Enchanting artifact weapons

[ddubois]

Is there a design discussion thread somewhere that covered the rationale for not allowing artifacts to be enchant scrolled? Coming from Nethack, I found it very odd to discover:
1) that the standard route to the best weapons for this game is to find a mundane weapon of the highest damage base type (albeit, preferably one with a good brand), and slam a dozen enchant weapon scrolls into it, and
2) that artifacts are so often terrible in comparison.

In other RPGs, usually any weapon good enough to have a proper name was something to get excited about. By comparison, reading through the List_of_unrands page on the wiki, it's a never ending stream of "only use this is the moon is full, it's a leap year, and your hair is on fire".

I haven't had a game yet where a artifact weapon was something I could even consider as potential endgame material. It's always "well, that interesting until I get a real weapon, but it's base type is mace, so it's garbage by level 12" or "I'll swap that for resists on my caster", or, by far the most often, "Ugh, I wasted a detect/remove/identify scroll again".

I'm guessing this has something to do with the design goal of "no no-brainers", but I feel like perhaps the pendulum has swung so much in the other direction, that not using artifact weapons is now the no-brainer.

[mageykun]

It's not so much a matter of no brainers as it is one as trade offs.

On the downside, artefacts cannot be enchanted up. So in the long run, any artefact below +9,+9 is inferior to what a normal weapon can reach. You also cannot chose your brand (either using enchant <foo> spells and + vorpal, or TSO/Kiku/Lucy).

On the upside: artefacts are immune to corrosion, so enchantment can't go down. Enchantment is fixed at generation, but can go as high as +12, better than the usual max. Finally, artefacts can come with a wide variety of additional properties (evocables, resistances, stat boosts, slaying, etc).

Make artefacts enchantable, and there really isn't a downside. (Well, aside from potentially bad added properties).

Getting an artifact is still worth getting excited about in crawl. You never know how good a randart is gonna be! And many of the unrands aren't as quite bad as that particularly pessimistic wiki-page makes out. Some are even game changing! And even the useless ones are still fun to collect, in if nothing else.

[Kate]

The list of unrands page is (of course) full of misinformation. Lots of the unrands are great, and randarts can be as good or better.

[pratamawirya]

Artefacts being occasionally bad (sometimes they're good/decent) is quite flavorful without sacrificing gameplay. Artefacts are really old stuff, so you can't expect them to always have sharp edges.

[ddubois]

Artefacts being occasionally bad (sometimes they're good/decent) is quite flavorful without sacrificing gameplay. Artefacts are really old stuff, so you can't expect them to always have sharp edges.

But my point was that they are not "occasionally" bad. They are almost always bad. I mean, they are great when you find them at level 5, sure, but even then they are almost always a holdover until you can get your demon X, and slam the pile of enchant scrolls you have in your stash into it.

There isn't an interesting tradeoff. There would be an interesting trade-off, if the player had to make a decision about whether to use enchant scrolls on that mace with 4 resists on it, or on his demon X, but he doesn't have that choice.

Go through the YAVP posts. Find someone using an artifact weapon. It'll take awhile. I found a few people using enhancer staves, a boatload of demon foo/lajatang/bardiche/eveningstar/etc (or the TSO-buffed equivalent), and one single solitary stabber using a dagger of Truth. It's far more common for someone to be wielding nothing than an artifact.

Maybe the designers were trying to avoid the Nethack-esque system wherein everyone was using +7/+7 Mjollnir and the endgame gear variety was laughable, and that's fine, I get it. But like I said, the pendulum has swung too far. Artifact weapons should be, if not awesome, then at least competitive, and they're neither.

[mageykun]

You're missing the point, I think.

Normal weapons are reliable. You will pretty much always be able to find your chosen weapon of all but the rarest base types, you will always be able to bless/corrupt it, and you will always be able to enchant it. You can decide on turn 1 what your endgame weapon will be, and you can get it. (Going unarmed is even more reliable, since you need nothing).

Artefacts, however, are subject to the whims of randomization. You trade reliability for unpredictability. Unpredictability that can, at times, generate results even better than the best reliable gear.

So why have more people gone with the reliable gear? Because it's reliable. Not because there's something wrong with artefacts.

[sardonica]

An important part of crawl philosophy is "no no-brainer choices"

if you can enchant artifacts, almost every single time that would be the "right" choice.

Unlilke in nethack where there is almost always a "right" or "wrong" choice, thats rarely true in crawl.

