reduce artifact drop rate

[Londorth]

I don't see how you can make that work. People are complaining that detection is too good already.

Apply it to drops, not floor items? I really like this idea, ploughing though mounds of dropped junk is just tedious. But this is a valid objection.

[Tiburon Silverflame]

Why not simply eliminate treasure detection altogether? Or if that's too much, then reduce the amount of information you get from it. So, say, clairvoyance tells you that you've got 5 rings out there, but not what they are.

Note too, the squelch-before-creation, as we've been discussing it, has been focused on weapons...in part because they have by far the broadest range of power, even within the notion of "ego items." If presquelch is done in a relatively restricted manner, there's still *how* many types that wouldn't be squelched? If all we ever squelch was, say, Slay [Orc, Animal, Troll, Dragon], and the [Fire, Electrical, Cold] brands...you'd still have 2 more brands, 4 slays, 8 *slays*, and 7 special types (HA, defender, etc). There's still going to be considerable ambiguity whether that BoC is really worthwhile.

[PowerDiver]

Apply it to drops, not floor items? I really like this idea, ploughing though mounds of dropped junk is just tedious. But this is a valid objection.

If the drops all fell in the same square, squelch on walkover would be pretty painless except in pits. Even in pits, it would be an order of magnitude better than current, which is much better than it used to be. But we are never satisfied.

I'd prefer full id of dice, plusses, and any previously learned flags on all items in LOS, but that's a radical and extreme change even compared to id on walkover.

[fizzix]

Applying the hard squelching only to drops (and items generated in vaults) is reasonable. Tons of useless weapons really only become a problem when you've cleared out a dragon pit, a bunch of demons or a greater vault. Trekking across the dungeon to ID yet another BoFF is probably something we want to keep.

[Derakon]

I'd rather see ID-on-walkover tried first, presumably as an unlock when you reach a certain character level (maybe instant pseudo-on-walkover at clvl 25 and instant ID-on-walkover at clvl 35). I think it'd solve the problem by leveraging the existing squelch system, without being as prone to per-character biases or induced drop rarity confusion.

[Magnate]

I'd rather see ID-on-walkover tried first, presumably as an unlock when you reach a certain character level (maybe instant pseudo-on-walkover at clvl 25 and instant ID-on-walkover at clvl 35). I think it'd solve the problem by leveraging the existing squelch system, without being as prone to per-character biases or induced drop rarity confusion.

I rather like this idea, as it's consistent with a fairly longstanding aim of takk's, which is instant pseudo at a certain clevel. I'm not sure about full ID: the whole ID issue is so fraught with subjectivity because we've all forgotten that when we first started playing it really was fun to ID stuff, and sometimes to have to wait to do so.

I think there are two separate things to do here: one is to revisit the last set of changes to randart gen chances - they were made in response to feedback that randarts were noticeably less common than standarts, and that seems to have reversed. (I really need to code in a reliable monte carlo function so I can reduce reliance on anecdotal feedback. So many aspirations, so little time.)

The second is to implement pseudo on walkover at cl25 and see if this, combined with squelch UI improvements (I am planning on stealing NPP's excellent ego squelching asap), starts to address the issues of deep junk drops. It shouldn't, initially, have any impact on treasure detection, but if we extend to LOS-based pseudo or ID we might need to look again at that.

[Laie]

[EDIT: the previous two posts didn't exist when I started writing this. So tis might be a bit repetitive... I wasn't aware that pseudoID-on-walkover has been seriously considered before. But please consider the bullet points at the bottom of this post.]

Let me recap: starting somewhere in mid-game, the character is wading knee-deep in dropped items. Squelch helps, but in the case of weapons and other equipment:

The problem remains, as it always has been, the need to identify to know what wieldables do before you can squelch them.

Yet at the same time...

People are complaining that detection is too good already.

The character has to sieve a lot of chaff in order to find the occasional worthy item. I don't see how this could be changed without turning Angband into a totally different game. So, we're looking for something that will make sieving less tedious yet doesn't immediately give away everything. Well, the latter requirement leads to pseudo-ID like tags, {good} and {cursed} and somesuch, so squelch system has something to work on. "Less tedious" means that one doesn't need to pick it up or cast a spell or anything... this pretty much means it should work on walkover, or possibly even on items in sight.

