Arcanaville

Answering multiple requests, I've updated the Guide to Defense to version 1.2. There appears to have been a couple of minor changes to Defense since the last version of the guide

Here's a list of the major updates:

* Some improvements in the tohit formula explanations
* Specifics of the level scaler
* Specifics of the Streakbreaker
* I7 Defense Scaler
* Purple Patch discrepancies
* More accurate numbers for some things
* Some formatting changes and rearrangements
* Lots of minor fix up and confirmation changes

Of course, additional suggestions always welcome.

Defense in CoH


Defense, and its relationship to the basic tohit equations, is one of the least understood mechanisms in City of Heroes. This guide will try to explain Defense, how it functions, how it relates to tohit probabilities, and how it interacts with the other elements related to Defense in the game.


DEFINITIONS AND THE BASIC TOHIT EQUATIONS

The basic tohit formula

NetToHit = BaseToHit - Defense

where:

NetToHit: the probability that one thing will hit another thing with an attack. If net tohit is 45%, then 45% of the time when A attacks B, A will hit B.

BaseToHit: the probability, associated with the attacker that represents the base probability that attacker has of hitting any target in general, before buffs, debuffs, and defense are taken into acount.

Defense: the ability, or power, to reduce the chances of an attacker from hitting you. Defense is normally expressed in percentage points, and is the number of percentage points that the defensive ability will reduce your chances of being hit by an attacker.


The advanced tohit formula

The advanced tohit formula (my terminology - there isn't really a term for it) takes into account accuracy enhancements, tohit buffs, tohit debuffs, and defense debuffs. It is:

NetToHit = (BaseAcc + AccuBuffs - AccDebuffs) * [ BaseToHit + ToHitBuffs - ToHitDebuffs - (Defense - DefenseDebuffs) ]


Tohit buffs and defense debuffs

One way to improve your tohit chance is to use, or have cast upon you, tohit buffs. Tohit buffs are, according to the devs, additive:

BaseToHit + ToHitBuffs

So if your base tohit is 75%, and you use or receive a 60% tohit buff, your modified tohit becomes:

0.75 + 0.60 = 1.35 = 135%

Note that this is higher than 100%: see tohit floors and tohit ceilings below. Tohit buffs and tohit debuffs are subtractive from each other, as the advanced formula shows.


Defense debuffs are subtractive from defense: if you have 40% defense, and you are hit with a 10% defense debuff, your effective defense becomes: 40% - 10% = 30%.


Base Accuracy and Accuracy enhancements

Both inherent Accuracy bonuses and Accuracy enhancements are Accuracy Buffs. Accuracy Buffs work differently than tohit buffs. As shown in the formula, accuracy buffs take effect after defense, while tohit buffs take effect before defense. The difference is that tohit buffs are much more effective than accuracy enhancements when defense is high. If your tohit on a target is 30%, a 33% accuracy enhancement SO will boost that percentage to 40% (30% * 1.33) regardless of what the defense of the target was (as long as the net effect of base tohit and defense was 30%).

All attacks have what is referred to as "Base Accuracy" or sometimes just "accuracy." As defined by the devs, Base Accuracy is the "inherent accuracy" of an attack power. A "normal" attack has base accuracy of 1.0, or 100%. Attacks that are less accurate than normal have base accuracy values less than 1.0, and attacks that are more accurate than normal have base accuracy values more than 1.0.

Note on accuracy buffs and tohit buffs: it seems that the general rule is that anything that buffs (or debuffs) the accuracy of an individual power is an accuracy buff, while anything that buffs or debuffs the player - and all attacks he or she performs - is a tohit buff (or debuff).

I am currently unaware of anything that is an accuracy debuff (and not a tohit debuff) but I'm told that they theoretically exist in CoH. In practice, the term AccDebuffs is almost always zero.

"Base Accuracy" and "Base ToHit" is very frequently confused. Base ToHit represents the intrinsic accuracy of an attacker: its the chance that he or she will hit a target, in general, assuming all other factors are absent (defense, buffs, debuffs, etc). Base Accuracy represents the intrinsic accuracy of an attack relative to other attacks, and is scaled to 1.0: "normal" attacks have Base Accuracy of 1.0, which means they have no effect on the overall accuracy of the player. Attacks inherently more or less accurate have Base Accuracy values of more or less than 1.0, which increase or lower the overall accuracy of any attacker using them.

The best analogy to distinguish Base Accuracy and Base ToHit is to consider two people shooting firearms. One of those individuals might be inherently a better shot: he will have higher Base ToHit than the other. Separate from that, both of them will have different accuracies when firing snug nose revolvers and sniper rifles: the actual weapons have an intrinsic relative accuracy separate from the shooter, and thus the sniper rifle would have a higher Base Accuracy than the pistol.

(Player) Inherent Accuracy modifiers

Player attacks can have inherent accuracy-related modifiers, which potentially affect Defense sets in PvP.

Certain player attacks have certain inherent accuracy bonuses or deficits. All attacks within an offensive set that require a weapon draw (i.e. katana, assault rifle) are supposed to have an inherent tohit bonus, said to be about 5%. In addition, all snipe attacks also have a tohit bonus, in a similar range. AoE control based attacks have an accuracy penalty, but a recent post by geko stated that normal AoE attacks do not have an inherent accuracy penalty by default. The devs have stated that the archery attacks have an inherent tohit bonus higher than the standard weapon-draw bonus, but the precise bonus has not (to my knowledge) been determined.

At one time, attacks launched while flying had a significant accuracy penalty (said to be about a -50% tohit debuff). This penalty was replaced by travel power suppression when suppression was added to all travel powers.



Floors, ceilings, and caps

There is a maximum net tohit value and a minimum tohit value honored by the CoH game engine. No power or set of powers can drive your net tohit higher than 95% or lower than 5%. In other words, there is always at least a 5% chance of hitting anything, and always at least a 5% chance of missing something.

The 5% minimum chance to hit something is referred to as the tohit floor.
The 5% minimum chance to miss something, or alternatively the 95% maximum chance to hit something, is referred to as the tohit ceiling.

It used to be thought that there was no cap on the amount of defense that a hero could achieve. It turns out there is, but it is very high and not normally applicable to most reasonable combat situations: its in the range of 300% defense at high levels (it scales upward with increasing combat level of your hero).

The *true* tohit formula

Taking into account accuracy, defense, buffs, debuffs, and floors and ceilings, its this:

Bounded[ (BaseAcc + AccuBuffs - AccDebuffs) * Bounded[ (BaseToHit + ToHitBuffs - ToHitDebuffs + Defense - DefenseDebuffs) ] ]

Where Bounded[x] is the result of setting x to be 5% if x is lower than 5%, and 95% if x is higher than 95%. Notice that bounds checking is done twice: first after tohit and defense are combined (cf: the simplified tohit formula) and then again after accuracy buffs and debuffs are factored in.

(In excel terminology, Bounded(x) is MIN( MAX(x,0.05), 0.95) )


Rank and Level scaling, and the I7 Defense Scaling Update

Base tohit of villains

The base tohit of villains is as follows:

minions: 50%
LTs: 57.5%
Bosses, Snipers: 65%
Monsters, Giant Monsters, AVs, Controller Pets: 75%

These numbers are for even level villains: villains equal to your own combat level.

This increase in tohit based on the type of villain is sometimes referred to as the rank bonus or more colloquially "higher ranked villains have better tohit."


Tohit based on level

Villains of higher level than you gain tohit bonuses and become more accurate. Villains of lower level than you become less accurate. The numbers do not follow an exact scale, but have been determined by other players. It is a fairly reasonable estimate to assume that its 9% increase per level (i.e. 50% tohit at +0 becomes 50% * (1.09) = 54.5% at +1).

The actual numbers out to +4 appear to be:

+0 1.00000
+1 1.09400
+2 1.18670
+3 1.26670
+4 1.36000


Base tohit of heroes (players)

The base tohit of heroes in PvE is 75%. In PvP (arena combat and player vs player fights in PvP zones) base player tohit was recently reduced (in I6) to 50%. This improves the performance of defense sets in PvP combat substantially, although tohit buffs (being additive) are still a significant issue. Tohit buffs maybe also have been affected or reduced in I6, but this has not been confirmed. The devs have stated that high tohit buffs severely impacting defense sets is a problem they are working on a solution for.

Question: is this a "nerf?" Answer: no, its a proper balancing of defense sets. Defense sets performance were balanced against even level minions, which have a base 50% chance to hit. Furthermore, it is just as reasonable to view this as a +25% buff to player defense across the board, instead of a base -25% tohit chance.

[Note: as of 11/16/05 a patch note was added which specifically stated this exact thing.]


*** The I7 Defense Scaling Change ***

In I7, the tohit increases that higher rank and higher level villains currently get will be replaced with accuracy increases. This will act to make defense just as effective against higher ranked and higher level foes as even level minions - which is what defense is balanced around. Note: this does not mean higher rank or higher level foes will not hit any more often: they *will* hit more often, but in the same proportional way that they would hit players without any defense more often.

The rank and level accuracy buffs will be scaled in such a way so that the net tohit "works out" in such a way so that a player with no defense sees no change. How can that be done?

