Jump to content

All enemies have global chance to proc +to-hit


Recommended Posts

37 minutes ago, ShardWarrior said:

You will note the words "diminishing returns" do not appear anywhere in that.

How about you try to state an affirmative point somewhere in here?  Do you have anything to actually say on this subject?  Because here's my assessment of your contribution to this thread so far:

 

1.  You completely failed to understand the proposal and responded to something that nobody suggested.

2.  When called on that, you did not acknowledge it and tried to claim that your point was valid.

3.  When explained again, you made sweeping statements about progression that still did not actually address anyone's point.

4.  When called on your bullshit you started claiming that you didn't say...  uh... anything?

5.  When things were painstakingly spelled out to you, you....  didn't address anything and went back to saying that you...  didn't say anything.

 

The only actual thing I've seen you say that doesn't appear to be based entirely on your own painful misreading is, "I don't see any need for a change to how defense works."  Do you actually have anything to say besides that?

  • Sad 1
Link to comment
Share on other sites

On 7/11/2020 at 1:47 PM, aethereal said:

This isn't a terribly serious suggestion.  More...  Something to think about.

Eightteen days and four pages later...

44 minutes ago, aethereal said:

How about you try to state an affirmative point somewhere in here?  Do you have anything to actually say on this subject?

 

  • Thanks 1
  • Haha 1

 Everlasting's Actionette 

Also Wolfhound, Starwave, Blue Gale, Relativity Rabbit, and many more!

Link to comment
Share on other sites

5 hours ago, aethereal said:

Do you have anything to actually say on this subject?

Yes.  I have already expressed my opinion on the matter, only to get attacked by you for some unknown reason.  I am sorry that I do not agree with your idea or concept.  Again, what I am reading here is a solution looking for a problem and attacking others is certainly not endearing you to anyone.  Your idea simply is not needed in my opinion.  Feel free to disagree. 

Link to comment
Share on other sites

Maths help a lot ether real.  Can you give us some charts and graphs to make your idea more clear?  I think some people donut understand what your meaning to explain in the idea for end goal and difference in how it will play.

  • Sad 1
Link to comment
Share on other sites

1 hour ago, Patti said:

Maths help a lot ether real.  Can you give us some charts and graphs to make your idea more clear?  I think some people donut understand what your meaning to explain in the idea for end goal and difference in how it will play.

We get the idea. We simply think it's a bad one.

Link to comment
Share on other sites

Ah, this thread again.

 

I'll just re-iterate and think that the idea in general (enemies get a random +ToHit when they attack) could/should be applied to a sub-set of enemies and certainly not all of them.

 

I've always preferred "challenge" to be engaging the player's mind, rather than just a number checking exercise. Things like Nemesis Lieutenants, Devouring Earth summons, which coyly wag their fingers at players trying to brute force their way through them by subverting the standard order of play. Not to say that this should be true of all enemy groups, but if we want people to challenged, let's do it smarter, not dumber. (funnily enough you don't see people lining up to do Nemesis or Devouring Earth farms, almost as if *gasp* they're too challenging)

  • Like 1
  • Thanks 1

Oh? You like City of Heroes?

Name every player character.

I'll be waiting in my PMs.

Link to comment
Share on other sites

So the math for this isn't so much hard as there's a lot of it.  Let's go through some of it to see how daunting it is to get a really good grasp of what's going on.

 

So first of all, chance to hit.

 

HitChance = Clamp( AccMods × Clamp( BaseHitChance + ToHitMods – DefMods ) )

Let's plug in some numbers:

 

+0 Minion fighting a character at 45 defense:

 

Accmods: Power Accuracy * 1 * 1

BaseHitChance: 50

ToHitMods: presumptively 0

DefMods: -45

 

HitChance = 5% for a power with 1.0 accuracy (note that even minions may have higher-accuracy powers so even against even-conn minions, you may actually have a higher than 5% chance of being hit at softcap)

 

+0 Minion fighting a character at 45 defense who has procced a +10% chance to hit, then, is all the same except ToHitMods is +10%, leading to 15% chance to hit.

