Stat balance (STR/INT/DEX)


Although the central place for design discussion is ##crawl-dev on freenode, some may find it helpful to discuss requests and suggestions here first.

Do you think that the 3 core stats (STR/INT/DEX) are balanced against each other?

Yes, everything is fine.
3
14%
No, I think this should be discussed.
19
86%
 
Total votes : 22
User avatar

Eringya's Employee

Posts: 1782

Joined: Friday, 7th October 2011, 19:24

Location: Athens, Greece

Post Saturday, 5th May 2012, 19:44

Stat balance (STR/INT/DEX)

I took the liberty of creating this poll so there's a single topic to discuss the 3 core stats, since there seems to be the general opinion that INT outweighs everything. Everyone feel free to chip in with your opinion and post suggestions on how they could be more balanced.

My take:

I'm not really sure if a caster would ever want to increase STR or DEX, under the current situation there's no point unless they don't have 9 strength for fire/ice/pearl dragon armour, really. And then most people would wear mottled dragon armour anyway, or rely on artifacts to get that extra needed strength, although that's a luxury.

This topic, however was created because of a topic on Cheibriados in the Tavern. It was pretty much stated that +15 to all stats (especially the STR/DEX part) isn't enough to make up for losing the ability to use haste (and let me add, all means of improving your movement speed). Tasonir said that perhaps STR should have a bigger impact on damage or INT should be nerfed. I'm not aware of the formulas that define the damage boost that STR grants, so I don't know if increasing the benefits of STR on damage would be too much.

IF STR and DEX need a buff to their importance for everyone, I think perhaps the following could be a nice idea.

How about, depending on weapon STR-weighting, in order to get delay down you'd have to have an appropriate amount of STR or DEX respectively?
This would split weapons in two categories (or even more, using a combination of STR/DEX depending on weighting). The minimum delay of all weapons would remain unchanged, of course.

A simple example: An executioner's axe reaches minimum delay at axe skill 26. What if you also had to have a minimum of 26 strength to be able to bring it to minimum delay? (so every 2 points of one stat would bring the "effective minimum delay" down by 0.1). Meaning that a 9 STR DEFE that wants to swing around said exec axe would only be able to bring it down to 1.55 (average) from 2.0 that is the base delay. 1.55 is pretty suicidal as an attack speed, of course, so said DEFE would have to stick with a lighter weapon (DEX based, if their DEX is decent).

This would mean that low STR/DEX casters would not be able to use the best weapons as efficiently as their fellow high STR/DEX adventurers(non casters, mostly).

This would also be a buff to Chei and the STR/DEX rings, since you'd be able to hit min delay without having to invest a single point in STR/DEX in many cases.
MuCK;
  Code:
612 | D:1      | Xom revived you
614 | D:1      | Xom revived you
614 | D:1      | Slain by a gnoll
User avatar

Eringya's Employee

Posts: 1782

Joined: Friday, 7th October 2011, 19:24

Location: Athens, Greece

Post Saturday, 5th May 2012, 19:57

Re: Stat balance (STR/INT/DEX)

You can always stick to pure casting, minmay. ;)

Perhaps my idea is incredibly bad, but this was just something that I thought. From a logical point of view, it doesn't make sense that a character with the strength of a pillow can wield a weapon as effectively as a 30+ str one.

Hope the topic doesn't die and that somebody is going to come up with a much more sensible solution.
MuCK;
  Code:
612 | D:1      | Xom revived you
614 | D:1      | Xom revived you
614 | D:1      | Slain by a gnoll

Ziggurat Zagger

Posts: 3037

Joined: Sunday, 2nd January 2011, 02:06

Post Saturday, 5th May 2012, 20:26

Re: Stat balance (STR/INT/DEX)

In general terms, the idea makes sense, even if this specific implementation would be the melee weapon equivalent of the 0.6 heavy armor nerf. It would also be nice if the eventual solution was paired with the removal of minimum weapon delay, probably using a limit function so that you can raise stats or raise skills to approach a minimum delay closer, but never actually reach it. That way, you could get large immediate returns for modest investment in either stats or skills, but eventually diminishing returns would encourage diversification. Different characters would hit diminishing returns at different points, and there would be no obvious breakpoint where further development would be significantly less rewarding.

That said, we're talking about an enormously complicated rewrite here, and I don't think it's particularly going to help for a bunch of non-devteam people to talk it out without the active participation of an interested member of the devteam. Maybe someday, though.

For this message the author KoboldLord has received thanks: 2
TehDruid, Tiber
User avatar

Ziggurat Zagger

Posts: 5832

Joined: Thursday, 10th February 2011, 18:30

Post Saturday, 5th May 2012, 20:30

Re: Stat balance (STR/INT/DEX)

I'm a fan of the Rock / Paper / Scissors methodology, where each of the three attributes enhances a particular factor towards a goal.
"Be aware that a lot of people on this forum, such as mageykun and XuaXua, have a habit of making things up." - minmay a.k.a. duvessa
Did I make a lame complaint? Check for Bingo!
Totally gracious CSDC Season 2 Division 4 Champeen!