Basically (and the basics have been dev-confirmed) rank and level buffs will be converted this way (by example):

AV's currently are +25% tohit (75% instead of 50%) relative to minions. That will be converted to 50% higher tohit (75/50 = 1.5). That will be the new Accuracy Buff for AVs: 50%.

Note that in the simplified case, against no defense AVs used to have 75% tohit, and now they will have (1 + 0.5) * 50% = 75% - the same chance.

Level buffs will be scaled in a similar fashion: +9.4% (at +1) will be scaled to +9.4% accuracy buff instead.

Discussing all of the ramifications of the change is beyond the scope of this guide: there are discussion threads on the forums currently hashing out all of the ramifications. However, there are fundamentally only two real possibilities for how the change actually works.

Possibility #1: Tohit formula basically remains the same, but rank and level buffs get added to the AccBuff term.

Possibility #2: Tohit formula changed to be:

Bounded[ (BaseAcc + AccuBuffs - AccDebuffs) * (RankBuff) * (LevelBuff) * Bounded[ (BaseToHit + ToHitBuffs - ToHitDebuffs + Defense - DefenseDebuffs) ] ]

Note the position of the Rank and Level buffs. Possibility #2 is the tohit formula change that best represents *all* of the properties that the devs have stated for the change, but possibility #1 is numerically similar, and would be difficult to tell the difference in, except for extreme cases.

Note also that Statesman said this would work "up to +5." Its unclear if he was simply picking a number by way of example, or if the change would alter its behavior at +6 and higher. If the defense scaling change does break down at +6, the most likely way that will happen is that starting at +6, villains will begin to receive tohit increases again. It is *extremely unlikely* that at +6, villains will *suddenly* receive the +6 tohit buff; the change is likely to be less abrupt (but I have no information one way or the other).

Its important to note that the defense scaler does not affect tohit buffs in any way (directly). Tohit buffs are just as dangerous for defense sets as before. What the change does is remove the tohit increases from higher rank and higher level foes: anything that nevertheless still possesses high tohit buffs is still a major threat to defense.

Its also important to note that because it changes villain tohit/accuracy, the defense scaler does not affect player's ability to hit critters or other players, in any way.


DEFENSE MECHANICS

Typed Defense and Defense Stacking

Every attack power is classified based on how the attack is delivered, and based on the type of damage it delivers, and every defense power has an associated type or types that represents what types of attacks that defense power is effective against. There are two basic classes of attack types, and many subclasses.

Positional, or attack vector classes

Every attack is classified as either a melee attack, a ranged attack, or an AoE (area of effect) attack. In general, melee attacks are attacks that are limited to melee range (typically 5 to 10 feet maximum). Ranged attacks are attacks that will work to larger ranges. AoE attacks are defined as attacks that affect multiple targets (AoE attacks are typically defined as either cones or (general)AoEs, but that is not generally important to the issue of defense). All attacks in CoH are classified as exactly one of these types: no more and no less (although there is a special case that muddies this a bit: autohitting attacks - see below).

There is a special case issue that comes up with these basic definitions: melee cones and PBAoEs (point-blank area of effect attacks). It seems that in general, melee cones and many PBAoEs are generally considered melee attacks, but other PBAoEs (i.e. PBAoEs of giant monsters) are considered AoE attacks. There is apparently no sure-fire rule guaranteed to predict what a melee-ranged AoE will be classified as, although if its net range is less than 10 feet, it is very likely to be classified as a melee attack for the purposes of defense.

Damage-type classes

Every attack has one or more (usually no more than two) components of damage: each component has a particular type of damage associated with it. The damage types in CoH are:

smashing, lethal, fire, cold, energy, negative energy, toxic, psionic, and untyped

An attack can be all of one type: electric attacks are generally all energy damage. An attack can have multiple damage components: most energy blast attacks have smashing damage and energy damage.

Damage-types are often thought of as coming in pairs: smash/lethal (physical damage), fire/cold (elemental damage), and energy/negative (energy damage), because damage resistances are often organized that way. Toxic and psi are both considered special cases (see the special note on toxic defense below).

*** SPECIAL NOTE ON TOXIC DEFENSE ***

If a power is listed as defense to All but psi, then (as of this writing) that does not include toxic. The explanation is long and historical, but there is no toxic-specific defense in CoH. This does not mean that no power provides such defense, but rather that no such defense can exist due to a complication in how defense was originally designed. If a power is listed as "melee defense" this implicitly works against all melee attacks, even ones with toxic damage, but a power listed as "all damage types" does not include toxic.

Untyped is another special case: there used to be a significant amount of "untyped" damage in City of Heroes, most of which eventually became toxic damage. The main source of untyped damage left in CoH is the damage dealt by Hamidon, and the Hamidon Mitochondria. Its unclear precisely how untyped damage works, and its still heavily debated - the debate is mostly moot given the fact that it now mainly exists only as a singular special case that makes it difficult to generalize. We'll ignore untyped damage in this guide unless specifically mentioned.

Defense types

Just as every attack is classified as melee, ranged, or AoE, and also, smashing, lethal, etc, each defense power is classified based on what type(s) of attacks the defense power is effective against. In City of Heroes, nearly all defense powers are either attack-vector typed (also referred to as positionally typed, or ranged-typed), or damage-typed, but not both. Thus, a particular defense power might be effective against melee, or melee and ranged, but not melee and fire, for example. The one exception appears to be Force Field bubbles, and these were recently changed to provide multiply-typed defense (damage typed and also vector typed).

For more information on the specifics of defense within particular power sets, see the section DEFENSE IN POWER SETS below.

Defense Stacking Rules

If you have multiple defenses running (either your own powers or defense buffs cast on you by other players), certain defenses stack. Defense in CoH stacks additively, which is to say, if you have Defense A, and Defense B, and they stack, your net defense is A+B.

Which defenses stack and which do not is slightly tricky. Fundamentally, the following is true:

* All defenses of exactly the same class stack (melee stacks with melee, fire with fire, etc).
* A single defense that protects against multiple classes of attack functions like multiple defenses, each of which protects against a single one.
* You are only allowed to use the best defense you have against an attack with multiple classes.

For example, if you are attacked with a power bolt (from the energy blast set), that attack is ranged, and has smashing and energy damage associated with it. You are only allowed to use the best of your net ranged defense, your net smashing defense, or your net energy defense.

There was an issue a while ago in which the game engine was considering, say, someone with smashing defense and energy defense to get smash+energy defense against energy attacks with both smashing and energy components. This was considered a bug by the devs and corrected.

Defense enhancements

Each defensive power can be enhanced using defense enhancements. Defense enhancements (along with resistance enhancements) are one of the few enhancements that do not follow the general 8.33%/16.7%/33.3% TO/DO/SO (training, dual origin, single origin) enhancement progression. Defense enhancements are worth 5% for training, 10% for dual origin, and 20% for single origin enhancements.

The way enhancements work in defense powers (and in powers in general) is that the power has a base defensive value. Enhancements increase that value by a percentage amount equal to their value. To be precise: if a defense power's base value is +5% defense, and an (even level) defense SO is slotted into it, the defense power's new enhanced value is 5% * (1 + 0.2) = 6%: the power is increased in value by 20% (and not 20 percentage points, which would be 5% + 20% = 25%).

Defense enhancements themselves vary in strength based on your hero's level relative to the enhancement: enhancements are 10% weaker than their base value for every level lower than your combat level they are, up to three levels lower, where they are 70% of effective strength (they are worthless if you are more than three levels higher). Conversely, they are 5% stronger for each level higher than your level they are, up to 15% stronger when they are three levels higher than you are (enhancements more than three levels higher than your character's level cannot be slotted). For example, a -2 defense SO (normally +20%) is +16% (20% * 0.8) while a +3 defense SO is +23% (20% * 1.15).


Inspirations

Luck inspirations appear to be +Defense to all, and additively stack with any other defense powers you might be running. This has not, to my knowledge, been conclusively tested, but appears to be either explicitly true, or very near true.

Insights, it has been confirmed, work like tohit buffs: they are exactly the opposite of lucks in effect.


***Enhancement Diversification***

New to I6 is a change in how enhancements work called "enhancement diversification." Basically, it works like this for defense SOs: however you slot, and whatever you slot, you get the first 40% benefit (i.e. 2 even level SOs) at full strength, any benefit above 40% and below 60% at 0.85 (85%) of their value, and any benefit above 60% at only 0.15 (15%) of their value.

This is tricky, so an example should help illustrate what's going on. You slot (presume even level SOs) one defense SO. Defense SOs provide 20% benefit, so you get 20% bonus to the defense power you slotted it in. The second SO adds 20%: you now have 40%. The next SO provides benefit above 40% (and below 60%), and so you get 85% of its value. 85% of 20% is 17% (0.85 * 20%). So SO #3 adds 17%, not 20%, and your net benefit is now 57%, not 60%. The fourth SO you slot is providing benefit above 60% (note, this is calculated based on the raw values of the SOs, not the reduced value). So you only get 0.15 (15%) of its value. 15% of 20% is 3%. So you now have 60% total benefit, instead of 80%. The fifth and sixth SOs would similarly provide 15% of the SO strength, so SO #5 brings you to 63%, and #6 brings you to 66%.

