Rework Character Generation

[ifreund]

Issues with the current system

The current point-based character creation system encourages min-maxing by
taking negative traits or difficult scenarios for the purpose of getting more
points to build the strongest possible character. This is undesirable because
starting traits are largely flavor-oriented and should be used as an opportunity
for role play as opposed to min-maxing. The same is true for starting scenarios,
and perhaps to a somewhat lesser extend stats and skills.

Proposed solution

I propose hiding the point values of scenarios, professions, and traits from the
player. These point values would still be tracked internally to give a general
impression of how easy/hard the player's current selection would make the game.
Also, these starting options should be made completely freeform, albeit with a
clear message stating "This character is a god among mortals" or "This character
probably won't survive 5 minutes" according to the magnitude of the
positive/negative selections made.

As of yet I am undecided as to whether this same freeform system should be
applied to stats and skills as well. There is a good argument to me made that
they are no different than the rest of character generation, and that players
should be free to set them as high as they wish. However, I'm of the mind that
it would be realistic to impose an upper limit on total player stats/skills. I
don't think that allowing players to start with 20 points in every stat fits
within this project's focus on realism. Selecting less than this cap would still
be possible, and feedback would be given in terms of textual messages with the
underlying points hidden.

To sum up, the proposed changes are:

• Hide the point system used during character generation
• Give textual feedback regarding how difficult playing the character will be
• Make starting scenarios, professions, and traits totally freeform
• Cap stats at the value used by the current multi-pool system (38 total)
• Cap skills at some reasonable limit

[I-am-Erk]

Kevin's been suggesting this for as long as I've been a contributor at least and I continue to think it's a good idea.

We may want to have some rough calculation of "skill survival value", where certain skills and stats may contribute more to difficulty calculations than others, to keep the report accurate. High intelligence is not equally valued to high strength eg.

[ifreund]

We may want to have some rough calculation of "skill survival value", where certain skills and stats may contribute more to difficulty calculations than others, to keep the report accurate. High intelligence is not equally valued to high strength eg.

Those are my thoughts as well. My plan would be to keep the current point values internally as an initial approximation to start with, and then tweak things from there.

[I-am-Erk]

Actually it'd be more trouble but quite interesting to estimate a few different survival qualities. Like "combat", "crafting", and "character interactions".

[Chambs-15]

With this system, skills and stats could be something that only your traits/profession gives, with no way to add more during character creation without picking related traits/professions.
This would give a soft cap for how much each skill/stat can be on character creation and would make whatever skills the survivor be based on it's "lore".
Another option would be giving a very small amount of points to allocate among skills/stats, something like 3 or 4.

[Night-Pryanik]

I'm vehemently opposed to the proposed ideas. I see nothing wrong in min-maxing. Why are you forbidding me in creating the character the way I like it? There's literally nothing stopping anyone from creating a complete roleplay character with no regards to any min-maxing by using freeform points pool.
The only way I'd be fine with proposed ideas is either creating a totally new points pool in addition to 3 existing ones (named something like "point-less pool") or reworking an existing freeform points pool.

[mrkybe]

Since this doesn't actually change how free-form works, could this 'difficulty estimation' just be added to the free-form picker?

[I-am-Erk]

I'm vehemently opposed to the proposed ideas. I see nothing wrong in min-maxing. Why are you forbidding me in creating the character the way I like it?

Who's stopping you from doing anything?

[Night-Pryanik]

I didn't mean "right now". I meant "when/if proposed changes will be implemented".

[KorGgenT]

the proposed implementation isn't going to stop you from building the same characters. they'll just shift how the points are presented to the player as difficulty levels.

[Night-Pryanik]

And yet I prefer to see exact numbers and not vague textual descriptions.

[I-am-Erk]

I don't think a preference to see numbers, especially when the numbers are arbitrarily assigned and not exact at all, is a reason to avoid reworking something like this. There's a solid argument for keeping numbers displayed, since they'd be something you could look up anyway, but it's pretty trivial overall. Is there anything else about the current system you're concerned enough about keeping to be so "vehement"?

[sayke]

The current system's points are an approximation but still extremely useful to get a sense for the rough magnitudes of impact traits are going to have in-game. They're also a familiar idiom for players and set up a relatively consistent default set of starting conditions.