 

+4 Boss fighting a character at 45 defense:

 

Accmods:  Power Accuracy * 1.3 * 1.4

BaseHitChance: 50

ToHitMods:  presumptively 0

DefMods: -45

 

HitChance = 9.1%

 

With +10% chance to hit, that raises to 19.1%

 

Okay, so one way to describe this change would be to say that the expected damage of an even-level minion, after the alpha strike, would be 15% higher (95% chance of 1x damage, 5% chance of 3x damage).  If a minion gets 3 total attacks off, then we'd say that their total expected damage would rise by 10% (since the first attack has no chance of proccing).  This is for a character right at softcap, exactly 45% defense.  A character with say 43% defense would experience only about a 5% expected rise in damage in the same scenario (3 total attacks), and a character with 47% defense would similarly experience about a 5% expected rise in damage.

 

A +4 Boss gets 2.12x damage against a character exactly at softcap from this change, and procs is 15% of the time, so about a 17% rise in expected damage after the first hit, and here we might reasonably say that a boss doesn't die after just a couple of attacks, so maybe we just ignore the first hit and say it's a 17% rise in expected damage altogether (for a character exactly at softcap.  Again, someone a few points of defense higher or lower will experience decreased damage changes).

 

But expected damage is maybe the wrong metric for defense calcs in general.  We plausibly care much more about what we in the software industry would call P90 or even P95 or P99 damage -- that is to say, the damage that you experience at your 10% unluckiest sets of rolls, or your 5% unluckiest, or 1% unluckiest.  That is to say, if you survive 9 spawns out of 10 fine, but the 10th spawn you run into kills you every time, my guess is probably players end up turning down difficulty there.  Do they turn down the difficulty if they die to one spawn in 20?  I'm not sure.  One spawn in 100?  Probably not, unless maybe it's a farming toon.

 

How do you calculate P90 or P95 damage?  Well, it's doable as a pure math calculation, sort of, but it's quite complex.  The way I'd do it is write a monte carlo simulation and then just order the results of a 1000 or so runs by the total damage the player experiences and choose the 100th from the end or the 50th from the end.  But what should we simulate?  Well...  shit.  I mean, really what you want to do is simulate an entire spawn.  Which isn't just generic minions or lts or bosses firing scale 1 attacks, it's...  a lot of stuff.  It's different mixes of enemy classes firing different attacks, and you also need to figure out how fast do they die.

 

So there are a lot of variables here.  A selection of them:

 

What level difference between player and critter

What mix of critter types

What attacks do the critters have

   - Do these attacks debuff defense?

   - Do these attacks have inherent accuracy or to-hit?

How long do critters of each class survive

Exactly what defense level does the player have (to each of the attacks the enemies have)

 

A really useful tool, I think, would be a simulator that allowed you to plug in even rough answers to these things.  Like where you could say, "Okay, here's a stylized +4/x8 spawn of Council, and here's some metric of how fast they'll die to the player attacks, let's run a few thousand combats."  It would be necessarily imperfect, but I think it would be a useful tool to try to attack the complexity here.  However, it'd be a hard tool to write, and it involves data that's not right at my fingertips, at least, about the details of enemy attacks.  Even if you had it, trying to really understand the spread of possible scenarios would be a big undertaking.

  • Thanks 1
Link to comment
Share on other sites

@aethereal Thank you for your post and clarification of your solution. The suggestion appears overbroad, to address of small set of issues, I.e. IO set bonuses, softcap targets, playing through more controllable environments (s/l enemies without accompanying debuffs of any serious nature). Certainly, market fixes have created more easily attainable “amazing” builds. Is your proposal 
 

Overall, the proposal is a widespread response to a minimal problem. First, the theoretical target player (or more precisely, character), is less prevalent than forums would suggest (while merely an anecdote, I’ve come across more masterminds than I care to count that don’t take the level 6 and 32 upgrade powers in-game. These players are not the target in your proposal. Nor are the number of melee ATs that are running with basic IOs or a smattering of frankenslotted sets). 
 

Second, the game itself provides sufficient levels of challenge and increasing difficulty for those who choose to play it. Ouro settings, spawn size and level selection, enemy selection, etc... Choice matters. When players are routinely farming Carnival groups, for example, there might be a cause for concern. Though I have seen one advertised on Everlasting, but now most “high challenge” groups seem to focus on AE 801 runs.
 