Notice the extremely sharp cut off in benefit after enhancement #3. The basic rule on ED is: do not slot more than three SOs *worth* of enhancements. Its based on *benefit* and not on the little round thing you slot. So if you slot three SOs, and then one DO, the DO gets hit by ED. If you slot 3 HOs (Hamidon enhancements) of */*/defense and then an SO of defense, the SO gets hit by ED, even though "I didn't slot more than three of anything."

What matters is *benefit* - you simply aren't going to get much more than 3 SOs *worth* of benefit on anything, no matter what crazy combination of enhancements you try to use to dodge it.

Other enhancements, such as accuracy and damage, follow a different, but proportional scale (i.e. 3 SOs worth is where ED kicks in, even though 3 SOs of damage is about +100%, and not +60%).

I6 introduces something else: you can now tell *exactly* how much benefit your powers are getting from enhancements, by hovering your mouse cursor over the blue bar of the power in the enhancement screen. Also, you can "hover" (without dropping) an enhancement over a slot, and see in a popup window what the net overall change to the attributes will be if you chose to slot there.

One final note on I6 change to enhancements: tohit buff enhancements were originally schedule A enhancements (i.e. one even SO was worth +33% buff). They are now schedule B enhancements (i.e. one even SO is worth +20% buff) just like defense enhancements are.


Resistance to defense debuffs

As of the writing of this guide, supposedly resistance to defense debuffs have been added to hero and villain sets. The powers themselves now state that they include such resistance in the in-game power descriptions. Castle posted that to the best of his knowledge, the debuff resistance is in. However, careful testing has not demonstrated conclusively that they are working properly, and several tests appear to suggest that it isn't working correctly, or is working at a much lower level than originally specified. The thread where the concept of resistance to defense debuffs was originally put forth by Statesman is here. Castle has also stated that the mechanism might not be as simple as we were lead to believe: sometimes the resistance reduces the duration of the debuff, and sometimes the magnitude (but most of the time, the magnitude - see this post for more info. I believe the jury is still out on whether or not this effect is really working properly.


THINGS RELATED TO DEFENSE

What is mez defense?

"Mez defense" is the generic term sometimes used to refer to powers that protect against mez. (Using terminology originally used by geko when explaining mez) there are two types of mez "defense" :mez protection and mez resistance. Neither of these is directly related to Defense in terms of damage mitigation, but its worth reviewing.

The basics of mez are: everyone has a threshold that mez effects must break through in order for the mez effect to take hold. Without any mez defense, everyone has a base mez level of -1. All mez powers have a mez magnitude. When a mez power lands, it adds its magnitude to your mez level. A hero with mez level of -1 that gets hit by a magnitude 3 hold has mez level of 2 (-1 + 3). Any mez level higher than zero means the target is mezzed. Mez protection continuously subtracts its associated defense magnitude from your mez level while the power is running. Someone running a mez defense power with mez protection magnitude 10 has a mez level of -11 (-1 - 10). If hit with a mag 3 hold, mez level increases to -8 (-11 + 3). It would take 3 more such holds for the mez level to reach +1.

Mez effects last for a certain period of time, then expire. Mez resistance allows a target to shake off mez effects faster. So instead of a mez effect lasting ten seconds, it might last eight.

Mez protection and mez resistance are not true Defense or Resistance, but its useful to understand and is often confused with true Defense and Resistance.

All mez protection powers in melee defense sets scale up with level, with tankers getting maximum protection at level 35, and scrappers at level 45.

*** Note: in I6 mez protection powers were reduced from their previous levels. Maximum protection for tanks and scrappers used to be about magnitude 15, which in effect means controllers needed 6 holds to break protection. In I6, this has been tested to happen at 3 or 4 holds, which implies mez protection has been roughly cut in half.



DEFENSE, ACCURACY, AND VILLAINS

Some villains possess tohit buffs (either inherent ones, like those attributed to rank and level, or power-based ones, like Behemoths that use invincibility), and some behave like their attacks are slotted with accuracy enhancements. Rularuu Watchers appear to have significantly higher than normal base tohit (the precise value is unknown to me). Malta gunslingers have an accuracy buff instead of a tohit buff on their pistol's cone attack. Its been approximately measured as about +65% - comparable to two accuracy SOs of accuracy boost.

Also interesting: Paragon Protectors that use MoG have *massively* higher defense than the ones that (apparently) use Elude. Its unclear precisely why the large difference exists.

Villains seem, over time, to be acquiring defense powers and abilities. That could be a large source of people believing that "accuracy" has been reduced, when accuracy is unchanged, but the defensive capabilities of the villains has improved.


DEFENSE IN POWER SETS

Its important to note that the information related to Defense in the printed manuals is, as with all things, both dated and often inaccurate. Again: this guide is not focused on the numbers, but as this information appears to be difficult to find, power set-specific Defense issues (especially what stacks with what) are listed here. Note: just as the manual is out of date, so to this guide might be out of date at the time its read. Force Fields, for example, had positional defense added literally a few weeks before this guide was finalized. For specific details, numbers, and other set information, consult the links provided at the end of this guide.

Super Reflexes defenses (scrapper and stalker) are all positional or ranged-typed. Every SR defense power is effective against one attack vector only: melee (Focused Fighting, Dodge), ranged (Focused Senses, agile), or AoE (lucky, Evasion). The exception is Elude, which is effective against all vectors. Because SR defenses are typed with positional types, SR defenses do not stack with any defense claiming to defend against a particular damage type or types. However, SR defenses do work against attacks that do toxic damage, because positional defenses work against all attacks within their range band, irrespective of damage type. SR defenses also stack with all power pool defenses, because all power pool defenses are melee/ranged (and therefore positional). However, SR defenses do not appear to stack with defense buffs such as fortitude (which appears to be damage-typed), and until recently they did not stack with force field bubbles (FF was recently changed to add positional typing on top of damage-typing to address this).

Ice Defenses are all damage-typed. Ice defense powers are generally typed against two damage types (as is generally true for many damage-typed resistances). Ice Defenses do not stack with power pool defenses, because all power pool defenses are positional, and not damage-typed. However, Ice defenses do work on any attack that has a component of damage within the defensive scope. For example, Frozen Armor provides smashing/lethal defense. Glacial Armor provides energy/negative defense. If attacked with an energy blast attack that does smash/energy, both defenses potentially apply. As with all damage-typed defenses that overlap, Ice tanks will always use the greater of the two - they do not stack together. Ice has one of only two "scalable" defenses in the game: energy absorption is a click power that boosts Ice tanker defenses based on the number of villains it hits with a PBAoE "attack." For more information, consult the links at the bottom of this guide.

Granite Armor has a power that functions differently from the printed manual. Rock Armor provides Defense, not Resistance. Granite Armor has four defense powers. Three are stackable defenses (in the sense that they can be run simultaneously - they do not stack defensively with each other): Rock Armor (smash/lethal), Crystal Armor (energy/negative), and Mineral Armor (psionic). One cannot be used with the others: Granite Armor, which has defense to all but psi (as well as resistance to all but psi). Granite Armors, like Ice Armors, do not stack with power pools (which are melee/ranged).

Force Fields used to be damage-typed. They are now both damage-typed and positional typed. Specifically, Deflection Field provides both smash/lethal defense, and melee defense. Insulation Field provides both fire/cold/energy/negative (energy/elemental) and ranged defense. Dispersion Bubble provides both defense to all damage types except psi (and of course, toxic - toxic-typed specific defense does not exist) and AoE defense. This means Force Fields will stack with anything (specifically, the right bubble will stack with any conceivable defense). It also means - although I have not specifically tested this - that Force Fields now implicitly protect against toxic attacks (since Deflection Field *should* protect against melee-based toxic, Insulation Field should protect against ranged toxic, and Dispersion Bubble should protect against AoE toxic attacks - which are essentially all of them).

Of particular interest to Force Field defenders and controllers is the fact that there are unusal side-effects due to the (new) way the various bubbles' defenses stack. In particular, dispersion bubble stacks with deflection shield by smash/lethal typing only and dispersion bubble stacks with insulation shield by energy/elemental only. This means if the FF defender or controller takes another defense buffing aura power - specifically maneuvers - that stacks positionally, instead of damage-typed, dispersion bubble and maneuvers won't stack with each other, even though each will stack with the smaller bubbles.

The net result is that because dispersion bubble is likely to be stronger than maneuvers, the net effect is for maneuvers to have no effect on team mates within the range of dispersion bubble (except for toxic/psi attacks, since dispersion bubble does not offer either toxic or psi defense specifically by type).

Invulnerability has two defense powers: invincibility and tough hide. It now appears (as of this writing) that invincibility (and tough hide) both provide defense to all (damage types) but psi (and not melee/ranged). At one time invincibility was thought to provide melee/ranged defense (and its possible it did, and that was changed recently). Invincibility is the other scalable defense that exists in CoH (the other being energy absorption). Invincibility, unlike EA, is a PBAoE aura that continuously surrounds the hero while its on, and buffs the defense of the hero using it based on the number of attackers that are in melee range. Its actual internal workings are quite complex and still subject to active discussion. Tough Hide is also defense to all but psi. Interestingly, Invulnerability also has a self defense debuff. Unyielding (the power originally called Unyielding Stance) originally rooted you to the ground when activated. It now has a self defense debuff, of about -5%. This defense debuff appears to be a -DEF to all attacks, but I do not know if this has been carefully tested.


