Improving Random Altars

[Tiktacy]

I sort of feel like they don't offer any kind of interesting choice to the game. The bonus to piety is a step in the right direction, but I think maybe having it be completely random pushes people away from rolling the dice(xom and chei are possiblible choices after all).

I really like the idea of them though and I think it has potential, but I suggest an alternative:

Instead of having the altar include all gods, have each altar include a selection of gods listed on the description of the altar. So, for example, if you pray at a random altar, it would give the message: "This altar appears to be fairly ambiguous, but you can make out a few important details. This altar could be an altar of any of the following gods: Trog, Ashenzari, Jiyva, or Qazlal."

This would make it a lot more likely that players might choose to use one of these altars instead of skipping them entirely based on the fact that they could potentially ruin their entire gameplan.

[Hurkyl]

A complementary alternative is to give a player a choice of three random gods on use (but still require one be chosen). Harder to flavor, though.

[le_nerd]

Flavor is easy to come by. "You find an ecumenical altar of advertising. On it pictures of three different gods show the boons they bestow upon the follower who falls to their knees. Do you wish you pray to Trog or Okawaru or Ashenzari?"

[byrel]

The only change I'd like to see is removing Xom from the roll list. He's unique in that he's (by design) worse than atheist for most characters. That means that rolling Xom is close to just having a 1/23 chance to auto-lose. Like, even rolling Trog on a FE gives you a reasonable chance to adapt to his playstyle and win. IT'll be rough, but on D:2 often doable. Xom is simply going to be worse than atheist the entire game.

[njvack]

If it's less randomness folks want, how about a portal vault? Say, you can't enter the portal if you're worshipping a god, and you must be worshipping one to get back to the dungeon.

Inside the vault, you find 3-5 altars; at least one will be allowed for your character.

[le_nerd]

Sounds phantastic, but might have communication problems, potentially leading to a stream of aggravated "I didn't want none of the gods".

[njvack]

Yeah. There would need to be a warning message like labyrinths used to (still do?) have.

[chequers]

I agree that if the terrible choices could be removed random altar would be better (eg tso for vm, gozag for tr). I'm not sure the solution is giving the player a choice -- part of what I like about the altar is that there's no player choice beyond deciding to use it.

Perhaps some conditions can be added for especially bad cases like the above. These conditions could either deduce or completely remove certain gods from the random pool.

This would be the most fragile solution from a code implementation perspective, but would change the user experience the least -- the altar would just seem to be "lucky" and never/rarely get terrible combos.

BTW, Xom will always be an option. That's definitely his thing.

[savageorange]

Hm. What if we invert the idea? ie. instead of 'you can get gods X, Y, or Z from this altar', specify 'You won't get gods X, Y or Z from this altar'. Then the player will choose based on whether that's an acceptable risk to them; it makes them slightly more informed, without constraining the result so much that it can be considered 'reliable'.

I feel that the player making an 'informed choice' (even if the information is extremely limited) is one of the things that can allow them to feel a random altar is acceptable. So I'm basically saying here : I disagree with the 'going in completely blind' aspect - let the player go in mostly blind, and judge whether their odds are good according to the limited information they get.

I agree in principle with adding more conditions, though it would definitely be fragile. Perhaps reducing weightings for a god based on 'spell known / XL' ratios (that is, the strength of the weight reduction would be reduced as XL goes up. For example, knowing 3 spells at XL 5 would pretty much rule Trog out. At XL 10, not quite so severe). Skill based weight reductions seem like they could potentially help, but how to implement them (besides the worst case: hardcoding) is far less obvious to me.

[chequers]

After thinking more, I don't think forbidding any combo is a good idea. There's no bright line between "should definitely be forbidden" and "probably shouldn't be forbidden".

For example getting TSO on a VM is a really interesting result. You could 1) give up VM (probably tough) 2) give up TSO and worship a non-evil god (lose bonus piety) 3) give up TSO and worship an evil god (TSO wrath!). All of these are interesting decisions.

Gh/Tr^Gozag is a pretty bad combo, but Gozag wrath is quite survivable.

