jeremygurr / dcssca

Dungeon Crawl Stone Soup: Circus Animals fork
11 stars 8 forks source link

Bring back HP boost of Fighting skill #662

Closed Sandman25DCSS closed 8 years ago

Sandman25DCSS commented 8 years ago

Yesterday I was playing OpTm and some spectator told something like "You really like hard combos" but the truth is that this is one of the easiest combos in CA. Transmutation spells don't require any investments (Spider form would be usable right at XL 1 if there was a way to memorize it), Fighting does not give any HP so I was training just UC and Dodging. With Fighting giving some HP boost I might actually train it... Species are balanced around having to train Fighting skill, without that they become too easy. For example, DECj might train just Conjurations and Dodging and be fine in Lair while in vanilla it would need to train spellcasting and fighting also.

jeremygurr commented 8 years ago

DECj still has a lot of reason to train spellcasting, otherwise their spellpower will be lacking and spells will cost so much they can only fire off a few before they are exhausted.

jeremygurr commented 8 years ago

Fighting is more important than ever for most melee weapon fighters because of the stamina cost reduction. But it is true for UC they can get by fine without it. So we need something to help with that a bit.

Sandman25DCSS commented 8 years ago

According to cheatsheet Spellcasting does not affect spell power.

I was talking about pre-lair. Currently 80 MP pool allows to spam low level spells almost indefinitely, spell power is much more important because there are monsters which should die fast

I didn't notice Fighting effect on SP cost of melee, my Naga was spending 2 stamina with just 7 Fighting. I am sure it makes no sense to train 18 Fighting on OgBe like I did when I didn't know about lack of HP boost and training Fighting to 18 before 1st rune is standard training for Og.

jeremygurr commented 8 years ago

Oh you're right. Spellpower wouldn't be affected. But spell cost is still quite significant. As is SP cost for most fighters.

Also, for lower level spells, the MP cost isn't very significant.

So I guess the issue is that it's too easy to train high levels of damage because you don't have to divide up between two major skills anymore. The same problem probably exists with spellcasters.

The fact that you always trained fighting on an Og to 18 before the first rune in vanilla is a symptom of the very kind of game design weakness I'm trying to eliminate. There should be very few exact recipies of how to play the game optimally. Optimal play should vary considerably based on the many randomized factors in the game, allowing players a lot of room to discover new and exciting strategies as they learn how to better handle the myriad challenges that the game throws at them.

So we need one more aspect that fighting and spellcasting gives that makes them worth training over just focussing on a specialist skill. But it shouldn't be so good that it's a no-brainer for players to always train fighting and spellcasting. It can't be damage / spellpower increase though, since I want it distinct from what the specialist skills provide. So maybe it needs to increase attack speed.

Since spellpower (basically damage) is determined by intelligence and spell school skills, weapon damage should be based on strength and weapon skills.

Spell casting speed should be determined by dexterity, spellcasting skill, and possibly the spell chosen. Likewise, weapon sttack speed should be determined by dexterity, fighting skill, and weapon being used. This both increases the value of a fighter training dexterity, and for a fighter training fighting in a big way. Probably making things more difficult than vanilla. That's okay, it's easy to buff things if needed.

Oh I like this. By having this strong parallel between fighters and spellcasters, it effectively cuts the complexity in this area in half. Once you know how a given factor works for fighters, you will also know how it works for spellcasters.

Yeah I know, many here will probably complain about the complexity and ask why we couldn't do it the way every other RPG does it. My purpose is not to do what everyone else has already done, but to go where no RPG has gone before, and see what happens...

Sandman25DCSS commented 8 years ago

The fact that you always trained fighting on an Og to 18 before the first rune in vanilla is a symptom of the very kind of game design weakness I'm trying to eliminate. There should be very few exact recipies of how to play the game optimally. Optimal play should vary considerably based on the many randomized factors in the game, allowing players a lot of room to discover new and exciting strategies as they learn how to better handle the myriad challenges that the game throws at them.

I am not sure I was understood so I will clarify. Ogres have +3 Maces& Flails, +3 Fighting, +20% HP (synergy with Fgithing aptitude), bad Dodging (-1 and low Dex), large size and cannot wear non-dragon armour (robes are almost useless) so it is a good idea to train Fighting much higher because your weapon is already at min delay and there are 3 types of defense: HP, AC and EV while Ogre has problems with latter two. Of course if the Ogre is lucky to find some dragon armour (even swamp one), then training Armour becomes a good idea. Of course I train some Dodging early no matter if I find dragon armour or not but getting Dodging/Armour to 18 is a very bad idea at this point anyway.