Stalkers

All stalkers have a power called hide. Hide appears to offer defense to melee/ranged/AoE. The defense appears to be about 5% to melee/ranged, and 37.5% to AoE (that's not a misprint: thirty seven point five percent) when hidden, and about 2.5% to melee/ranged/AoE when hide is suppressed (the 5%/2.5% number is one of the numbers I've seen: there have been lots of other numbers quoted, from 5%/2.5% up to 7.5%/3.75%. I cannot say with certainty what the precise value is. The 37.5% AoE defense, however, has been red name confirmed directly). Hide also provides the highest -perception (i.e. stealth) of any power, and while hidden stalker attacks critical (double damage) and assassin's strike powers do six times bonus critical damage.

The Ninjitsu stalker set has positional defenses similar to SR. Ninja Reflexes is similar to Focused Fighting (melee), and Danger Sense is similar to Focused Senses, but it has both ranged and AoE defense.

The Energy Aura stalker set has damage-typed defenses. Supposedly, the energy aura version of hide offers defense to all but psi, instead of the positional hide everyone else has, but I haven't confirmed that yet (if energy aura has the same hide as everyone else does, it wouldn't stack with its own defenses because hide would be positional, and energy's defenses would be damage-typed). Kinetic Shield offers defense to smashing, lethal, and (to a lesser extent) energy. Power Shield offers defense to fire, cold, energy, and negative. Overload offers defense to all damage types except psi (remember, "all but psi" excludes toxic) [note: Overload also has a dull pain component].



*** New for I5/I6 ***
Power Pool defenses are now supposed to offer defense to all, to guarantee that they stack appropriately with any defense that might be possessed by a hero/villain from their primary and secondary sets. This change was made to ensure that power pool defenses did not discriminate for or against any particular defense sets. Originally, most power pool defenses offered melee/ranged defense, and for a short while power pools offered melee/ranged and smash/lethal to try to address some stacking issues. They were changed to defense to all when it became clear that limited typing was not going to fully address the stacking issues, and was going to make stacking highly complex.

Power pool powers with defense components:

Concealment/Stealth
Concealment/Grant Invisibility
Concealment/Invisibility
Fighting/Weave
Flight/Hover
Leadership/Maneuvers
Leadership/Vengeance
Leaping/Combat Jumping



What other powers provide defense?

The following additional powers provide defense:

Devices/Cloaking Device (melee/ranged)
Illusion Control/Superior Invisibility (melee/ranged)
Illusion Control/Group Invisbility (melee/ranged)
Dark Miasma/Shadow Fall (melee/ranged)
Empathy/Fortitude (apparently all damage types)
Storm Summoning/Steamy Mist (melee/ranged)
Dark Armor/Cloak of Darkness (melee/ranged)
Regeneration/MoG (all but psi)
Cold Mastery/Frozen Armor (smash/lethal) [note: this power also has cold resistance]
Force Mastery/Personal Force Field (base defense)
Warshade/Shadow Cloak (melee/ranged ?)
Katana/Divine Avalanche (melee/lethal)
Broadsword/Parry (melee/lethal)

Note: Hasten used to have defense; it was removed in I5


Special Note on Stealth

Stealth powers generally break their concealment component when you either attack or are attacked. When the stealth is broken, most stealth powers that have a defense buff component will have about half their defense also suppressed while the stealth component is broken.

The following stealth powers appear to suppress a portion of their defense when the stealth is broken:

Devices/Cloaking Device
Illusion Control/Superior Invisibility
Illusion Control/Group Invisbility
Ice Control/Arctic Air
Concealment/Stealth
Concealment/Grant Invisibility
Concealment/Invisibility


According to Statesman, stealth powers in "Primary Defensive Sets" do not suppress their stealth when concealment is broken. The following stealth powers appear to not suppress any of their defense even if concealment is broken.

Dark Miasma/Shadow Fall
Storm Summoning/Steamy Mist
Dark Armor/Cloak of Darkness
Warshade/Shadow Cloak


Special Note on Power Boost

The power Power Boost (both the blaster energy manipulation version, and the epic power pool version) boosts defense powers while power boost is active. The boost is equal to the base value of the defense power being boosted. For example, if you have hover running (2.5% defense) and you trigger power boost, hover gains 2.5% additional defense. If hover was 5-slotted with defense SOs (net 5% defense) the boost would still be 2.5% (to 7.5% total defense).


What is the Streak Breaker?

The streak breaker is a bit of code within the tohit calculator that is designed to prevent very long strings of misses. There is a lot of misunderstanding about how the streak breaker works, so I'm going to be very specific in terms of detailing how I know what I know about the streak breaker.

First, the streak breaker only breaks streaks of misses, not hits. Confirmed by my own testing, dev postings, and red name PMs.

Second, the streak breaker affects both heroes and villains. Confirmed by my own testing, dev postings, and red name PMs.

Third, the streak breaker "decides" to break a string of misses when the string of misses exceeds a particular value. That value is dependent on the tohit probability between the attacker and the target. Here is Weirdbeard's specific statements on how the streakbreaker works:

[ QUOTE ]

Final to-hit : misses allowed
>.9 : 1
.8-.9 : 2
.6-.8 : 3
.4-.6 : 4
.3-.4 : 6
.2-.3 : 8
0 -.2 : 100

Auto-hit powers are not included in the system.

Critters get the benefits of the system as well.

The system does not track each power individually; instead it tracks every miss you make in a row, regardless of power (or target). Otherwise you could have nine different powers, each with a 0.95 to-hit, and if you executed them all in a row you could miss each attack (note a caveat at the bottom of the post regarding this).

AE attacks are considered distinct sequential attacks on indivudual targets for the purpose of the system (so if you AEd two targets and had 0.95 to-hit for both, you be guaranteed to hit one of them).

To determine the to-hit used in the table above, you take either the current to-hit, or the worst to-hit in your current miss series, whichever is lower.


[/ QUOTE ]

This basically matches all the testing I've done to measure the streakbreaker, correcting for some errors in my testing methodology that Weirdbeard was able to detect in my discussions with him.


DEFENSE ISSUES

These are some of the issues related to how defense and tohit works in City of Heroes


Autohitting attacks

There are attacks that automatically hit, bypassing the tohit floors and ceilings. Typically, these things are damage auras, such as the aura emitted by Circle of Thorns Death Mages, or patches, such as the damage due to caltrops. No amount of defense reduces the damage of autohitting attacks. Note: some people used to think burn (firey aura) was autohit, but in actual fact it simply has a very high accuracy.

*** Update from version 1.1 ***

It seems that autohitting attacks are being slowly removed from CoH, to address this issue. In fact, it appears that the damage aura from Death Mages is now considered an AoE attack, defendable with AoE defense. This agrees with dev statements that autohitting *damage* (but not necessarily autohitting debuffs) were being toned down or removed from CoH in the long run.


Special Note on defense debuffs

Although defense debuffs were covered earlier, its important to note that the subtractive nature of defense debuffs makes them extremely dangerous. Up to the writing of this guide, defense sets did not have any resistance to defense debuffs (such resistance is currently being added in some form). Their only means of defending against them was defense itself. This creates a problem whereby any defense debuff that manages to land decreases defense and makes the hero both more vulnerable to damage, and more vulnerable to more defense debuffs - a spiralling downward situation.

This is significant because resistance does not work that way. All resistance powers have an inherent resistance to resistance debuffs. When someone with 40% defense is hit with a 10% defense debuff, defense is reduced to 30%. When someone with 40% resistance is hit with a 10% resistance debuff, 40% of the debuff is resisted, and actual damage resistance drops to 34%, not 30%. Furthermore, the resistance to debuffs remains 40%. If hit with another 10% resistance debuff, resistance drops to 28%, not 20% (like defense would be) and not 27.4%, which would be the case if resistance was truely dropped to 34%.

*** Update from version 1.1 ***

There are more specific statements about Defense Debuff Resistance, which are covered in this guide. However, I have yet to be able to construct a test to measure the Defense Debuff Resistance, nor has anyone else posted tests confirming its effect (that I'm aware of). So I think the jury is still out here.


Quartz eminators, quicksand patches

Quicksand patches are autohitting slow and defense debuff patches. These were highly lethal to defense sets, because their defense debuffs couldn't be defended against or otherwise avoided, and once hit, the slow made it difficult to escape (Super Reflexes has a resistance to slow, but it didn't fully mitigate the -fly -jump which could trap a scrapper between villains and friends alike, and it didn't necessarily allow for quick escapes from the patch). Quicksand was also spammed by Earth thorn casters - a CoT minion - in CoT missions from levels 35 to 39. Although this was supposedly fixed (by lessening the frequency of earth thorn casters as well as reducing their propensity to cast quicksand) its still an example of a highly powerful defense-unfriendly power that has few analogs for resistance or regeneration.

Quartz eminators - the eminators dropped by DE LTs - is even more exceptional. Quartz eminators emit a tohit buff to all DE within its buff radius. The tohit buff eminated from quartz eminators is extremely large - by some estimates several hundred percent. To put Quartz eminators into perspective, I3 SR scrappers running perma-elude and the toggles combines were running with more than 150% defense - and still being easily hit by Quartz-eminator buffed DE minions. Once again, there is no analog to the quartz eminator for any other form of damage mitigation, such as resistance and regeneration.