There's a reward for using random altar and that is a boatload of bonus piety. Some results might be terrible, but that's the drawback. This might be exposing issues in inconsistent punishment from certain gods that can be cleared up, so that abandoning any god in the early game is (at least occasionally) survivable.

[WingedEspeon]

I agree that if the terrible choices could be removed random altar would be better (eg tso for vm, gozag for tr). I'm not sure the solution is giving the player a choice -- part of what I like about the altar is that there's no player choice beyond deciding to use it.

Perhaps some conditions can be added for especially bad cases like the above. These conditions could either deduce or completely remove certain gods from the random pool.

This would be the most fragile solution from a code implementation perspective, but would change the user experience the least -- the altar would just seem to be "lucky" and never/rarely get terrible combos.

BTW, Xom will always be an option. That's definitely his thing.

If Xom is an option then the reform will be pointless.

TSO on Vm is fine because TSO has no wrath on abandonment. Gozag on ghouls is really bad but still (somewhat workable) I have won Gh^Gozag and it was definatly better than Xom. I'm not sure about Gozag on trolls. If you want to do a troll of gozag you probably can't use auto explore do maybe forbid it because it is super tedious. Gozag on Vp is workable but you will need to go stabber to get food. Trog is stupidly strong to the point that my DEWz^Trog runs were actually fine. Zin on transmuters is like TSO on VM. Quaz is bad but at least his downside is predictable unlike Xom. I feel like I will start a flame war if I say anything about chei (although I generally agree with tabstorm about the god). Honestly no Xom/Chei would make me try the altar with a lot more characters and just no Xom would have me trying it with strong characters.

[byrel]

I think the line between intentional troll god and non-troll gods is a pretty clear one. Xom isn't intended to be a viable god choice. I don't see any reason he should be treated the same as those who are.

I agree that the 'suboptimal' non-troll gods lead to interesting situations. TSO on VM, Trog on caster, etc. lead to interesting decisions and play as you adapt. Xom just doesn't force the same sort of interesting play on you. He makes you either want to quit the character, or play it for a lark to see what happens. Either way, it doesn't introduce new tensions to the game like a random non-troll god does.

I don't support removing Chei, btw. He's a god who's intended to have a positive power level. I know people debate about whether he actually does (and I have no opinion) but if he doesn't, he needs fixed, not excluded from ecu altars.

[PleasingFungus]

worth noting that xom wrath is one of the weakest wraths.

[ololoev]

An altar of intervention!
When you worship on it, a random god pops up and suggest to join his ranks. You may accept or decline, but if you decline, then you worship another random god.
This will make some interesting decisions whether player wants to worship a given god, or reject ant get something else (maybe better or worse).

[savageorange]

An altar of intervention!
When you worship on it, a random god pops up and suggest to join his ranks. You may accept or decline, but if you decline, then you worship another random god.
This will make some interesting decisions whether player wants to worship a given god, or reject ant get something else (maybe better or worse).

Exactly the right level of choice, IMO. Just needs a bit of flavour explanation worked out.

[njvack]

In general, I like the "press your luck" mechanic. Of course, you should only get the bonus piety if you accept the first god; no one likes being second choice.

[MainiacJoe]

There are 23 gods with Pakellas. Xom happens therefore 4.3% of the time. There are probably a handful of other gods that you really don't want based on your character. So let's be generous and say that 17% of the time you'll be disappointed, that's roughly 1 in 6--but it's probably closer to 10% or even just Xom. So a random altar gives you at least 5/6 chance of getting an acceptable god, instant bonus piety, and also a couple extra floors to build piety with the old-fashioned way. That sounds like a good risk to me.

[byrel]

It's actually a bad risk in a roguelike. Generally, any action which is entirely non-forced and has a 5% chance of maiming your character should never be taken. Mechanically (assuming perfect play) you have a 80%+ chance of winning. Making a single choice that lowers your best-case chance of winning to 77% is a bad deal.

[chequers]

This chance is also on d:2 or d:3, so if you do get a God that you consider game-ending it's not the worst thing to start again.