Sandman25DCSS commented 8 years ago

I am glad fighting will be more useful if I read your comments right.

jeremygurr commented 8 years ago

Ok the melee attack code is even more convoluted than the spellcasting code. So for now I just made attack speed come equally from dex and fighting skills, and I will put off the other changes until I have more time.

Sandman25DCSS commented 8 years ago

So for now I just made attack speed come equally from dex and fighting skills

Does it mean weapon skill has no effect now?

jeremygurr commented 8 years ago

Right. Weapon skill is only for increasing damage and accuracy, like spell skills are for damage and reduced failure rate.

Sandman25DCSS commented 8 years ago

So now weapon skill is useless instead of fighting :( Casters still have one less skill to train than melee characters in CA, in vanilla casters need to train both Fighting and Weapon, in CA they need just Fighting.

Sandman25DCSS commented 8 years ago

Though I suspect it can be fun to use bardiche, exec. axe and triple sword at min delay via training fighting to 27, it does not look very balanced.

jeremygurr commented 8 years ago

Why is weapon skill useless when it is the most significant damage and accuracy factor? Are those no longer important? It looks like 25 levels of weapon skill gives you 50% more damage. And every 30 levels of fighting apparently gives 50% more damage too. I'll have to fix that. I'll take out the fighting damage for now, and increase the amount of boost that weapon skills give to damage, since fighting no longer contributes, but this whole mess of code is due for a cleanup.

I think casters still have more to train, even though they don't need fighting as much. Fighters rarely need to train more than a single weapon skill. Casters almost always need to train multiple spell schools. Add to this the fact that I've already made things harder on casters in a few other ways, and it seems highly unlikely that casters pull ahead of fighters in how quickly they can train essential skills.

Sandman25DCSS commented 8 years ago

Yes, 50% more damage is insignificant for casters because they use weapon for popcorn only, speed is important. Melee characters won't train Weapon to 25 to get doubled damage, there are usually better skills to train. Compare to casters, would you really train Conjurations to 25 to get double damage for Magic Dart?

Yes, casters need more skills to train, that's their challenge in vanilla. In CA they need one skill less. Basically you buffed casters (they usually have high Dex) and nerfed fighters in heavy armour (they usually have high Str and low Dex). But ok, let me try that in actual game, maybe my assumptions are wrong.

Sandman25DCSS commented 8 years ago

It looks like 25 levels of weapon skill gives you 50% more damage. And every 30 levels of fighting apparently gives 50% more damage too. I'll have to fix that. I'll take out the fighting damage for now, and increase the amount of boost that weapon skills give to damage, since fighting no longer contributes, but this whole mess of code is due for a cleanup.

Yes, I know about 25-30. This code can be complicated but it results in good game balance. Early caster can use any weapon without training because Fighting still gives some accuracy and damage. Weapon skill is always more useful than Fighting for damage and even more so as long as it improves speed.

jeremygurr commented 8 years ago

Yes it is possible to balance a mess of code, but it is very costly to maintain and very fragile. The tiniest changes make the whole thing come crashing down. That's one of the reasons vanilla devs are so reluctant to make big changes in this area. I prefer code I'm working in to be agile and fluid, accepting change with the minimum of cost and pain.

As far as spellcasters needing more things to train, keep in mind that you were just telling me yesterday how worried you were that spellcasters would ever be able to pull off level 9 spells. If they didn't have enough to train, and they advanced too quickly in skills, we would be having the opposite problem. Spellcasting is in a good place now. Not because they are nicely balanced, because they probably are far from it still. But because the code that governs the core mechanics is so much cleaner and easier to tweak parameters for, minor adjustments to improve balance are very quick, which is pretty important, because there's probably a lot of tweaking still needed there.

The same kind of balancing can easily be applied to fighters, once I get time to streamline their code also. For now, hopefully I didn't break things too badly.

Sandman25DCSS commented 8 years ago

I prefer code I'm working in to be agile and fluid, accepting change with the minimum of cost and pain.

I didn't know you were going to change the code. The code was already present and working, it could continue working without any problems.

As far as spellcasters needing more things to train, keep in mind that you were just telling me yesterday how worried you were that spellcasters would ever be able to pull off level 9 spells. If they didn't have enough to train, and they advanced too quickly in skills, we would be having the opposite problem. Spellcasting is in a good place now.

I don't take my words back, the same problem is still present. Low level spells are too easy to cast and you have almost unlimited MP, high level spells are almost impossible to cast and it does not make much sense to go for them.