Make all stats useful for all chars

[bananaken]

72 (this number is the reverse of 27 you will notice)

Ok so it IS capped, but that's high enough that I suspect I'm not going to hit that mark anytime soon.. Up until now I've piled on bonuses to a single stat undeterred (this is regardless of the actual benefit, of course). It's something possibly.. shallow to enjoy? But whatever.

Back to the topic, I always end up looking for advice on how to distribute my stats because it's very confusing. I'm all for making it easier to determine what to pick when leveling up as long as the player is actually given more insight on what the tradeoffs are. If achieving this while retaining stats is too hard of a problem to solve, I could live without stats, though I'd be sad to see the BIG NUMBERS go (just look at that sexy 50 INT).

[archaeo]

Whatever floats your boat. Meanwhile, I am playing this game where I push buttons to move an @ across a 80x24 console screen, trying to turn alphanumeric symbols into full stops.

And I'm playing on a web-based interface with good-looking and informative tiles. Your point? Just because crawl has old-school aesthetics doesn't mean its gameplay needs to be stuck in the past as well.

What in this world makes you so entitled? This is a free, open source game sitting in a tiny niche -- it's not even part of any video game market --, and there is no absolute minimum at all. If you feel so strongly about Str/Dex/Int, send us a patch or devise a help screen. (That's exactly what I did when I wanted to help Crawl and couldn't code.)

Entitled? I just want this excellent game to be more excellent. What a thing to say!

What I really don't like about your argument is how you construe this idea that people have to "sift through inconsistent information" or "code dive" in order to play the game. Yes, Crawl's information politics is medieval, there are good reasons for this, and if you want to play optimally, you have to do something like this (perhaps relying on learndb suffices, I wouldn't know). If you just want to play the game, including winning it, you don't have to. The game makes it roughly clear what Str/Dex/Int will do, and an observant player can actually figure out on her own. (Note: I expect players to be either literate or observant, if you're neither, perhaps play Angry Birds instead?)

Thanks, dpeg, for implying I'm some kind of sub-literate moron just because I think the game could be more clear and doesn't need to have "medieval" "information politics," whatever the hell that means.

You are telling me it's bad. I don't follow. Would you like to lead us to enlightenment?

PS: I am getting too angry over this, so I'll just make a note about coarser stats and call it a day.

We've been telling you how it's bad, over and over, throughout the entire thread. How many different ways can I say "stats hide all their complexity for no reason, duplicate the better skills system for no reason, and force the players to make decisions they often can't fully understand"?

I would apologize for making you angry, but I literally can't tell what we've done to make that happen. Sorry for disagreeing with you, I guess?

[mps]

I'm going to go out on a limb and say "1980s game design."

[archaeo]

I'm going to go out on a limb and say "1980s game design."

For what little it's worth, I meant this only in the context of "information that is important for the player to know is only in the manual." While current AAA games take their tutorials way too far, there's something to be said for making sure that the best way to learn how to play the game is by playing the game, and not reading a .txt file hidden in the package.

I think crawl does a pretty good job offering in-game help, and it has only been doing a better and better job since I started playing. We should extend the same to the stat choice prompt if stats are retained.

[duvessa]

It's worth noting that Crawl's manual has also been filled with flagrantly wrong information for pretty much its entire history.

[all before]

PS: I am getting too angry over this, so I'll just make a note about coarser stats and call it a day.

I'm confused about dpeg's anger, but more confused because no one in this thread has advocated making stats coarser. If anything in the thread has broad consensus, it's that stats should be more transparent and distinct in their effects if they are going to exist.

[archaeo]

I'm confused about dpeg's anger, but more confused because no one in this thread has advocated making stats coarser. If anything in the thread has broad consensus, it's that stats should be more transparent and distinct in their effects if they are going to exist.