Team scalers and difficulty sliders

Important to note for defense sets: the difficulty slider (also known as the reputation slider) increases the level of villains within your missions, and therefore increases the base tohit of those villains (it doesn't generally increase the ranks of villains, except for the fact that heroic suppresses bosses). The team scaler increases the difficulty of missions based on the number of heroes on the team, and it increases rank and level and numbers of villains. With much lower defenses in I5 than earlier issues of CoH, high level missions can be less than friendly to defense-oriented sets, moreso than other damage mitigation sets.

*** Update from 1.1 ***

In I7, this problem should vanish when tohit increases for both rank *and* level basically go away.


Is Defense really inferior to Resistance?

Not especially. Defense and Resistance both have pros and cons in terms of their inherent effects. Defense's main problems are three-fold:

1. There are sets that rely heavily on Defense, but most other protection sets do not singularly rely on a single mitigation effect.

This is not a critical issue, but it amplifies the others.

2. Defense is - in the opinion of many - scaled too low.

The argument goes that because Defense avoids status effects, Defense has an inherent advantage that more than balances the fact that the damage mitigation of sets that rely on Defense is significantly lower than other sets. Most testing, analysis, and review of a transparent nature (i.e. open to review) suggests this is false. The devs, who do not generally reveal their own analysis, testing, or reviews, disagree.

3. What Defense is most vulnerable to, is plentiful in the CoH environment.

The most common secondary effect in CoH besides DoT (damage over time) is defense debuff. Defense debuffs are more common than resistance debuffs and regeneration debuffs combined. And Defense debuffs are undoubtably more dangerous to Defense sets than resistance debuffs and regeneration debuffs are to resistance and regeneration sets, respectively (regeneration debuffs would be significantly more dangerous to regeneration sets if they prevented things such as reconstruction and dull pain from functioning). Defense is also vulnerable to tohit buffs, and every single villain of higher rank than minion, and every single villain higher in level than even con, has an effective tohit buff.

To say that Defense is inferior to Resistance, given the large environmental disadvantages that Defense faces in CoH, would be comparable to changing all the damage dealth by villains to toxic and psi, and then claiming that Resistance was inferior to Defense.

What's up with tohit buffs?

Good question. Very high tohit buffs are, at least, uncommon in PvE. They are very common in PvP, because high order tohit buffs are extremely common in player power sets.

The two most common tohit buffs are build up and Aim, and both are high order tohit buff (Build Up is a 60% tohit buff, and Aim is a 100% tohit buff). Virtually all blasters, most defenders, almost all scrappers, and most tankers have access to either Build Up or Aim, and many blasters have access to both. Only controllers as a class lack BU or Aim (and pets have a tohit bonus).

If you are relying on defense in the arena, here's the score. If you have SR or Ice, and a couple stacked bubbles, someone who elected to 6-slot Aim with tohit buff enhancements to kill defense sets will hit you no matter what defense level you think you have. Realistically, that one power, and 5 extra enhancement slots, can effectively nullify an entire team's worth of defense buffs (7 stacked bubbles will beat Aim, of course, but all reasonable levels of defense and most unreasonable ones are going to be beat by 6-slot Aim). Without significant buffs, anyone with either build up or Aim will hit you.

It is unclear why Defense was lowered as part of the Global Defense reductions in I5, but tohit buffs were (apparently) not. If they were, this fact was not reported, nor has it shown up yet in anyone's testing.

*** New for I6 ***

One change made to tohit buffs in relation to balancing them with defense is that tohit buff enhancements are now schedule B (like defense: +20% for even SO) instead of schedule A (like damage: +33% for even SO). This at least places tohit buff enhancement and defense enhancement on a relatively even footing, although for high tohit buff powers, the net benefit of even an equal strength SO will be higher than a similar enhancement in a lower numerical strength defense power.


UNANSWERED QUESTIONS ABOUT DEFENSE

(At least, I don't know the answers)

1. When a click defense is triggered, does the defense stack immediately, or only after the activation is completed, or something in-between?

Although I never fully tested this (and now its almost impossible to do so) I (and many other SR scrappers) experienced an alarming sense that we were being hit in mid-backflip while cycling elude more often than chance would suggest, even though the protection of elude should have been fairly continuous. It was conjectured that when elude was cycled, the original cast of elude was momentarily dropped, and the new cast of elude was significantly (by a second or two) delayed. This would have made elude comparable to powers like grant invisibility, which if you refresh it, causes the targetted player to become momentarily visible again.

Whether this is true, and how this affects other defense powers, like parry/DA, is unclear.

2. Just exactly how does invincibility work?

Much more industrious people than me are continuing to investigate invincibility, most recently Stargazer. Invincibility was originally thought to have a base defense, plus an additional amount of melee and ranged defense per villain in melee range. Havok concluded that the original belief that invincibility had a base defense was false, and attempted to correct that mistaken belief. Much more recently, Stargazer seems to have done fairly convincing tests that lead one to believe that invincibility is not offering melee/ranged defense, but rather damage-typed all except psi defense. Whether invincibility was always like this, or changed to be this, is not clear to me, given the complex history of invincibility testing.

Additionally, further testing by others have hinted that invincibility might be offering twice the defense the developers quote for it because (like all auras) it "pulses" to generate its effect, and the pulses might be coming twice as fast as the actual pulse duration, in effect causes invincibility to stack with itself.

Testing of invincibility might be the longest running attempt to define how a power works in CoH by the player community.

3. Are hits and misses "streaky?" Is the random number generator in City of Heroes "broken?"

Its possible the random number generator has some sort of flaw, but in my opinion, whatever flaws it has, they are unlikely to be causing major problems in the game. However, its possible there are other systematic errors in the game related to how random numbers are actually used. There are some instances where it is blatantly obvious that the tohit calculators are doing something weird, but across a wide range of other cases, the randomness of hits and misses appears to be fairly random. Its important to note that "random" does not mean "not streaky." True random numbers are inherently streaky to a degree: the question is whether or not the hits and misses in CoH obey statistical norms of streakiness. This is still an open question, because its such a difficult thing to test for and because few people are able to test it precisely.

(New)4. How do radiation attacks work? How exactly do they "bypass defense?"

The most logical way for radiation attacks to "bypass" defense is for them to have inherent tohit buffs. If that is the case, radiation attacks would be the only exception I am aware of to the rule that *attack specific* accuracy increases are accuracy buffs, not tohit buffs.


THE CANONICAL LIST OF DEFENSE-RELATED COMPLAINTS REGULARLY DISCUSSED ON THE FORUMS

In no particular order (and without commenting on validity):

* SR underperforms other scrapper sets
* SR is a "one trick pony" that has only defense
* Ice tanks uderperform other tanker sets
* Ice tanks performance is too similar to SR scrappers for a tanker
* High defense is too frustrating in the arena
* Low defense is too frustrating in the arena
* There are too many defense debuffs in the game
* Defense debuffs are too strong
* Tohit buffs in the arena are too strong
* Defense requires you to be lucky
* Defense is inferior to Resistance in all respects
* The SR set is too reliant on power pool defenses
* The Force Fields set is insufficiently strong as a buff set
* Resistance buffs are more appreciated than Defense buffs
* The SR (and to a lesser degree Ice) set can be too easily simulated with a few luck inspirations
* Lucky and Evasion are in the wrong order in the SR set
* Invincibility is too powerful a defense power for Invulnerability given that it can outperform the supposedly "defense-oriented sets"
* The -DEF in Unyielding should be removed given the overall reductions to the invuln set
* SR passive defenses are too inefficient to slot
* There are too many autohitting attacks
* There shouldn't exist autohitting defense debuffs
* Quartz eminators
* Defense stacking is too complicated, unfair to some sets, and creates problems in improperly selecting and slotting certain defense powers

Be forewarned: this stuff has been debated to death. Also, while I strongly encourage people to post their ideas, observations, comments, and suggestions on defense-related issues, bear in mind that if you post a message stating, essentially "I have the answer to everything" one of two things is extremely likely to happen: the message will be ignored, or the suggestion in the message will be heavily critiqued. Be prepared for both.


THE PURPLE PATCH

Get asked about this all the time. Here is what happens when you try to attack something much higher than you are, in terms of your powers effectiveness going down, and in terms of your base tohit going down also. Note: this affects players attacking higher leveled foes. Low level villains attacking a higher level player are not affected by the purple patch. These numbers come from a Geko post from the distant past.

[ QUOTE ]

Foes your level have not changed. You have a 75% chance to hit and your powers are 100% effective.
Foes 1 level above you - No Change. You have a 68% chance to hit and your powers are 90% effective.
Foes 2 levels above you - No Change. You have a 61% chance to hit and your powers are 80% effective.
Foes 3 levels above you - You have a 55% chance to hit and your powers are 65% effective.
Foes 4 levels above you - You have a 48% chance to hit and your powers are 48% effective.
Foes 5 levels above you - You have a 41% chance to hit and your powers are 30% effective.
Foes 6 levels above you - You have a 34% chance to hit and your powers are 15% effective.
Foes 7 levels above you - You have a 25% chance to hit and your powers are 8% effective.
Foes 8 levels above you - You have an 11% chance to hit and your powers are 5% effective.
Foes 9 levels above you - You have a 6% chance to hit and your powers are 4% effective.
Foes 10 levels above you - You have a 5% chance to hit and your powers are 3% effective.
Foes 11 levels above you - You have a 5% chance to hit and your powers are 2% effective.
Foes 12+ levels above you - You have a 5% chance to hit and your powers are 1% effective.


