Smaller levels

[Derakon]

Doesn't ToME 2 already have infinite randarts alongside the standard set?

[Therem Harth]

Yes, that's probably the code I'll end up looking to. (Neoband is based on V.)

And yes, I repeat, I do want infinite randarts in Neoband.

[fizzix]

I don't think Scatha's solution works. I also think sil could use squelch too. Although it's not in much of a need of it because the level size and length of game is shorter.

I've never seen the problem with squelch. I wish more games had it.

[Derakon]

Some thoughts on squelch:

* Do we want to have Scrolls of Trap Detection? Because those are junk as soon as you get a Rod of Trap Detection, but useful earlier.
* Ditto Scrolls of Magic Mapping, Detect Treasure, Detect Doors/Stairs, Trap/Door Destruction, etc. Some of these are probably eligible for outright removal, but not all of them, and they all get obsoleted by staves or rods eventually.
* Potions of Boldness? Cure Light Wounds? Neutralize Poison? Resist Fire/Cold? Some of these have early niche uses and get obsoleted, some just become pretty marginal later in the game. I usually don't squelch CLW but I hardly ever use it past the early game. Elemental resistance potions are potentially useful game-wide for warriors, and early on for everyone else.
* Scrolls of Trap Creation? Staves of Summon Monster? Spellbooks for the wrong class? Some things just aren't ever going to be useful; are you suggesting that they not be generated? That gets into Magnate's concern about adapting the game to the character, which I agree seems wrong for hard-to-articulate reasons.

Equipment quality squelch is maybe a different matter...but I don't think it's reasonable to say that squelch is categorically bad when it has obvious uses for items that eventually become obsolete, but at dfferent times for different characters.

[AnonymousHero]

Doesn't ToME 2 already have infinite randarts alongside the standard set?

I'm sorry, I'll try expanding the phrasing if that helps:

I do not think you need infinite randarts as in ...

Why is everyone thinking that I support infinte randarts?

[buzzkill]

Some thoughts on squelch:

* Do we want to have Scrolls of Trap Detection? Because those are junk as soon as you get a Rod of Trap Detection, but useful earlier.
* Ditto Scrolls of Magic Mapping, Detect Treasure, Detect Doors/Stairs, Trap/Door Destruction, etc. Some of these are probably eligible for outright removal, but not all of them, and they all get obsoleted by staves or rods eventually.
* Potions of Boldness? Cure Light Wounds? Neutralize Poison? Resist Fire/Cold? Some of these have early niche uses and get obsoleted, some just become pretty marginal later in the game. I usually don't squelch CLW but I hardly ever use it past the early game. Elemental resistance potions are potentially useful game-wide for warriors, and early on for everyone else.
* Scrolls of Trap Creation? Staves of Summon Monster? Spellbooks for the wrong class? Some things just aren't ever going to be useful; are you suggesting that they not be generated? That gets into Magnate's concern about adapting the game to the character, which I agree seems wrong for hard-to-articulate reasons.

PSSS tackles all these with zero additional input from the user, though soft squelch could be hastened (or retarded) with deliberate action.

Equipment quality squelch is maybe a different matter...but I don't think it's reasonable to say that squelch is categorically bad when it has obvious uses for items that eventually become obsolete, but at different times for different characters.

Weapons and armours, because of their various attributes, is a trickier situation, but I don't see why a further refined PSSS couldn't tackle such items in a reasonable manner as well.

[Derakon]

PSSS tackles all these with zero additional input from the user, though soft squelch could be hastened (or retarded) with deliberate action.

Okay, and an addendum: potions of stat gain? When you've maxed the stat in question they are no longer useful except for stat restoration -- so every once in awhile I'll find myself turning off squelch to see if any are lying around.

[buzzkill]

Okay, and an addendum: potions of stat gain? When you've maxed the stat in question they are no longer useful except for stat restoration -- so every once in awhile I'll find myself turning off squelch to see if any are lying around.

No squelch solution is going to be perfect. Angband can't read the players mind, nor does it currently respond to a character's current status ailments. There's no reason PSSS must result in item destruction or non-creation. It can operate with the exact same visible/invisible toggle that is the status quo and also in conjunction with status quo squelch.

