Fixing Melee Regeneration set

[Galaxy Brain]

Awesome post @Koopak! My only caveat is when comparing the, well "comparative" value of more buffs on Regen vs the "absolute" on others. Like yes, a Regen character going from 9% to 32% resistance is a big leap compared to 23% -> 62%  (3.5x vs 2.7x), but end of the day the 62% is a much more powerful raw amount of mitigation as it relates to incoming damage. 

 

Given you have similar values for Defense in both examples we can put those aside, and lets look at say 400 incoming DPS vs those resist values after defenses:

 

400 vs 62% = 128/s

400 vs 32% = 272/s ( over 2x the incoming damage )

 

To sustain against this, Regen will need to heal >2x as much HPS as Rad to not die. Granted, Regen has more HP to chew through but if it's not regenning a constant 272hps it will eventually fall where Rad has a smaller goal to hit. 

 

In your examples, you had effective regen of 127.55 for Rad and 152.2 or Regen outside of IH which has an atrocious uptime even at the fastest which makes it unreliable... one of these sets is gonna walk away with a bigger bruise given rad will be taking an effective 0.45 dps vs Regen's 119.8 dps.

 

If we go by just their raw effective HP outside of res/def:

 

Rad = 2,865.16 / 0.45 = 6367s

Regen = 2,894.17 / 119.8 = 24.15s

 

 

Granted if both characters can "Win" the fight in 10s this doesn't really matter, but it does show a difference when it comes to the absolute values.

 

Part of the issue though is that Regen isn't really a unique thing. As shown by Rad being able to get a healthy amount of HPS, green insps can be made/queued up easily, healing procs, and even just natural regen rates (given: Rad is a bad example as it does have +Regen and self heals but still) most all characters have ways to sustain their HP that in other games would be considered a healing factor. Its not like an invincibility character in CoH can be chipped away over time while a Regen can't.

 

 

However, I 1000% agree that the way Regen works should be preserved. A set where you have to be on your toes and actively maintain your survival is super cool and unique among the cast of powers. The issue is that what it does either requires a lot of hoops that other sets do not need / get more absolute benefit from in a way that is not exactly equitable to the effort.

[Koopak]

16 minutes ago, Galaxy Brain said:

Given you have similar values for Defense in both examples we can put those aside, and lets look at say 400 incoming DPS vs those resist values after defenses:

 

400 vs 62% = 128/s

400 vs 32% = 272/s ( over 2x the incoming damage )

 

To sustain against this, Regen will need to heal >2x as much HPS as Rad to not die. Granted, Regen has more HP to chew through but if it's not regenning a constant 272hps it will eventually fall where Rad has a smaller goal to hit. 

 

In your examples, you had effective regen of 127.55 for Rad and 152.2 or Regen outside of IH which has an atrocious uptime even at the fastest which makes it unreliable... one of these sets is gonna walk away with a bigger bruise given rad will be taking an effective 0.45 dps vs Regen's 119.8 dps.

 

If we go by just their raw effective HP outside of res/def:

 

Rad = 2,865.16 / 0.45 = 6367s

Regen = 2,894.17 / 119.8 = 24.15s

You made a small error here, you used the effective regen on the effective dps, this compounds the effects of resistance inaccurately. The 127.55 ehp/s regen of rad, would be scaled against 400dp, not agains 128dps as that 128 is after resistance, and that regen is after resistances. Accurate numbers would be:

Rad
400dps vs 127.55 ehp/s

net 272.41 dps
2,865.16 / 83.22 = 34.42s


Regen
400dps vs 32% = 272 dps vs 66.66 hp/s

net 205.34 dps

2,894.17 / 205.34 = 14.09s

 

Which while far from dismantling your point, does reduce the severity of the disparity painted. I should also note that what is show here is the "immortal or dead" issue of regen. 400 dps is ALOT of incoming damage for CoH, not that such situations don't exist but its definitely in the big encounter range.

That said i realized my effective hp values are WAY off and i horribly miscalcualted so give me a bit

 

[Night]

I once tried to put forth my take on a Regeneration buff, but it's buried who knows where in this forum. And while I'm not great with numbers, I tried to adjust and add to the powers to not change their basic premise, but more adding bits to them to fit with how the game evolved after the live devs nerfed and forgot about the set. Mainly an inherent recharge resistance to a few powers, slows are a big hit against the set, and trying to incorporate some absorb as overheal, since it was incorporated well for Sentinels, and a change to Revive to be usable while alive just as Unrelenting from the Presence Pool.