Crypt Cleanser

Posts: 726

Joined: Friday, 11th February 2011, 18:46

Post Saturday, 5th May 2012, 20:38

Re: Stat balance (STR/INT/DEX)

KoboldLord wrote:In general terms, the idea makes sense, even if this specific implementation would be the melee weapon equivalent of the 0.6 heavy armor nerf. It would also be nice if the eventual solution was paired with the removal of minimum weapon delay, probably using a limit function so that you can raise stats or raise skills to approach a minimum delay closer, but never actually reach it. That way, you could get large immediate returns for modest investment in either stats or skills, but eventually diminishing returns would encourage diversification. Different characters would hit diminishing returns at different points, and there would be no obvious breakpoint where further development would be significantly less rewarding.

That said, we're talking about an enormously complicated rewrite here, and I don't think it's particularly going to help for a bunch of non-devteam people to talk it out without the active participation of an interested member of the devteam. Maybe someday, though.


A somewhat simpler, yet similar solution would be to treat weapon delay, or even damage, with a step down function similar to how spell power is determined. Optionally, take it even further and do both, with dex used in the delay function and str used in the damage function.

For this message the author Tiber has received thanks:
TehDruid

Ziggurat Zagger

Posts: 3037

Joined: Sunday, 2nd January 2011, 02:06

Post Saturday, 5th May 2012, 20:56

Re: Stat balance (STR/INT/DEX)

Tiber wrote:A somewhat simpler, yet similar solution would be to treat weapon delay, or even damage, with a step down function similar to how spell power is determined. Optionally, take it even further and do both, with dex used in the delay function and str used in the damage function.


Yes, stepdown functions accomplish the same task. They're clumsy and make the code hard to read, though, and are themselves hard-coded breakpoints, providing players spoiled on the inner workings of the code with a small but perceptible advantage. Given the choice between a stepdown and a limit that are functionally very similar, I'd prefer the limit function, but a stepdown function is better than an easily-accessible and easily-calculable hard cap.

Dungeon Master

Posts: 3618

Joined: Thursday, 23rd December 2010, 12:43

Post Saturday, 5th May 2012, 21:05

Re: Stat balance (STR/INT/DEX)

It is widely realised that the stats are not equal at all. A common proposal is to ditch them altogether but I feel that the three stats can be made meaningful.

Some wiki pages relevant to the matter:
https://crawl.develz.org/wiki/doku.php? ... tion:stats
https://crawl.develz.org/wiki/doku.php? ... n_reform&s[]=stats#nilsidestep

And a history lesson: carrying capacity used to depend a lot less on Str than it does now. Also, some weights have been adjusted. Body armour uses Str in a strong way now. One, even if minor, reason to push constriction was about stat differentiation: two nagas should play differently (also in the late game) if they have high or low Str. However, constriction is something like a sore spot for me -- I have no idea whether that ideal is realised or not.

One of my longterm plans was that combat moves (like Constriction) would make use of stats (Str or Dex, as appropriate) rather than XL or skill, say. I don't know if there are any plans to add another combat move anytime soon.

Regarding Str/Dex demands of weapons: That is not out of the question; for example, pre-DCSS 4.2 (the mainline Crawl version that got abandoned) had it. Several good comments have already been made: best weapon performance (e.g. weapon speed) should be approximated but never actually be reached. I am not sure if this is the best area for stat differentiation, though. For one, the weapon skill must be strongly coupled to performance, so stats can only play a minor role. Secondly, most weapons have some "real life" bias towards one of Str or Dex (the example of an executioner's axe you gave makes one think of Str), so that is less room for diversification than one would hope.
[This is a digression, but I believe that weapon differentation could be approached like this: high end (i.e. late/end game) weapons make use of a secondary skill. That skill should not just add damage or similar; rather, it should trigger visible and strong effects -- those could be rare. One goal is to create more skill decisions for pure melee fighters; another is more tactical diversity among such fighters. However, all of this seems pointless until melee does not feel like a handicapped playing style anymore, discounting Trog.]

For this message the author dpeg has received thanks:
TehDruid
User avatar

Dungeon Master

Posts: 4031

Joined: Thursday, 16th December 2010, 20:37

Location: France

Post Saturday, 5th May 2012, 21:20

Re: Stat balance (STR/INT/DEX)

KoboldLord wrote:In general terms, the idea makes sense, even if this specific implementation would be the melee weapon equivalent of the 0.6 heavy armor nerf. It would also be nice if the eventual solution was paired with the removal of minimum weapon delay, probably using a limit function so that you can raise stats or raise skills to approach a minimum delay closer, but never actually reach it.

That's exactly what I had in mind.

KoboldLord wrote:That said, we're talking about an enormously complicated rewrite here, and I don't think it's particularly going to help for a bunch of non-devteam people to talk it out without the active participation of an interested member of the devteam. Maybe someday, though.