The existing system works quite well, IMHO. Is "min-maxing" really such a huge problem that the points system needs to be overhauled? Saying "starting traits are largely flavor-oriented and should be used as an opportunity for role play as opposed to min-maxing" (emphasis added) suggests that you don't see min-maxing as fun, but I think many players do see it as a fun part of the game. I do. Has there been any significant player feedback complaining about having too much opportunity for min-maxing?

I play with several traits that are a fun (and not very min-maxed) combination: "PACKMULE", "SLOWREADER", "EAGLEEYED", "ADDICTIVE", "SQUEAMISH", "GLASSJAW", and "SLOWHEALER". Storage management is annoying hence packmule, slow reader and slow healer bring things closer to realistic settings, glassjaw makes headshots hurt more, and addictive/squeamish make those mechanics more intense and harder to work though. I use the default evacuee scenario.

I suspect that, without the points system incentivizing them, very few people (and basically no new people) would play the harder scenarios. If you want to encourage people to take "flavor-oriented" starting traits that have significant role-playing potential without trying to min-max, you could just tweak the point costs for them... Or you could just turn on the freeform points pool. That functionality already exists, but the current incentives work.

I appreciate the thought but in a situation where we have lots of serious issues that need lots of serious work, this seems like the perfect becoming the enemy of the good to me.

[Night-Pryanik]

Is there anything else about the current system you're concerned enough about keeping to be so "vehement"?

"You're-playing-it-wrong" rationale behind proposed changes is making me so vehement.

Min-max is bad, you're saying? Well, and I'm saying it's not. Let people create the strongest possible character if they want to. Those who want to create a roleplay character could use existing freeform points pool.

[I-am-Erk]

I'm not sure that 'min maxing is a bad way to play' is the thing to take home from a proposed system change. The trouble with systems built around minmaxing is that they are less accessible to new players, it's not a problem that affects people that know the system intimately. A system that focuses more clearly on what your character is going to find easy or hard makes character creation easier to get into, and provides a sliding scale of scenario type challenges for people. EG: "my third day playing a character with Impossible Combat Difficulty!". I think it's self-evident that this would be a positive change overall.

Absolutely no character you could presently create is more difficult to create under a system that judges character strength/weakness rather than point pools. However, a system like this would present more useful feedback to the player and, ultimately, encourage much more complex characters.

[Kelenius]

I'm not sure that 'min maxing is a bad way to play' is the thing to take home from a proposed system change.

??????

The current point-based character creation system encourages min-maxing by
taking negative traits or difficult scenarios for the purpose of getting more
points to build the strongest possible character. This is undesirable because
starting traits are largely flavor-oriented and should be used as an opportunity
for role play as opposed to min-maxing.

[I-am-Erk]

I didn't write that.

Edit: sent early. I didn't write that, but my point is that there's more to it than stopping minmaxing.

[kevingranade]

My take on this isn't that min-maxing is a bad way to go about assembling a challenging and interesting character to play, my issue with the current system and all variations of it is it's ludicrously imbalanced, and trying to tune it is a never ending tarpit of bikeshedding. It's not min-maxing that's wrong, min-maxing is a symptom of the system as a whole being broken.

The concept of "various traits are worth different numbers of points, either positive or negative, and adding those numbers together gives you a number that tells you how hard challenging that character is to play" is simply incoherent outside of an extremely constrained system which we don't have. All the skills are slightly different in value, all the stats are VERY different in value, and traits have shot-in-the-dark values, and to complicate things further, the actual values of all of these things actually depend on what other traits you have, so simply summing them together gives you an almost completely useless number.

The point of the proposed system is to flip things around such that you pick whatever traits you want (which includes stats and starting skills), and instead of having fixed numbers assigned to everything, the game does whatever it can to give you feedback about how survivable that makes your player.

One of the most obvious things is it immediately eliminates things like dump stats and dump traits. Take all the negative social traits? Go ahead, but that won't make your combat feedback budge since those traits are irrelevant for combat. If you want to make a character that is challenging to play, you're going to have to actually give them drawbacks in various categories instead of shuffling all your "points" from areas you don't care about to areas you do.