 

[America's Angel]

When assessing the HPS of regen, you must always include the following:

• Moment of Glory (reduction of incoming DPS via res & def divided by recharge time)
• Reconstruction (heal divided by recharge)
• Dull Pain (heal divided by recharge)
• Dull Pain boosting native regen
• Dull Pain boosting Instant Healing +regen
• Resilience (reduction of incoming DPS via res)
• (And native regen, Fast Healing & Health, obviously.)
   

When assessing the HPS of regen, it's advised to also consider the following:

• Tough (reduction of incoming DPS via res)
• Weave (reduction of incoming DPS via def)
• Maneuvers (reduction of incoming DPS via def)
• Combat Jumping (reduction of incoming DPS via def)
• Rune of Protection  (reduction of incoming DPS via res divided by recharge)
• Unrelenting (heal divided by recharge)
• Aid Self (heal divided by recharge)
• Field Medic (+heal to the unresisted heals in the sets, divided by recharge)
   

Other things probably worth considering:

• Power Transfer Proc (x2) in the epics.
• Power Transfer Proc (x2).
• Regenerative Tissue, Numina, Panacea, and Impervius Skin procs.
• Superior Entomb proc.

Edited May 19, 2021 by America's Angel

[Koopak]

Okay I went back and cleaned up the math, I do agree with @America's Angel in that to many of these discussions, even my giant ass post, omit ALOT of details. I did try to narrow it down to the standing mechanics however, the things likely in play in 90% of fights. MoG, while excellent due to the fact that Regen can actually benefit fro ma giant resist/defense T9, is something i try not to have to account for due to the complexities of working it into the math.

At some point I need to sit down and do that, likely dividing the effective resistance (likely the 75% cap minus the resistance when tis off) across its full activation+recharge period) but that would be an even deeper dive than the one I just wrote up.

I will note that after cleaning up my math, the numbers look shaky for regen, despite my confidence I have the math right. (gonna go back and edit in the formulas real quick too)

[Koopak]

10 minutes ago, America's Angel said:

 

MoG would be:

(IncomingDPS*(Res/100))+(IncomingDPS*(Def/100))/Recharge

Its a bit more complicated than that, if yer calculating effective hp/s  the formula shifts to begin with. Personally I consider that the optimal way to quantify things because otherwise you have to assume a given dps value. More importantly though that equation doesn't account for the 15 seconds of active time. What you'd want is something like...

( ( ( HPs / ( 1 - ( 0.5 + Defense ) ) ) / ( 1 - Resistance ) ) * 15 ) / Recharge

Since that's a lot of parens thats...

HP per second, divided by the percentage chance of a hit getting through
That divided by the percentage of the damage you would take
That gives you the effective regen while MoG is up.

You then take that and times it by 15 seconds, the duration of Mog
Then finally take that value and divide it by the Recharge.

However this doesnt factor in existing resistance and defense, so you need to FURTHER modify it like so

( ( ( HPs / ( 1 - ( 0.5 + ( MoG Defense  - Base Defense ) ) ) / ( 1 - (MoG Resistance - Base Resistance) ) ) * 15 ) / Recharge

This accounts for how much MoG is ACTUALLY giving and you obviously have to clamp those values to their caps.

Edited May 19, 2021 by Koopak

[America's Angel]

26 minutes ago, Koopak said:

( ( ( HPs / ( 1 - ( 0.5 + ( MoG Defense  - Base Defense ) ) ) / ( 1 - (MoG Resistance - Base Resistance) ) ) * 15 ) / Recharge

Yeah this is similar to what I've got in my spreadsheet. (Serves me right for trying to type it up freehand from memory!)

 

Still doesn't account for autohit attacks which ignore the defense part of MoG, and it assumes all enemies have a 0.5 hit chance, but it's near enough.

[Koopak]

57 minutes ago, America's Angel said:

Yeah this is similar to what I've got in my spreadsheet.

Soooo.... tell me more about this spreadsheet....

[Galaxy Brain]

3 hours ago, America's Angel said:

When assessing the HPS of regen, you must always include the following:

• Moment of Glory (reduction of incoming DPS via res & def divided by recharge time)
• Reconstruction (heal divided by recharge)
• Dull Pain (heal divided by recharge)
• Dull Pain boosting native regen
• Dull Pain boosting Instant Healing +regen
• Resilience (reduction of incoming DPS via res)
• (And native regen, Fast Healing & Health, obviously.)