Actually, it was mentioned in the second post I made to the thread: "At one point, wasn't there a proposal to effectively cut stats in half while doubling the effect of each individual point? Losing a bit of granularity in exchange for clarity doesn't seem like a bad idea." It's not my idea, as I said, but I think it's one that devs have bandied about a bit in the past.

e: obviously I don't think this thread provides any evidence of consensus for making that change, though the devs neither want nor need us to have a consensus on anything at all to make changes.

[Siegurt]

We've been telling you how it's bad, over and over, throughout the entire thread. How many different ways can I say "stats hide all their complexity for no reason, duplicate the better skills system for no reason, and force the players to make decisions they often can't fully understand"?

So this is the point you and others have been making and the people who disagree with you are pretty much saying "No, stats don't hide all their complexity, at least not in any significant game-impacting way, and are not a duplicate of the skill system, but rather a compliment to it, and the players can fully understand the decisions (or at least as well as they can understand the decision about what skills to train)"

Point by point:

1. The effect of stats are at least as well documented as the effect of skills, the "obscureness" of the effects of strength and dex and int are no more "obscure" than the effects of say, stealth on stabbing, or how spellcasting relates to spell power or hunger level.

2. Stats and skills complement each-other, you get *most* of your stats at the beginning of the game, and the type of stats you get depends on your race and class, that makes for a large differentiation in the early game, and even a fair amount of differentation late into the game, depending on how you split this up, this *could* be replaced with some other system, but you'd have to create something entirely new, and there's no reason to suspect it would be any less complex or any less obscure. To be clear *there is no existing mechanic that provides this type of differentiation* Skills and aptitudes don't do it, mutations don't either (They are closer, but don't serve the same purpose), you could create something new to replace it, or decide that the provided differentiation wasn't worth the complexity, or even both, but suggesting that an existing mechanic could simply and directly take the place of stats is simply false. Skills are something that you invest in, Stats are something that you have (intrinsically or extrinsically), with no investment, the two are compliments of each-other. Mutations are closer (Since racial mutations are things you start with and have an immediate effect on gameplay), but there's no mutations presently that have the sort of granular effect that Stats do (You'd need more levels of mutation at a minimum) and there's simply not enough room in the current design of mutations to have something that has a significant effect early, which you can invest in over time to increase it's effect. There's room to design something either building on the existing system that would replace the role that stats play, but until you have something that's clearly and unqualifiedly better from a game design standpoint, there's no point in talking about it.

3. The information about what stats do, to at least the degree needed to make intelligent decisions that will help (or at least not hinder) your gameplay are in game, and can be accessed, if this information needs to be more prevelant, then that's something that can and perhaps should be addressed, however there's no reason to suspect that a replacement for stats would inherently be any better documented, or inherently less obscure, than the current system.

If I said "Some spells aren't optimally designed and are poorly documented or have strange mechanics, so let's just remove all spells and replace them with with mutations, except we'll need to create like a bunch of levels of mutations that replace the current function of higher and lower level spells, and you'll have to be able to find them in game, then invest XP into making them work, but it will be better because mutations work well, and spells don't always" I think I might get some disagreement there as well.

[archaeo]

1. The effect of stats are at least as well documented as the effect of skills, the "obscureness" of the effects of strength and dex and int are no more "obscure" than the effects of say, stealth on stabbing, or how spellcasting relates to spell power or hunger level.

Except that when I'm in the skill menu, I can hit ? and immediately receive information, though I think even this isn't enough (I'd like to see, for example, a menu that better communicated how you need increasing amounts of XP to get each additional level of a skill). When I get the option to raise stats, there is no such menu.

I can hit ? C, which will show me the pertinent part of the manual, but the cursory information therein a) doesn't discuss the myriad other functions of stats and b) makes no mention of the fact that you often want to choose a completely counterintuitive stat to gain the bonus you want. I'm sure that I could construct any number of examples in which it's better to take Str over Int, or Dex over Str.

but there's no mutations presently that have the sort of granular effect that Stats do (You'd need more levels of mutation at a minimum) and there's simply not enough room in the current design of mutations to have something that has a significant effect early, which you can invest in over time to increase it's effect. There's room to design something either building on the existing system that would replace the role that stats play, but until you have something that's clearly and unqualifiedly better from a game design standpoint, there's no point in talking about it.