[ghost]

Recently, there was a Reddit post, where people were criticising the point cost of various professions, the crazy cat lady came up as being "ludicrously high cost" I think it costs 4 points or something.
The crazy cat lady comes with 1 points in skills, and 20 cats as starting pets.

How do you quantify the point cost of 20 cats? they dont particularly do much, but they are flavour, maybe you can kill them and eat them for some early food but thats a bit weird, maybe when NPCs can trade livestock and pets, they will have more value...

But still, people proceeded to debate and argue about what the "best cost" was for various professions such as that, and it can be entirely subjective - a lot of the professions lump skills together that would cost more to take individually, this is a quagmire of balancing as Kevin says.

Also , I've got the PR coming so that professions can start with vehicles too, what point cost would you give to a hatchback versus a helicopter?

I'm not particularly attached or opposed to this change, I just thought I'd add that as a way that points costs can be a rabbit hole of opinion about balance, I can see some benefits arising from doing away with this, but sure I can see the point of people who appreciate seeing numbers and the meta of working within constraints of the games systems.

[ashGlaw]

The point of the proposed system is to flip things around such that you pick whatever traits you want (which includes stats and starting skills), and instead of having fixed numbers assigned to everything, the game does whatever it can to give you feedback about how survivable that makes your player.

Thanks, Keven, now I get what you’re going for. Great clarify

[I-am-Erk]

So revisiting the primary topic: I think we should basically refine point cost to an array of different point values and an option to either add/subtract from the points total, or to multiply the points total. I'd suggest the categories be "combat", "toughness", "wilderness", "crafting", and "social".

Some things might impact both of these, like Dodge skill or points in strength would impact both Combat and Toughness. Some aspects would be a multiplier, like taking a profession that starts with a weapon and a high level in the melee skill associated with that weapon, or something like Quick traits that flat speed you up in all ways.

[shadowsapex]

I'd just like to say that without point cost minmaxing, plenty of people would still play challenge scenarios. it's ridiculous to think that almost no one would play challenge scenarios without points. people love challenging themselves in games

[Night-Pryanik]

min-maxing is a symptom of the system as a whole being broken.

So you're saying that literally TONS of games and points-based gaming rules systems (with inherent min-maxing) are broken?

How do you quantify the point cost of 20 cats?
what point cost would you give to a hatchback versus a helicopter?

Somehow? Somehow as we quantified all other things like starting gear, mutations, cbms etc? If people blaming profession in having a "ludicrously high cost", then it's a reason to balance the profession, not to hide points altogether.

In general I find all this points hiding idea pointless. We're not removing points system, we're just hiding it, right? We're hiding points from the player, but we still use the same points system to calculate so-called "combat difficulty" or how would you call it. And now we're giving player some vague textual feedback instead of exact numbers, both based on the same points system. What's the point in this? Hiding for the sake of hiding?

Okay, for the purpose of discussion let's presume the whole points-less character generation system is better than points-based one. Then let's enhance the existing freeform mode. Remove all indications of any points from this mode, and it's good. Make all the changes to this mode you need. Even make it a default mode, if you think that

The trouble with systems built around minmaxing is that they are less accessible to new players.

But please leave old points-based modes for those who, slightly rephrasing, "appreciate seeing numbers and [love] working within constraints of the games systems."

[Aluminumfoil]

What about having a handful of fully fleshed out, ready-to-play benchmark characters at each "competency" level, which the player could then modify in a freeform manner to help players who are interested in such things keep their character at the intended level of challenge?

For example:

Powerful Characters: these archetypes are drawn from the top 10% of survivors. They have the range and depth of abilities that make them likely to survive and prosper. Characters based on them have a good chance at coming out on top of challenge scenarios and are likely to have a relatively easy time in non-challenge scenarios - although they aren't without their weaknesses.

• the Athletic Mechanic
• the Army Engineer
• the Combat Medic

Competent Characters: these archetypes represent the bulk of characters competent enough to have survived the initial events of the Cataclysm. They are effective at the sort of tasks expected of the average competent adult, and have one or two stand-out abilities that will help them survive. Characters based on them have a good chance of surviving non-challenge scenarios, but may struggle in challenge scenarios.

