Physical Power and Physical Save

[edge2054]

Saves are worthless. I hear this again and again. Stuns are far too lethal to rely on saves. Con is a crap stat. Etc. etc.

Weapon Mastery talents give 10 physical power per talent point at mastery 1.0. For contrast Arcane Power grants 6 on the first talent point with quick diminishing returns that peak at 25 spellpower with five talent points invested, but at 1.3 mastery instead of 1.0. The Solipsist equivalent, gives half a Mindpower per talent level but is capped at the sum total of all tiers of psionic or nature powered gear you wear, divided by 2. Okay that example is a little wonky but it illustrates how hard it is for non-physical classes to stack power, especially early game.

What this results in is a huge power disparity, especially early game. I'm not talking about damage wise. I'm talking about Power (i.e. Physical Power and Spell Power).

Accuracy and defense have some similar issues (and accuracy and physical save for that matter) though in some ways it's even a larger gap because defense is really hard to stack compared to physical save for many classes and Combat Accuracy gives even more accuracy than weapon power does. And again I hear similar arguments. Defense sucks. It's not worth investing in. In fact, players are discouraging me from bothering with it at all on with the Chronomancer rewrite.

To give you all an idea just how bad this is a skeleton warrior starts off with 24 physical power and 19 accuracy (min level). A starting skeleton bulwark with two points in Shield Wall, probably the tankiest level 1 character in the game, only dodges this attack 45% of the time and never saves against it. Thankfully for the Bulwark he has a bit of stun immunity to fall back on. Not every class has this option. And keep in mind this is a starting monster with just one point in each of the offending talents.

So here, as a class designer, I'm caught in this position of power creep. Do I give the class evasion? Do I just give them enough defense for it to matter? Then what? Chronomancers out defense the defense class just to have enough defense for it to matter?

Instead I'm going to offer a very simple solution. Cut these values in half. At the time Sus rewrote Weapon Mastery and Combat Accuracy the game was not in the state it is in today. Egos still needed a lot of work. Artifacts were lackluster. The player had fewer options for getting Accuracy and Physical Power out of gear. That's not the case anymore. If players want to stack these stats, let them, but make them gear for it like everyone else does.

If halving combat accuracy makes it too crappy then give it another bonus. A bit of extra crit would make sense but it's hardly the only option. As to weapon mastery, I think the damage multiplier alone makes it still worthwhile, even if we cut the physical power bonus down to more reasonable numbers.

And what's this all accomplish? Well it makes physical save and defense matter more because they're not playing catch up with two talents that are disproportionate to the rest of the game.

Anyway thanks for reading.

*steps off his soapbox*

[Fortescue]

This is a great post. Would be interesting to see you test your theory since you didn't propose a huge amount of changes. Modify those values yourself and try playing?

[donkatsu]

Saves and defense are the way they are because their actual mathematical effects depend on how much power or accuracy your opponent has. Most defensive stats are not like this. Going from 20%->40% evasion or resist or status immunity has the mathematical effect against every opponent, barring specific counters. But if I go from 40 to 60 defense, the effects are wildly variable depending on my opponent. It could mean I've gone from dodging attacks 50% of the time to dodging them 95% of the time, a 900% increase in the number of attacks I am expected to be able to tank. Or it could mean I've gone from dodging attacks 5% of the time to 50% of the time, a 90% increase in the number of attacks I am expected to tank. Or, more realistically, it does absolutely nothing.

A defensive mechanic that makes you untouchable against weak enemies but is worthless against strong enemies, is just worthless. That's the opposite of the kind of defense that you want. Power/save tiers were implemented to dilute the arms race, and we can also adjust how much of it you get from talents or stats, but no matter how you tweak the numbers, the fundamental mechanic is always going to be, good against weak, useless against strong.

Here's a crazy idea.

What if defense/accuracy followed the resist/resist penetration model? 50 defense means you avoid 50% of attacks. 50 accuracy means your target's defense is cut by 50%, so instead they avoid only 25% of attacks. Keep the tiers, they will function as a sort of soft cap on defense as a substitute for resist's hard cap of 70%. Remove the Combat Accuracy talent altogether. Remove accuracy from Dexterity, it'll still be a good stat because now the defense it gives is that much more valuable. Just limit how much accuracy you give out on anything, like how resist penetration is limited. This won't fix defense, but it will give it the POTENTIAL to be fixed by adjusting numbers on talents. This system has a precedent for working, because resistance is already a working, valuable stat. Current defense can never be fixed just by adjusting numbers, as explained in the previous two paragraphs.