Third, like challenge selection, build choice matters. A great strength of this game is the presentation of meaningful choices. This includes building for defense softcap, resistance At-specific hard caps, sufficient recharge for either attack chains or “important” long-recharge powers, while balancing HP goals, regeneration, debuffs, procs, and endurance. (I recall the discussion earlier in-thread re diminishing returns for recharge/side issue. In a vacuum you’re certainly correct. I’d suggest that recharge is only valuable in tightening an attack chain or making permanent a long click power. Any recharge over those goals is wasted - like extra defense - unless building for protection against slows. Again, much like added defense can substitute, albeit less effectively, for DDR). 
 

May I recommend you check the detailed work going on the Tanker (and by extension, scrapper) forums to assess survivability metrics. Much of what your suggesting/discussing is being deconstructed by the active posters in those threads. As one very small example (which also addresses the relative value towards survivability when one exceeds an identified softcap as a way to approximate DDR):

 

 

  • Like 1
  • Thanks 2

The Splintered Soul Project: (Nyght****) 21 and counting (18 max). 

 

DSorrow: “Give a man a build export and you feed him for a day, teach him to build and he's fed for a lifetime.

Link to comment
Share on other sites

27 minutes ago, Nyghtmaire said:

@aethereal Thank you for your post and clarification of your solution. The suggestion appears overbroad, to address of small set of issues, I.e. IO set bonuses, softcap targets, playing through more controllable environments (s/l enemies without accompanying debuffs of any serious nature). Certainly, market fixes have created more easily attainable “amazing” builds. Is your proposal 
 

Overall, the proposal is a widespread response to a minimal problem. First, the theoretical target player (or more precisely, character), is less prevalent than forums would suggest (while merely an anecdote, I’ve come across more masterminds than I care to count that don’t take the level 6 and 32 upgrade powers in-game. These players are not the target in your proposal. Nor are the number of melee ATs that are running with basic IOs or a smattering of frankenslotted sets). 

 

Players who aren't pretty near softcap would almost certainly not notice this change.

 

So let's say you're at 30% defense.  You're fighting a +0 minion.  Its base chance to hit you is 20%.  If it procs the to-hit bonus, its chance to hit you is 30%.  Let's say that, because you are a less optimized player, you let a minion get four attacks off before disposing of them instead of three.  If each attack does 100 damage, without the proc you expect to take 80 damage.  With the change you expect to take 83 damage.  Yes, again, expected damage isn't really the right metric here, but just to be clear, it's a tiny change for anyone who isn't right at the softcap.

 

Which, again, was what this was actually intended to do -- not add a ton of challenge, to soften the incentive to be exactly at the softcap, no more, no less.  Give some more value to above-45% defense, give some more value to below-45% defense.

Link to comment
Share on other sites

6 hours ago, aethereal said:

So the math for this isn't so much hard as there's a lot of it.  Let's go through some of it to see how daunting it is to get a really good grasp of what's going on.

 

So first of all, chance to hit.

 


HitChance = Clamp( AccMods × Clamp( BaseHitChance + ToHitMods – DefMods ) )

Let's plug in some numbers:

 

+0 Minion fighting a character at 45 defense:

 

Accmods: Power Accuracy * 1 * 1

BaseHitChance: 50

ToHitMods: presumptively 0

DefMods: -45

 

HitChance = 5% for a power with 1.0 accuracy (note that even minions may have higher-accuracy powers so even against even-conn minions, you may actually have a higher than 5% chance of being hit at softcap)

 

+0 Minion fighting a character at 45 defense who has procced a +10% chance to hit, then, is all the same except ToHitMods is +10%, leading to 15% chance to hit.

 

+4 Boss fighting a character at 45 defense:

 

Accmods:  Power Accuracy * 1.3 * 1.4

BaseHitChance: 50

ToHitMods:  presumptively 0

DefMods: -45

 

HitChance = 9.1%

 

With +10% chance to hit, that raises to 19.1%

 

Okay, so one way to describe this change would be to say that the expected damage of an even-level minion, after the alpha strike, would be 15% higher (95% chance of 1x damage, 5% chance of 3x damage).  If a minion gets 3 total attacks off, then we'd say that their total expected damage would rise by 10% (since the first attack has no chance of proccing).  This is for a character right at softcap, exactly 45% defense.  A character with say 43% defense would experience only about a 5% expected rise in damage in the same scenario (3 total attacks), and a character with 47% defense would similarly experience about a 5% expected rise in damage.