• the Brawler
• the Nurse
• the Police Officer

Challenge Characters: these archetypes represent characters who have only gotten this far by pure luck or the kindness (or nefarious designs) of strangers. Although they were valuable members of society while society existed, they have no great strengths for surviving the apocalypse. Characters based on them can eventually overcome their weaknesses if they're lucky and careful, but are unlikely to survive challenge scenarios.

• the Unathletic Student
• the Stay-at-home Mom/Dad
• the Elementary Educator

Each archetype character having a full set of stats, advantages, disadvantages, and skills reflecting the totality of their character, down to hobbies and near universal abilities like 2-point competence in cooking and driving, etc.

[CrankyUnicorn]

I agree specially in the part of the TRAITS limit. I see a lot of abuse on converting "Negative Traits" into overpowered characters in "Let's Plays". I think that you should not get points from "Negative Traits" at all since in a way it breaks the game a bit. About Positive Traits hum it should be more limited also. I like the pool system but I would suggest randomize the positive traits available. Example divide the "Positive Traits" into Tiers and allow x of each Tier making it impossible to pick only the finest in a character. I think for player that want versatility just let the "Free form".

[kevingranade]

This issue has been mentioned on Cataclysm: Dark Days Ahead. There might be relevant details there:

https://discourse.cataclysmdda.org/t/profession-point-costs/23788/2

[hexagonrecursion]

kevingranade commented on Feb 10
My take on this isn't that min-maxing is a bad way to go about assembling a challenging and interesting character to play, my issue with the current system and all variations of it is it's ludicrously imbalanced, and trying to tune it is a never ending tarpit of bikeshedding. It's not min-maxing that's wrong, min-maxing is a symptom of the system as a whole being broken.

The concept of "various traits are worth different numbers of points, either positive or negative, and adding those numbers together gives you a number that tells you how hard challenging that character is to play" is simply incoherent outside of an extremely constrained system which we don't have. All the skills are slightly different in value, all the stats are VERY different in value, and traits have shot-in-the-dark values, and to complicate things further, the actual values of all of these things actually depend on what other traits you have, so simply summing them together gives you an almost completely useless number.

The point of the proposed system is to flip things around such that you pick whatever traits you want (which includes stats and starting skills), and instead of having fixed numbers assigned to everything, the game does whatever it can to give you feedback about how survivable that makes your player.

One of the most obvious things is it immediately eliminates things like dump stats and dump traits. Take all the negative social traits? Go ahead, but that won't make your combat feedback budge since those traits are irrelevant for combat. If you want to make a character that is challenging to play, you're going to have to actually give them drawbacks in various categories instead of shuffling all your "points" from areas you don't care about to areas you do.

I think I get it. You are proposing a paradigm shift. Right now char gen is a game mechanic, one can play it. One can take negative traits and dump stats to get points to buy other stats and positive traits. You are proposing to demote it to a tunable difficulty setting. There are many mechanics in cdda that appeal to players who like making trade-offs and min-maxing, but your opinion is that char gen should not be one of them.

[TobalJackson]

What I can say about this after having read this entire thread is this:

Every character I've made in this game has been with through the "default" system (multiple pools) since I've always assumed this as the intended (suggested, balanced, etc) way to create your character.

Implicit in my choosing this is the assumption that this choice balances the games' difficulty (tuned through scenario and profession).

Right now my characters almost always start with all 12 points of negative and positive traits (far/near sighted, lactose intolerance, truth teller, trigger happy && quick, light step, parkour expert, etc) simply because I want these positive traits and have take on the negative ones as the cost.

I don't particularly want my character to be near sighted and far sighted, or any of the other negative traits, but I take these options because the point cost of the positive traits forces me to.

I know that I could cheat this and change some settings around to create a godly character, but I'm operating under the assumption that the devs have enforced these pluses and minuses around a "balanced" difficulty playthrough.

It seems as though the changes proposed will do away with the default mode of enforcing the points constraint, and instead give the player feedback that will indicate whether or not their chosen build will make for an easy, difficult, challenging, nightmarish, etc playthrough.

If this change were implemented such that the "baseline" difficulty I'm accustomed to playing were exactly reimplemented with the description conveying such, then despite having the "under-the-hood" point assignment restrictions lifted or relaxed, I could still build my character in pretty much the exact same way such that the advertised difficulty matched what is currently enforced under the default multiple-pool character creation.