[/ QUOTE ]


*** New for version 1.2 ***
Testing seems to indicate that the base tohit of players might not follow this progression precisely. Additional testing seems to show, and the devs seem to have confirmed, that the tohit decrease follows this progression instead:

-4 .95
-3 .90
-2 .85
-1 .80
+0 .75
+1 .65
+2 .56
+3 .48
+4 .39
+5 .30
+6 .20
+7 .08

It is unclear what the source of the discrepancy is.


Things on the horizon

* Statesman has suggested that high tohit buffs are being looked at, but no solution has been put forth by the devs.

* More details about the specifics of the Defense scaler will hopefully allow us to definitively state the modified tohit formula, with all scaling buffs in their appropriate locations.


SOURCES OF ADDITIONAL INFORMATION

I kept the numbers out of this. If you want them, here they are:

Buffy's Scrapper Guide and Tanker Guide is an excellent source of defense (and resistance and regeneration) numbers.

An additional source of numbers, that include power pool values and tohit bases, is The Scrapper Defense Values Site. Note that power pools for tankers are higher than for scrappers (scrappers generally have 75% of the value of tanker numbers).

RedTomax is working on a web-based guide to all of CoH: it contains Defense information including defense types and values in tabular form here.


Use of this Guide

Anyone compiling information for use by players of City of Heroes and City of Villains has permission to reproduce this guide whole or in part, so long as some form of attribution is maintained. But if you make a ton of money off of it, and I find out about it, I'm going to come looking for my cut.

[ QUOTE ]

if the +DEF limit has anything to do with the resistance limit. Which it clearly doesn't since damage mitigation as an /SR, /EA, /Ninjitsu, etc, can all currently exceed 90%.


[/ QUOTE ]

It doesn't *now*. The point is that the I7 change much more closely couples them. You can't argue that it doesn't make sense to change the system to enforce a defensive mitigation maximum, simply because the current system doesn't enforce one now. Defense is broken now.


And in fact, the PvP accuracy change does for PvP precisely what this change does for PvE: in PvP, there is a similar limit on defense mitigation, and the accuracy enhancements players slot buys a higher floor in precisely the same way that the rank and level accuracy increases similarly buy a higher floor, and capitate defensive mitigation.

[ QUOTE ]
[ QUOTE ]
I would not be opposed to making the tohit floor for everyone about 10%, SR scrappers about 7.5%, and Ice tanks 5% (I actually suggested doing this at roughly the same time I originally suggested balancing defense and resistance with accuracy), or numbers thereabouts.

[/ QUOTE ]

Can you clarify what you are arguing about, exactly?


[/ QUOTE ]

If you're asking what that statement above means, then what it means is that since the ratio of the tohit floor and the base tohit of villains (now locked at 50% in I7) represents the maximum mitigation defense sets can get, since there is a difference between the maximum mitigation resistance offers for scrappers and tanks, it isn't totally out of the question for the same to be true for scrappers and tanks. But there are complications in setting the scrapper maximum the same for both SR and non-SR scrappers, and it has to do with how the set itself actually functions. You really have to dig into the nitty gritty of it, but the bottom line is that the tohit floor has a net stronger impact on SR than, say, the resistance cap has on invuln. Setting the SR tohit identically to all scrappers would create some complex balancing issues for SR, but setting sufficiently low(er) would counteract that.

But the *biggest* problem to tweaking tohit floors, based on my previous experience with floating ideas in the forums, is that I do not believe at this time that there is *any* chance for reasonable concurrence for who should get what floor. FF and radiation defenders could make a case for getting lower floors, for example. It opens the floodgates for practically every single AT that uses either defense or tohit buffs to claim that their usage of them is primal, and therefore ought to buy lower floors. I would not relish the idea of moderating such a debate.

[ QUOTE ]
Heck, and I just realized that powers like Fearsome Stare which has a -ACC effect (something I even slot for) is hit three times.

Increasing mob accuracy and increasingly resisted by mobs that you have to hit against their increasing defense.

[/ QUOTE ]


Just to be clear: Fearsome stare is hit twice in I7: it has to hit with a lowered base player tohit, and its effects are resisted. Its numerical effect will be proportionalized in I7 just like defense is.

It *is* hit thrice now, the third hit being the same one that hits straight defense now.

[ QUOTE ]

I'm not forgetting anything.


[/ QUOTE ]

The thing you are forgetting is that at higher ranks and levels, a resistance based set can never get penalized more than +90% in accuracy; from base 50% for even minions to the 95% ceiling. However, a defensive set can be penalized more. There will always be edge effects, but they are not exclusively in Defense's favor.

You're also forgetting that while defense "deflects" more secondary effects than resistance, the devs *explicitly* make many secondary effects and debuffs autohit, specifically to compensate for that. In fact, they've gone so far as to state that while much of the autohitting *damage* is likely to go away, autohitting *effects* are likely to stay.

You're also forgetting that the #1 most common secondary effect is defense debuff, something that is still very powerful against Defense in I7.

If we are going to balance defense and resistance, we have to start somewhere: pick a specific metric that we are going to balance around initially, and then consider all other imbalances as requiring secondary adjustments after. The primary metric I choose to use to balance the two is damage mitigation. The devs seem to think thats a good place to start also. After that, all other differences can be looked at secondarily, but to my mind, I can still make a stronger case that on the whole they still hurt defense more than resistance when taken as a whole.

I believe that because my SR scrapper cares a lot more about "secondary effects" than my invuln scrapper. If it doesn't end drain me to zero; if its not a sapper or a Ring Mistress, my invuln doesn't really care. And if its something like a single Ring Mistress, unstoppable makes me not care. Only things like malta sappers scattered throughout a mission cause me any concern.


[ QUOTE ]

If we are going to talk about how +DEF should scale like resistance, then if scrappers can only reduce damage by 75%, then they should only be able to reduce MTH by the same amount....75%. This means the minion floor should not be lower than 12.5%.


[/ QUOTE ]

Actually, I would agree except that SR's net maximum mitigation would suffer. A set like invuln's has dull pain: maximum net overall mitigation with dull pain: ~84%. The passive resistances are not that strong, and even if they were, they are also highly variable. And I'm not counting the self heal.


I have actually been in favor of increasing the tohit floor, not specifically to balance defense and resistance in general, but rather to more specifically balance SR scrappers and Ice tanks. If they are the defense specialists, they should have lower tohit floors than everyone else, and Ice tanks should have a lower floor than SR scrappers. Also, if non-defense specialists had higher tohit floors, then things like power pool defenses might not need to be so vanishingly small, on the fear that they might be used to drive tohit to very high levels of net mitigation.

I would not be opposed to making the tohit floor for everyone about 10%, SR scrappers about 7.5%, and Ice tanks 5% (I actually suggested doing this at roughly the same time I originally suggested balancing defense and resistance with accuracy), or numbers thereabouts. But doing so has significant collateral issues that would need to be worked out, because they affect all sorts of things, including and especially defense buffing.

And it would start an argument over who should get lower floors. Lets face it, that thread would make the Hamidon Enhancement thread look like a love-fest.

[ QUOTE ]

I repeat my request for you to share what Castle said about how things worked previously with the flooring logic and why they changed it and when.


[/ QUOTE ]

What I said was what I was told, and all I was told: that the tohit formula had been changed (at least) once before, that it was arena combat that had prompted the change. Beyond that, I can infer that the change was related to the tohit floors and ceilings, because that is the specific issue I was asking about (to wit: did the second floor/ceiling check always exist, because I had conducted tests that suggested that at least at one time it might not have, and if it did not, at what time was it put in).


[The rest is strictly coming from me only, not Castle]

The second (interior) check is meaningless when there are no accuracy buffs/debuffs (because the outer accuracy term is 1.0) and while players have high accuracy (due to slotting) villains typically do not; thus, the absence of the second check has limited effect on the game - prior to arena combat, when players face players, and high accuracy faces high defense. Thus, it made perfect sense to me that the absence of the second check might survive in the game *until* arena combat forced the issue. Clearly, if the devs responded to arena combat (aka high accuracy/high defense combat) by instituting the second check, that strongly signals how they expect defense to "scale" (honestly, I'm not fond of the term except colloquially).


[ QUOTE ]

How defense should work and how to-hit floors should work are not logically connected.


[/ QUOTE ]

They are if your reference point for defense performance in the game is resistance performance. If you don't feel that defense performance and resistance performance need to be parallel, then the tohit floors need not float - but then again, defense need not scale, either.

(And defense-oriented sets like SR and Ice really ought not exist)

[ QUOTE ]

I understand the argument here, but I think it's revisionist


[/ QUOTE ]

I don't understand how it could possibly be revisionist, because:

a) I've always felt that way, and I was expressing my opinions, not what I thought the devs thought.

b) The post you quote was a direct response to someone suggesting that the tohit floor - whatever it is - ought not to move because the tohit floor was the true measure of maximum mitigation, which isn't strictly true. The math is significant because its on point to the question, and is not offered to make any specific assertion of how the math originally worked. The math originally worked differently, but thats irrelevant to the point.

c) None of this is either relevant nor significant, because the devs have made sufficient statements about how they *want* defense to work in I7 as to make any statements about how it was *originally supposed to work* largely meaningless, especially since they are acknowledging, by way of making a change, that at least *at this time* they believe the way it used to work wasn't correct.


