So what else should we revert?

[bron]

1) I support the idea of reverting/modifying the item-level calculation. I will note however that this has a big effect at the deep end: there are no level 80+ "ordinary" monsters (only uniques), and only a handful of level 70+ monsters (and over a third of those don't drop anything - e.g. various Quylthulgs and hounds). This makes certain valuable things (e.g. rings of speed, amulets of trickery) a *lot* more rare. And the deep Dragon Scale Mails (not just Power) would almost never be found. Whether this is good or not depends on your point of view.

[Slightly off topic: I would support a somewhat more complicated version of the item-level calculation to fudge this a little. Say for example if the monster level is less than the dlevel, boost it by 10% rounded up (but not to more than the dlevel), and average that boosted value with the dlevel. This would have almost no effect early on, but would help towards the bottom. Or one could just make the item depths shallower, possibly in conjunction with removing levels 60-90.]

2) I think Rings of Escaping should just be removed, rather than trying to "fix" them. Rings of Teleport serve the "escape" function well enough.

3) CLW/CSW/CCW are overpowered; their effect should be capped (at say, three times their corresponding minimum).

4) I admit to being a minority (possibly of one), but I did not like the change that gave priests a combined "detect traps" and "detect doors and stairs" ability. I always liked the "flavor" of the fact that there were a
variety of detection spells for priests and mages (e.g. detect evil vs. detect monsters) and each one was just a little bit different. But if this change is going to stick, you should at least also combine the Rods into one.

[Magnate]

Greetings,

things I would like to revert :

1. Kobolds, really. They irk me.
2. New 'n' repeating code, it is functionally a step back.
3. Fricking game code files in subfolders in src
3b. I dont mind OS specific files in subfolders
4. Hengband help on enter, begone, if they aren't gone already.

I've checked the list of changes out, I have 2 remarks :
1. Wow, that is a boatload of good stuff
2. There is no consistency on using the [Regression] tag, all in all I feel that items like
* 'n' no longer repeats last action
* Casting resistance/resist poison crashes the game
* infinite loop
* Silver jellies aren't absorbing turn of light when they attack.
should have been tagged with [Regression] and if certain pieces of code cause too many regresssions, chuck it out instead of trying to fix the faulty re-implementation.

Finally,
the link to the coding standards is broken :

http://ajps.mine.nu/angband/wiki/CodingGuidelines

I am fairly curious as to what they are.

http://rephial.org/wiki/CodingGuidelines

Thanks for the thoughts about regression - we don't use that keyword at the moment, but it's an interesting idea. We'd need to distinguish between regressions due to coding errors and those due to duff ideas though!

[Timo Pietilä]

You are so far off the mark here it isn't even funny: go here and see the 227 tickets in 3.1.0, 205 in 3.1.1 and 178 since. A lot of those were bugfixes, of course, but there are several dozen significant changes in there as well.

You should define "significant" here. To me only additional thing that did stick out from that lot was new mushrooms, which need tweaking. Especially the duration of the bad effects for items that are supposed to be mixed blessing (like terror lasting forever and ever). There are obviously other changes too, but most of them are not either a) significant or b) broken so they do not raise discussion.

I think most of the "broken" parts of the game are result of overall cumulative small changes over a very long period of time. Some of them have roots at changes made over decade ago.

[PowerDiver]

Now this is why I started this thread - this is an excellent idea. At the moment with less-than-max CHA it's possible to sell stuff for the price the shopkeeper will then sell it on for - i.e. s/he makes no profit. That's silly, so I would support a significant increase in the buy:sell ratio as a move towards no shopping.

But that's a change, not a reversion!

To be clear, my problem with value pricing less about the economics than the gameplay. Consider

cost to buy MB8 / price received selling an average ego weapon

cost to buy boots speed+10 / price received selling avg ego weapon

Those ratios have dropped enormously because of value pricing. If you reduce selling prices, that would at least address this issue.

[Tiburon Silverflame]

So the approach isn't perfect - it's particularly difficult to conceptualise defensive qualities (like AC, resists, hold life etc.) in terms of "damage per blow". There are hacks needed to make the formula-based approach work in "real life" (by which I mean in the game) - as well as telepathy and speed, other examples are torches of Brightness and off-weapon blows/slays/shots.

So, in other words, the pricing scheme is broken. I gave *examples* Magnate, of obvious potential problem areas, to point out that reconsideration is a better approach than simply throwing it all out. And if anything, you've convinced me completely, that there *is* a better way. You took a one-size-fits-all model (equate to damage per blow) and try to apply it across the board. That almost never works completely satisfactorily.

And hey, I'll be the first to admit, this is a tricky subject. Is telepathy worth 35 damage? *On a weapon*...probably not. I don't use the Theoden axe very often, if ever, because it's so poor in combat at its primary job. On an amulet or ring or helm, it's awesome...given no other source of telepathy, do you prefer to use Thranduil, or the Crown of Numenor? Might depend on character type, but I think for a goodly percentage of people, telepathy trumps almost every other defense.

