Pre-enchanted armours encourage scummy behaviour

[BlackSheep]

The current system requires that you risk your ?remove curse to find out the enchantment level of armor. Six robes, six chances to have to burn a scroll.

The proposed system still requires you to risk the exact same amount of remove curse scrolls. The exact reason why is not intuitive, but I will continue to try to make it easier to comprehend.

You've yet to address this point at all. Unless your proposal is meant to work like your pseudocode in which case you're taking the decision to use scrolls out of the hands of the player.

[some12fat2move]

The current system requires that you risk your ?remove curse to find out the enchantment level of armor. Six robes, six chances to have to burn a scroll.

You realize that under fergy's proposal all cursed armours would be unidentified alongside all the armours better than yours? You'd still be looking at armour that may or may not be cursed and may or may not be better than your current armour, so there's no way you could try on potentially better armour without risking RC.

I did not start this thread, I only created a solution to a (very minor) problem. I'll let others argue on how big of a problem it actually is, and whether or not it's worth the time to code.

Thanks for the suggestion, but galehar already said it probably isn't worth it, and I happen to agree because trying on armours takes very very VERY little time.

in my mind, it's balance neutral

It's actually not balance neutral because of corrosion - you could wind up knowing about armour better than yours for free just because at one point you had armour better than it.

[BlackSheep]

You realize that under fergy's proposal all cursed armours would be unidentified alongside all the armours better than yours? You'd still be looking at armour that may or may not be cursed and may or may not be better than your current armour, so there's no way you could try on potentially better armour without risking RC.

And again, you'd be giving players information about the non-cursed items without them risking anything. I don't care if you'd have to burn through the same number of scrolls to get the same information. The player doesn't know which items will require a scroll until they require a scroll.

[some12fat2move]

And again, you'd be giving players information about the non-cursed items without them risking anything.

Does it matter? Wouldn't that information be useless barring corrosion?

[Fergy]

that looks pretty messy, if it's just suppsoed to represent the decision making tree there is no point in writing it in pseudocode.

all of that could be summed up with:

if you have un-ID armor, your current armor isn't as good as you'd like it to be and you still have remove curse scrolls equip-ID one of those armors.
you like the result? no? un-curse if necessary, rinse, repeat.

The pseudocode on a possible decision making tree may be completely pointless. However, understanding game theory is very difficult. I'm running out of ideas into how to express that this proposal is balance neutral... I think my comparison to a game of chance was the easiest to comprehend, but it didn't work either.

Imagine a bag of marbles of know quantity and type. The bag contains three types. Each turn you can pull a marble out of the bag.

• Blue: You win, game over.
• Red: You must use a "remove curse scroll", if you have 0 remove curse scrolls, you lose.
• White: Neutral, try again.

Example 1: Imagine the bag has 3 marbles: 1 blue, and 2 red. The player has 1 scroll. What are the overall chance of winning?

Round 1:
Chance of blue: 1/3
Chance of drawing red: 2/3 (must use scroll)

Round 2a: (only if drew red on round 1) Bag contains 1 red and 1 blue.
Chance of blue: 1/2 you win
Chance of red: 1/2 (no scrolls, so you lose)

Overall chance of winning = 1/3 + 2/3 * 1/2 = 2/3

Imagine the same bag, but now it contains white marbles as well. Should this change the overall odds of winning? (answer = no, proving why is harder)

[BlackSheep]

The proposal ignores the fact that order matters when identifying a set of armors with a limited pool of remove curse scrolls. If you have only one scroll and are unlucky enough to try on the cursed armor first, you have to wait until you find more or else risk the next piece being cursed too.

[some12fat2move]

The proposal ignores the fact that order matters when identifying a set of armors with a limited pool of remove curse scrolls. If you have only one scroll and are unlucky enough to try on the cursed armor first, you have to wait until you find more or else risk the next piece being cursed too.

Which also happens under fergy's proposal in exactly the same way?

[Fergy]

in my mind, it's balance neutral

It's actually not balance neutral because of corrosion - you could wind up knowing about armour better than yours for free just because at one point you had armour better than it.

You are correct, it's not exactly balance neutral. I will have to restate it as: So close to balance neutral, that it might as well be...

[BlackSheep]

Which also happens under fergy's proposal in exactly the same way?

The proposal says you get to know about all armors that are worse than yours automatically.

[mumra]

Just because something is not broken, doesn't mean it cannot be made better. The only risk is the cure is worse than the annoyance. This is a very mild annoyance.