Yes, I agree. The stats balance need to occur as a part of a much bigger weapon reform which will address speed formulae and maybe weapon type differentiation too. I'm considering adding such a thing on the agenda, and there's at least one other dev interested in working on it. I haven't had much time lately, but I'll try to initiate the project on the wiki soon. It probably won't be mature before a few versions though.

KoboldLord wrote:Yes, stepdown functions accomplish the same task. They're clumsy and make the code hard to read, though, and are themselves hard-coded breakpoints, providing players spoiled on the inner workings of the code with a small but perceptible advantage.

Not anymore.

dpeg wrote:One of my longterm plans was that combat moves (like Constriction) would make use of stats (Str or Dex, as appropriate) rather than XL or skill, say. I don't know if there are any plans to add another combat move anytime soon.

I think we should consider adding them in the project yeah. I wouldn't call them "move" though, since they are passive special effects. We already have stabbing for short blades and adding reaching to polearms is a success. Let's keep going that way.
<+Grunt> You dereference an invalid pointer! Ouch! That really hurt! The game dies...

Dungeon Master

Posts: 3618

Joined: Thursday, 23rd December 2010, 12:43

Post Saturday, 5th May 2012, 23:51

Re: Stat balance (STR/INT/DEX)

galehar: Yes, as usual, the label ("combat *move*") is a bit of a misnomer. However, if you think about it, it does capture what's going on in the dungeon: the @ is skillfully evading, or constricting an enemy, etc. The issue is just "move" already being a technical term in a game like this (with discrete turns).

I absolutely support adding those things one by one. Reaching and stabbing are fine, constriction is perhaps a semi-success, and the wiki has some decent ideas.
User avatar

Dungeon Master

Posts: 332

Joined: Friday, 15th July 2011, 22:43

Post Sunday, 6th May 2012, 00:09

Re: Stat balance (STR/INT/DEX)

Perhaps minotaur retaliation can be tied to Dex, rather than dodging?

Dungeon Master

Posts: 3618

Joined: Thursday, 23rd December 2010, 12:43

Post Sunday, 6th May 2012, 00:24

Re: Stat balance (STR/INT/DEX)

roctavian wrote:Perhaps minotaur retaliation can be tied to Dex, rather than dodging?

My memory on that is a bit shaky but if I recall correctly, retributtal is not related to Dodging (the skill) but to evading attacks -- in other words, Dex helps, not Str. One could argue that using Dex explicitly is the more conceptual approach but I feel that "Mi evades an attack and butts his horns into the attacker" is so clear that no further explanation is needed -- which is always desirable.

Halls Hopper

Posts: 81

Joined: Saturday, 18th December 2010, 07:08

Post Sunday, 6th May 2012, 00:30

Re: Stat balance (STR/INT/DEX)

dpeg wrote:A common proposal is to ditch them altogether but I feel that the three stats can be made meaningful.

For what it's worth, my (eventual, not in the upcoming version) plan for Light is to replace the numerical stats with three-tier mutations ("You are [very|extremely] intelligent/dopey.") with greater effects per level than you currently get per stat point. Some races would get these mutations intrinsically, of course.
Just a well-spoken spambot.

Ziggurat Zagger

Posts: 3037

Joined: Sunday, 2nd January 2011, 02:06

Post Sunday, 6th May 2012, 05:14

Re: Stat balance (STR/INT/DEX)

galehar wrote:
KoboldLord wrote:Yes, stepdown functions accomplish the same task. They're clumsy and make the code hard to read, though, and are themselves hard-coded breakpoints, providing players spoiled on the inner workings of the code with a small but perceptible advantage.

Not anymore.


I know that most people probably won't notice the difference, but I think that fix is great.
User avatar

Spider Stomper

Posts: 231

Joined: Saturday, 1st October 2011, 18:32

Post Thursday, 10th May 2012, 12:20

Re: Stat balance (STR/INT/DEX)

dpeg wrote:
roctavian wrote:Perhaps minotaur retaliation can be tied to Dex, rather than dodging?

My memory on that is a bit shaky but if I recall correctly, retributtal is not related to Dodging (the skill) but to evading attacks -- in other words, Dex helps, not Str. One could argue that using Dex explicitly is the more conceptual approach but I feel that "Mi evades an attack and butts his horns into the attacker" is so clear that no further explanation is needed -- which is always desirable.

This google doc has the formula for Minotaur retributtal chances, although I am unsure as to when it was made.
Also (just sayin'), when LearnDB gives this advice on which stat...
"It's lucky to be smart, but smarter to be lucky."
- sirlaser the Eclecticist (Human Wanderer), worshipper of Ashenzari

Return to Game Design Discussion

Who is online

Users browsing this forum: No registered users and 9 guests

Powered by phpBB © 2000, 2002, 2005, 2007 phpBB Group.
Designed by ST Software for PTF.