What I see as the potential benefit of this change is in allowing for the difficulty to be tuned across the character creation pages (scenario, profession, stats, traits, skills) rather than the point pools being more or less exclusive to themselves.

It's my understanding that the other Points options (single pool, freeform) allow this to some degree (I've never tried them) but that this new system becoming default would in a sense "softly" enforce the multiple pools restriction of character traits by indicating resulting difficulty with the freedom of "freeform"'s mechanic of assigning player pluses and minuses.

If I'm seeing the implementation correctly, I can see myself having the freedom to create much more interesting characters while staying within the bounds of what is still a "challenging" playthrough.

I think the perceived problem with the default character creation is essentially how I personally go about creating my character; I stay with the default since I'm under the assumption that the character creation pages are each balanced with each other and within themselves. Due to this, I tend to min-max while staying constrained by the points which results largely in the same character every time (with respect to traits and stats), varying mostly in starting scenario and profession.

If I'm envisioning this proposed change correctly, I'd likely be creating wildly different characters, going off of the interplay of difficulty modifications resulting from changes to each of the character creation tabs rather than constraining my character to having only different scenario and profession for each playthrough.

I think the one point brought up by those against this change is valid: namely how to correctly gauge the impact of each character scenario/profession/stat/trait/skill/items spawned with/starting location such that the displayed difficulty is more or less indicative of true game difficulty, and of a "baseline" difficulty.

On the other hand, this might be simpler than anyone's assuming.

In either case if that problem could be more or less solved, I'm for this proposed change since I think I'd be making more interesting characters.

[Balgamtr]

I don't think character generation has anything to do with min-maxing at all. Most of the players ( at least from what I see) roleplaying using default point system/debug/freeform anyways.

From my point of view seeing actual points in stats serves as a good measurement of character instead of vague flavor text. You can start any roguelike game with a stat system and within 1-2 hours you can pretty much understand which numbers mean what and how well your character does a giving action. I see nothing wrong with that. If you're assuming random player with no gaming history suddenly gonna start playing cataclysm and get confused with the current point system that's making changes for %1.

Also for experienced players, I don't even think character selection has anything to do with balance issues and/or min-maxing. You can start with default 8/8/8/8 character without and traits or skills, find a working vehicle, find a soldier site, run over soldiers get a weapon, crash a lab door, kill turret at night, get unlimited ammo. This alone trivializes %95 of content if you're not roleplaying. Takes 1 or 2 days max.

[anothersimulacrum]

From my point of view seeing actual points in stats serves as a good measurement of character instead of vague flavor text.

I strongly disagree with this, because that flavor text would be provided by a much better system than the incredibly arbitrary systems that points are.

One point in driving, and one point in dodging are hugely, incredibly different in value to your character, but they still cost the same. If that flavor text is provided by the same system, yes it's worthless. But as has been discussed in this thread, the idea is to add some better/different systems for evaluating these.

I don't even think character selection has anything to do with balance issues and/or min-maxing. You can start with default 8/8/8/8 character without and traits or skills, find a working vehicle, find a soldier site, run over soldiers get a weapon, crash a lab door, kill turret at night, get unlimited ammo. This alone trivializes %95 of content if you're not roleplaying. Takes 1 or 2 days max.

I don't think we should be catering to people trying not to have fun :P
Sure, there may be a specific way to play the game that makes it so you can totally ignore your starting position, but for most of the ways to play the game, it's simply not true that your start has no impact. And even if that is your plan, unless you've chosen a nice safe start, it does matter what your starting character is.

I don't think character generation has anything to do with min-maxing at all. Most of the players ( at least from what I see) roleplaying using default point system/debug/freeform anyways.

I've had a different experience.

[Hulkninja]

So, will the old points based system be removed entirely, or will this new system replace it as the default one, without removing it?

I'd love to at least be able to choose. Obviously a significant minority of the players don't want to be forced to play the game with this new system. I'm not 100% sure where I stand on this, but if I try it out and don't enjoy it as much as the old one, I'd at least want to have the option to revert, without also having to downgrade to an earlier release...