The risk is that someone might spend time non-trivial time discussing, designing and then coding something only to find that it's worse, the same, or only marginally better; when they could have been creating fixes for actually broken things, or even developing entirely new content ...

[dpeg]

By the way, all of this would be (even more) moot if Crawl's curses were a little more meaningful. I had some idea, but I don't feel good enough about it to propose it (galehar and evilmike were very reluctant, to put it mildly).

In any case, curses are one of the few things where I suggest not to take a page out of Brogue's book. Perhaps a decent idea will emerge over time... (For context, look here.)

[Fergy]

Which also happens under fergy's proposal in exactly the same way?

The proposal says you get to know about all armors that are worse than yours automatically.

You get to know only about non-cursed armors that are worse than yours. These non-cursed weak armors are like white marbles in my previous example. Do you agree that the addition of white marbles does not affect the odds of winning that game?

The jump from marbles to armor is a difficult jump. It involves changing the known composition of the bag to an unknown one.

Current system:

Total Marbles = Unknown blue + Unknown red + Unknown white
And a Known amount of remove curse scrolls

My proposal changes the game to:

Total Marbles - Known white = Unknown blue + Unknown red
And a Known amount of remove curse scrolls

The overall odds of winning these games are exactly the same.

(I'm not trying to be condescending. I apologize if I may sound like I am. I'm only trying to find out where the disconnect is. Game theory is not easy math.)

[TwilightPhoenix]

You get to know only about non-cursed armors that are worse than yours.

There lies the problem. You're essentially telling the player "These pieces of armor are both worse than yours and are not cursed. Therefore, as your armor improves, each unidentified piece is more likely to be cursed rather than useful." So, there's even less incentive to try out new armor the better yours gets because the odds of that piece being cursed are much higher. Under the current system, all unidentified armors have the exact same chance of either being an improvement, cursed, or both regardless of what enchantment level your armor is at (max enchantment aside). In your system, that no longer holds true.

And again, we already have a feature in the game for easily identifying armor that's either cursed or better than yours. It's called Ashenzari.

[some12fat2move]

There lies the problem. You're essentially telling the player "These pieces of armor are both worse than yours and are not cursed. Therefore, as your armor improves, each unidentified piece is more likely to be cursed rather than useful." So, there's even less incentive to try out new armor the better yours gets because the odds of that piece being cursed are much higher. Under the current system, all unidentified armors have the exact same chance of either being an improvement, cursed, or both regardless of what enchantment level your armor is at (max enchantment aside).

Actually, that "as your armour improves" bit is exactly how it works right now. Supposing you have +3 plate armour, you'll have to wade through a lot of cursed suits of plate before you find a better one - in the current game as well as in fergy's proposal. There is no change in this respect.

[TwilightPhoenix]

Not quite. Currently, if you see a piece of plate mail on the ground, it might be a potential upgrade or it might be cursed. The fact that it might be an upgrade may encourage you to go check it out. There's low risk (unlikely, but possible, to be cursed) but the potential for reward is low too (unlikely to be an improvement).

Under the proposed system, if the plate is auto-IDed as worse than yours, you ignore it. If it's not auto IDed, then it's either cursed or an upgrade. It's far, far more likely to be cursed than an upgrade. So, it ends up being a much higher risk vs still a very low reward, assuming it's unidentified to begin with.

Basically, it's handing out free knowledge (which is what Ash is in the game to do) as the player automatically now knows which pieces of armor is worse than theirs and also which pieces are risky to try. In addition, it's also significantly skewing the risk vs reward on what's not a major decision. It's not a huge risk, I'll be honest, but it's still there and it's technically increasing.

Also, by the way, how does this system handle corrosion? Does auto-IDed armor become unIDed if your armor gets corroded to a point worse than it? Does equipment worn by monsters get IDed if it gets corroded to a point that it's worth than yours? Can you ID armor based on the highest level enchantment you had on the armor or the level it got corroded down to?

[Fergy]

Not quite. Currently, if you see a piece of plate mail on the ground, it might be a potential upgrade or it might be cursed. The fact that it might be an upgrade may encourage you to go check it out. There's low risk (unlikely, but possible, to be cursed) but the potential for reward is low too (unlikely to be an improvement).