 

Agreed, tho MOG's uptime makes it very, very dubious to include. If we assume normie slotting that gives it ~1/2 rech we are looking at an averaged 8.91% Def/Res, at capped rech it is a more respectable 17.81%, tho still in reality there are really big gaps with either a 12.5% or 25% uptime. 

 

Reconstruction grants between (brute numbers) 25 ~ 50 HPS at norm/max rech (and max heal outside Heal buffs)

 

Dull Pain grants between ~40% max HP avg and 80% max HP, as well as 6.6 ~ 13.3 HPS at norm/max rech

 

Resilience lowers all damage taken by 14.86%

 

Instant Healing provides between 17.26/34.53 HPS over time, or 24.15/62.07 with Dull Pain

 

Base HPS all added is 26.78, with DP it boosts to between 37.46 / 48.14

 

 

So, added up:

 

Normal Rech Regen w Dull Pain = 93.12 HPS / 23.77 Res (except Psy) / 8.91 def (except Psy) / 2096.9 avg HP

 

Max Rech Regen w Dull Pain = 173.23 HPS / 32.67 Res / 17.81 def / 2694.8 avg HP

 

This is averaged out with all pistons going to to speak.

 

 

3 hours ago, America's Angel said:

When assessing the HPS of regen, it's advised to also consider the following:

• Tough (reduction of incoming DPS via res)
• Weave (reduction of incoming DPS via def)
• Maneuvers (reduction of incoming DPS via def)
• Combat Jumping (reduction of incoming DPS via def)
• Rune of Protection  (reduction of incoming DPS via res divided by recharge)
• Unrelenting (heal divided by recharge)
• Aid Self (heal divided by recharge)
• Field Medic (+heal to the unresisted heals in the sets, divided by recharge)
   

Other things probably worth considering:

• Power Transfer Proc (x2) in the epics.
• Power Transfer Proc (x2).
• Regenerative Tissue, Numina, Panacea, and Impervius Skin procs.
• Superior Entomb proc.

 

This is valid, though we can only fit so many into a given build. With many of the powers / bonuses giving the same thing we can honestly just slap in X% (Stat) to the equation too.

[RogerWilco]

21 hours ago, America's Angel said:

I would advise against changes to the game that can be solved with a tutorial. Regen is fine. If players don't like the reactive nature of it, they can play Willpower.

 

That said, if you're looking to tweak things, then I suggest:

• Adding [more/some] -recovery, -end, -recharge, and -regen resist.
• Have the heals activate at the start of the animation.
• Cut the MoG animation in half.
• Have the MoG Animation Time Before Effect the same duration as the Animation Time.
• Have MoG and Resilience give resist to all.

I get that minor tweaks aren't sexy/exciting to dream up and implement. But there's no real need for anything beyond this.

 

The only power it's safe to really play with is Revive. Maybe doing something like giving it a 0s recharge, removing the 15s untouchable, and having it restore lost toggles, and have it remove a deaths-worth of XP debt could be the way to go? It would suit, thematically. (Revive is a big part of Wolverine's HPS chain.) And I like the idea that regen does not experience death the same way the other armor sets do. It would reward the "knife's edge" playstyle by making death less of a burden.

I like the direction you're thinking.

 

I like the resists. One of the fights I have had the most difficulty with, is one that gave me -50% Endurance.

 

I like the MoG animation, but it makes it hard to time correctly, together with it's short duration. I don't know the details well enough, but keeping the animation would be nice.

The interesting animations is one of the things that make CoH fun to me. I will pick powers based on what looks cool, although Titan Weapons is pushing my patience.

 

I like the 15s untouchable, It allows for some interesting choices, and don't mind the cooldown on Revive. Some toggles are really slow to activate though. I have Integration, Focussed Accuracy, Manoeuvrers, but FA seems to be much slower than the other two? This means that it takes quite a while before you can rejoin the fight.

I would like to see it not have XP debt. I end up using it quite often when fighting enemies I don't know, before I am able to figure out which enemies I need Instant Healing and Dull Pain against.

 

I will stay away from discussing the math in this tread for now, as that needs a much clearer head and attention to detail. I do have the feeling that people are trying to argue based on fixed values, with a lot of assumptions, where the dependency on underlying values is a function and would be better represented with a graph or something. Like a graph of survivability vs. incoming dps.