Others have explained to you the benefits and maluses of arts

SO if you aren't going to use any, can I borrow them? Haha

[crate]

Artefacts are often strong if you find them early, or eventually tso/oka might gift you one that is good. I dont think it would be terribly uncommon to find games that have wielded an artefact weapon for some portion of the game. You cannot really figure this out from morgues of winning games though, since by the end of the game--when you have found enough ?EW scrolls--most of the time a normal, highly-enchanted weapon is better.

That said I can think of a few of my games where I was wielding an artefact at the end of the game as I won (and I have a game going on a robin right now with an artefact demon whip of freezing as our weapon).

[pratamawirya]

Winning the game with a good artefact ranged weapon is more likely than with a good artefact melee one, since ranged weapons aren't as tiered as melee ones (there's only one kind of sling and one kind of crossbow, and only two kinds of bow, whereas we got up to five kinds of axe).

[Jeremiah]

I think it's kind of ironic that the characters most likely to be wielding an artefact weapon are those who have no intention of actually fighting anything with them - eg primary spellcasters using them for their resistances or Int bonuses.

It's also a bit weird that artefact light armour has a better chance of being useful than artefact heavy armour. If a robe is only +1 but has some good powers or resistances, then it's quite usable because an ego robe can't get much better enchantment than that anyway. But a +1 artefact plate is junk because on heavy armour the AC value is a big part of how powerful it is.

I feel that the number of "good" artefacts is about right, but there are too many junk ones as well. What if the overall frequency of artefact generation was reduced, but those that were generated had a higher chance of getting a good enchantment value or a good base weapon type?

Also:

Artefacts are really old stuff, so you can't expect them to always have sharp edges.

But artefacts "cannot be changed by magical or mundane means," so how did they lose their sharp edges?

[Jabberwocky]

Also:

Artefacts are really old stuff, so you can't expect them to always have sharp edges.

But artefacts "cannot be changed by magical or mundane means," so how did they lose their sharp edges?

Probably because the smiths who made these artifacts were out drinking the previous night. It's not that they dulled, it's just the spells were cast over them before anyone thought to sharpen them, and once the spells to prevent them from changing were sung, there was no going back.

[XuaXua]

I have no problem saving that early game artefact -3/-5 dagger if it comes with blink+ or rc++, till I find something to replace it or if it has recurse property.

[smock]

That a branded weapon can generally become more powerful than randarts is OK by me, but I think the game would be just a bit more interesting if, say, one in five games your best weapon were an artifact.

I think it's a bit sad that hardly anyone would use the singing sword past the midgame. Even if you have a heavy-armoured long blade specialist, the first branded scimitar you find is likely to be better. By way of example I mean to suggest that bettering the base type on some fun unrandarts might help them see play. If the singing sword were a double sword then it would certainly see more (any?) play. Likewise on randarts. I'm not sure what is coded now, but perhaps generating more randarts with a "good" base type later in the game would help.

[One-Eyed Jack]

Artefact weapons are generally unimpressive, but non-artefact weapons are generaly shitty too. So I don't see a real problem there.

If you made artefact weapons scrollable, the only difference between them and glowing weapons would be the possible egos. This isn't enough to justify a new type of item, in my opinion.

I do, however, dislike the asymmetry between heavy and light artefact armours. I can't really think of a elegant way to fix it, though.

[cerebovssquire]

I think it's a bit sad that hardly anyone would use the singing sword past the midgame. Even if you have a heavy-armoured long blade specialist, the first branded scimitar you find is likely to be better. By way of example I mean to suggest that bettering the base type on some fun unrandarts might help them see play. If the singing sword were a double sword then it would certainly see more (any?) play. Likewise on randarts. I'm not sure what is coded now, but perhaps generating more randarts with a "good" base type later in the game would help.

Do you know of the "Screaming Sword" property? I think it makes the Singing Sword better than a scimitar than a normal brand (obviously speed, vamp, pain are probably better).

[mageykun]

I think it's a bit sad that hardly anyone would use the singing sword past the midgame.

I have sworn an oath that as soon as I get a game where the Singing Sword actually generates, I'm going to use it for the rest of the game. I don't care how bad it breaks my current build or plans. =_=

How about increasing the odds of the interesting unrands actually generating? For instance, I'd love to see more flavored vaults built around housing specific artefacts. The broken throne room of dyrovepreva. The Singing Citidal. The Hall of Prunes! Etc. Or how about the chance for a noisy artefact weapon to spawn as a unique in blade? Provide some taunting, obnoxious speech and you've made a dull branch somewhat more amusing. Obviously, there's a small chance it'll be the singing sword. Which will be worse. Not just because of the sonic damage thing. But because the insults will have to rhyme.