Fair point, insofar as I'll cheerfully admit that I have neither the skill nor the inclination to do this. But why shouldn't I talk about it? I think the game would be better without stats, and I hope that, by saying so, a dev will agree and design a new set of mechanics to take their place. I'd be happy to help with such a design, but I doubt any of my ideas would be better than what the devs could come up with, if they agreed with my basic proposition that stats are useless.

If this is really such a problem, then we should just remove GDD altogether and require that anyone who wants to make a suggestion code it themselves and submit it to mantis.

As for mutations not being better than stats, even without having specific examples, it seems completely obvious to me that mutations, which are named and given a descriptive phrase in the % and A menus, respectively, are inherently clearer than stats, which are just three numbers. I don't think that granularity is necessary for something that you intrinsically have as a species, anyway. Indeed, I like the idea of flat mutations that differentiate a character's species, with all subsequent growth represented through the skill system.

If I said "Some spells aren't optimally designed and are poorly documented or have strange mechanics, so let's just remove all spells and replace them with with mutations, except we'll need to create like a bunch of levels of mutations that replace the current function of higher and lower level spells, and you'll have to be able to find them in game, then invest XP into making them work, but it will be better because mutations work well, and spells don't always" I think I might get some disagreement there as well.

Yes, you probably would. It's a good thing we're talking about stats and not this silly straw-man idea, isn't it?

I do understand and respect the fact that this entire thread is very much making perfection the enemy of the good. If I was the God-King of Crawl for a day, I would much rather remove hunger or the ID minigame than revamp stats. But I nevertheless think that the game would be better without them, even if nobody particularly wants to do the work to design a perfect replacement.

[Wahaha]

If you're going to argue for a system that replaces stats then why not just fix stats? If your complaint is that the information about what stats do is not clearly available then surely it is smarter to make it available than to remove stats.

it seems completely obvious to me that mutations, which are named and given a descriptive phrase in the % and A menus, respectively, are inherently clearer than stats, which are just three numbers

How is that different from stats if the three stats are given a description of their effects? "Stats are abstract numbers" is a bad complaint because no one older than 10 has difficulty with the concept of stats as numbers. The majority of RPGs use stats as numbers and everyone understands them just fine.

It's a good thing we're talking about stats and not this silly straw-man idea, isn't it?

It's the same logic. The point being that it's easier to fix stats than to make a new system that would hardly be better than something as simple as stats.

The one other thing in this thread that I agree with, aside from making stat effect info more accessible, is that stats could lose their niche effects like bonuses to shields and aux attacks. That hurts Chei though.

[archaeo]

If you're going to argue for a system that replaces stats then why not just fix stats? If your complaint is that the information about what stats do is not clearly available then surely it is smarter to make it available than to remove stats.

Mostly because I haven't heard any suggestions that retain the basic idea of stats while correcting its myriad problems. I can't think of a way, but I'm pretty dim.

How is that different from stats if the three stats are given a description of their effects? "Stats are abstract numbers" is a bad complaint because no one older than 10 has difficulty with the concept of stats as numbers. The majority of RPGs use stats as numbers and everyone understands them just fine.

Most RPGs also don't have just three stats that are so tangled up together that it's difficult to see which stat has the biggest effect on an individual mechanic. I don't have an issue with stats as a concept, I just think the crawl implementation is so hopelessly messed up that removing it wholesale is likely going to be easier than trying to bend it back into some kind of reasonable shape.

Even if you disagree with my core proposal, I think this thread has generated plenty of good evidence that the current implementation of stats needs work, as well as a wealth of good ideas for fixing the problems. Even if we can't agree on one thing, hopefully we've come up with enough material that the devs can do something productive with it.

[bananaken]

Kind of related I guess: I like how SPECIAL stats worked in the Fallout series. This is more the SPECIAL system than Fallout itself, but the primary stats are nicely divided so that it's enough to give a short description on each to let you know what they do and how they're used as a base for any derived stats.