I might think of a clever idea for saves later, but I just wanted to get this idea out there.

edit: come to think of it, this is how crit/ crit avoidance works too, and I haven't heard anyone say that crit is useless

[edge2054]

I like your proposal Don. I'd go a step further and say that with some consideration the same thing could work with saves (saves reduce duration by a percentage based on their power, power works to negate a percentage of saves). Also evasion could be rolled into defense at that point. (i.e. temporary huge defense buffs).

It would be a fairly big overhaul though. Which isn't to say that it's bad or not worth doing. Just saying that in the interim, if there is an interim, I'd like to see something done about weapon mastery and combat accuracy.

[Davion Fuxa]

donkatsu's idea doesn't really sound crazy to me, but I can only tell you how frustrating it is in some games to have a situation where you miss an enemy too much. To bring up Stun and Physical Save for example, the reason Physical save is useless is because you can't be entirely sure you'll avoid Stun with it - and Stun can kill you. I do think Combat Accuracy has a place to help players in avoiding the situation where you have a really dodgy enemy and he just flattens you because his defense is so high that the RNG decides to screw you over. The nice thing about Tales of Maj'eyal right now is that if you die you can be pretty sure it was you own fault - not the RNG's fault.

Saves would likely still be useless if all they did was reduce duration. They are useless now because you can't rely on them to get rid of that nasty Stun or annoying Disarm - you need a Wild or Immunity for it or you are likely toast. If they did something like reduce the effectiveness of Stun for example - like reducing the Slow and Damage Dropoff you get then that might work however. Apply to other talents like Impending Doom and Saves can really matter for something.

[Nori]

I like donkatsu's idea a lot too.
Reduced dependance on generics and dex for accuracy is awesome since i've always hated that.
Melee characters have to spend 10 generics just to be decent in melee.

Some stat (talent even) that you can invest points in though would have to increase accuracy. It makes no sense to rely on a weapon for accuracy when you consider that it is you (the character) actually swinging the weapon.

Duration reduction from saves is actually nice. I've had quite a few times when I had a long, long status effect. I might be able to survive 2-3 turns of it but not 6-8 for instance.

[SageAcrin]

I don't like blanket changes that totally skew how the entire game plays, though.

In particular, the proposals here end up on the lines of melee nerfing. You can justify that and rebuild the game to fix it, but... is this something that really is required as a fix for this problem?

Here's a much simpler solution that ups the impact of saves.

Firstly, up all negative status durations by 1. This is a small rebalance for the next impact. (Not necessary for extremely long, 10+ duration status though.)

Secondly, make all three saves reduce all relevant negative status durations by 10% for each tier they go up, to a minimum of half the original duration, rounded up. This does not check powers in any way.

And I do mean all. Burning? Sure, you can reduce that now-four-turn burning effect to 2, if you have tons of physical save. Confusion? Sure, you can still Confuse Atamathon, but it's not going to lock him down.

Defense is a little more complex, but I once suggested that Defense be usable as general damage reduction vs various powers, with a lesser effect. This ups the impact of Defense, because it means that you could, say, have a Defense equal to a target's Magic Power, and avoid say 20% of the damage(If you had 20 more, you could, say, avoid 40% of the damage, and cap off. If the enemy had 20 more of a power, no damage evasion.).

These changes have the benefit of not massively altering the game, but still granting very meaningful impacts that anyone can parse at any time, to these statistics. At this point in the game, I feel like turning things upside down is something to be avoided-too much code is based on too many fundamental concepts to really be totally overhauling the mechanics, in my opinion.

[donkatsu]

It would most certainly be a massive overhaul. However, defense in ToME has been broken since the dawn of time. There have been all of these little changes: tiers, cross-tier effects, changing Combat Accuracy to 5 points instead of 10, but none of them have worked. None of them have even helped all that much, really. By letting the fundamental problem fester and plastering over it with ineffective bandaids for fear of rocking the boat too much, I feel like there has been more wasted effort than there would have been if the problematic root was just torn out from the beginning, and that this pattern will continue to hinder design until it is dealt with definitively.

There might be a better way to tackle it, but for example, that last solution you proposed keeps the problem of defense being meaningless against strong enemies. My defense is 60, this vampire has 20 spellpower, so it reduces damage by 40%. But then I fight a rare that has 80 spellpower, and my defense is still just as useless as ever, if I am understanding your proposed calculations correctly. It also becomes completely detached from its narrative justification, since defense's theme is dodging stuff and you can't dodge an explosion.