[dpeg]

Many good points already made.

Bear in mind that with a randart, you get the whole package in one go. Whereas you slowly work up your way with mundane items (first have to find one, then possibly brand it, collect enchant scrolls along the way), there is nothing like this with artefacts.

Suppose you found a +12,+12 weapon of a good base type on D:3. Cool? Perhaps. But also boring. Now assume you found a +5,+5 artefact scimitar on D:4. Will you use it? Sure thing. But there remains the question of when to switch. Decisions are good, for they provide opportunities for mistakes.

What has been rightfully mentioned is that junk artefacts have no place. There should simply be less of them. And vaults dedicated to unrandarts already exist; more are clearly welcome. By the way, I'd love to hear reports on how the new Singing Sword works out. I came up with the concept and a lot of the new speech, but I have no idea if it viable in end or even midgame.

[njvack]

What has been rightfully mentioned is that junk artefacts have no place.

Well... they do have one purpose: they're like little roulette games randomly scattered around. Finding junk artefacts makes finding good artefacts more rewarding.

Psychologically-speaking: It's rewarding when the actual utility of an outcome is greater than the expected utility. Removing junk artefacts increases the expected utility of identifying an artefact, thus decreasing the difference in utility when you find a good one.

[galehar]

Removing junk artefacts increases the expected utility of identifying an artefact, thus decreasing the difference in utility when you find a good one.

He didn't suggest that we remove junk artefact, just that we reduce their frequency. I agree with it, but it's not a trivial change.

[dpeg]

nvjack: Sure, and they shouldn't all go, especially as many might have fringe uses. But I believe there are too many pointless artefacts. Simply reducing their number a bit would help. For example, an artefact could be downgraded into its mundane base item with a certain chance depending on how bad it is.

[njvack]

Sorry -- removing didn't mean "completely removing." Reducing the probability that an artefact is useless also increases the expected utility of an artefact. It's also worth noting that the important thing is the distance from the average utility -- if you decrease the number of "meh" artefacts, you can compensate by adding more very bad artefacts. {faith +mut curse}, anyone?

There's definitely a balancing point there, and determining the utility of artefacts isn't broadly possible anyhow, since it depends on character build.

[Blade]

As a quick example, things like "The +1 leather armour of Insufficient Utility {rC+}" should not be possible. I may be misremembering, but I am fairly certain that I have gotten artefacts that are essentially unenchantable branded items before, without any other traits. Those are by far the most problematic randarts; at least things like "the +1 leather armour of Insufficient Utility {rC+ Int-2} aren't functionally identical to normal items.

When I first saw an artefact in Crawl, I expected it to be something significant, and was disappointed when I found out that it was useless. The frequency of artefacts should be greatly reduced in general, with certain minimum parameters (so, as other people have mentioned, they could be very good or very bad, but not boring and forgettable). This might be swinging the pendulum a bit too far in the other direction, but I would be pleased to have games where <10 artefacts generated in a standard three-runer, as long as none of them were like my previous examples.

[Galefury]

One problem with regular junk is that equipment-wearing monsters need some equipment. Which has to be mostly junk, because otherwise there would be a ridiculous amount of good stuff. Also, what is and isn't junk is highly dependent on the character. Even a +9 vampiric bardiche will be junk to a lot of chars.

[XuaXua]

Throwing it out there:

We can enhance most weapons (aside from enhancer staves and ammo), and armour (including accessories - gloves, boots, cloak, etc.)
We can't enhance jewellery.
Randarts come as weapons, armour, and jewellery.
I generally find randart jewellery and accessory armour to be more valuable than randart weapons or body armour, and these seem to consist of less junk.

I'm wondering if the scale can be tipped to favour non-weapon, non-body armour randarts if the non-junk generation is reduced.

[ddubois]

I don't really see a difference between junk artefacts and junk non-artefacts.

Maybe logically and from a game-mechanics standpoint, there is none, for a crawl player. But as a game player, with past experiences with other RPG games shaping my expectations, there is a big difference.

