Rebalancing device difficulty

[backwardsEric]

After 4.2.3 was released, Nick proposed rebalancing how difficult magical devices are to use, https://angband.live/forums/forum/an...481#post225481 . A first draft of that rebalancing is up in the second comment on https://github.com/angband/angband/issues/6179 . The intent is to include the balance changes in 4.2.6 whenever that is released.

If the changes listed there go too far, not far enough, or don't include a device that you feel also needs to be rebalanced, please reply to this thread or leave a comment on that GitHub issue. As it currently stands, the draft has few changes to wands, staves, rods that damage, heal, banish, or smash the dungeon so comments on those would be especially welcome.

Balancing the activation difficulty for randarts will need more than changes to the data files (thus https://github.com/angband/angband/issues/6187 ), so that will probably be done separately and may not make it into 4.2.6.

[Nick]

Just noting, this was a lot of work from backwardsEric

[MITZE]

Going entirely from memory of FA2 (which it's been some time since I've played, and I acknowledge that it may not be relevant since this is a V topic), the main thing I remember is thinking that -Curing was difficult to use for anything that wasn't a device race/class, and didn't seem to be relevant anyway because by the time it was showing up you were probably stocking CSW or CCW on your person—the exact situation I think I'm remembering is Green-Elf Ranger and having like only 60-70% success rate on it, finding it sometime in Amon Rudh? Not positive, like I said it's been some time.

Oh, and _RemoveCurse being identical to the scroll but being a dungeon staff and decently deep. Just FWIW.

[Estie]

Rod of curing becomes easier to use, so that is fine; also, Eric went through all the artifacts individually A lot of work indeed.
I haven?t been a fan of dragon armor; with the improved activation chance, Ill give them a fresh try.

[PowerDiver]

Rod of curing becomes easier to use, so that is fine; also, Eric went through all the artifacts individually A lot of work indeed.
I haven?t been a fan of dragon armor; with the improved activation chance, Ill give them a fresh try.

Rod of curing is supposed to cure confusion, so IMO it's activation chance needs to be balanced assuming you try to activate it while confused. If it can't be used when confused in battle, it is not worth a slot.

[MITZE]

Another thing I've remembered is that _RemoveCurse (a decently deep dungeon staff) is identical in effect to the scroll (which is sold in town)—giving you no reason to use it, except in the extremely niche case that you find it before finding a scroll.

I've always thought there should be some mid-tier curse removal between ?Remove and ?*Remove*. Again, this is going by FA2 memory, so it might be outdated or not applicable to V.

[backwardsEric]

Rod of curing is supposed to cure confusion, so IMO it's activation chance needs to be balanced assuming you try to activate it while confused. If it can't be used when confused in battle, it is not worth a slot.

With the proposed changes it has a difficulty of 40 so a device skill of 56 or better should bring the failure rate to less than 10%. If the character is confused, the level the player would have to be to get to that device skill is (assumes no more intelligence than what the character got with the default point buy):

	Warrior	Paladin/Blackguard/Druid	Ranger	Priest	Rogue	Mage/Necromancer
Half-troll	-	43	41	40	37	28
Half-orc	-	41	37	36	34	24
Human	-	38	34	34	32	22
Half-elf	-	35	32	31	29	21
Dunadan/Kobold	-	34	31	29	27	19
Elf	-	33	30	29	26	17
Dwarf	-	32	29	27	25	17
Hobbit	45	24	21	19	17	11
High elf	42	22	19	18	15	10
Gnome	40	21	18	16	14	9

[PowerDiver]

Keep in mind that the rod of curing isn't worth carrying after the char has poison resist and confusion resist. I can't really judge those char levels without playtesting.

I'm sure you're moving in the right direction, but if you don't think you are overdoing things, I doubt you are going far enough. I'd just remove the rods. It's better to use a staff instead, unless you make the rod as easy to activate as the staff or you cannot afford the 4 lbs extra weight.

[agoodman]

It sounds like this is almost all configuration (gamedata) changes. Is there any code piece I should be keeping an eye out for reflecting in the borg code?

[backwardsEric]

The changes proposed in the original post are all to the activation difficulty, the "level" parameters in object.txt and artifact.txt. Internally, that corresponds to the level field in struct object_kind or struct artifact. They will affect the failure rate and the amount of damage boost. There's no changes to the base effects of the magic devices. What I'm working up for randart activation difficulty will add a level parameter in activation.txt and in struct activation. A randart that gets an activation will use the level value from that activation. A randart that inherits the effect of its base kind will use the level value from that kind. All other randarts, those without activations, will have level set to their minimum depth.

Mitze's suggestion to differentiate the staff of remove curse from the scroll of remove curse is up as a separate pull request, https://github.com/angband/angband/pull/6195 . If that's accepted, the curse removal power of those staves will between the effect of a scroll of remove curse and a scroll of *remove curse*.

EDIT: Autorepeat on device failure (only happens if the repetition has a chance to work; so failure rate less than 100% and device either still has charges or is not recharging) has already been checked in to the post 4.2.5 releases. It works like the autorepeat for digging or unlocking. That could impact the borg, especially if it was experimentally determining the failure rate .

[backwardsEric]

Autorepeat on device failure (only happens if the repetition has a chance to work; so failure rate less than 100% and device either still has charges or is not recharging) has already been checked in to the post 4.2.5 releases. It works like the autorepeat for digging or unlocking. That could impact the borg, especially if it was experimentally determining the failure rate .

[backwardsEric]

Proposed activation difficulty settings for the randart activations are up as the second comment at https://github.com/angband/angband/issues/6187 . There's been some slight changes to the proposed difficulties for the standard items in the second comment at https://github.com/angband/angband/issues/6179 :

1. rod of curing is at 35 rather than 40 based on the feedback above
2. staff of curing is at 20 rather than 25 since the difficulty on the rod was lowered
3. increased the difficulty for Th?oden from 20 to 30 since it has a stronger effect than the activation on Turmil which has a difficulty of 20

[Nick]

Thanks for doing all this; I reckon give people a couple of weeks and then we pull it in.