Fixing Max Resistance imbalance.
|
The resist system is terrible because it allows for very little granularity between baseline adequacy (75%) and total immunity (100%). That's 25 units. They've had to tiptoe around this since the game was released to the public, and they continue to struggle with it as can be seen in the huge disparity between Atziri build capability and the recent attack on max res from uniques and the tree. Any system in which 100 marks an absolute will have these issues. Block and reservation are similarly problematic--useless for low values and overpowered for high values because of increasing returns per next-point-invested. This is why, by and large, systems in RPGs tend to be designed around diminishing returns.
It will never be fixed because the itemization is implicated. A full conversion to a resist rating system would likely mean rerolling every item in the game. It would probably be possible to hamfist a system to use the existing 1-45 rolls and bend the formula around that, but it probably wouldn't be ideal. If anything about this were going to change, it would have been done in the CB->OB transition to minimize damage. It wasn't. | |
" No, resists are not linear, they have well known increasing returns as you approach 100% resistance. That is the fundamental flaw that PoE inherited from Diablo II, and GGG's inconsistent dabbling with max resist boosts has only made it worse. Armor is a completely different story, a case where diminishing returns was applied in an overly zealous manner. A diminishing returns curve should be adjusted to make it easy to get to 50%, but difficult to acquire more than about 75% effectiveness in the attribute it affects. The armor curve is soft-capped way too low for the extremes of physical spike damage in PoE. Defensive physical damage conversion Uniques like Lightning Coil or even Cloak of Flame should not outpace the effectiveness of armor, they should provide a trade-off rather than an outstanding advantage. And no, these Uniques should not be nerfed, armor's diminishing returns should be boosted up to their competitive level. " No, it can in fact be retroactively fixed, and that's what makes diminishing returns formulas well-suited solutions for PoE's fundamental balancing issues. Applying these soft-cap formulas to the core combat engine would avoid the havoc and turmoil of attempting to revise the numerical stats on existing items and Skill Tree nodes. Diminishing returns allows players to continue to stack their stats as high as they want, while the underlying formulas soft-cap their combined effectiveness to reasonable limits. As a result, it would take fewer nodes and items to reach best-bang-for-the-buck levels of effectiveness, allowing wider options for more flexible builds. At the same time, players who want to specialize in particular stats would still have the option to optimize them, but with decreasing effectiveness with each additional point. Last edited by RogueMage on Oct 11, 2014, 2:30:11 PM
| |
|
There is a transition from gearing early to gearing later where you initially want lots of resists on things, and later you look for something else because your resists are maxed. A rating system loses this transition.
What good is a transition? It keeps the game from getting stale for longer. The transition means you don't evaluate gear in the same way at level 40 as you do at level 80. It also gives you a particular goal that you can reasonably accomplish. Resist rating trades all that for mathematical purity, which is a nice thing to have, but it does not create those dynamic goals or completable achievements that we currently have, either. Last edited by PolarisOrbit on Oct 11, 2014, 2:07:45 PM
|
|
" So are we saying something like softcap resists at 30%, remove the resists penalty, balance around 50%, and then make it so that every point after 30% only gives 1/6%, and every point after 60 only gives 1/12%? That works, but you run into other major problems. For one, as people have said, there's never any end to resist stacking. Normally this might be good, but given the hard-on for elemental spike damage recently, it would make 40%+ all gear worth much more than well-rolled armor/eva. The other is that all uniques with +max resistance would have to be redesigned. Not legacied and re-divinable, but totally scrapped and redone like Voll's was, because the +max stat would no longer have a function in PoE. Again, this isn't a realistic change. It's too much to expect. We've been making this argument for years now and we've seen absolutely zero impetus from GGG to redo the mechanics. It's like Pain Attunement--everything they've done surrounding the broken system has just further entrenched it. Last edited by Uvne on Oct 11, 2014, 4:20:11 PM
| |
" Astute observation, but flawed conclusion. The very same transition would occur regardless of whether resists are hard-capped (now) or soft capped (via diminishing returns). At some point you look at your gear and say "I have enough cold resistance, these extra thirty points will only give me an additional 1% resistance." You may be lured into stacking resistances ad infinitum to reach as close to 99.99% as possible, but you would be wrong to do so at the expense of other potential benefits. That is, for all intents and purposes, the very same thing you argue on behalf of. Edit: Also, cross post from a duplicate thread. I don't know what diminishing returns formulas people would suggest to apply, but I kind of accidentally stumbled upon a methodology that would (could?) work with everything in game exactly as it is now. " Only posting it because the number match up pretty well with what expected resistance values we might see in game currently, given difficulty penalties, mob curses, and map modifiers. If values were rounded down (which they are not in the quoted post), then a hard-cap would occur around the 99th iteration (some few thousand resistance rating) at 99% resistance. Devolving Wilds Land “T, Sacrifice Devolving Wilds: Search your library for a basic land card and reveal it. Then shuffle your library.” Last edited by CanHasPants on Oct 11, 2014, 4:42:52 PM
|
|
|
My solution would not require any change on any item by making resist a flat value as with all other stats in the game and the current percent affixes scale that value. Give all players a flat elemental damage resist. This can be done by player level or passives replacing the current % resist nodes on the tree.