 

I also don't know the damage and hit calculation mechanics well enough to understand if one mob doing 100 dps, or 10 mobs doing 10 dps, makes a difference. I think there are some things like the auto-hit and miss {?} mechanics that might make the math more complicated?

 

To me it seems that Regen is mostly fine, but might lack especially in its ability to tank large groups because of the way it (doesn't) scale.

Given that CoH is more focussed on making characters interesting to play than trying to balance everything, I'm quite fine with that.

I left World of Warcraft for CoH, because it was simplifying and reducing the complexity in the name of (PvP) balance.

I used to play Druid in WoW, so I seem to like "HP sponge" type tanks.

[Leo_G]

16 hours ago, Galaxy Brain said:

Don't get what you mean by player control over damage scaling, unless you mean difficulty options but that is a different thing.

 

Buffs that mitigate damage for you are an outside X factor that all sets can enjoy, its not a uniquely Regen thing. Sure, it may help Regen more comparatively but in total a more mitigative set + buffs would generally fare better.

 

The last point here is basically saying "don't play regen" 😞

 

 

Yes, I'm talking about the notoriety settings.

 

And I disagree. A team that caps def and resistance, the Regen will be better. A team that boosts rech reasonably will benefit Regen more than sets like Inv, WP or SR.

 

As for changes to Revive, I'd just add the functionality of being usable while alive.  A cool factor that might have to be adjusted is Revive instantly recharging all your Regen powers (Reconstruction, Integration, Dull Pain, Instant Healing and Moment of Glory).  If used while alive, it recharges all your regen powers but has a 3x penalty to Revive's recharge (so 900sec instead of 300sec).  If used while dead, it recharges all your regen powers with the standard Revive recharge (300sec).

Edited May 20, 2021 by Leo_G

[Galaxy Brain]

8 hours ago, RogerWilco said:

I also don't know the damage and hit calculation mechanics well enough to understand if one mob doing 100 dps, or 10 mobs doing 10 dps, makes a difference.

This is always an interesting topic. It does and it doesn't.

 

With 1 mon doing a ton of dps, odds are they are a boss and will stick around longer in a fight and thus have more opportunities to hit you and deal their big damage. The 10 x 10 crowd are likely small fry, and as you fight them their DPS plummets as each one you defeat lowers incoming damage by 10%. That said, with 10 attacks at once any debuffs they carry will stack fast and can be deadly.

[Koopak]

16 hours ago, RogerWilco said:

I will stay away from discussing the math in this tread for now, as that needs a much clearer head and attention to detail. I do have the feeling that people are trying to argue based on fixed values, with a lot of assumptions, where the dependency on underlying values is a function and would be better represented with a graph or something. Like a graph of survivability vs. incoming dps.

 

I also don't know the damage and hit calculation mechanics well enough to understand if one mob doing 100 dps, or 10 mobs doing 10 dps, makes a difference. I think there are some things like the auto-hit and miss {?} mechanics that might make the math more complicated?

This is actually why I so aggressively push the idea of EHP and EHP/s. These two values cover the entire scope of an engagement by giving you a flat value that the raw incoming dps must exceed before a character even begins to fall, and then a total that can divided by that value to give a survival duration. This greatly simplifies the conversation by eliminating assumptions.

That said some assumptions are still in play yes, namely that no auto hit powers are in play (to be fair those aren't super common else defense sets would be worthless), and that the damage is coming in a consistently, or rather, at least not in waves sufficient to take large percentages of your health all at once before your regen can work on the taken damage and even out the curve.

In the end these factors don't come into play THAT much, as you still have EHP to measure against alpha situations, and if its just a case of the mob doing 10% of your hp with every attack, then you are only losing a survival time of one attack rotation at most. That is usually below 5 seconds, and well if you aren't going to last long enough for plus or minus 5 seconds to be an issue, yer gonna have a bad times regardless.

As for the multiple mobs, that only applies to the above, a single mob doing 100 dps with a bunch of 10 damage attacks, and 10 mobs doing 10 dps each with slower attacks,

are comparable to the point of irrelevance. The only reason "one mob" might matter is because USUALLY that one mob is doing one slow big hit which introduces the above.

 

16 hours ago, Galaxy Brain said:

 

Agreed, tho MOG's uptime makes it very, very dubious to include. If we assume normie slotting that gives it ~1/2 rech we are looking at an averaged 8.91% Def/Res, at capped rech it is a more respectable 17.81%, tho still in reality there are really big gaps with either a 12.5% or 25% uptime. 