Under the proposed system, if the plate is auto-IDed as worse than yours, you ignore it. If it's not auto IDed, then it's either cursed or an upgrade. It's far, far more likely to be cursed than an upgrade. So, it ends up being a much higher risk vs still a very low reward, assuming it's unidentified to begin with.

Your logic is poor. The wear-id game will be played until one of 3 possibilities happen:

1. Find an armor better than current armor.
2. Find a cursed armor.
3. Run out of unidentified armor.

Under the first two conditions, the game will be continued if the player wishes to push his luck. The game always ends when there is no armor.

You keep assuming that the wear-id game is x% chance of finding a better armor, but this is not how the game is played. There is not a cost associated with playing. There is only a cost associated with losing.

The actual wear-id game is z% chance of finding a better armor before finding a cursed armor. All armors equal to, or weaker than the current armor are not factored into the game. They are merely tedious actions that must be done.

Example 32: You have found an amazing stack of 100 armors. 1 is cursed, 1 is better than what you have, and the other 98 are rubbish.

What are the odds of finding the better armor before the cursed armor?

Spoiler: show

50%
If you want to work out the actual math, its: Chance of better armor/(chance of better armor +Chance of cursed armor)

sumfrom0to98(1*(98-n)/(100-n)^2)) / (sumfrom0to98(1*(98-n)/(100-n)^2)) +sumfrom0to98(1*(98-n)/(100-n)^2)))

This can be simplified to: # of better armor / ( # of better armor + # of cursed armor)
or 1/(1+1) = 1/2

Example 33: You have found a small stack of 2 armors. 1 is cursed, 1 is better than what you have.

What are the odds of finding the better armor before the cursed armor?

Spoiler: show

50%

Basically, it's handing out free knowledge (which is what Ash is in the game to do) as the player automatically now knows which pieces of armor is worse than theirs and also which pieces are risky to try. In addition, it's also significantly skewing the risk vs reward on what's not a major decision. It's not a huge risk, I'll be honest, but it's still there and it's technically increasing.

Basically, it's actually allowing the player to get back to playing the game instead of trying on worthless armor. My suggestion is nearly equivalent to making the memorization of spells binary.

Also, by the way, how does this system handle corrosion? Does auto-IDed armor become unIDed if your armor gets corroded to a point worse than it?

Once identified, always identified. To do otherwise would encourage scummy behavior (keeping track of formerly id'ed armor outside the game).

Does equipment worn by monsters get IDed if it gets corroded to a point that it's worth than yours?

Only armor in the player inventory, or on the tile directly below the player gets ID'd. Equipment worn by other monsters remains an Ash ability.

Reasoning for only IDing armor in inv or under player: Imagine you see a piece of UN-id armor on the other side of a trap. Would you still go after it if you knew it was weaker than what you are currently wearing?

Can you ID armor based on the highest level enchantment you had on the armor or the level it got corroded down to?

The current enchantment level. Although IDing armor based on the highest level enchantment would make sense from a RP perspective, it is a different system than what the game currently has. My goal is not to change the system, my goal is to make the current system less tedious.

[BountyHunterSAx]

Fergy is right. If *AND ONLY IF* cursed armours (regardless of their enchantment) remain un-id'd and white just like non-cursed upgrades then risk/reward is preserved. I'd like to point out that this was not at all made clear by the original poster in his first post or in more than a few of his followup posts where he basically said "lower level armours are ID'd" - regardless of their curse status.
But with that change, the difference is minimal and pretty much a straight improvement. Yes, the number of times you need to make a risk/reward decision changes, since you're (effectively) ignoring a large number of armours now. But the number of times that decision MAKES A DIFFERENCE - for good or for ill - is identical. If anything, it makes the choice MORE interesting, not less, by removing the times where the answer is: "Trick question! Guess again."

That said, more accurate is also what others (including himself, iirc) said earlier: this is too small an issue to really care about. It's a trivial number of extra turns spent, generally ego-items are more favorable. Still, I'd love to see a non-dev write a patch for it as a side-project and then see what the community thought of it in practice.

-AHMAD

[mumra]

Also, by the way, how does this system handle corrosion? Does auto-IDed armor become unIDed if your armor gets corroded to a point worse than it? Does equipment worn by monsters get IDed if it gets corroded to a point that it's worth than yours? Can you ID armor based on the highest level enchantment you had on the armor or the level it got corroded down to?

This is an extremely good point, and additionally none of the reductionistic examples involving marbles take into account that there are loads of different brands of armour and of course artifacts.