So, the value of any ability isn't a simple function. It's hugely dependent on several factors:

--How common is the ability in question?
--How valuable is it?
--What slot is it in, and is the ability complementary or separate from that slot's fundamental role? What other good options are there for that slot? How many other slots interact with this one, in impacting this slot's fundamental role? This is why off-weapon extra swings, for example, is hugely valuable...because only 1 slot strongly supports the fundamental role of doing melee damage, and that's the weapon slot. (The fundamental roles, IMO: melee damage, ranged damage, defenses, and spellcasting. Speed's a category unto itself, because it's such a strong role enhancer.) OTOH, armor, shield, boots, and cloak are all strongly tied to the Defense role, with gloves/gauntlets kinda split half and half between melee damage and defense. So a shield that gives a nice juicy melee damage bonus, is VERY attractive...because I've got other ways to cover up any defensive deficiencies.

[Derakon]

I don't think it's necessary that items get a thoroughly detailed pricing scheme that precisely measures how valuable a particular item is to the player both right now and in the endgame. If nothing else, that would lead to oddities like shopkeepers dropping the price of items as the player acquires abilities the items provide ("Oh, you have Thranduil; guess I'll give you a discount on this Cap of Telepathy then"). Honestly, I think the current algorithm is not fundamentally flawed; it just needs to be tweaked for some of the abilities whose value can't easily be calculated from the datafiles.

That said, there might be some value in having a price factor that devalues abilities when they're readily available on multiple slots.

[Antoine]

I am curious to know which other changes you think should be reverted or revisited in V.

Haradrim +1 blow/+1 shot?

A.

[Derakon]

I would also say that any buffs that have been applied to the "core set" artifacts should be reverted. Here's the core set of armors as I would call them (with no indication of which, if any, have been buffed; I haven't tracked artifact changes):

Phial/Star/Arkenstone/Palantir
Narya/Nenya/Vilya/One Ring
Isildur/Rohirrim/Celeborn/Belegennon/Arvedui/Caspanion/Thalkettoth
Thorin/Celegorm/Anarion
Thranduil/Holhenneth/Gondor
Colluin/Holcolleth/Colannon/Luthien
Cambeleg/Cammithrim/Fingolfin
Dal-i-thalion/Thror

As I see it the game was feasibly winnable without relying on any buffs that these may have gotten, so buffing them, if it was done, was unnecessary. The same probably holds true for weapon artifacts, but I'm not about to list all of the weapons that any given player would commonly use in a single game.

[miyazaki]

Confusion and Paralysis effects:

I think paralyzation should be returned to its past lethal levels. It's way wimpy now. The game was better when floating eyes could kill you.

And the status healing powers of !CMW and !CSW are too much, especially in regards to confusion.

Before those changes finding early FA was like winning the lottery! And ResConf was the holy grail.

[fizzix]

The more I think about paralysis the more I think it should revert also. There are enough sources of FA. Make FA mandatory to fight paralyzers. At worst, we should have something in the middle between the old insta-death and the current 'meh'.

I still think !CCW should cure confusion. Confusion is still dangerous, losing a turn to quaff a potion can be deadly. Furthermore you can't melee hit-to-confuse monsters without rconf.

[Hariolor]

not sure what the actual frequency is on the random teleportation rings/amulets, but it seems to me that if they had a rapid enough activation rate (maybe 10% chance per turn), they could be a decent early stand-in for FA. If someone *really* feels the need to melee a paralyzer without FA, having one of these on would at least make the prospect survivable.

[Pete Mack]

ESP isn't worth 35 damage/blow on any equipment. (Though it may be worth it as a swap.) I would certainly use Ring of Acid + Glaive of Pain vs Sauron--+65--, even if I had to give up ESP. I would probably use a Ring of Slay Evil--+32--as well.

[Derakon]

My experience with the new paralysis code is basically as follows:

* One encounter with a druid where he cast a paralyze spell and I died before it wore off.
* One encounter with a paralyzing melee monster where it paralyzed me and I died before it wore off.

Now, I'll freely admit that I play a bit recklessly, but I wasn't in that bad shape when these events happened. They've basically reinforced, to me, the fact that FA is non-negotiable; you need it to survive, even after it's been nerfed to not stack. The only change is that they are now theoretically survivable, which, as I said before, means that newbies have a chance to discover why FA is so important without losing a character to paralyzers.

Have other players found that the new paralysis behavior allows them to forgo free action for longer while surviving paralyzing attacks? Or is it simply that they feel that paralysis is now less scary, and are thus more willing to forgo FA and never actually happen to encounter paralyzers without it? I mean, I used to try to have basic four resists before 1000'; now I rarely have all four covered before 2000'. The attacks aren't any less deadly; I've just changed how I think of them.

[PowerDiver]

Have other players found that the new paralysis behavior allows them to forgo free action for longer while surviving paralyzing attacks? Or is it simply that they feel that paralysis is now less scary, and are thus more willing to forgo FA and never actually happen to encounter paralyzers without it? I mean, I used to try to have basic four resists before 1000'; now I rarely have all four covered before 2000'. The attacks aren't any less deadly; I've just changed how I think of them.

I was careless once and was happy to survive because of the change. I was unlucky once and died to paralysis without believing I had made a mistake. The change has not altered my play. If I have FA I wield it, but if I don't have it I don't obsess over it.

[ewert]

The new paralysis is IMHO better, I have died once and got scared the bejeezuz out of me once when going to stat gain depths without FA. I totally prefer something not near 100% instadeath versus an nearly 100% instadeath.