 

Reconstruction grants between (brute numbers) 25 ~ 50 HPS at norm/max rech (and max heal outside Heal buffs)

 

Dull Pain grants between ~40% max HP avg and 80% max HP, as well as 6.6 ~ 13.3 HPS at norm/max rech

 

Resilience lowers all damage taken by 14.86%

 

Instant Healing provides between 17.26/34.53 HPS over time, or 24.15/62.07 with Dull Pain

 

Base HPS all added is 26.78, with DP it boosts to between 37.46 / 48.14

 

 

So, added up:

 

Normal Rech Regen w Dull Pain = 93.12 HPS / 23.77 Res (except Psy) / 8.91 def (except Psy) / 2096.9 avg HP

 

Max Rech Regen w Dull Pain = 173.23 HPS / 32.67 Res / 17.81 def / 2694.8 avg HP

 

This is averaged out with all pistons going to to speak.

 

 

 

This is valid, though we can only fit so many into a given build. With many of the powers / bonuses giving the same thing we can honestly just slap in X% (Stat) to the equation too.

I want to note I feel those numbers for recharge are a bit off base, but it shouldn't effect your point to much.

I'll agree MoG tends to be so reactive as to make assessing it as a distributed factor is a bit dubious due to its uptime, the build i used in my example accomplishes a ~17% up time, plus or minus some depending on Force Feedback procs. I also think you just treated it as a 75% resisst 45% defense when the effective values of those will be lower.

That said, I don't think the uptime and recharge are the main reasons to not assess it, as I feel Instant Healing SHOULD be assessed in that way. The reason being IH should be used on cooldown much more often, and while MoG is easily one of the best T9 armor powers there is, its still an oft unused T9 Armor.

 

I'll try to find some time later this week to mock up a full assessment with recharges and such considered.

 

Edited May 20, 2021 by Koopak

[Galaxy Brain]

7 hours ago, Koopak said:

I want to note I feel those numbers for recharge are a bit off base, but it shouldn't effect your point to much.

I'll agree MoG tends to be so reactive as to make assessing it as a distributed factor is a bit dubious due to its uptime, the build i used in my example accomplishes a ~17% up time, plus or minus some depending on Force Feedback procs. I also think you just treated it as a 75% resisst 45% defense when the effective values of those will be lower.

That said, I don't think the uptime and recharge are the main reasons to not assess it, as I feel Instant Healing SHOULD be assessed in that way. The reason being IH should be used on cooldown much more often, and while MoG is easily one of the best T9 armor powers there is, its still an oft unused T9 Armor.

 

I'll try to find some time later this week to mock up a full assessment with recharges and such considered.

 

You know, I may have done the max rech a bit off when looking at it. But still, rech-cap values wouldn't be much higher than this.

[BrandX]

Resilience: Resist All w/ Extra Toxic Resist (What?!  I just feel Regen should be very Toxic Resist-y concept wise!)

MoG: Resist/Defense All

Integration: Throw in exotic Resists (Slow/Debuff/Whatever)

 

 

[Replacement]

On 5/19/2021 at 3:35 PM, Koopak said:

To be fair, ALOT of suggestions for "how to fix regen" center around the rough argument of "I cant react fast enough!" or "I don't like clicking defense powers, I just want to hit things!" or some variant there of. Obviously I'm strawmaning, but the point stands, a common amount of the argument that regen is somehow bad is centered on it not working the way other power sets do and some players not liking that. In those cases "just don't play regen" is a valid counter point, since I, and a non zero number of others, consider it a highly enjoyable and rewarding set of mechanics to engage with.

The issue with that is, we keep seeing "you don't like/ know how to play the set" in response to being fundamentally dissatisfied with the math.  It's not the end of world to have a linear system (healing) compete with multiplicative systems (defense and resist), but it's a lot of extra work the original devs never followed up on - being told "don't play Regen" isn't at all an honest response or even acknowledgement of what's being discussed. 

 

I like how Regen plays, with 4 non-congruous Clicks to juggle to stay alive (which is a large part of why I dislike the common suggestion to turn IH into a toggle).  

 

But gameplay and performance are not synonyms.

 

(In case it's not clear, this isn't an attack on you - just an explanation)

Edited May 21, 2021 by Replacement

[ABlueThingy]

I wouldn't change Recon like that, it works fine as it is I think.