[byrel]

Fair enough. Though roguelikes have a tradition of looking down their noses at start-scumming, the main point of playing is to have fun. I still stand by it being a bad risk from a character-survival perspective, but it's not like that's the only legitimate grounds to make a decision on.

[Sprucery]

It should be pretty clear that the random altar is just a challenge for those who wish to take it. If you're concerned about winning, you should never take it.

But it's a fun addition imo. I haven't used it yet but I most certainly will at some point when I play a non-book start with a species and background I've already won (separately; I never play a won combo again).

[beakerthefrog]

Perhaps instead of giving the player some semblance of choice (which is exactly the opposite of what the altar is intended for) the player could be given some other sort of boon instead? Say, "You come across a worn altar with a strange artifact embedded in its base. Perhaps if you proclaimed your devotion, the god would gift it to you? Worship the unknown god? Y/N" It gives the player a reasonable um... reason, for accepting the offer, but maintains the randomness and risk. Obviously the artifact would always be something extremely positive.

[Siegurt]

Presently there is a reward for worshipping the random good in the form of bonus piety, if that isnt communicated well enough it probably should be.

Do you feel that the piety bonus isn't sufficient incentive to make it a real choice?

[TeshiAlair]

Isn't the whole point to have a fun random possibly detrimental thing at an early level so that rerolling is not the end of the world?

Because if you want "Random but not bad and how about only out of a few guys" just go to the temple and roll a d6.

[beakerthefrog]

Presently there is a reward for worshipping the random good in the form of bonus piety, if that isnt communicated well enough it probably should be.

Do you feel that the piety bonus isn't sufficient incentive to make it a real choice?

I don't believe it is. Piety for the vast majority of gods is easily attainable, and the majority of immediately useful invocations only require one pip (Berserk, kiku book, etc). A single floor is usually enough to achieve that. Unless you were going to die right there in that room, next to the altar, and only one of the random god abilities was going to save you, it isn't all that useful. Giving the player something tangibly useful offers more incentive to go down that route.

That being said, I don't feel like the altars NEED a change. I just feel like they're kind of boring, and people don't really use them unless they're doing a mutation roulette, wanderer, random god run, in which case, they will most likely worship the first god they come across, rather than waiting for an unknown altar to show up. I'd like to see them inspire a fun, if not viable playstyle niche. Right now, they really aren't all that fun.

[WingedEspeon]

No it isn't. Especialy as the piety bonus Doesn't to anything because you will roll Xom.

[beakerthefrog]

Isn't the whole point to have a fun random possibly detrimental thing at an early level so that rerolling is not the end of the world?

Because if you want "Random but not bad and how about only out of a few guys" just go to the temple and roll a d6.

I partially agree with you. The god choice itself should be completely and utterly random, with no holds barred. That was the whole point of the altar. I just believe that currently there is no point in using them. Even just for fun. Anyone going for a random god run will just worship at the first altar available. Making the altar redundant (the only gods not attainable by the method I described are Lugonu and Jivya, and even then, only in most world generations. It's still possible for them to show up on D:3 for the lulz)

The altars need to do something other than just give you a god in order for them not to be redundant. The extra piety doesn't really cut it, since by the time you use those benefits, you would have already acquired them via normal piety gain.

[Siegurt]

So my first question is, dies there exist an amount of piety that would make random altars a reasonable choice? Tweaking the number is easy after all.

I penalties feel like getting to * immediately is pretty good if you get a decent god, on par with being a zealot, and it isn't hard to see that a gladiator who takes the randaltar and ends up with trog on level 2 is better off than starting as a berserker.

Obviously the worst case is pretty bad (dith on a defe would suck, xom on anybody sucks, sorta), but the best case is pretty good, and while it is never a good idea to take risks in crawl, there is certainly not zero temptation to play the lottery, I would put the risk/reward on par with !Mut.

[Quazifuji]

random god run, in which case, they will most likely worship the first god they come across, rather than waiting for an unknown altar to show up.