This doesn't seem to be too far-fetched, if you ask me. For one thing, a seasoned character probably has seen so many enchanted items that she could quickly decide whether something is worthy of a closer look. Furthermore, it seems much more reasonable if pseudo-ID worked in a way similar to searching and perception. I never quite understood how it does you any good to carry something around on your back -- but standing among the plunder and having a look you might well notice the more interesting stuff.

A few more points:

• It shouldn't be 100% effective. If you get, say, half of your finds pre-sieved this will help a great deal. Should possibly depend on character level.
• {good} and {cursed} may be alright early on, but later you're possibly only interested in items that are {great} or better. Eventually, this level of detail should be available for every character class. Oh, and you also want to know whether an item is {plain} or unknown.
• no false negatives. While i can well imagine that someone would mistake an artifact for a simple item (like a plain, unadorned ring...), this should never happen.
• false positives might be cool, though. I think that everyone should ID a weapon of morgoth at least once.

[Nomad]

I rather like this idea, as it's consistent with a fairly longstanding aim of takk's, which is instant pseudo at a certain clevel. I'm not sure about full ID: the whole ID issue is so fraught with subjectivity because we've all forgotten that when we first started playing it really was fun to ID stuff, and sometimes to have to wait to do so.

I'd vote for instant pseudo on walkover combined with more subdivisions of pseudo ID tags (possibly also unlocked at a certain clevel?). Break it up so there are more categories of pseudo feelings, like, I don't know:

{powerful} = slay
{enchanted} = branded
{mighty} = *slay*
{superior} = Defender, Westernesse, etc
{unique} = artefact

That would aid squelch-by-category, too.

[Tiburon Silverflame]

I'm not sure about full ID: the whole ID issue is so fraught with subjectivity because we've all forgotten that when we first started playing it really was fun to ID stuff, and sometimes to have to wait to do so.

And by the time we reached DL 40 for the first time, we got over that.

VERY few of these changes would even be seen by newbies. Presquelch, if done, wouldn't start until at least DL 20; auto pID is being suggested for no earlier than CL 25, which is probably later than DL 20. And full ID is CL 35; no one's a newbie at that point.

BTW, I'll also vote against full ID, because I think it's just too much.

Nomad's scheme fails, IMO, because I may keep certain Slays LONG after the junky ones would be tossed aside, and we've pointed out that Acid brands also have value much longer than any others. The key point is that, within any of the major categories (slay, greater slay, brand, superior), there are still *massive* discrepancies between types. Hey, what would you rather have: Slay Evil or *Slay Orc*, at DL 50?

Heck, looking at these...many of the *Slays* are really damn marginal. Comparing the regular Slay, to the supposedly superior *Slay*...
Evil: Blessed, small Wis
Orc: sustain Dex, small Dex
Troll: regen, small Str
Giant: sustain Str, small Str

The stat boosts are 1 or 2 only.

Certainly, the sustains are useful, and I like regen...but those are things I can easily do without. These *Slays* DON'T appear to do more damage; they use the same SLAY_X flag as the Slay type does.

[PowerDiver]

I think there are two separate things to do here: one is to revisit the last set of changes to randart gen chances - they were made in response to feedback that randarts were noticeably less common than standarts, and that seems to have reversed.

The second is to implement pseudo on walkover at cl25 and see if this, combined with squelch UI improvements (I am planning on stealing NPP's excellent ego squelching asap), starts to address the issues of deep junk drops. It shouldn't, initially, have any impact on treasure detection, but if we extend to LOS-based pseudo or ID we might need to look again at that.

I think standarts still drop more than randarts. Certainly, randarts are not sufficiently more common to matter in my experience.

You cannot sieve items based on pseudo. You still need a wield at a minimum, and cannot actually do things properly without seeing dice and plusses unless you are playing a non-melee character. This has been exaggerated by off brands and slays and *slays. In these days of acid rings, or more particularly randart armors with brands or slays, a good katana 5d5 may well be more valuable than a katana 3d5 of *slay undead.

I can't be bothered to id everything, so I currently squelch the good katana, but if it is time to rework the system then this needs to be fixed.

[Timo Pietilä]

Heck, looking at these...many of the *Slays* are really damn marginal. Comparing the regular Slay, to the supposedly superior *Slay*...
Evil: Blessed, small Wis

This one has also random ability. Might be ESP, and after my proposal for moving some of the resists to abilities also blindness or confusion. Might really affect your game. Or not. Feather fall or slow digestion or extra lite wont affect the value much.

