experimenting with minimal id

[Antoine]

Second question:
1. Make the sale price of a wand or staff not depend on whether it is identified

uh, providing the item is of a known flavour.

A.

[takkaria]

"What can we do to help you implement your thoughts on improving ID?"

Eddie is doing the work, really. He is playing with ID more actively than I have time for and is releasing his changes publicly. I'll then nick a subset of his changes.

[takkaria]

1. Make the sale price of a wand or staff not depend on whether it is identified [This means I don't have to ?-ID wands which I plan to sell]

I'm in favour of displaying charges of wands without ID, which amounts to the same thing.

2. Once a weapon or armour is pseudo-id'd, make the component of its sale price which relates to hit, damage and AC plusses not depend on whether it is identified [This means I don't have to ?-ID 'good' items which I plan to sell]

I'm not a massive fan of this, really. The information leakage bothers me—it means stores are buying based on what the storekeeper can discern rather than what the player knows, and if this were to be adopted as game philosophy, I'd want it to take account of ego items the storekeeper would know too.

3. Make shopkeepers refuse to buy unidentified potions, scrolls and mushrooms [This means I have to use-test consumables]

It's an idea, but is the opposite approach Eddie is taking (that you can sell for no money for instant-id). I can see the benefits of either.

4. Remove !Death and !Ruination and all other extremely adverse consumables [This means I feel I can use-test consumables]

Already done in the dev branch.

5. When I hit a monster and do extra damage from a slay or brand, change the message accordingly - "You smite the Hill Orc!" for Slay Orc weapons, "You burn the White Wraith" for fire branded ammo. [This helps me use-id ego weapons and ammo, and adds flavour]

Yeah, this is good.

6. When I wield an unidentified weapon or armour, if it gives visible bonuses that can only come from egos or artifacts (like stat bonuses), give it the {excellent} pseudo-id status. [This helps me use-id ego items]

That works too.

7. Halve the sale price of all items, and double or triple all gold drops. [I just thought I'd try to sneak this on the end]"

Why this over halving the sale price of all items and decreasing item costs by two or three times? What's the problem this suggestion is trying to address?

[Antoine]

>> [suggestion that identification should not change the sale price of wands]
> I'm in favour of displaying charges of wands without ID, which amounts to the same thing.

Even better.

>> [suggestion that identification should not change the sale price of 'good' weapons and armor]
> I'm not a massive fan of this, really.

Fair enough. Do you agree, though, that there should be some way around the need to ID 'good' weapons and armor purely for sale purposes??

An alternative approach would be to show the plusses of pseudo-id'd items once you wield them, or hit a monster with them N times... as per Eddie. Or for AC bonuses of armor, I suppose that's 'get hit by monsters N times'.

>> [suggestion that you should not be able to sell unidentified potions, mushrooms and scrolls]
> It's an idea, but is the opposite approach Eddie is taking (that you can sell for no money for instant-id). I can see the benefits of either.

Yah. If you want to favor use-id over shop-id, then I'd say go with mine (especially now you've taken out !Death - nice).

>> [flavor text for successfully using a brand or slay]
> Yeah, this is good.

Want a patch?

>> [wielding an unidentified ego item with obvious bonuses causes 'excellent' pseudo-id]
> That works too.

Again, want a patch?

> Why this over halving the sale price of all items and decreasing item costs by two or three times? What's the problem this suggestion is trying to address?

Yeah, that would work too. Saves fiddling with gold drops I guess.

I guess it's a step towards Eddie's no-selling-to-shops, which I like. Primary effect is to incentivise players to pick up less junk for purposes of sale, secondary effect is to reduce the amount of Id needed to identify said junk.

A.

[PowerDiver]

>3. Make shopkeepers refuse to buy unidentified potions, scrolls and mushrooms [This means I have to use-test consumables]

It's an idea, but is the opposite approach Eddie is taking (that you can sell for no money for instant-id). I can see the benefits of either.

The main benefit being avoiding mobs of angry *banders.

Originally I had no selling at all, but I changed to selling for 0 because some people said they would refuse to play otherwise.

[takkaria]

>> [suggestion that identification should not change the sale price of 'good' weapons and armor]
> I'm not a massive fan of this, really.

Fair enough. Do you agree, though, that there should be some way around the need to ID 'good' weapons and armor purely for sale purposes??

An alternative approach would be to show the plusses of pseudo-id'd items once you wield them, or hit a monster with them N times... as per Eddie. Or for AC bonuses of armor, I suppose that's 'get hit by monsters N times'.

All I can really say is that what Eddie's working on sounds promising. These are the kinds of things that need playtesting to figure out the best solution, and I've not done any of that in this area so I haven't got a firm opinion.

>> [flavor text for successfully using a brand or slay]
> Yeah, this is good.

Want a patch?

>> [wielding an unidentified ego item with obvious bonuses causes 'excellent' pseudo-id]
> That works too.

Again, want a patch?

Both would be lovely. If you could do them against latest SVN, then you get serious brownie-points.

> Why this over halving the sale price of all items and decreasing item costs by two or three times? What's the problem this suggestion is trying to address?

Yeah, that would work too. Saves fiddling with gold drops I guess.

I guess it's a step towards Eddie's no-selling-to-shops, which I like. Primary effect is to incentivise players to pick up less junk for purposes of sale, secondary effect is to reduce the amount of Id needed to identify said junk.

Actually, fiddling with gold drops is probably easier than the other option, but I was just wondering what the rationale behind the change was. Gold could really do with a proper rebalance, with some serious thought going into how much gold should be available per dungeon level, how this should relate to consumables necessary for survival, and how to price things that aren't necessary, and on top of this all, how much things should sell for relative to their starting price. I've not done said serious thought yet, though.