 

A +4 Boss gets 2.12x damage against a character exactly at softcap from this change, and procs is 15% of the time, so about a 17% rise in expected damage after the first hit, and here we might reasonably say that a boss doesn't die after just a couple of attacks, so maybe we just ignore the first hit and say it's a 17% rise in expected damage altogether (for a character exactly at softcap.  Again, someone a few points of defense higher or lower will experience decreased damage changes).

 

But expected damage is maybe the wrong metric for defense calcs in general.  We plausibly care much more about what we in the software industry would call P90 or even P95 or P99 damage -- that is to say, the damage that you experience at your 10% unluckiest sets of rolls, or your 5% unluckiest, or 1% unluckiest.  That is to say, if you survive 9 spawns out of 10 fine, but the 10th spawn you run into kills you every time, my guess is probably players end up turning down difficulty there.  Do they turn down the difficulty if they die to one spawn in 20?  I'm not sure.  One spawn in 100?  Probably not, unless maybe it's a farming toon.

 

How do you calculate P90 or P95 damage?  Well, it's doable as a pure math calculation, sort of, but it's quite complex.  The way I'd do it is write a monte carlo simulation and then just order the results of a 1000 or so runs by the total damage the player experiences and choose the 100th from the end or the 50th from the end.  But what should we simulate?  Well...  shit.  I mean, really what you want to do is simulate an entire spawn.  Which isn't just generic minions or lts or bosses firing scale 1 attacks, it's...  a lot of stuff.  It's different mixes of enemy classes firing different attacks, and you also need to figure out how fast do they die.

 

So there are a lot of variables here.  A selection of them:

 

What level difference between player and critter

What mix of critter types

What attacks do the critters have

   - Do these attacks debuff defense?

   - Do these attacks have inherent accuracy or to-hit?

How long do critters of each class survive

Exactly what defense level does the player have (to each of the attacks the enemies have)

 

A really useful tool, I think, would be a simulator that allowed you to plug in even rough answers to these things.  Like where you could say, "Okay, here's a stylized +4/x8 spawn of Council, and here's some metric of how fast they'll die to the player attacks, let's run a few thousand combats."  It would be necessarily imperfect, but I think it would be a useful tool to try to attack the complexity here.  However, it'd be a hard tool to write, and it involves data that's not right at my fingertips, at least, about the details of enemy attacks.  Even if you had it, trying to really understand the spread of possible scenarios would be a big undertaking.

Close.

 

The short version? Without dumping a statistical inference model (not difficult, but it would certainly not be something most here would get and would be time consuming to produce), is that this would be a huge impact to defense. 17 MOBs with chance to proc as you have described would most certainly cause a very aggressive lowering of defense, bypass defense and render defense dependent sets at a large disadvantage. Further, AoE would be multiplicative. 

 

This would require so many dependency clauses that preclude situations not happening (such as cool down, follow up hits, con, type exclusion) that it would be highly unlikely to be helpful, resulting in a net loss in terms of time investment to outcome potential.

 

So.... No.

Link to comment
Share on other sites

2 minutes ago, SwitchFade said:

Close.

 

The short version? Without dumping a statistical inference model (not difficult, but it would certainly not be something most here would get and would be time consuming to produce), is that this would be a huge impact to defense. 17 MOBs with chance to proc as you have described would most certainly cause a very aggressive lowering of defense, bypass defense and render defense dependent sets at a large disadvantage. Further, AoE would be multiplicative. 

 

This would require so many dependency clauses that preclude situations not happening (such as cool down, follow up hits, con, type exclusion) that it would be highly unlikely to be helpful, resulting in a net loss in terms of time investment to outcome potential.

 

So.... No.

 

I can't tell if you're imagining that this change would itself debuff people's defense, or if you're just worried about the interactions of this change with already-existing defense debuffs.  I agree (and said back to someone else a page or so ago) that one possible negative effect of this would be that existing minor defense debuffs on like gun/bladed-weapon mobs that currently are too small to cause cascading failure would, with this change, become more significant and start to cause cascades (at least on non-SR sets).