 

Revive is the power no one takes.  Rename it to "Second Life" and make it so you can use it while alive for a big Absorb shield.  Maybe make it give you a that if you use it while you're dead.  Maybe have it auto recharge Recon or something too.

 

And throw in a ton of -regen and -rech protection. 

 

There, anyone what feels Regen is lacking can pick up "Second Life" and everyone that thinks the set is fine doesn't have to change anything.  And Regen gets it's most glaring and unnecessary holes patched.  And with the ability to soak an Alpha strike via Absorb it would be safer to port to Tanks.

 

[BrandX]

2 hours ago, ABlueThingy said:

I wouldn't change Recon like that, it works fine as it is I think.

 

Revive is the power no one takes.  Rename it to "Second Life" and make it so you can use it while alive for a big Absorb shield.  Maybe make it give you a that if you use it while you're dead.  Maybe have it auto recharge Recon or something too.

 

And throw in a ton of -regen and -rech protection. 

 

There, anyone what feels Regen is lacking can pick up "Second Life" and everyone that thinks the set is fine doesn't have to change anything.  And Regen gets it's most glaring and unnecessary holes patched.  And with the ability to soak an Alpha strike via Absorb it would be safer to port to Tanks.

 

 

Revive (and possibly all Self Rez Powers)

 

Passive

When you hit 1 HP or Going to die on hit.  You automatically regain 50% of your health (You can enhance it for better health return).  Power then goes on normal cool down (300 seconds in the case of Revive).  If dead it can be used as a click power to self rez at the same amount of health (50% unless enhanced).

 

Basically one big passive heal that activates when you're about to die!

[Galaxy Brain]

17 hours ago, Koopak said:

I want to note I feel those numbers for recharge are a bit off base, but it shouldn't effect your point to much.

 

I knew I goofed lol, I divided by 4 instead of 5 for max rech (recharge floor is 4x, so I always mirror it to 1/4th at fastest). Lets keep that result (400% added rech), and lets also toss in say... 250% rech (150% global) along with the capped 500% rech

 

New numbers:

 

+100% Normal Rech Regen w Dull Pain = 93.12 HPS / 23.77 Res (except Psy) / 8.91 def (except Psy) / 2096.9 avg HP

 

+250% Rech Regen w Dull Pain = 151.91 HPS / 30.45 Res / 15.59 def / 2545.33 avg HP

 

+400% Rech Regen w Dull Pain = 173.23 HPS / 14.86 Res  / 2694.8 avg HP 

 

+500% Max Rech Regen w Dull Pain = 218.44 HPS / 37.13 Res / 22.27 def / 2993.75 avg HP

 

 

 

 

Same but ignoring MoG since at tippity top capped rech it still has less than 1/3 uptime:

 

+100% Normal Rech Regen w Dull Pain = 93.12 HPS / 14.86 Res  / 2096.9 avg HP

 

+250% Rech Regen w Dull Pain = 151.91 HPS / 14.86 Res / 2545.33 avg HP

 

+400% Rech Regen w Dull Pain = 173.23 HPS / 14.86 Res  / 2694.8 avg HP 

 

+500% Max Rech Regen w Dull Pain = 218.44 HPS / 14.86 Res  / 2993.75 avg HP 

 

 

 

As we can see, Regen relies a LOT on Recharge for effective anti-DPS. Lets put it through the ringer against a handful of enemies though.... lets say 5th Column / Arachnos / Incarnate Banished Pantheon? 

 

Luckily I have been able to get data on the damage spread of these three groups at different intervals of difficulty:

 

 

5th column
Mob Size	x1	x4	x8
Avg DPS	52.37	256.42	481.53
Smashing	80.00%	67.00%	70.00%
Lethal	7.00%	18.00%	15.00%
Fire	2.00%	3.00%	2.00%
Cold	0.00%	0.00%	0.00%
Energy	0.00%	1.00%	1.00%
Negative	11.00%	12.00%	11.00%
Psionic	0.00%	0.00%	0.00%
Toxic	0.00%	0.00%	0.00%

 

Arachnos
Mob Size	x1	x4	x8
Avg DPS	67.3	277	554.57
Smashing	11.00%	15.00%	14.00%
Lethal	39.00%	33.00%	37.00%
Fire	11.00%	4.00%	5.00%
Cold	0.00%	0.00%	0.00%
Energy	18.00%	22.00%	23.00%
Negative	0.00%	0.00%	0.00%
Psionic	17.00%	13.00%	11.00%
Toxic	4.00%	15.00%	12.00%