[Bandobras]

> > 4. Remove !Death and !Ruination and all other
> > extremely adverse consumables
> > [This means I feel I can use-test consumables]
>
> Already done in the dev branch.

Ah. Why not an approach similar to S or Un, where you don't die if use-iding at more than half HP and any other bad effects (e.g. stat drains) are temporary, but very pronounced? I think it adds to the excitement of use-id and presents the unique opportunity to a bored overpowered player, to play a mage with INT halved, a ranger that is better with wands than bows or a tank warrior suddenly finding himself with 1 HP and halved CON in the middle of the level, for a couple of thousands of turns...

[takkaria]

> > 4. Remove !Death and !Ruination and all other
> > extremely adverse consumables
> > [This means I feel I can use-test consumables]
>
> Already done in the dev branch.

Ah. Why not an approach similar to S or Un, where you don't die if use-iding at more than half HP and any other bad effects (e.g. stat drains) are temporary, but very pronounced? I think it adds to the excitement of use-id and presents the unique opportunity to a bored overpowered player, to play a mage with INT halved, a ranger that is better with wands than bows or a tank warrior suddenly finding himself with 1 HP and halved CON in the middle of the level, for a couple of thousands of turns...

I guess it's a matter of taste what to do with them. I went through the item list and tried to reconsider it from the ground up, and I just didn't think including the potions was a good idea.

[Donald Jonker]

I'm not a massive fan of this, really. The information leakage bothers me—it means stores are buying based on what the storekeeper can discern rather than what the player knows, and if this were to be adopted as game philosophy, I'd want it to take account of ego items the storekeeper would know too.

And this is also really slippery territory. You end up with a sort of pseudo-id where you infer the value of an object based on how much a storekeeper offers, without actually having to sell the item. You end up with a nethackish selling id subgame as an unintended consequence.

I think as an all-around note of caution it should be kept in mind that id by trial is the nethack paradigm, simply by virtue of the fact that ?id is so rare in that game.

The mission, I think, is to make the identification of objects a fluid part of the combat system - not to eliminate ?id in favor of trial simply for the sake of it, and not to create an id subgame. The opposite, in fact, is true: we want to eliminate all id subgames. At least, that's my idea in all of this. Correct me if I'm wrong.

[Antoine]

If you could do them against latest SVN

When trying to checkout the Angband trunk, I get "PROPFIND of svn/trunk: could not connect to server (http://dev.rephial.org)"

am I doing it rite?

A.

[Antoine]

The mission, I think, is to make the identification of objects a fluid part of the combat system

The problem with this is that most weapons and armour that you pick up are never wielded. Rather, they are taken back to town and sold. I'd like to not have to use ?ID on these sale items, but I don't want to have to use-id them either. Hence the idea of making it profitable to sell them in an unidentified state.

Better still would be to cut the Gordian knot by disallowing selling to shops...

A.

[PowerDiver]

6. When I wield an unidentified weapon or armour, if it gives visible bonuses that can only come from egos or artifacts (like stat bonuses), give it the {excellent} pseudo-id status.

I added an extra pseudo level {splendid} to deal with this. I now wonder if that was a mistake. It is very useful for quality squelch to have this, but that is only because of the particular definitions of egos in vanilla combined with another change to make brands obvious on wield.

Should I get rid of quality squelch entirely? If you take the viewpoint that the char should not know the list of possible egos, it would never be safe to squelch all excellent items of any particular type.

I am near to getting a generalized squelch system where you can squelch anything and then any other item dominated by things you have squelched disappears. However, setting up your squelch in a game to accomplish the equivalent of saying "once I have elvenkind armor all lesser body armors are pointless" seems very involved as opposed to simply squelching all non-splendid body armor.

[Bandobras]

we want to eliminate all id subgames

I think it would be throwing the baby out with the bath water. Angband is actually just a blob of various subgames intricately and randomly mangled together. If we start weeding out subgames, instead of improving them, we may soon run out of the matter Angband is made from.

I'd rather strive to make interesting

1. the id mini-game taken separately
2. the impact on id on the main game

This means that

1. because a single Angband game can take hundreds of hours, the id game should not get boring as the character progresses
2. since Angband is so replayable and is intended to be played all over again many times, every single phase of the id game should be replayable

The example of the id mini-game is the management of unknown potions at the start of the game, especially if trying to minimalize game turns. In V, this may involve buying and selling 1 of each good potion from the stores and selling potions from dungeon to the stores. It influences the main game only by wasting gold and pack space. Arguably, this is not boring wrt a., but somewhat repeatable and artificial wrt b. In S and Un the mini-game is extended with id by use, which speeds the mini-game (diminishes 1), and makes it much less predictable by having potentially large, but random (and not necessarily adverse) impact on the tactical part of the main game (enlarged 2.).

IMHO, the S and Un approach is a good design choice, but of course there are many local maxima of a. and b., and sometimes they may even favour 1. over 2. (though I doubt these would be global maxima --- separate minigames suck).

[zaimoni]

When trying to checkout the Angband trunk, I get "PROPFIND of svn/trunk: could not connect to server (http://dev.rephial.org)"

am I doing it rite?

A.

The server in question is historically unstable, and is not serving webpages either. I wouldn't be too concerned.

[Antoine]

The server in question is historically unstable, and is not serving webpages either. I wouldn't be too concerned.

I'm not concerned as such, but my checkout failed - is there another way to get the development trunk files?

A.