster out there (provided he has the basic resists + poison covered). He can easily die to 2 or 3 hits, whether from the same monster with high speed or from multiple monsters at once. But these are things than can be avoided through careful (and tedious) playing. A warrior is much tougher and can take multiple hits from almost anything, making him much more forgiving to play.
I would also say that taking occasional calculated risks is more fun (and much faster) than trying to play with zero risk, especially when playing mages.

[fizzix]

I'm going to make three completely disparate points that this thread has reminded me of.

1) Spell-class specialization. Many other roguelikes force their spell-dealing classes to choose some form of specialization. For example, you can choose to cast fire spells or ice spells, but not both. This is what allows distinctions between different runs of the same class. Angband doesn't do this. The only difference between successive runs of a mage class is the order in which you find the dungeon books. This is why I like the randomized book idea. It brings in some difference in the way a mage run will play out.

2) Multi-targeted damage spells are for convenience only. There are very few games that make good advantage of the many vs. one situation in roguelikes. For the most part, they all revolve around trying to separate tough enemies from each other and handle them individually. Angband is probably on the extreme end of desiring 1v1 engagements. That's why single target damage spells (bolts) are just as good (and sometimes better) than multi-damage spells (balls). This is also why Raal's seems underwhelming, and the best spell in it is probably Rift, and not the many ball spells it has. Your acid bolt is probably just as damaging as most of the Raal's spells.

3) Angband sucks at debuffs. While Angband is pretty good with its buff spells (haste, bless, resistance, prot. from evil) it's pretty awful with debuff, or status effect spells. They cost too much and are too unreliable. I'll make use of wands of confuse or slow monster in the early game, but I don't think I'll ever cast the spell. When we're thinking of improving mage (and priest) classes, this seems like a prime target area.

[Derakon]

3) Angband sucks at debuffs. While Angband is pretty good with its buff spells (haste, bless, resistance, prot. from evil) it's pretty awful with debuff, or status effect spells. They cost too much and are too unreliable. I'll make use of wands of confuse or slow monster in the early game, but I don't think I'll ever cast the spell. When we're thinking of improving mage (and priest) classes, this seems like a prime target area.

I'll just leave this out here...

[Bogatyr]

I'm going to make three completely disparate points that this thread has reminded me of.

1) Spell-class specialization. Many other roguelikes force their spell-dealing classes to choose some form of specialisation. For example, you can choose to cast fire spells or ice spells, but not both. This is what allows distinctions between different runs of the same class.

Great reason not to play the others. I like the choice and the ability to adapt to the situation (fire vs. ice) in vanilla angband.

Angband doesn't do this. The only difference between successive runs of a mage class is the order in which you find the dungeon books. This is why I like the randomized book idea. It brings in some difference in the way a mage run will play out.

Hmm, Maybe this won't be terrible. But then one should continue to find randomized books all through the game, and maybe able to eventually drop 2 books and replace them with one book that has what you actually use. Of course the downside is that you never may be able to build up the current level of spell books in the same number of slots. Mordinkanen's is a uniformly very useful book, I'd hate to have to piece it together out of 3 or 4 books. If so, then inventory size would need to be looked at. edit: make it a birth option, like randarts. Every randart game is a different game. Each randspellbook games for casters would be similarly different.

This is also why Raal's seems underwhelming, and the best spell in it is probably Rift, and not the many ball spells it has. Your acid bolt is probably just as damaging as most of the Raal's spells.

Raal's is underwhelming because of the immense mana cost of the spells. That can be adjusted to make the spell book live up to its name.

3) Angband sucks at debuffs. While Angband is pretty good with its buff spells (haste, bless, resistance, prot. from evil) it's pretty awful with debuff, or status effect spells. They cost too much and are too unreliable. I'll make use of wands of confuse or slow monster in the early game, but I don't think I'll ever cast the spell. When we're thinking of improving mage (and priest) classes, this seems like a prime target area.

Totally agree here (it was my initial point in this thread), slow/sleep/confuse begins and ends with "o"'s and "p"s generally. I'd like to see these spells remain useful up to and through the endgame.