I'm still unsure what to do about saves.

Oh, and when I proposed taking out Combat Accuracy, that was with the stipulation that classes would get accuracy in their own talent toolboxes, just like they do with resistance and armor penetration. But that's just keeping with the established pattern from those other systems; you could just keep Combat Accuracy and it wouldn't really be a balance problem as long as NPCs didn't casually reach 100 accuracy like it was nothing.

[SageAcrin]

Well, if you want a flat rate bonus, Defense could actually do something like a flat rate 2.5% damage reduction(cap at 10%) per tier, against all damage.

It's not necessarily against the narrative; Think of it along the lines of the logic D&D uses for Saving Throws. You're getting partially, but not entirely, out of the way, or shielding yourself, or crouching behind a rock, etc.

[donkatsu]

I think that would be approximately equal to removing all instances of defense and replacing them with resist all, with very quickly diminishing returns.

I'm not entirely against a complete removal of defense either.

I really like the idea of lowering damage across the board though, and turning defense into resist all would certainly accomplish that.

[notacorporal]

I am not in favour of overhauling the fundamental mechanics. At least I wouldn't dare do this myt self until I got a decent idea what behaviour is intended and what is causing problems. At the moment I am willing to treat anything and everything as a "feature" untill proven otherwise.

If you were going to change the opposed roll mechanics I favour:

Code: Select all

                     Attack  
  p[Hit] =     -------------------
                 Attack + Defence 

Where Attack and Defence are some function of the relevant stats. This is guaranteed to give a smooth curve. With a little algebra, it can be made to produce any kind of behaviour you want.

Regarding the OP, it seems problematic that physical power serves to seperate purposes. It can account for most the weapon damage. This is particularly significant for daggers if the wielder is not pumping Strength. The base power will not be great, but this also means that the contribution from dagger mastery will be that much greater. It also means that, even with Lethality, Strength is the most significant stat for daggers until you get a few levels in Knife Mastery.

With this in mind, it looks at though edge is absolutely right in his analysis of Weapon Mastery. As for combat accuracy, I still don't know what people are hoping to achieve.

[HousePet]

One thing to consider is that debuff duration reduction from saves was corrected in 1.2. So old information about the usefulness of save is out of date.

Current formula for attack versus defense is 50 + 2.5 * (atk - def). This doesn't seem too bad. Though it does mean your def needs to be more than 20 less than their accuracy to have an effect.
And the site is offline, so I can't do a quick check of late game stats to compare. It could be that there is a disparity between achievable defense and accuracy.

[edge2054]

The old hit style was a bell curve and really I think a lot our problems came about when we switched over. The old system had the problem of being too hard to explain. I mean we basically had 50/50 hit chance when values were equal and it approached a 5/95 split as the values became more distant but the further away you moved from 50% the more of a gap you needed for that next percent.

As to accuracy and defense. It's a similar issue. Accuracy currently outpaces defense to such a degree that unless you can really stack defense it's not seen as valuable. This makes defensive stats, i.e. defense and saves, far less valuable than they should be in my opinion. So what we end up with instead is a bunch of other defensive stats that bypass the hit mechanics (immunities, resists, crit reduction, etc. etc.)

All of these values, accuracy, defense, saves, and power, use a similar hit system to hit (2.5% per difference up until you hit 100%).

Here's the old hit calculation for anyone that's curious. It's still in the code because it's still currently used for stealth and invisibility.

Code: Select all

function _M:checkHitOld(atk, def, min, max, factor)
	if atk < 0 then atk = 0 end
	if def < 0 then def = 0 end
	print("checkHit", atk, def)
	if atk == 0 then atk = 1 end
	local hit = nil
	factor = factor or 5

	local one = 1 / (1 + math.exp(-(atk - def) / 7))
	local two = 0
	if atk + def ~= 0 then two = atk / (atk + def) end
	hit = 50 * (one + two)

	hit = util.bound(hit, min or 5, max or 95)
	print("=> chance to hit", hit)
	return rng.percent(hit), hit
end

[HousePet]

More defense from dexterity!
0.3 is crap when it gives 1 accuracy.

[edge2054]

More defense from dexterity!
0.3 is crap when it gives 1 accuracy.

If it loses the crit reduction and that moves over to Con I think that would be a nice idea.

But I don't think these things exist in a vacuum. Either way combat accuracy and weapon mastery seem, to me at least, to be a very obvious place to start.