V 3.5 now in feature freeze

[Timo Pietilä]

Is it a bug that if you choose location targeting, but a monster happens to be on the location, the targeting reverts back to monster targeting? It seems to me that when I choose to target a location rather than a monster, that I really mean to target the location. As it currently stands, if a monster happens to be there when I target the location, the targeting reverts to targeting the monster and when I kill it, I have to retarget the space. Anyone else find this frustrating?

+1 from here. Location targeting should remain location targeting. It isn't player error to target location instead of monster if the monster is in that space.

[takkaria]

Is it a bug that if you choose location targeting, but a monster happens to be on the location, the targeting reverts back to monster targeting? It seems to me that when I choose to target a location rather than a monster, that I really mean to target the location. As it currently stands, if a monster happens to be there when I target the location, the targeting reverts to targeting the monster and when I kill it, I have to retarget the space. Anyone else find this frustrating?

Well, this was an intentional change. But maybe it shouldn't have been. Someone wanted it to be the case that if they targeted a grid with a monster on, it targeted the monster and that seemed like a good idea to me because sometimes you want to target a monster the game won't let you, so you have to target the square instead. But, I can change it back. Seems most people were happy with the previous behaviour.

[Ingwe Ingweron]

Well, this was an intentional change. But maybe it shouldn't have been. Someone wanted it to be the case that if they targeted a grid with a monster on, it targeted the monster and that seemed like a good idea to me because sometimes you want to target a monster the game won't let you, so you have to target the square instead. But, I can change it back. Seems most people were happy with the previous behaviour.

Thank you! I know I was happier when it was my choice whether to target monster or location. Besides, didn't the original behavior target the monster by default and only if you pressed "o" or "p" did the targeting change to location (starting at last monster target position "o" or at player position "p"? I know I liked having the choice.

[Beirlis]

My plan for Pyrel is for Teleport Other to be able to land you any distance up to the maximum distance, rather than for it to try to force you to be at least some minimum distance away.

I wouldn't want to see that spell (and effect) become so unreliable. I would prefer to see it weighted towards teleporting you far away. So something like distance = max - rand_range(0, max * max - 1) / max where the median distance is 75% of your max distance. So basically if your max distance is 100, you'll have a 30% chance to be at least 90 spaces away, 50% chance to be at least 75 spaces away and only a 10% chance of being less than 20 spaces away. It doesn't have to be weighted that heavily, but I would hate to have it become highly unreliable.

Also, why are they removing charisma? I know that it had limited utility and I'm sure this part has been discussed elsewhere, but why not add more usefulness to charisma instead? (perhaps affecting the chance of spells like slow, confusion, sleep, etc.)?

[MattB]

why not add more usefulness to charisma instead?

Why indeed. But equally, why?

[Mikko Lehtinen]

It would require big changes and new game mechanics to make Charisma equally good with physical stats. There isn't any consensus among Angbanders about what those effects should be. I guess removing Charisma was the easier option, and perhaps the best one.

On the other hand, I like pretty much all the wildly different takes on Charisma introduced in variants. (At least Un, Iron, PosCheng, and HoM come to mind.)

[Timo Pietilä]

It would require big changes and new game mechanics to make Charisma equally good with physical stats. There isn't any consensus among Angbanders about what those effects should be. I guess removing Charisma was the easier option, and perhaps the best one.

On the other hand, I like pretty much all the wildly different takes on Charisma introduced in variants. (At least Un, Iron, PosCheng, and HoM come to mind.)

OTOH, one "junk" stat wasn't that bad. Every time lose one -gain one hit CHR that was free stat boost, nexus scramble hit when CHR was already max could give you sudden huge boost to one stat.

In old days shops also were affected by CHR as well as race quite a bit.

[Mikko Lehtinen]

OTOH, one "junk" stat wasn't that bad. Every time lose one -gain one hit CHR that was free stat boost, nexus scramble hit when CHR was already max could give you sudden huge boost to one stat.

There's still one or two junk stats left, INT and/or WIS depending on class. Thankfully there's only one non-spellcaster class in Angband, so usually you only have one junk stat.

[Derakon]

I wouldn't want to see that spell (and effect) become so unreliable. I would prefer to see it weighted towards teleporting you far away. So something like distance = max - rand_range(0, max * max - 1) / max where the median distance is 75% of your max distance.