[Sprucery]

When I get the option to raise stats, there is no such menu.

I hope that one concrete thing that results from this discussion is that such a menu is created.

[Sar]

I like how SPECIAL stats worked in the Fallout series.

SPECIAL is also notoriously imbalanced, with AGI reigning supreme (in 2D ones, where it determines AP), INT being close second (you like getting skill points, right?) and CHA being literally useless in the first game (okay, that was fixed in 2).

[bananaken]

I like how SPECIAL stats worked in the Fallout series.

SPECIAL is also notoriously imbalanced, with AGI reigning supreme (in 2D ones, where it determines AP), INT being close second (you like getting skill points, right?) and CHA being literally useless in the first game (okay, that was fixed in 2).

It actually depends. If your goal is just to win easily then yes, you're right. There's some things to keep in mind though:

• You can't just level up and put all points into AGI/INT. You do that on character creation and even when you max them out you're going to have to think about where to put the rest of those SPECIAL points, and these are meaningful choices.
• It's not optimal to max them over other stats depending on what your goal is. If you're speedrunning for example, perception is just as important in Fallout 2 (sometimes more!) than intelligence and agility.
• Other combat-oriented builds can also have just as much emphasis on AGI but for EN or other stats.

Also, the strength of a particular stat is linked to the game it's on. You said it yourself with AGI being the best in Fallout 1/2. In New Vegas I've played with 1 AGI and it was fine. You don't use VATS (don't need to), you reload slower, but you run just as fast and that's fine with me.

I feel "notoriously imbalanced" is very strongly worded when SPECIAL does its job and it allows for variation. I'm not saying TURN CRAWL INTO SPECIAL STATS, but maybe we could glean some of what they got right. Mainly, I like how these go from primary stats (i.e. SPECIAL) to derived stats (Speech, Small Guns, Barter, etc.) while describing this succinctly. I think a more worrying flaw would be somehow incorporating spoilery knowledge requirements of the older games (i.e. you open new dialogue paths by having Perception 7 but have no way of knowing this).

[Wahaha]

A more appropriate example of the SPECIAL stats thing might be Dungeons of Dredmor, because it is a roguelike. When I played that game, at first it was hard remembering what stats affects what substats. It might've been because all of the stats have silly names. It's the same thing as listing the effects of each stat in the stat's description. It's just a different way of presenting it. Unless it is possible to boost sub-stats directly, in which case it's different, but I don't think crawl wants to go that way.

[dpeg]

Even if you disagree with my core proposal, I think this thread has generated plenty of good evidence that the current implementation of stats needs work, as well as a wealth of good ideas for fixing the problems. Even if we can't agree on one thing, hopefully we've come up with enough material that the devs can do something productive with it.

I disagree with the bolded citation (I would agree with "the current implementation can be improved, which is a tautology, however).

Lemma: repetition does not create truth. This is about game design, not theology.
You are stating opinions, please stop pretending you have proved anything. (Likewise, I state opinions. Hopefully I didn't claim to have proved something here.)

Myself, what I have learned from this thread is that stats are actually in a decent place, as far as their relevance to gameplay is concerned. (Thanks to Sandman and Siegurt for their postings, they were very useful to me!) Moreover, I acknowledge three problems: (1) the effects could be trimmed down (Int is a good example), (2) there should be some feedback during the SDI level-up prompt, (3) the numbers could be scaled better.

Discussion on these would probably help to actually achieve something, use another thread on this, and perhaps different threads for different (aspects of) stats. I'd definitely object to stat removal: it'd be an actual gameplay loss; it'd cause real work; a replacement system has not even been devised (and we already have stats). If someone feels like it, dtsund's Crawl Light might be a good place to test statless Crawl.

[archaeo]

Never mind, after a cigarette, I decided that the post I just made was too heated and there's no use having a slap fight over this. Clearly my proposal was a bad idea; I'll look into writing something up for the help screen I suggested.