As to the tohit floors being "artifical" they are no more so than the resistance caps, but the math involved in explaining precisely how the resistance caps relate to the tohit floors and ceilings within the context of the current tohit formula is involved, and I suspect insufficiently convincing. It is reasonably transparent to anyone that has studied defense in CoH, though, and probably completely transparent to anyone judging the merits of the design from a designers point of view.

[ QUOTE ]
I agree with the spirit of what you're saying Arcanaville, but I do not agree that I5 and this change are linked. At the very least, I do not agree that the must have been linked in order for defense to be "fixed" in the way this change does help it.

The shift of higher level and rank foes from toHit scaling to accuracy scaling was feasible at the game's release. Nothing whatsoever required I5 for that to be possible or desirable.

[/ QUOTE ]

At release, SR scrappers had similar defense to what they do now. They could have put this change in at release, because defense was even lower than now. Its Issue 1 through Issue 4 that's the problem: the boosted I1-I4 SR numbers would have allowed SR to floor everything in the early teens with this change, and its unlikely they would have ever given us that level of performance.

(As a separate issue, a lot of things had equal or better defense than SR, and power pool defenses were much too high relative to SR defenses; that would have had to be dealt with eventually.)

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
Honestly, I can't see this being related to I5 or ED. Defense has always been inferior to other mitigation methods. If I5 and ED had never happened this would still have needed to be done to balance it.

[/ QUOTE ]

But with the old defensive numbers for powers I'd say this particular change would have made defense the obviously superior form of damage mitigation. Nerfs would have definitely followed or accompanied this change had it been done prior to I5.

[/ QUOTE ]

Remember, everything was basically halved. I'm not saying defense numbers wouldn't have had to come down to prevent them flooring everything, but I5 as such and especially ED are not logical follow-ons of this change.

[/ QUOTE ]

I consider them linked in this sense: the devs stated that I5 "calibrated" defense numbers to have the desired effectiveness against base 50% tohit (even level minions). That is logically tantamount to saying that defense is *not* working at the desired effectiveness against anything that has higher base tohit, unless that thing is deliberately designed to affect defense more than other mitigation mechanisms.

Therefore, it makes no sense to balance defense around even level minions, and then have all other ranks and all other (higher) level foes be imbalanced against defense. Contrawise, it makes no sense to adjust all (most) villains to base 50% tohit if defense *isn't* balanced around that number.

Thus, if you care about the game being balanced, the I5 defense adjustments (regardless of whether or not you think they were *quantitatively* correct), and the defense scaler, are intertwined. If you were designing the game from scratch, you'd want to balance defense around a "normal" tohit number, and then you'd want to give most things that base tohit. That's what the I5 reductions and the defense scaling change combine to do.

They only look like buffs and nerfs in isolation: if they had been launched simultaneously, I would have considered the two changes to be a single design change, and neither a buff nor a nerf explicitly (specifically referring to Defense).

[ QUOTE ]

The maximum mitigation becomes a sliding scale, depending on foe rank or level.

Because, with this system, you can, in fact, mitigate a +0 minion to 5%. But your best mitigation vs. a +5 is 10%. Everyone in between has some proportional value.


[/ QUOTE ]


The 5% floor is not a proper measure of mitigation, at least comparing it to resistance. Lets go through it step by step:


We define the mitigation benefit of a particular amount of defense or resistance to be the amount of damage that you avoid, relative to not having that defense or resistance active. The comparison is on verses off. And its expressed as a ratio of mitigated damage to standard damage (without mitigation).

Its a nice way to measure mitigation, because resistance is trivially easy: the mitigation ratio for a resistance power is exactly equal, numerically, to the resistance percentage. That's not a coincidence: its true because resistance itself is defined in exactly the same way. A 75% resistance power deflects 75% of all incoming damage - that's the same way we define "mitigation."


Not so easy for defense. The mitigation ratio or value of a defense power is defined in the same way: amount of damage avoided (on average) when the power is on, verses when the power is off. So lets look at minions and AVs.

Suppose you have 30% defense. That is 60% mitigation verses minions. Why? Because on average, *without* the power, you'd be hit 50 times out of 100, and with the power you'd be hit 20 times out of 100. The power allows you to avoid 30 out of every 50 hits on average: 30/50 = 60%.

Suppose you floor the minion. That would be net 5% tohit, relative to 50%. Net overall mitigation would be 90%, because you'd be avoiding 45 out of every 50 hits, relative to having no defense. 45/50 = 90%

Now AVs: flooring an AV in this new scheme will only floor the AV to (1.5) * (5%) = 7.5%. In other words, out of 100 swings, the AV will hit you only 7.5 times. BUT, how many times would he have hit you if you had no defense? That's right, 75 times (1.5 * 50% = 75%, or 75 times out of 100). So for AVs, the act of flooring the AV - the best you can do with defense - will allow you to avoid 67.5 out of every 75 hits (75 - 7.5). Thats a mitigation ratio of 67.5/75 = 90%. The same 90%.

Far from making mitigation a sliding scale, the change *freezes* the mitigation ratio for all ranks and all levels (that the change operates on).

[ QUOTE ]

I understand why the inner quantity is bounded. But why is it bounded by 0.05 and 0.95? Wouldn't it make vastly more sense for the inner quantity to be bounded by 0 and 1, leaving the outer one bounded as it is now?

This would avoid the oddity of raising the floor by bounding the low end twice.


[/ QUOTE ]

I don't consider it an oddity, I consider that change to be the *proper* behavior.

Consider: the 90% resistance cap is unaffected by level or rank. You cannot get better than 90% resistance against a minion, or an AV. The *proper* thing to compare that to in the defense world is *also* the mitigation ratio, not the tohit floor. The maximum mitigation ratio possible for defense is the floor over the base tohit of the foe; Floor/Base. As base goes up, so must the floor.

Now, the base tohit of an AV is going to be adjusted down to 50%, but all this doesn't take accuracy into account. The standard chance to hit (because I can't use the terms accuracy or tohit without being confusing) of an AV against something with no defense is (1.5) * (0.5) = 75% - same as now. The absolute best a defense set ought to be able to do against that is 7.5%, or 90% total mitigation; the same maximum mitigation they have against minions.

That defense now has the same maximum mitigation against AVs as they have against minions is exactly what we mean when we say "defense will now work just as well against minions and AVs."


Another way to put it: if the math didn't work that way, I would make it work that way with some form of adjustment. When I first suggested altering defense and tohit in this way, I wasn't even aware that accuracy and tohit worked differently or that the floors and ceilings were checked twice: in my suggestion, I *made* accuracy work this way, and I *made* the equation check floors and ceilings twice, because that was the only way to make it fair, without realizing that in fact it always worked that way. cf: The Accuracy vs Defense thread.

(Note: Castle has pointed out to me that it *didn't* always work that way, in terms of checking floors and ceilings twice, and that it was partially the arena that caused them to specifically add that check in. I imagine that is because it was first in the arena that the issue of extremely high accuracy buffs first became prevalent. Outside of PvP combat, ultrahigh accuracy was uncommon, and therefore the issues surrounding the doublecheck on bounds was also uncommon)

[ QUOTE ]
Arcanaville: [ QUOTE ]
Using your numbers for DN, and rounding things off nicely, that 30% debuff is going to be slotted to 60% (we're dropping ED for now).

[/ QUOTE ]
But that number is wrong, since castle said 3-slotted DN is -30%, not 60.

[/ QUOTE ]

The number isn't important: I was responding to your concern about "negative tohit" which in the current system, and the new system, can't happen, because of the double-flooring that exists that you were not taking into account.


So using no darkest night numbers at all the intermediate tohit will be 50% - DN%, and that will be bounds checked, which means it can never be lower than 5%, and then that number will be multiplied by the (in this case even level) AV rank scaling accuracy buff factor of 1.5, and the net number will either be 7.5%, or some number higher than 7.5%, but in no case negative, and in no case less than 7.5%.

[ QUOTE ]

I did miss that. Darkest Night was a base -30% ACC DEBUFF and -35% DAM DEBUFF. But you could only slot it for -ACC.

As those are "A" type enhancers, they are 33% each.

Using these numbers, Darkest Night will actually be giving a base negative to hit number that is then multiplied by the mobs rank.

Something does *not* sound right there. I somehow don't think that AVs are actually going to have a -15% ACC under this system.


[/ QUOTE ]

Using your numbers for DN, and rounding things off nicely, that 30% debuff is going to be slotted to 60% (we're dropping ED for now).

Then the actual calculation will go like this:

Net tohit = (BaseAcc + Accbuff) * (BaseToHit - TohitDebuffs)

(I've dropped the non-essential terms from the equation)

BaseToHit is 50%, the Tohit Debuff is 60%, so that term is -10%. *However* that intermediate term is checked against the 5% floor and 95% ceiling (see the Ask Geko thread). So that -10% tohit is floored to 5%.