It seems to me that the complexity of designing a completely satisfactory solution massively outweighs the triviality of just trying things on. And let's not forget, some people actually enjoy the part of the game that this proposal attempts to eliminate.

Still, I'd love to see a non-dev write a patch for it as a side-project and then see what the community thought of it in practice.

I said this somewhere else; if someone, particularly a non-devteam member, is going to the time and effort to write a non-trivial patch, then I'd far rather it was fixing a real issue of more universal benefit; or even developing new, quality content with a good chance of getting accepted ...

[galehar]

Still, I'd love to see a non-dev write a patch for it

Please don't. Seriously don't. It would never be accepted anyway. Many things have been overlooked.

First, how do you explain it in the manual? For such a small change, if you can't explain it clearly with a simple sentence, then you should realize that something is wrong. When players start asking "what's up with the identified armours?", what do you tell them? When bugs about "most armours are generated identified past D:4", how do you answer them?
Also, telling if an armour is worse than the one you're wearing isn't always as easy as in the given examples. What happens if you're wearing a +0 scale mail and you find a ring mail? At what enchantment is it considered better than your armour? Does it depends on your armour skill? What if your scale mail has rN+? And what if it has rF+?

Nobody wants to have to answer all those questions. And the (theoretical) gain is so small. This is a bad idea.

[mrbobbyg]

I'm stunned it has lasted as long as it has.

[Fergy]

Please don't. Seriously don't. It would never be accepted anyway. Many things have been overlooked.

First, how do you explain it in the manual? For such a small change, if you can't explain it clearly with a simple sentence, then you should realize that something is wrong.

Maybe say:

Wearing an uncursed armor identifies the enchantment level of all other armors that are also: uncursed, of the same type, unbranded or the same identified ego, and of an equal or lower enchantment.

Or if you want to change the suggestion a little, you could say:

Identifying armor will also identify all subsequent armor that is exactly the same and of an equal or lower enchantment. Cursed, unknown ego, and artifact armors are never identified in this fashion.

This would make armor identification work just like potion / scroll id. IMO, it's even more intuitive than both what is in the game, and what I suggested. I did not suggest this because: it changes the current balance, and sounds harder to code.

Also, telling if an armour is worse than the one you're wearing isn't always as easy as in the given examples. What happens if you're wearing a +0 scale mail and you find a ring mail? At what enchantment is it considered better than your armour?

These armors are of a different type. Wearing one will never identify the armor.

Does it depends on your armour skill?

No.

What if your scale mail has rN+? And what if it has rF+?

Wearing an uncursed "+2 scale mail of positive energy" would Id:

An uncursed +0 to +2 scale mail of positive energy
An uncursed +0 to +2 scale mail

Wearing an uncursed "+2 scale mail of positive energy" would not Id:

Any cursed armor
Any armor that is not scale mail
Any armor greater than +2 enchantment level.
Any armor of a different ego
Any armor with an unknown ego (even if its positive energy)
Any artifact

(I was unsure if these were rhetorical questions that someone might ask, or if you were legitimately asking a question. I decided to answer anyway.)

Nobody wants to have to answer all those questions. And the (theoretical) gain is so small. This is a bad idea.

Every improvement, no matter how insignificant, is still an improvement.

IF this idea is deemed so complex that you are worried about some newb getting confused,
AND IF on the obscenely rare chance that someone bothers to make a patch for it,
AND IF on the odd chance that the community likes it,
THEN don't implement it directly. Instead make it super secret squirrel feature that must be turned on in the settings.

[Grimm]

Fergy, it's time to let go.

[Fergy]

Fergy, it's time to let go.

I understand, and have already done so mentally before I even suggested it. My posts in this thread generally address on of three issues:

1. This idea changes balance.
   
   This is a failure on my part to accurate describe why the idea works nearly the same as the current system. Clearing up confusion must happen before any meaningful discussion can take place.
2. This idea is too complex.
   
   This is a failure on my part to succinctly explain the idea.
3. This idea is too insignificant of an improvement to be implemented.
   
   This is a failure of others on how they perceive the world. Progress does not need to happen in leaps and bounds, it merely has to be incrementally better than the status quo. Eventually these small changes add up to something profound.

[TwilightPhoenix]

Let's put it this way. The amount of coding it would need to result in a small, if any improvement, makes the opportunity cost of adding this proposal to the game too high to be considered when there are lower opportunity cost proposed features that'd give much larger returns for the same, or less, effort.