To me, this is the biggest issue. I think it's perfectly okay if worshiping at a random altar is never the optimal choice to maximize your odds of winning, but given that praying at the first altar you find is already an option if you want to play a game worshiping a random god that possibly fits your character poorly, I think the random altars would be much more interesting if they offered a stronger incentive to use them. It doesn't need enough incentive to make them optimal, but it should be enough incentive to make them tempting.

Basically, I think finding a D2 random altar should be like finding a D2 potion of mutation. I know it's not optimal, but the possible rewards should make me kind of want to try it anyway. Right now, the extra piety is small enough that it barely even figures into the decision for me at all, and the only temptation is the entertainment value.

[edgefigaro]

Isn't the whole point to have a fun random possibly detrimental thing at an early level so that rerolling is not the end of the world?

Because if you want "Random but not bad and how about only out of a few guys" just go to the temple and roll a d6.

I partially agree with you. The god choice itself should be completely and utterly random, with no holds barred. That was the whole point of the altar. I just believe that currently there is no point in using them. Even just for fun. Anyone going for a random god run will just worship at the first altar available. Making the altar redundant (the only gods not attainable by the method I described are Lugonu and Jivya, and even then, only in most world generations. It's still possible for them to show up on D:3 for the lulz)

The altars need to do something other than just give you a god in order for them not to be redundant. The extra piety doesn't really cut it, since by the time you use those benefits, you would have already acquired them via normal piety gain.

This is totally false. An early piety boon is not trivial with many gods. Get the altar on d2, pull trog, and berserk that adder. Other candidates for fight swinging * abilities (especially at xl 3) include oka, fedhas, ely, makh. There are more gods where early piety is great as well (ru comes to mind.)

I don't go for "random god runs", and you are presupposing I have an idea what the character is going to develop into at creation. I rarely do unless I am making a zig buster.

[Siegurt]

Basically, I think finding a D2 random altar should be like finding a D2 potion of mutation. I know it's not optimal, but the possible rewards should make me kind of want to try it anyway. Right now, the extra piety is small enough that it barely even figures into the decision for me at all, and the only temptation is the entertainment value.

I would say that the random altar temptation on D2 is roughly on par with a D2 potion of mutation for me as things stand right now, personally. (This may be because I am less inclined towards !mut than you are as easily as it could be that I'm more inclined towards randaltar)

[Quazifuji]

I would say that the random altar temptation on D2 is roughly on par with a D2 potion of mutation for me as things stand right now, personally. (This may be because I am less inclined towards !mut than you are as easily as it could be that I'm more inclined towards randaltar)

I'm usually not particularly inclined towards either. But mostly, the upside of a random altar seems very, very small compared to the downside. Potion of mutation is as likely to do something good as it is to do something bad.

[dpeg]

So my first question is, dies there exist an amount of piety that would make random altars a reasonable choice?

That's a good question, and the answer is yes: getting access to a god with ***** piety is extremely strong. So like you say, there should be a sweet spot to make random altars interesting. (With the current mechanics, it is not possible to make random altars optimal, in my opinion.) I am not convinced that the current piety boost is too small -- it is noticeable.

I think random altars are fun, and can get you play a game never intended, if you're into that sort of things. It's alright to make them stronger, the two obvious methods would be increased piety boost and some brake on early wrath. Restricting the random selection (e.g. get one god out of three) defeats the point: this would be still very much into "what kind of build do I pre-plan".

[savageorange]

Restricting the random selection (e.g. get one god out of three) defeats the point: this would be still very much into "what kind of build do I pre-plan".

So what about my inverted idea ("one god not out of these three")? Like I said, it's a little silly-seeming and difficult to flavour, but avoids the build preplanning idea, instead assuring the player about a few gods they will -not- get
(getting a really annoying god is one of the primary, and IMO justified, concerns expressed in OP; some people consider Xom or Chei 'worse than atheism', and getting Trog on a character that has thus far focused on spell casting can reasonably be considered a 'character is over, time to quit' moment.)