And what exactly do you (@anothersimulacrum) mean by:

I don't think we should be catering to people trying not to have fun :P

Fun is entirely subjective. Nobody plays the game trying not to have fun. Surely you don't mean to imply that if someone is enjoying the game in a way that wouldn't be enjoyable to you, then they're doing it wrong?

[anothersimulacrum]

That was mostly-jokey, but what I meant was that if people found that playing that way was making the game less fun (e.g. "95% of the content being trivialized"), then they shouldn't do it.

The old points system would be removed entirely, because it just doesn't really do anything we want from a point system. It does a bad job of providing feedback, it does a bad job of creating interesting characters, it just incentivizes creating characters with the same-ish set of traits again and again because that's how you get the points you want to do other things with the character.

[dissociativity]

to be fair, a difficulty option right at the start that is always indicated somewhere in the status bar (to name and shame) that decides how many points you're allotted could be an option, or simply making strength and combat based positive traits, skills and stats cost more earlier than non-combat related.

Actually it just occured to me, could traits start requiring stats both in the positive and negative?
you can only have fragile if your strength is below 9, so you can only benefit from the points if you're already limited there.
You cannot have parkour expert with dexterity below 10? no deft below 8 dex? no fast learner below int 10? no near/far sighted with perception above 10?, no truth teller with perception and/or int above 10?

These are obviously arbitrary numbers but you get the idea, people want certain traits, including negative traits, if there's limitations on stats to reach said traits, they'd think again.
If intelligence unlocks traits that str based chars would take beforehand, people would be less likely to use it as a dump stat and it's inherently more useful.

Perhaps this should also be expanded to some professions, no doctor is going to have low int, soldiers are required to pass certain standards to be soldiers in the first place, ect, ect.

[stale]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions. Please do not 'bump' or comment on this issue unless you are actively working on it. Stale issues, and stale issues that are closed are still considered.

[kevingranade]

This issue has been mentioned on Cataclysm: Dark Days Ahead. There might be relevant details there:

https://discourse.cataclysmdda.org/t/what-are-proficiencies-trying-to-model/24808/8

[glimpsebeyond]

As a new player I feel that both pool options encourage min/maxing to an uncomfortable level. This is a pretty difficult game in some ways and I keep dying, so naturally I want to build the strongest character possible to have a better chance at making it through the early game for once. When I found out stats/traits are more difficult to get through gameplay than skills I stopped taking skills. Multi-pool forces me to take skills so I stopped using it. Seeing that negative traits give bonus points makes me want to take as many as I can manage for no reason other than more power. 8 point starting scenarios make the others feel worthless, so I'm compelled to take Really Bad Day and reroll 30 times until I luckily survive once.

Inspired by this thread I created my most recent character in freeform, took everything I would expect an average human to have, then added some additional run-specific attributes fitting a bionic athlete. There was no temptation to pump attributes that didn't make sense in context despite having as many points as I want. I got to make the character I wanted without having to jump through a bunch of hoops. So far this is the most fun I've had yet.

[ghost]

I've been thinking about this lately and I believe removing the trait limit has hurt the game's replayability. I can only speak for myself, but I find many of my characters are very similar even when I feel like I'm trying new things.

Since removing it was my idea, I feel obligated to point out the option of reinstating a limit for how many traits a character can have. Previously, it was 3 good and 3 bad. A combined total might make more sense.

Pretty sure this will be unpopular.

[Svatopluk1]

After reading through the thread:

• The current chargen min-maxing "minigame" is very shallow. It does not take a genius to figure out that e.g. the Quick trait is almost always a powerful pick, while things like Weak Stomach or Truth Teller give you points with little downside. The fun of exploiting an easily-abusable points-based system is not a serious argument for keeping it.
• Calculating a "difficulty value" would be incredibly difficult (the value of a skill/trait can strongly depend on all other picks). Replacing a simple-but-imbalanced system with a complicated one for the sake of better balance is not a serious argument for changing it when the more complicated system would be extremely difficult to balance.
• Players regularly settle in Evil biomes in Dwarf Fortress, and land on Ice Sheets in Rimworld. These difficult options do not give the player any extra points for picking them: people just do it for the challenge. In fact, getting extra points for a difficult starting location could make "hard" starts easier, which would defeat the whole point.