Like others have said, I don't have a problem with the artifact generation rate on jewerly. The junk:useful rate is great there. It's about 50:50 that you want to wear/carry some artifact jewelry you find. But for weapons, once Trog has given me something, I don't even buy them for $200 from the store anymore. I know it has base damage 8, so I don't care what else it has. It'll just be a waste of a scroll.

Maybe if the game mechanics weren't such that base type was the be-all-and-end-all (like if the base damage contribution was additive instead of multiplicative), then these artifacts could find some use. But that's a big and separate discussion.

[One-Eyed Jack]

I don't really see a difference between junk artefacts and junk non-artefacts.

Maybe logically and from a game-mechanics standpoint, there is none, for a crawl player. But as a game player, with past experiences with other RPG games shaping my expectations, there is a big difference.

Good.

Edit: That is, it's fine if you feel intuitively some way or another about a certain element of gameplay, but if you recognize that this has nothing to do with crawl itself then you shouldn't use that feeling as evidence for changing anything.

[ddubois]

Good.

Edit: That is, it's fine if you feel intuitively some way or another about a certain element of gameplay, but if you recognize that this has nothing to do with crawl itself then you shouldn't use that feeling as evidence for changing anything.

Well, I don't agree with that at all. Why shouldn't external biases and player expectations impact game design?

Let's hypothesize crawl was written so that enchant weapon scroll made your weapons' numbers go down. But that was actually good, because weapons got more powerful with lower numbers, so by the time you hit -7/-7 on your sword, man, you were really rocking. A new player might come in and go "Oh, wow, negative numbers are better? That's totally the opposite of what I'd expect. Every other game, bigger numbers are better. You guys should change this." You'd have to be some kind of crawl super fanboi to defend that -7 system with a statement like "this has nothing to do with crawl itself then you shouldn't use that feeling as evidence for changing anything".

[One-Eyed Jack]

Some people prefer THAC0.

[sardonica]

Good.

Edit: That is, it's fine if you feel intuitively some way or another about a certain element of gameplay, but if you recognize that this has nothing to do with crawl itself then you shouldn't use that feeling as evidence for changing anything.

Well, I don't agree with that at all. Why shouldn't external biases and player expectations impact game design?

Let's hypothesize crawl was written so that enchant weapon scroll made your weapons' numbers go down. But that was actually good, because weapons got more powerful with lower numbers, so by the time you hit -7/-7 on your sword, man, you were really rocking. A new player might come in and go "Oh, wow, negative numbers are better? That's totally the opposite of what I'd expect. Every other game, bigger numbers are better. You guys should change this." You'd have to be some kind of crawl super fanboi to defend that -7 system with a statement like "this has nothing to do with crawl itself then you shouldn't use that feeling as evidence for changing anything".

This is not nearly the same situation. We can define -7 or 7 with relations to personal experience. Have you ever seen an artifact weapon in real life? If so, I want to move to your place.

[bobross419]

I think the argument for improving artifact weapons in some way stems from the 'no brainer' aspect of crawl that gets thrown around a lot...

As it stands it is a no brainer to get the Exe Axe, Eudemon/Trishula/WhipThing, etc. with a good brand and enchant it up all the way.

Is there a chance that a better artifact will spawn? Sure there is, but its such a small chance that you're better off just going with the nonartifact weapon + EW.

This is just a very basic example of how small the chance is to get a randart Exe Axe better than +9/+9...

Chance of getting any axe: 1/6 (just based on number of weapons)
Chance of that axe being an Exe: 1/5 (again based on number of axe types)

Been a while since I've had to do stat/probabilities, but thats what, 1/30 chance of even getting an Exe Axe when an artifact weapon is generated.

When you factor in the chance of the enchantment being even close to say +8/+8 (i'm not source diving for the chance, sorry guys) the likelihood of getting an axe better than +9 exe is even worse. Not to mention the chance for getting bad modifiers (-TELE or *Rage or something that can't really be quantified). And Axes has fewer available types than any other weapon class (save Staves), so the chances of getting an equivalent Demon Sword/Eudemon are even worse.

I'm sure I'm not the only player that just ignores any artifact weapon that isn't of the same type I'm wielding (which in effect is like saying I just ignore artifact weapons in general since I've never seen a Blessed Trishula artifact which is my melee of choice).

Again, my numbers aren't firm, I didn't dive the source to check out how Crawl determines what item base type to generate... The numbers I gave are just for example purposes.