[WhiteShark]

Lemma: repetition does not create truth. This is about game design, not theology.

This is a rather silly thing to say. Nobody thinks this, theologians included.

edit: and to add something to the discussion, I like the idea of all the things stats affect being documented somewhere in game.

[dowan]

Yeah, no question that a ? when leveling up and choosing a stat would be a really good thing. Asking a player to make a choice, and giving him no opportunity to get information about his choice is obviously problematic.

What might help get us all to the same place on stats might be some example characters, with similar skills, but differing stats, to back up the claim that stats actually do create meaningful and valuable differentiation and character development options.

The most basic example is probably an OgHu, where putting all stat points into dex is probably the most common choice, but putting all points into int gives you a less agile(lower ev) character, but requiring less XP to be spent on spells to get them decently castable and powerful. By doing this, the XP you saved on your spells could be spent elsewhere, for example toward fighting, armour, a weapon skill, or even dodging(Although I can't imagine spending that XP on dodging would give better EV than just getting dex).

Unfortunately I can't give actual numbers here, as I don't have access to crawl ATM

[Sandman25]

What might help get us all to the same place on stats might be some example characters, with similar skills, but differing stats, to back up the claim that stats actually do create meaningful and valuable differentiation and character development options.

I have already provided such example in this thread. DEBe vs DEWz, 140% damage in melee, 45% spell power and +18 EV with identical skills.

[and into]

Never mind, after a cigarette, I decided that the post I just made was too heated and there's no use having a slap fight over this. Clearly my proposal was a bad idea; I'll look into writing something up for the help screen I suggested.

Well, I think better documentation and clarity, and trimming down the number of very marginal things that stats affect would all be a serious benefit. Coarser stats (similar to what happened with SH) might also be interesting. Maybe those things were "obvious," but it is still helpful for them to come up in GDD from time to time.

There are a lot of things about stats that I do not like, so I appreciated chequers and yourself and others bringing these things up.

As far as I can tell, many times that a new feature was added that involved the player's character and has some component of chance or damage, it was made to check dex, int, str, or some combination thereof. So the number of things—including very marginal things—that stats technically influence is very bloated. At the same time, some things that it makes sense for stats to affect (melee damage) are also weighted down by things that probably are not worth the added complexity. The strength vs. dexterity damage weighting comes to mind—which is something I don't pretend to understand, btw, and I seriously doubt my ignorance here has once impacted a character I played.

So, my two cents: If the marginal effects of S/D/I went away, and the remaining effects were clearer and better documented, I would be happy with stats in DCSS. What counts as "marginal" is up for discussion, but I think a good standard is, "If you basically never put a point into str, dex, or int in order to improve the effect in question, it is marginal and could be safely handled by something other than stats." For myself, this would leave something like the following:

Strength:
* Reduces penalties from armor encumbrance
* Boosts damage of melee and ranged attacks

Dexterity:
* EV
* Boosts accuracy of melee and ranged attacks
* Keep the stealth bonus, too, maybe?

Intelligence:
* Boosts spell power, success rate
* Reduces hunger costs from spells


Nearly everything else I can think of can just be moved over to skills and/or character level or simplified in some other way (I think it would be fine if SH depended entirely on the type of shield, enchantment, and skill).

I do think most of the above listed "major" effects of stats can actually influence character development and species differentiation in a non-trivial way, and thus are worth keeping (in a way that the other effects are not IMO and thus add needless complexity).

[Quazifuji]

So I just tried to look up a full list of the effects of each stat in the learndb, and couldn't find it. I'm not saying it's there, but I couldn't find a section labeled "attributes", "stats", or "strength". I looked on the Wiki, but we all know about the issues with that (the article on dexterity hasn't even been updated since 0.12).

So anyway, given my lack of reliable information on what stats actually do, here's my thought: in addition to some of the other suggestions in this thread, is there anything wrong with removing some of the more obscure effects of stats? Things like (according to the wiki, don't know if this is still in the game or not) dexterity helping you close doors more quietly.