Link to comment
Share on other sites

On 7/26/2020 at 11:25 PM, aethereal said:

 

It's not really that much difference because that only applies to even-conn minions, and other enemies have a higher than 5% chance to hit you at soft-cap.  But sure.  The 5% of even-conn minions who procced this to-hit bonus would have triple the chance to hit you on their second and third and maybe fourth attacks, if they survived that long.

 

If ten such minions attacked you four times each for 100 damage each, without this change you'd expect to take 200 damage (two hits).  With this change, you'd expect to take about 250 damage.

 

It's not totally insignificant, and if you were riding the ragged line of your challenge rating, you'd expect to have to lower your challenge rating.  But it's a long way from giving everyone +5% to hit, especially when you consider that most minions and lieutenants do not survive long.

Just want to note here that higher-conned enemies only get Accuracy buffs, not ToHit (up to +6, anyway).  So yes, a higher conned minion will have higher than a 5% net chance to hit you when you're at the soft cap, but its net chance to hit you will always be 10% of its initial value, absent other buffs/debuffs.  In other words, 45% DEF is always 90% mitigation against the generic (and non-Incarnate-content) NPC.

 

For example, if we give a minion a 50% Accuracy bonus, then its default chance to hit rises to 50 * 1.5 = 75%.  If the minion's target has 45% Defense, this drops to (50 - 45) * 1.5 = 7.5%, or 1/10th of the initial value.  Therefore when a player is at the soft-cap, losing 10 points of DEF results in a three-fold increase to damage taken, regardless of the enemies' relative level:  (50-35) / (50-45) = 3 = ((50-35)*1.5) / ((50-45)*1.5) = 3. 

 

We often default to discussing even-con minions for the sake of simplicity, but nothing significant changes against higher cons in the generic case. 

 

What does throw a wrench in all of our DEF discussions is that we don't often fight "the generic case."   Many NPC factions have either ToHit buffs or DEF debuffs.  Some of these effects are too small or sporadic to make a huge difference most of the time, but they can add up.  Others can be mitigated through target selection (e.g. killing Nemesis Lieutenants last, or Tarantula Mistresses first).  Others can be managed through smart positioning.  Still others are absolutely devastating and without much in the way of a useful counter (e.g. PPD peacebringers, Rularuu).  And then of course we have Incarnate content, which not only raises the soft cap by 14 points, but is also chock full of NPCs with nasty buff/debuff effects.

 

In short, there are already a lot of NPCs out there that counter DEF, to a greater or lesser degree.  Soft-capped characters can find challenge even now, and they don't have to look particularly hard for it.  (This goes double for soft-capped characters without high DEF-Debuff Resistance, i.e. pretty much all builds that derive most of their DEF from IO bonuses, which draw the most ire from forumites.) 

 

Adding another anti-DEF mechanic across the board seems excessive.

Edited by Obitus
  • Like 6
Link to comment
Share on other sites

If this idea were implemented it would make the ITF nearly impossible for a large swath of players. 

 

You can't look at it in a vacuum of ONE enemy having the proc go off. Because you are almost never fighting ONE enemy. 

 

I used the ITF as my example because it is the nightmare scenario of this idea in effect. Every enemy can proc a higher chance to hit you, and every enemy has defense debuffs in pretty much all their attacks. If your resist based tank who has gotten significant amounts of defense through pool powers and IO sets dives into a mob of Romans they're going to be faced with cascading defense failure almost immediately. 90% resistance is nice, but when you have 17 dudes beating on you with a 95% chance to hit it's going to drop you in a hurry. And that's what would happen to a TANK. Scrappers would be virtually unplayable in an ITF if enemies got random to-hit buffs. 

 

It would have the net effect of making some high level content almost unplayable and would force lower level non-minmaxed characters to play at the lowest difficulty settings all the time, especially if they're a new player who doesn't know what they'll be fighting in any given mission. 

 

Gonna have to jranger this one. 

 

 

  • Like 2
Link to comment
Share on other sites

Part of my issue here is basically if done enough to 'work' what happens is the goal posts are just moved.  Instead of 45% being the cap is folks when ever possible will now build as if the cap is 50 (with a hypothetical +5% to hit buff proc).  Anything less than treating it as a new cap and situations like @Claws and Effectdescribes can and will happen.

Edited by Doomguide2005
Typos
  • Like 3
Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...