And for the record, I don't really care one way or the other... I'm fine with artifact weapons the way they are... I just consider them floor trash and don't ever bother with them. No biggy. I do agree though that non-weapon artifacts are great as is and I never really have a no-brainer (unless its just an unbeleivably good randart).

[galehar]

As it stands it is a no brainer to get the Exe Axe, Eudemon/Trishula/WhipThing, etc. with a good brand and enchant it up all the way.

Why do you only consider the end-game weapon? You are going to use many weapons before you find and enchant that perfect weapon, some of them might be artefacts.

[crate]

An early good weapon has a tremendous impact on the game, probably more of an impact than upgrading from glaive to bardiche at Crypt:5 can ever hope to have. Like I said earlier it is true that most artefacts are not endgame weapons ... but they are often pretty good early on.

[bobross419]

As it stands it is a no brainer to get the Exe Axe, Eudemon/Trishula/WhipThing, etc. with a good brand and enchant it up all the way.

Why do you only consider the end-game weapon? You are going to use many weapons before you find and enchant that perfect weapon, some of them might be artefacts.

Why don't you consider the end-game weapon? You are going to work towards one perfect weapon after you use all the others, but it won't be an artifact.

Like I said, I couldn't care less. If I find a good weapon on the way, I'll use it, but I'm always just waiting to find a Demon Trident that will eventually become a Trishula... Once I get the DT I'll switch to it... And I usually pick one up at the end of Lair since I seem to always get the Hell entrance. If not I'll be getting one by the end of Vault from Hell Knights. No brainer right there.

[TwilightPhoenix]

There's more to Crawl than the end-game, especially since 99% of all characters never see the end game.

[XuaXua]

If you've already beaten the game by going TSO with a trishula, wouldn't you want to try something different the next time?

[Midpoint]

This is probably a terrible distraction, but the thought occurred because of this thread.
Why not introduce a "scroll of artifacture" or some such. Similar to a scroll of acquirement, but you apply it to an item in your inventory. Introduce probabilities for rebranding, bonus/malus, additional properties. Balance would have to be a combination of the probability of completely ruining a decent weapon/armour/jewelry, and extraordinary rarity (i.e., generate only in branch ends, or specific mini-vaults with depth restrictions, maybe portal vaults). Only really a help for randart usage (at least it's base type is something you wanted to use), and risks making it useless.

I'm unsure how to discourage scummy behavior of dumping enchant scrolls first, other than having the final enchantment ignore initial enchantment.(Maybe with a higher chance of highish final enchantment, +5 or greater, over low bonus/malus?)

I like the idea of having to decide when a base type is good enough to risk it on, but not game breaking if you lose the item due to +Mut or bad maluses.

It's not a complete answer, but like OP, I often find myself in late game wearing a set of armour and jewelry artifacts, and only rarely swinging artifact weapons (and then, more likely a randart than a unrandart). The heavy armour issue can probably be fixed by changing the enchantment bonuses on armour as a proportion of the range of enchantment (maybe this is already being done, I haven't looked at artifact generation code yet). I have found in midgame that I'm more likely to wear artifact armour of a crappy base type, sacrificing GDR for +AC IF it's enchanted over the standard limits.

Here's another thought - part of the endgame balance is sometimes the artifacts you've found are mostly crap. How you decide which to equip and which to leave on the floor is the strategic decision (AC? EV? Resists? Evokables? Stat bonus/malus? Slaying? GDR?). I really don't relish having superawesome artifacts remove that choice, which is reliably more variable than many choices I have to make. I want artifacts to force me to make a choice, not make the choice for me.

[BlackSheep]

"Artifacting" a weapon sounds like an interesting idea for a one time only god ability.

[rchandra]

The code for artifacting is already sort of ingame, there's a wizard mode command to convert an item into a randart.

[dpeg]

I don't think a scroll of artifacturing will be a good addition. If it creates random artifacts out of mundane items, it's like a poor man's version of the scroll of acquirement. If it always produces useful (but not necessarily extremely good) artifacts, the result will still often be a dud because you use it precisely to go for an endgame item. If it always produces highest tier artifacts, it is too strong.

However, I believe artifacturing would make a great Xom power. I've been discussing Xom a bit with jpeg and we agreed that Xom needs a strong early game appeal. You want a real chance of something *awesome*, like a good demon companion, a great mutation, a cool gift or the like. Artifacturing your worn item is great -- even if it is just good, that'll be very helpful in the early game.