Right now, stats, or at least strength and dexterity, seem like they're loaded with obscure side effects. And a lot of these don't really seem necessary. It feels like in the vast majority of the discussion in this thread, and the vast majority of discussions of stats elsewhere (whether from a game design standpoint or just giving advice on what stats to focus on for inexperienced players), the stat system is described roughly as such:

Strength: Helps with heavy armour and damage with large weapons.

Int: Helps with spells.

Dexterity: Helps with EV and Stabbing.

So what about just simplifying the stat system so that something along these lines could actually be an accurate description, and not a "well, here's all you really need to know" sort of thing? Are things like the effects of stats on aux attacks really necessary, let alone the volume of door closing?

2. Stats and skills complement each-other, you get *most* of your stats at the beginning of the game, and the type of stats you get depends on your race and class, that makes for a large differentiation in the early game, and even a fair amount of differentation late into the gameQuestion (this is not a suggestion, just a thought for the sake of discussion): what would happen if we kept starting stats in the game, but removed all choices involved in them (stat gains on level up become purely deterministic, and no items with stat mods chane)?

Question/Thought Experiment (not a suggestion, just for the sake of discussion): what would happen if we kept starting stats in the game, but removed all choices involved in them (stat gains on level up become purely deterministic, and no items with stat mods)? So stats still serve to differentiate between between races and classes, but that's it. How much would that hurt the game?

It seems to me like many of the arguments in favor of stats focus on their ability to distinguish between races and classes in a way that skills and mutations can't. Meanwhile, the arguments against stats seem to mostly revolve around the choice of what stat to raise every 3 levels being opaque and arguably meaningless. This means we're discussing two very different things, that happen to revolve around the same mechanic.

[quote="bananaken"I like how SPECIAL stats worked in the Fallout series[/quote]

If we're on the subject of games that successfully implement transparent stat systems, there's Pillars of Eternity. Every attribute's effect on combat can be expressed as a numerical bonus to two or three different values per stat point. Or, for a non-RPG example, but one that also only has str, dex, and int as its stats, there's Dota. I wouldn't want DCSS to adopt any of the details of Dota's stat system, of course, but it's an example of how you can have only three stats, each with very simple, transparent bonuses, where the stat system isn't completely redundant or meaningless.

[Sprucery]

While constriction is a special case wrt. melee fighting, in my opinion it's quite logical (as in "easy for the player to understand") that strength affects constriction damage as well. And currently the effect of strength is not insignificant. That said, I don't know if anyone raises Str as Na or Op just to increase constriction damage...

[dowan]

Here's how I think of the stats:

STR: Gives me better EV and spell success if I'm wearing heavy armor. #1 pick if my str is significantly less than the EVP of my body armor (or whatever it's called these days)
DEX: EV. #1 pick if I don't cast damaging spells, and I'm not wearing or plan to wear armor with EVP significantly higher than my STR.
INT: More spell power and spell success. #1 pick if I am using damaging spells, or if I am playing a species with bad spell apts (Mi, Tr, Og) and I want to cast spells anyway.

To me, stats are how I select the balance between my character's potential for armor heaviness, dodging ability, and spellcasting ability.

A HE with 8 str should really never wear plate armor, no matter what your armor skill. A Mi with 20 str should wear plate, even with 0 armor skill.

[all before]

STR: Gives me better EV and spell success if I'm wearing heavy armor. #1 pick if my str is significantly less than the EVP of my body armor (or whatever it's called these days)
DEX: EV. #1 pick if I don't cast damaging spells, and I'm not wearing or plan to wear armor with EVP significantly higher than my STR.
INT: More spell power and spell success. #1 pick if I am using damaging spells, or if I am playing a species with bad spell apts (Mi, Tr, Og) and I want to cast spells anyway.

This post shows how stats' primary effects are unintuitive, not just ancillary functions like constriction. Try explaining to a new player why investing in str will give better ev (not ac?) and spell success (not int?). Or why a MiBe should invest in dex.