Certainly, the sustains are useful, and I like regen...but those are things I can easily do without. These *Slays* DON'T appear to do more damage; they use the same SLAY_X flag as the Slay type does.

Damage is same, yes. Other than *evil* I agree. Especially when playing priest, blessed-flag alone can be very valuable (Blade of Chaos *slay* evil with ESP. Nice)

[Nomad]

BTW, I'll also vote against full ID, because I think it's just too much.

Nomad's scheme fails, IMO, because I may keep certain Slays LONG after the junky ones would be tossed aside, and we've pointed out that Acid brands also have value much longer than any others. The key point is that, within any of the major categories (slay, greater slay, brand, superior), there are still *massive* discrepancies between types. Hey, what would you rather have: Slay Evil or *Slay Orc*, at DL 50?

Hmm... How about ego-ID, then? Not quite full ID, but you recognise what type of ego it is on walkover. So you can see it's a *slay evil* immediately, but still have to pick it up and wield it or ID it to discover the enchantments, level of stat bonuses, and what the random ability is.

Coupled with the capability to set a separate squelch setting for each ego type, that would get rid of junk as soon as you step on it, but still leave you a bit of ID work to do with the good stuff.

[Laie]

Hmm... How about ego-ID, then? Not quite full ID, but you recognise what type of ego it is on walkover.

I think you're taking it too far.

I'm currently on level 66. There was a (small) room without doors, many items and a colorful mix of mosters, is that a vault? Also Scatha the Worm and one Ringwraith. A couple of Ancient D's all over the dungeon, and I milked an Enchantress for a few more. Make that five Ancient Dragons. Finally a Great Wyrm of Thunder.

That was quite a few drops, not to mention the stuff in that vault. I didn't count, of course, but have the impression that this was a rather juicy level. Now that I'm through with ID, I only see five items that were a slay or ego or armor-of-something.

A whopping five. Out of I don't know how many. Now, this was just me and a single level, but generally I don't have the impression that there's that many brands and slays. Somewhere between 70-80% of all items seem to be of lesser magic or completely unenchanted or even cursed. This is the true chaff, that's what makes ID so tedious. Once all of these are sorted out, there's not much left -- I don't think it's demanding too much if the player has to ID all of the rest. A dozen items per level, if he's lucky.

[Rizwan]

How about capping the number of ego items like artifacts. Say there are 150 something artifacts so cap the egos to hmm double that say 300 ego items? O r may break it down by category and say ok there will be 100 ego swords 100 ego launchers and so on?

[Magnate]

I think standarts still drop more than randarts. Certainly, randarts are not sufficiently more common to matter in my experience.

You cannot sieve items based on pseudo. You still need a wield at a minimum, and cannot actually do things properly without seeing dice and plusses unless you are playing a non-melee character. This has been exaggerated by off brands and slays and *slays. In these days of acid rings, or more particularly randart armors with brands or slays, a good katana 5d5 may well be more valuable than a katana 3d5 of *slay undead.

I can't be bothered to id everything, so I currently squelch the good katana, but if it is time to rework the system then this needs to be fixed.

Very ironic that it should be you who suggests this, because ... doesn't this lead us towards pseudo-by-object-power? What you're effectively saying is that you're more interested in the base damage of the weapon than in its other properties (for melee chars at least), so {good}, {excellent} and {splendid} should essentially represent tiers of damage (either absolute, or relative to the plain unenchanted base item - I'd favour the latter). A 5d5 katana with +15 to_d should pseudo higher than a 3d5 weapon of *slay orc* with only +7 dam, right?

For nonweapons of course there is no such simple measure ... which is precisely why the object power rating was invented. Of course the power rating is not perfect and is gradually improving, but is this what we want in principle? There's never really been much logic behind {good} and {excellent}, merely an assumption that ego items are always more desirable than non-ego items, which is of course incorrect (dagger of slay orc vs. SoS).

I can imagine something like:

{average} as now
{good} means power 101-150% of base item
{excellent} means power 151-250% of base
{splendid} means more than 250%

You would still use {cursed} if it is cursed and {special} for artifacts, but no reason why you can't combine tags: {good, special} or {splendid, cursed} etc.

Of course, we could always let the user set the squelch levels himself, defining the percentages above in the squelch menu. We could even let him choose whether the tiers were relative (percentages) or absolute (actual power rating, possibly more intuitively shown as gp value). With sensible defaults for newbies, everyone would then be free to optimise his or her squelching experience.