Yes, I realize this is still 'restricting the random selection'. It's just that we have enough gods that removing a few from the list is not going to be enough to allow any kind of meaningful build preplanning, it's just going to move it from 'russian roulette' to 'I get something that might or might not be helpful but won't inspire me to immediately quit the character.'

I agree that piety gain and wrath mitigation would help... for Chei. I don't think Xom can be helped. Wrath would have to be pretty much totally negated for a few character levels to make the Trog scenario into anything reasonably doable.

[HardboiledGargoyle]

If you totally negate wrath of the forgotten god (not sure what you mean by "for a few character levels" and wrath already scales with XL) then it's hardly any different from just having its altar there.

The "any god except X, Y, and Z" idea is so expert-centric it's disgusting, since it's meaningless noise except for the very experienced.

Might work as a temple altar though - take on a random god that didn't generate in the temple. You know that feel when you get to the temple, wanting to worship a certain god, and its altar is not there?

[savageorange]

If you totally negate wrath of the forgotten god (not sure what you mean by "for a few character levels" and wrath already scales with XL) then it's hardly any different from just having its altar there.

Dunno who you meant here -- I said 'for a few character levels' but not 'of the forgotten god' -- I was thinking of scenarios like primary spellcaster randomly gets Trog. Chances are you're just straight up hosed through having no reliable offense, unless penance is temporarily/partially suppressed (yeah, wrath != penance, sorry for that vagueness.)

The "any god except X, Y, and Z" idea is so expert-centric it's disgusting, since it's meaningless noise except for the very experienced.

... so, like current random altars, except less noisy?

I don't see how a uniform random selection between (N minus M) gods is expert-centric. You don't need to be expert to see potential problems with Trog, Chei, or Xom -- you just have to look at the ingame help for gods (which you should be using anyway especially if you're new.). The key facts:
* Trog hates spellcasting.
* Chei slows you proportional to piety
* Xom does random stuff to you at random times
are dead easy to work out; they are central themes for the given god. It also should be equally obvious that if a player hates Chei, and Chei is blocked on a random altar, then that particular altar is quantifiably better for that player. If you don't understand that Trog might be a problem for a primary spellcaster, or why move speed is a big deal, or being able to reliably predict what will happen in the next turn is a big deal... that's not being a non-expert, it's failing to think things out at all.

[edgefigaro]

If a player doesn't like a god that much, they have the choice to just eat the wrath, go back to d1 and leave, or play it out. This choice is a feature of the random altar IMO, not a bug. All of these options still allow going to the forum and complaining they got the god in the first place.

Working as intended?

[dpeg]

edgefigaro: Yes, I agree.

Also, if your DEFE gets Trog, the game is not over: pick a reasonable weapon and learn to rage. It might be the strangest deep elf of your Crawl career, but you didn't lose yet by getting Trog. If you get Xom, play it -- the god is fun. Or you can abandon Xom, still experience his effects and take on a new god. Again, nothing is lost.

If you knew before game start that your DEFE was supposed to cast Firestorm after Lair, then by all means don't take the random altar. I think there are at least two types of comments here: (1) players who don't like current random altars and who might try out a version with less variance; (2) players who do use random altars already (perhaps only sometimes), and who would like buff it.
Either way the comments are fine, but since random altars are in use, and fun to some, I suggest to wait more, and see if the piety boost feels alright. The basic mechanic (completely random god) is alright for now, I'd say.

[Jarlyk]

To be fair, when my DEFE got Dith, that was a tad disheartening; Trog would've been much easier. I still think random altars are great, though, and I don't mind getting a hopeless situation every now and then, since more often than not, you'll end up with something interesting even if it's not a spectacular synergy. Plus you have a chance to get a really early Jivya.

[Siegurt]

My suggestion is this:

There is no wrath for random altar gods, unless you select a new god before the old one is mollified, or you have gained the next pip of piety before abandonment.

This effectively makes taking a random altar, and getting a god you are unwilling to play cost a delay in getting your preferred god (unless you are willing to deal with their wrath)

If that isnt sufficient, maybe you can accrue a pretty "debt" that needs to be worked of before you can start gaining piety with the new god.