Better documentation will certainly improve the situation, including which stats a race "naturally" develops, though I still wish (if selected stat growth is going to exist at all) that what stats did was more straightforward (ie. improving hitting things, not getting hit, and casting spells).

[Lasty]

This post shows how stats' primary effects are unintuitive, not just ancillary functions like constriction. Try explaining to a new player why investing in str will give better ev (not ac?) and spell success (not int?). Or why a MiBe should invest in dex.

"The most important aspect of strength is to allow you to wear heavy armor comfortably. If you are too weak, armour will interfere more with your ability to perform precise movements, impacting spellcasting, evasion, accuracy. On a character like MiBe, you will naturally develop as much strength as you'll ever need for armor, so you should improve another stat instead. Intelligence serves no purpose since you cannot cast spells, so you should improve your dexterity to get the associated evasion and stealth bonuses. You could increase strength anyway to get more damage bonuses, but the defensive bonus of dexterity tends to lead to better durability."

None of that seems hard to explain or confusing once explained to me.

[Sandman25]

Also if new players expect Str to increase AC, there is something wrong with those games where the players are coming from.

[Sandman25]

Discussion was useful.

http://git.develz.org/?p=crawl.git;a=co ... 336929e647

[archaeo]

None of that seems hard to explain or confusing once explained to me.

That's because your (finely written) explanation speaks very generally about a problem the game treats granularly and specifically. It isn't at all obvious to a casual player which species have what kind of stat growth, and because players deal with stats on a point-by-point basis, it's tempting to want to get the "most" out of that level-up bonus, a task that requires leaving the game window to search through a bunch of inconsistent, poorly maintained information on the wiki and learndb. e: one can also be the kind of literate, observant player dpeg prizes in his post above, of course, and exhaustively experiment with dex/str/int+ artefacts to see how things change, of course.

Ideally, hitting ? at the stat prompt would give you the following information:

* Your species' natural stat growth pattern and the levels at which you have an opportunity to choose a stat point

* The manual's description of each stat

* A succinct description of the interplay between encumbrance values, str, int, and dex. This could be as easy as saying "As your armour becomes heavier, your strength has a greater impact on your ability to dodge and cast spells," but I'm sure someone else could come up with a better description of what's going on.

Of course, all you can really give players is general advice for granular character-building things, since mechanics like encumbrance are pretty opaque in-game. Short of accepting a bunch of Sandman25 patches and showering the player with numbers and graphs though, I'm not sure how you fix that.

[Lasty]

It isn't at all obvious to a casual player which species have what kind of stat growth[.]

Ideally, hitting ? at the stat prompt would give you the following information:

* Your species' natural stat growth pattern and the levels at which you have an opportunity to choose a stat point

* The manual's description of each stat

* A succinct description of the interplay between encumbrance values, str, int, and dex. This could be as easy as saying "As your armour becomes heavier, your strength has a greater impact on your ability to dodge and cast spells," but I'm sure someone else could come up with a better description of what's going on.

I was under the impression that the point about getting help text from the stat gain prompt was well-received and generally approved, as was the concept of improving the in-game documentation. I'm in favor of both. I was only responding the point about the difficulty of explaining the concepts.

[archaeo]

I was under the impression that the point about getting help text from the stat gain prompt was well-received and generally approved, as was the concept of improving the in-game documentation. I'm in favor of both. I was only responding the point about the difficulty of explaining the concepts.

Well, but that's kind of the thing to me: it's often easy to explain stats in general (and you did a v. good job), but as this thread has shown, the specifics are bedeviling and practically impossible for players to learn without turning to outside sources. But I do take your point, and I didn't mean to sound hectoring or anything like that.

e: it's also worth noting that MPA is removing a lot of those bedeviling things, which is a distinct improvement.

[dowan]

I can totally get behind removing those weird random effects that were seemingly randomly based on some stats. It's not like anyone in the history of crawl has improved dex because they were concerned about squeaky doors and blade traps, or improved int to improve sif's channeling success chance (What?!?)