 

Banished Pantheon
Mob Size	x1	x4	x8
Avg DPS	45.45	242.43	495.49
Smashing	16.00%	17.00%	20.00%
Lethal	0.00%	0.00%	0.00%
Fire	0.00%	0.00%	0.00%
Cold	32.00%	21.00%	20.00%
Energy	18.00%	29.00%	27.00%
Negative	34.00%	25.00%	26.00%
Psionic	0.00%	8.00%	8.00%
Toxic	0.00%	0.00%	0.00%

 

With this handy-dandy spread, we can calculate the survival rating of Regen at different stages of difficulty / recharge rates vs these 3 mobs. Given that MoG does not "scale" well over time, I'll omit that as basically you're gonna be adding +15sec as a flat rate when you pop it. Reconstruction's Toxic Res will also be calculated as it is relevant for Arachnos.

 

Lets start with the basic rech on 5th Column real quick to show how it adds up:

 

 

 

I have a cut-off at 300s survival time on the yellow Results row as otherwise it gets silly. As you can see, each damage type present within the mob is represented by the real-world data in how they spread. For regen this doesn't really matter except for Toxic Damage, and Psionic if we include MoG. The next step is gathering that yellow bar and looking at the "Score" it has vs that enemy type:

 

 

Looking at the Time to Defeat per difficulty rating, on the right hand side we can evaluate the performance of the set across the board. The total is just all the times added up, the avg is the average of all the times, and the Harmean is the "Harmonic Mean" or the Avg of the Avg, which boils down to an Average score with outliers taken out. Lets do this for all 3 vs the baseline 100% slotting:

 

 

 

Funny enough, it does best vs Banished Pantheon (before stuff like -Rech kick in). Now, to compare this to say.... Willpower and Invulnerability (also on Brutes), with the proper scaling of RttC and Invincibility per avg mob size (3 / 7 / 16 [10 cap]), also including the -ToHit from RttC:

 

 

 

 

 

Now lets average these results and compare:

 

In terms of overall performance, what does this mean? 

 

WP has the highest total and avg "score" on average, though keep in mind it has nothing that scales with recharge available so this will be it's static value compared to Inv and Regen which both have at least 1 power that gets better with Rech. 

 

Regen having the lowest, and Invuln the highest Harmean stats shows a key difference in how their mitigation scales tho:

 

 

(Lol Arachnos vs Invuln)

 

Invulnerability has a much smoother performance curve as difficulty (and thus damage) increases (Omitting the /1 stats as even tho Invuln has a lol vs Arachnos they're all basically immortal to it):

 

 

Its notable that Regen starts WAY higher, but then is consistently below Invuln per tier. Reminder too that this is with IH being averaged out over time. With the 100% rech slotting it only has an uptime of ~28%, which means for ~72% of the time your regenerator will perform much worse than Invuln which is more static. 

 

Lets redo this comparison with the 250% rech:

 

 

And 400%:

 

 

 

And capped 500% for good measure:

 

 

 

Sooooo, what does this show though? Regen does well when you have boatloads of Recharge? Yes, depending on where you look. In a team setting, you are likely only going to focus on the x8 values shown here (as well as at end game). This also only shows a sampling of enemy groups / etc for comparison, and to be fair we didn't give anything else to Invuln (blame time constrains to recalc it's Dull Pain with lots more rech). 

 

This said, this is all averaged out. For WP and Invuln, they are more or less always at peak-ish performance + have Debuff resistances that Regen does not have. If Regen is debuffed, or caught between clicks then the performance you see above will p l u m m e t.

 

 

 

[aethereal]

Wow!

[America's Angel]

Two main issues with the above analysis:

• MoG needs to be included. It is one of Regen's chief mitigation tools.
• Standing still in mobs as a DPS sponge is not representative of how the game is played, and is therefore not representative of how mitigation works in practice. What this means is you need to pause incoming DPS at regular intervals relative to killspeed/travel speed, in order to reflect moving between groups. (You also need to show DPS dropping as the fight continues due to enemies being killed.)

Edited May 21, 2021 by America's Angel

[Galaxy Brain]

52 minutes ago, America's Angel said:

Two main issues with the above analysis:

• MoG needs to be included. It is one of Regen's chief mitigation tools.
• Standing still in mobs as a DPS sponge is not representative of how the game is played, and is therefore not representative of how mitigation works in practice. What this means is you need to pause incoming DPS at regular intervals relative to killspeed/travel speed, in order to reflect moving between groups. (You also need to show DPS dropping as the fight continues due to enemies being killed.)