The difficulty is that when the game tries to place you far away, it has very few valid locations where you can land. That's what creates the problem with the current system. The spell needs to accept closer targets.

I suspect when we test it that we'll find that the spell is basically as reliable as it used to be, i.e. you shouldn't cast it when you're 1 turn from death anyway. My hope is that it will fix the problem where you get bounced between 2 or 3 bad situations when there's plenty of dungeon left that you could land in.

[takkaria]

There's still one or two junk stats left, INT and/or WIS depending on class. Thankfully there's only one non-spellcaster class in Angband, so usually you only have one junk stat.

I think it would be good to use the non-spellcasting int/wis stat as the fail rate stat. I'm not going to lobby for this to go into V but it would make sense.

[Mikko Lehtinen]

On the other hand, I like pretty much all the wildly different takes on Charisma introduced in variants. (At least Un, Iron, PosCheng, and HoM come to mind.)

I had to check what Ironband did with Charisma... "Charisma has been replaced by four new stats - Agility, Stealth, Perception, Luck." So Ironband doesn't actually have any Charisma mechanics, sorry!

To me, both Stealth and Perception make more sense as Angband stats than Charisma.

How about this:

Strength
Dexterity
Constitution
Power (magical ability)
Cunning (Perception + Stealth)

Power should contribute heavily to the Magic Device skill. Especially Warriors with low Power stat would be completely useless with devices. Power would also increase your spell save.

[Derakon]

We've had this conversation a number of times before. A couple of cautions:

* Warriors don't need nerfing, especially with magic devices. They're totally reliant on detection items to be playable at all; meanwhile, they don't get much use out of the others (attack items, rare rods, artifact activations, etc.) already. So the current balance is fine, really.

* Stealth, as currently implemented, only has 32 possible values; it uses exponentiation (every +3 stealth makes you about twice as stealthy). As a result, there's not much room to "play". This is fine currently because stealth is just race modifier + class modifier + equipment modifier. But if you make Stealth a "first-class" stat, then that implies to me the existence of Potions of Stealth that permanently make you more stealthy. I mean, what's the difference otherwise between the first-class stats and the others?

Assuming you fix that, if you want stealth to have a meaningful progression, then you end up with practically everyone aggravating monsters in the early game and being nigh-totally silent in the late game. Fixing that would require scaling monster alertness with depth, so that e.g. great hell wyrms would be preternaturally alert compared to novice rogues. And at this point I'm really not certain what we've gained.

I'm not saying these problems can't be overcome, but you will need solutions to them, and I'd say you also need to demonstrate that the new system would be superior to the old.

[Mikko Lehtinen]

We've had this conversation a number of times before. A couple of cautions:

Yeah, I know. The removal of Charisma made me happy, and that inspired some new brainstorming.

Thanks for reminding me that stats max out in Vanilla, and that limits design possibilities. I tend to forget it since I mostly play my own variant where stats almost never hit the maximum.

[Mikko Lehtinen]

Warriors don't need nerfing, especially with magic devices.

I was just proposing that you tie Warrior's Magic Device progression purely to his Power stat -- or let's say Intelligence -- and not to gaining warrior levels. Other classes would also gain bonuses to Magic Device from experience levels, but not warriors.

You could do the same thing with Wisdom and Spell Save. Make the stat much more meaningful, and reduce the relevance of gaining levels. Late game would be identical to what it is now, but early game would be more varied.

[Beirlis]

The difficulty is that when the game tries to place you far away, it has very few valid locations where you can land. That's what creates the problem with the current system. The spell needs to accept closer targets.

I suspect when we test it that we'll find that the spell is basically as reliable as it used to be, i.e. you shouldn't cast it when you're 1 turn from death anyway. My hope is that it will fix the problem where you get bounced between 2 or 3 bad situations when there's plenty of dungeon left that you could land in.

Ahh, I get it. Well I can think of a few ways to do it so that it still solves that problem, but is still weighted towards sending you further away. So iterate over all of the spaces that are max_distance away and make an array of open spaces, but exclude those that are less than x distance (maybe less than 20% of your max?). Then, you know the exact number of places you can go and pick a random number as an index into that array. But this would involve allocated and freeing the array.

The second solution I can think of (which isn't as pretty) is just keep choosing numbers between 0 and max * max * 4 and then check the location of that offset until you find a valid location.

The only other problem I can see is when you get into those maze levels where the distance from one end to the other isn't that far. So maybe the min distance should be sensitive to the level size.