And a ? with exactly what archaeo said above when choosing a stat to increase seems like a really good idea.

[partial]

Stat problems:
1. A complex mechanic that exists.
2. Misleadingly prominent on the main display

Some would disagree that complex mechanics are negative thing by nature. In my eyes they are, fundamentally. The game should be as easy to understand as possible, but no easier. Mechanics that don't appropriately add depth for how complex they are should be culled, any complex mechanic should only survive if it's worthwhile. To me this is universal - every mechanic should be considered for the chopping block and every mechanic should have a clearly defined purpose. I don't know if this applies to stats or not, I'm undecided, but it's a clear negative in that it's a complex thing in the game and as such needs to have a defined purpose or be removed. Having to justify removal of anything with this same argument is kind of tiring. Do you really think "a set of numbers that adjusts a grab-bag of character abilities in small ways" is as fundamental to crawl as moving?

How are stats complicated?

[njvack]

How are stats complicated?

... they do a few main (pretty comprehensible) things, and then a bunch of little (kinda random) things. The commit logs from the last few days show a bunch of little things being removed.

Mod edit: split a derail out to CYC.

[WingedEspeon]

I think stats are at a good place as they are. However, I would be in favour of removing the stat boosts from the backgrounds in a way they isn't a straight up (heavy) nerf to the player so that backgrounds are just starting equipment. The dex to EV stepdowns should also be tuned down IMO-my HeGl work hard to get 44 dex without Chei!

[goldstarbrother]

Per the title of the thread: I think there should be a concept of "critical hit" for melee and ranged attacks, and your chance to land one is affected by int. A critical hit would be something like a 2x multiplier on damage.

[Mulzaro]

I think stats are at a good place as they are. However, I would be in favour of removing the stat boosts from the backgrounds in a way they isn't a straight up (heavy) nerf to the player so that backgrounds are just starting equipment.

Please no. Backgrounds should affect playstyle more, not less. When you hit midgame, the effect of your background becomes quite obsolete already unless you are a Be or something.

[Pollen_Golem]

* Int affects evocation success/strength (problems: ???, changes Ar/Nem radically)

Please don't. One of the fun perks of evocation is that it is attribute-independent.

[Quazifuji]

Per the title of the thread: I think there should be a concept of "critical hit" for melee and ranged attacks, and your chance to land one is affected by int. A critical hit would be something like a 2x multiplier on damage.

Adding a weapon damage boost to int could be interesting. Right now the defensive components of dex and str tend to be considered much more important than the offensive components of them, so letting characters who use weapons as their main source of offense having the choice between getting more defense by raising dex or str or getting offense from int could be interesting. But I see a few problems:

1. Int giving crit, or any other form of weapon offense, is somewhat unintuitive. Dex or similar stats are much more commonly associated with crit in RPGs, and it's weird for int to give a bigger damage boost to weapons than str or dex.

2. Int is already very strong for spellcasters, and spellcasters still melee sometimes. Adding crit makes it even stronger. If you make int very strong for weapons, you risk making it a no-brainer for anyone who does any spellcasting at all. If you make crit too weak to motivate you to take it if you're not a heavy spellcaster, then it doesn't really create any interesting decisions, just makes int-boosting equipment a tiny bit better for Trog-worshipers.

3. Crit is a pretty uninteresting mechanic unless you add other interactions with it - it's just an unreliable damage boost. It's interesting in games with a lot of emphasis on theorycrafting, but DCSS tends to specifically avoid promoting theorycrafting. Crit can also work well when other mechanics interact with it (Path of Exile's a good example), but doing that would turn crit into a major mechanic and not just an excuse to reward Trog-worshipers for getting int bonuses, which is a big can of worms to open up.

[Sandman25]

Str increases weapon damage significantly.

[Lasty]

Please no. Backgrounds should affect playstyle more, not less. When you hit midgame, the effect of your background becomes quite obsolete already unless you are a Be or something.

Backgrounds being purely starting package is intended.