And so, a modest proposal:

• Make Freeform the default.
• During Freeform character creation, hide all point values, but:
• Display textural feedback regarding stats (e.g. "Your strength is above average").

If a freeform character gets the feedback "You have the strength of an Olympic weightlifter, the dexterity of a world-class musician, the intellect of a brilliant scientist, and the perception of an elite sniper", it will be clear that this is not a balanced character. If somebody picks 8 positive traits and 0 negative ones, it is already abundantly clear that they are not making a balanced character.

So that's it. This is a simple proposal that is easy to implement, discourages the abuse of the current easily-exploitable system, and encourages players to make the game as difficult as they want it to be.

There is no real need for multiple categories of nightmarishly difficult-to-get-right "character survival quality calculations". In fact, making a really unusual character and then finding out firsthand how easy or hard it is to survive can be fun. More importantly, a difficulty estimate could even discourage new players: getting told that "Your character is very strong and will find it easy to survive" and then quickly dying could be demoralizing. Promoting Freeform is a good idea, but a difficulty estimate has more downsides than upsides.

[I-am-Erk]

@Svatopluk1 That is, in essence, what is being suggested here. However, 'stats' are not the thing we'd give feedback on, but a set of gameplay elements. So, something like "melee combat: extremely strong" or "melee combat - very easy" depending on the exact formatting and phrasing we choose.

These numbers could be calculated, or could be based on JSON data fields, or could be based on JSON data fields that are normalized by CI testing to ensure we have a roughly accurate number. Splitting the current "point cost" fields into a number of json areas already makes it easier to normalize the data because we can ensure that being quick and being pretty do not impact the estimate in the same area, which is already better than what we have. It would also be easier if those json fields didnt accept numbers, but rather around four text inputs, eg. "difficulty_impact": [ "melee": "easy", "social": "hard", "wilderness": "very_easy" ] to avoid people thinking this is a highly granular system.

[Svatopluk1]

I had pointed out some issues with both the current system and the proposal, and suggested a slightly modified version that avoids those and would be easier to implement.

However, 'stats' are not the thing we'd give feedback on, but a set of gameplay elements.

Obviously - you just reiterated the proposal. Do you want me to reiterate the main arguments against it?

1. Calculating a reasonably accurate "difficulty value" would be really hard because tons of things are strongly context-dependent: a Cutting Weapons skill of 6 is going to be extremely useful if you start with a katana surrounded by zombies, but much less impactful (heh) if you start with a police baton. Yet in both cases a high Cutting Weapons skill would nominally make "melee" similarly "easy". This is a critical point: the valuation of many scenarios, skills, traits and stats would depend on each other. You could either have special rules for every potential combo (gl), or have a simpler linear system that is wildly misleading (and so not much better than the current wildly imbalanced point-based system).

2. One of the main points of the original proposal is to discourage minmaxing, and encourage building interesting characters instead. A "difficulty estimate" would encourage a sort of minmaxing where players would try to make characters as powerful as possible while still keeping the difficulty estimate at a certain threashold (e.g. "hard" in all gameplay element categories), by hunting for undervalued combos (see point 1). So a change to a multi-axis difficulty-meter would pile on complexity without addressing the fundamental issue with the current point-based system.

Not getting textual feedback on stats during character creation is a major oversight. Does a value of 8 correspond to an average person, or is it 10? Is 15 godlike, or merely "really good"? There is a similar issue for new players regarding skills (does "6" correspond to a decent amateur, or an expert in the field?). If these were addressed, then players would already have a very good idea about far on the Average Joe -> Demigod scale they are. At that point, a multi-axis difficulty estimator would add very little but the aforementioned downsides.

[LyranRenegade]

hey, I must admit i didn't read it all but seems an interesting discussion.

I just had to think about ADOM's character creation.. where you answer a bunch of worldly questions and get rolled a character w stats
of course you could min max that too, but that was very hard for some outcomes were random.

So, that could be an entirely optional game-start, to keep the rest intact, just this one let's you rollplay for the first second, with still some decisions to be made which suggest a more or less survivable character.

[NetSysFire]

Closing since @bombasticSlacks addressed this with #64697 recently.