(base player resist + max player resist) *(total % increase resistances) example (20 + 6)*(1+1.2) = 57.2% elemental resist This would get rid of the max resist stat without butchering the current currents too badly. Then cap all elemental resist at 75% or whatever and scale all damage down for the known limit. The base resist value can be adjusted and tested to find a nice scale at which resist is not easy to cap but not impossible. GGG can then balance endgame with 75% elemental resist in mind and forget any ridiculous possibilities like 90%+ resistances. Honestly anything higher than 75% less damage might be too much you are already taking anywhere between 25% to 100% of the damage which is a very large range. Oh and remove immunity from IC. |
|
" Under a system like your talking about the max resists wouldn't need removed or even drastically changed, just changed to raise resists at a equal level to their max resist value after normal resist rating is calculated. Using your example of 1 point of resist for every 12 points of resist rating after 60 resist, then +5 max resist on gear would convert into a value equal to 60 resist rating since it would add a flat 5 to your resists after the rating is figured out and ignores diminishing returns since it would be a direct increase to the end result itself. | |
" I did a poor job explaining if you think the "very same" transition occurs. With a soft cap there is always an efficiency calculation: are resists more valuable or is it affix X? Concretely speaking, I expect life and resists would be pooled together into an EHP calculation. Depending on how things are balanced, the formula may tell us the optimal choice varies depending on the situation. When the formula switches between the various answers, that is an abstract and ephemeral transition because we are still using the same formula regardless of what it tells us to do. I am interested in when we change formulas. A hard cap that is important to meet changes the way gear is evaluated before and after the cap is met. If we still want more EHP, we have to get creative and search for more exotic alternatives. Before the hard cap it is easy to keep pushing with the readily available mod, but there is a decisive point where change is needed. This structure is accessible so players always understand their goals without much decision paralysis (where we can't tell if something is better without leaving game to consult the spreadsheet oracle). Under a soft cap, players should maintain an optimal ratio of the factors under consideration. The more variables that get locked into these efficiency calculations the more complex gearing becomes, but it doesn't add any real depth until we add a new formula or change an existing one. The gearing process gets stale because soft cap doesn't do either of these things. In fact it does the opposite-- it gobbles up more variables into one calculation. You're never really "done" with any particular mod, which not only robs you of that sense of completion, but also enables more things to be compared to a gold standard. Intangibles become tangible, exotic defenses are on the same ground as just pushing the soft cap, choices which aren't currently comparable become competitors. Soft cap has a sinister side, and you would do well to be wary of it. I should write a conclusion, but I'm all typed out. |
|
|
Okay stop arguing about this.
D3 has the exact same system you are suggesting, Literally it does nothing late game because you stack resists extremely high regradeless. If they magically change resists to stack differently that doesn't mean whatever the new number of in a resist that equals to 75% isn't needed to all resists to do the same content you were before. If it does 1000 fire resist to get to 75% fire resist I'm going to do it anyway. |
|
" Is too linear take someone with 50, 75 and 95% resists something hits for 10K 50% takes 5000 dam 75% takes 2500 dam 95% takes 500 dam exactly linear. dam = input dam x (1- resistance) " Those 100% require massive sacrifice (23 purities, inner force, 50% buff auras, saffels, flask nodes etc) which is why most ppl dont do it and deal with monsters spells in about 10 different other ways available. (spell block, spell dodge, huge HP, kill monster before they get started, huge hp instant leech, etc) Game is not supposed to be easy eg hurp derp get 75 resits face tank atziri. would trivialize content and throw theory crafting and sacrificing out the window. Basically turn into a lame arcade game. And besides that you're not totally immune with 100% anyway. Some monster has penetration. Here is 108% (95% + 13% flask) on atziri and I still take 1.5K damage from atziri double flame blast. https://www.youtube.com/watch?v=UiVFnBniuK8&list=UUfr_OnbD7g8uPk0s_mjzZLQ Git R Dun! Last edited by Aim_Deep on Oct 12, 2014, 3:43:07 PM
|