Hard squelch settings will sync with PSSS settings. Hard squelch an item and it's PSSS rating get set to 0, hidden upon creation 100% of the time. Unsquelch and it reverts to 50, normal generation.

As I see it, hard squelch rather than being a separate entity, would in fact serve PSSS. It would be active, direct user configurable, component of an otherwise passive, unseen squelch system.

[Magnate]

I don't think it's too hard to produce a quality measure which gets things roughly right. Some value for the base item type, some values for modifiers. We did this in Sil, for smithing difficulties -- we ended up using it to improve artefact balance too.

Well I still have a long way to go before I'm happy with V's obj_power as a measure of quality, despite working on it for over a decade. It's pretty good for weapons (its original aim), but doesn't yet work as well for armour and not at all for consumables.

But yes, I take both your points - you can cut down junk without removing it altogether - arguably the single focus of the entire 3.x series of Vanilla was about doing this, with some success I hope.

And yes, you could deliberately not generate lower quality items later in the game. We haven't tried that yet, but it has been discussed.

[Magnate]

Yes, that's probably the code I'll end up looking to. (Neoband is based on V.)

And yes, I repeat, I do want infinite randarts in Neoband.

Which version of V are you based on? I have some time over the holidays, and it would be pretty simple to use V's randart generator to generate infinite randarts, alongside any proportion of the standart set. I could offer up a pull request over the holidays - even if takkaria/Nick don't want it, it would be useful to you - but only if your codebase is sufficiently modern ...

[Carnivean]

Well I still have a long way to go before I'm happy with V's obj_power as a measure of quality, despite working on it for over a decade. It's pretty good for weapons (its original aim), but doesn't yet work as well for armour and not at all for consumables.

But yes, I take both your points - you can cut down junk without removing it altogether - arguably the single focus of the entire 3.x series of Vanilla was about doing this, with some success I hope.

And yes, you could deliberately not generate lower quality items later in the game. We haven't tried that yet, but it has been discussed.

How about grouping certain objects together (ie CLW, CSW, CCW, Healing) and give them a level band? If you're < CL10 you get 94% CLW, 4% CSW, 1% CCW, 0.01% Healing. If you're > CL40 you get 1% CLW, 4% CSW, 75% CCW, 20% Healing.

This would eliminate a lot of the "junk" items.

Also, I would like more fine controls on squelch. I can squelch up to the level below artifacts and still get dozens of weapons that I would never use. Blessed weapons are on that level, but are always terrible for my characters. My player knowledge knows about them, so theoretically I should be able to ignore them. I know that the code doesn't work like that, but if it gets changed, I'd like that in the options.

[Magnate]

Also, I would like more fine controls on squelch. I can squelch up to the level below artifacts and still get dozens of weapons that I would never use. Blessed weapons are on that level, but are always terrible for my characters. My player knowledge knows about them, so theoretically I should be able to ignore them. I know that the code doesn't work like that, but if it gets changed, I'd like that in the options.

NPP has ego-type squelching and it works well. We merely need someone to port that code back to V.

[Therem Harth]

@AnonymousHero: Sorry, I misread what you said there.

Which version of V are you based on? I have some time over the holidays, and it would be pretty simple to use V's randart generator to generate infinite randarts, alongside any proportion of the standart set. I could offer up a pull request over the holidays - even if takkaria/Nick don't want it, it would be useful to you - but only if your codebase is sufficiently modern ...

Thanks, that'd be cool. Neoband is based on 3.2.0 though, not sure if that's considered modern at the moment.

[Magnate]

Thanks, that'd be cool. Neoband is based on 3.2.0 though, not sure if that's considered modern at the moment.

That's modern enough - I've rewritten obj-power since then, but not randart.c IIRC. The patch might take a little effort to apply, but it should work.

Interestingly, according to #angband-dev, takkaria thinks myshkin might have done this already (devs often have not-widely-known stuff hanging about in their repos), so that would be good.

[Therem Harth]

Cool, glad to hear. I'm hoping to get the next release done by Christmas Day, so if you have any tips now on a good way to do it...