 

1) I mentioned I omitted it for a reason, given it's maximum uptime of 32% is less Strength of Will, If I include MoG I am giving WP that too. Unstoppable can even get a similar if not better uptime. The tricky bit with these types of powers (even IH) is that their low uptimes make yo-yos of performance that is not as easily calculated as things like Reconstruction or Dull Pain. In the specific case of MoG, it makes you so essentially invulnerable for 15s that I may as well add 15s x Usage to the metrics for how often it can be cycled, which in most cases is only about once. 

 

2) This was done via calculation as otherwise I am not gonna do every permutation of attack set mitigation / movement as that is lunacy. This is more a representation of how well it *could* fare over time given things being averaged per cycle / etc. In a specific moment, such as the 90s window IH is active or DP + X is active the results would be better, but throughout a mission it won't always be on, as alluded to with the gaps between fights.

 

3) You are welcome to do everything I did, it's all available to parse via in-game logs for enemy DPS, armor values over time, etc

Edited May 21, 2021 by Galaxy Brain

[Haijinx]

<.<

 

You know ... if you hard capped def at 50%, Regen wouldn't need buffed ...

 

>.>

 

/runs

 

[ScarySai]

Inb4 40% defense hardcap proposal that has nothing to do with the thread.

[America's Angel]

1 hour ago, Galaxy Brain said:

 

1) I mentioned I omitted it for a reason, given it's maximum uptime of 32% is less Strength of Will, If I include MoG I am giving WP that too. Unstoppable can even get a similar if not better uptime. The tricky bit with these types of powers (even IH) is that their low uptimes make yo-yos of performance that is not as easily calculated as things like Reconstruction or Dull Pain. In the specific case of MoG, it makes you so essentially invulnerable for 15s that I may as well add 15s x Usage to the metrics for how often it can be cycled, which in most cases is only about once. 

 

2) This was done via calculation as otherwise I am not gonna do every permutation of attack set mitigation / movement as that is lunacy. This is more a representation of how well it *could* fare over time given things being averaged per cycle / etc. In a specific moment, such as the 90s window IH is active or DP + X is active the results would be better, but throughout a mission it won't always be on, as alluded to with the gaps between fights.

 

3) You are welcome to do everything I did, it's all available to parse via in-game logs for enemy DPS, armor values over time, etc

 

1) 

Given that MoG is a huge part of Regen's mitigation, any test that doesn't include MoG it is not accurately testing the mitigation of the set. If the test is not accurately judging the mitigation of the set, then its results don't have real-world application. (Koopak linked the formula for calculating MoG HPS , above. It's easy to factor it in.)

 

Also worth pointing out that MoG benefits you not just through the additional HPS from the defense and resistance, but also through the 15s breather it gives your healing cycle. (Each time you use it you're essentially cutting 15s off-of your heals' recharge times.)

 

2)

You can calculate real-world play via calculation, you only need three variables:

1. Length of fight per group
2. Rate of incoming DPS decrease during fight with the group
3. Length of time between fights

Length of fight (1) is easy to figure out, just work out your DPS vs the enemies HP and res. (Using War Mace as a baseline because Powerhouse has said that's the best balanced melee set.)  You've already calculated incoming DPS, so the rate of DPS decrease(2) would use that relative to the length of the fight, which you just calculated. Length of time between fights(3) can just be set to 5s. Easy.

 

So hypothetically speaking lets say you have 300DPS coming your way, you can kill your enemies in 30s, every 5s the incoming DPS drops by 100, and the rate at which you can move from one group to the next is 5s.

 

This would be how the incoming DPS varies during the fight:

 

1s-5 - 300 DPS

6s-10s - 300 DPS

11s-15s - 200 DPS

16s-20s - 200 DPS

21s-25s - 100 DPS

26s-30s -  100 DPS

31s-35s - 0 DPS

36s-40s - 300 DPS

41-45s - 300 DPS

46s-50s - 200 DPS

51s-55s - 200 DPS

56s-60s - 100 DPS

61s-65s -  100 DPS

66s-70s - 0 DPS

 

...suddenly that incoming 300 is ~175.

 

3) 

Considering that the dev team look at your results when making balance decisions, I'd like to do what I can to help make them as accurate as possible. I'm more of a tester/exploit hunter than a model maker.