Then accuracy is factored in: (BaseAcc + AccBuffs) * (5%)

In this case, we assume the AV is using "normal" attacks, so BaseAcc = 1.0. The AV accuracy factor is 0.5, so:

(1 + 0.5) * 5% = 7.5%

This is dev-confirmed.

[ QUOTE ]

The main reason most people had problems w/ defense scaling was that the other aspect the purple patch (the mob buffing part) meant defense failed just when it really counted, against +level foes. However, that's exactly the problem debuffs face, they fail just when they're really needed. Basically, I'm not seeing why debuffs should scale, but buffs should not.


[/ QUOTE ]

Its more specific than that. At such a high level, one could equally say "my offense fails when I need it the most" as well. In fact, you can say "I need X the most when facing the hardest enemies, and that happens when they are higher level," fill in the X with anything, and "justify" changing the purple patch to not affect anything at all.

But at a more detailed level, the devs added (in effect) tohit buffs to higher level and higher ranked villains, relative to even level minions. Now, we know what the net effect of that was: it served to make those villains more accurate, but also it hurt defense sets more than resistance sets. Now, the question is, was that intended or not.

Was it intended for defense sets to be hit harder than resistance sets by the level scaler and the rank scaler? If the answer is "yes" then there is no problem. If the answer if "no" then there is a problem, because the tohit buff addition (I call it an addition in the sense of the game design adding the effect for anything above even minions: I'm aware some form of this scaler has essentially always existed) has an unwanted side effect.

Now, the level scaler specifically reduces the effects of all debuffs (among other things). That cannot be an accident, because its explicitly designed. So the question is, did tohit debuffs get "caught in the net" when it is very obvious that the devs *did* want to reduce the effect of other debuffs.

So what is the justification for separating tohit debuffs from the rest of them as "special?" One argument is that tohit debuffs were meant to provide to debuffers essentially the same benefit as defense does to high defense ATs. However, there is no evidence that is true. It *might* be true, or it might not. It isn't written anywhere that radiation defenders use of RI is supposed to be *just as effective* as defense is to an Ice tank, say. If you play a rad, that might be your position (I play a rad: its not mine), but there isn't any evidence of it.

Let me suggest an analogy. Suppose SR was the set with dull pain, and invuln was not. SR had defense and +health, and invuln just had resistances. The burden of proof on people who thought defense didn't scale well would be much higher than it actually was, because it would have been so much trickier to construct an argument comparable to the one I *did* construct. And it strictly speaking did not have anything to do with "the main reason most people had problems w/ defense scaling was that the other aspect the purple patch (the mob buffing part) meant defense failed just when it really counted, against +level foes." It was this:

SR has only defense
Defense doesn't scale
If Defense doesn't scale, SR will always lag all other mitigation forms
The devs could not possibly have designed SR to deliberately lag, and if they did, they should say so.
If not, then defense *was* meant to scale, and since it doesn't, that points to a problem.


That's a pretty strong argument, in my opinion. Ironically, the fact that SR was heavily reliant on defense, and that mitigation didn't work right, gave me a stronger argument than if that were not true. With multilayered defense, I would be exposed to this counterargument:

Defense doesn't scale
But no one relies on defense only
Defense breaks down at higher levels, but you fall back onto other mitigation
Unlike other sets like invuln, that only have one mitigation - resistance - that *absolutely* must work, since it is all they have.

(I'm ignoring Ice tanks for the moment here)


I would have been left with a much weaker argument:

Tohit buffs are extremely prevalent
They serve to counteract defense
For the environment to counteract defense in all aspects of the game devalues defense as a mitigation mechanism
Why do that when there are better ways to parcel out accuracy, while still allowing defense to have a greater impact on the game experience: why not *use* defense, since you went to the trouble of creating it and all.


Now, as I said, in my opinion the burden is on people who think tohit debuffs should avoid the purple patch (I do believe the burden also was on those of us that advocated "fixing" defense - but I also think we satisfied that burden long ago). But let me emphasize its not my place to say which ATs "deserve" a buff and which do not: "deserves" has nothing to do with it. I simply believe that there are valid game design reasons why tohit debuffs should be "exposed" to the purple patch, but even if I didn't, the point is that there isn't a good counterargument yet that I've heard or thought up that strongly states tohit debuffs should be exempt.

"Tohit debuffs should work as well as defense, because tohit debuffs are used like defense" is simply not compelling. Controllers use holds like defense, but hold duration also decays with the purple patch. For an argument in favor of tohit debuffs to be valid, in my opinion, it has to be an argument that cannot easily be used for *all* effects, because if it can, then its really an argument to remove the level scaler altogether - which is a completely different thing.


Let me add something that seems like a quibble but I feel is actually significant. Defense was *never* harmed by the purple patch. It was harmed by the tohit level scaler and the innate higher base tohit of higher ranked villains. That has nothing to do with the purple patch: those effects *predate* the purple patch.

Why is that significant? Because at release, there were tohit increases for higher level villains and higher rank villains, and to the best of my knowledge those were never increased. The (relative) effect on defense sets never got worse: the devs never *explicitly* took any action to exacerbate that situation.

*But* they *did* increase the effect of the level scaler on degrading the effects of debuffs with the purple patch. The purple patch, in effect, is direct evidence that the devs *explicitly* wanted to weaken debuffs at higher level (among other things), and weaken them a lot. It is *that* that represents part of the reason why I feel there is a heavy burden to wanting to separate tohit debuffs from the purple patch: the devs didn't just originally design the game with an effect that hurt them, they came back for more. That to me signals that the devs did it deliberately. That doesn't mean they were right, but it does mean anyone wanting to convince them to change it ought not to take the position that the effect on tohit debuffs was some kind of accident. The effect of tohit buffs on defense appears to be an accident (at least in terms of the full appreciation for the magnitude of the discrepancy), but the effect of the purple patch on tohit debuffs appears to be deliberate.

We're getting into very subtle and complex issues of game design, which do not have simple answers - and actually, I don't have *any* "answers", just judgement and educated guesses. As it stands now, though, I consider tohit debuffs to be a type of debuff, for the purposes of thinking about how they should function, and not as a pseudo type of defense. I can be convinced otherwise, but I've already suggested what types of arguments are going to tend to be less impressive to me.

[ QUOTE ]
[ QUOTE ]
Its unclear to me if there are any specific instances where tohit debuffs are explicitly meant to be as strong as pure defense.


[/ QUOTE ] This is a curious statement. RI and DN seem to be better than pure defense as they have a higher base value than say /SR's toggles and they require less slots for completely comprehensive acc reduction..i.e. three slots emulating six powers in /SR in the non-resistance days.

[/ QUOTE ]

Rephrase: its unclear to me that tohit debuffs as an effect were meant to be equally scalable with level as defense by being immune from the effects of the (purple patch enhanced) level scaler. Although tohit debuffs function similar to defense mathematically its unclear they were necessarily meant to function identically with respect to not being degraded with level.

This says nothing specifically about whether power X was supposed to have a stronger or weaker effect than power Y.

Why "specific instances?" Because the devs use effect for different purposes, and the specific purpose can logically allow us to infer how the power is reasonably supposed to scale. For example, its reasonable to assume, without evidence to the contrary, that the slow in lingering radiation should be affected by the purple patch: the purple patch was designed to affect debuffs, including slows, so there's no specific reason to presume it ought to be exempted. The default assumption is the debuff was always meant to be degraded by level.

However, the slow put into Ice tank armor was there partially to offset the problems of defense. There, its apparent that the slow *might not* have been "meant" to be affected by the purple patch: it doesn't make logical sense that the devs would give Ice tanks a power that is most effective when they did not need it to be (against even level critters that have low base tohit), but least effective against higher level critters that have higher base tohit. That's bass-ackwards. However, the devs were confined to what the game engine supported, and their affection for the slow effect as a good overall effect to have for Ice outweighed the scaling issue with the slow. At least as I see it.

And that's why I say I can't think of *situations* where tohit debuffs appear to have intended on being exempt. In general, tohit debuffs are debuffs like all other debuffs, and the *default* assumption is that they *all* should be affected, since that is what the scaler is directly intended to do. But its *possible* that specific circumstances lend themselves to reconsidering that base assumption, for that specific circumstance only. I just can't think of any right now.


Put into the context of the defense scaler: the tohit buff benefit that higher ranked villains and higher level villains was explicitly put in there to make those villains hit us more often: that's obvious. However, its *not* obvious that they were intended to penalize defense sets more than resistance sets: that is *not* obvious, and based on balance statements from the devs, the most logical conclusion possible is that the nature of tohit buffs (in terms of penalizing defense sets more than other sets) was *unintended*. So the logical conclusion then becomes the devs should find a way to boost the accuracy of the villains in a way that's fair to defense sets: long ago I suggested a way: swap tohit for accuracy. It seems the devs are thinking the same thing: that the villains were supposed to become more accurate, but not in an unbalanced way from defense to resistance.

If a similar argument can be constructed for tohit debuffs, something *a lot* stronger than "tohit debuffs are kinda like defense, so why should they be treated any different" then I think a case can be made to consider a similar adjustment for them.

But I don't think such an argument can be made at the moment, and the burden of proof, so to speak, is always on the people who want a change. The burden of proof was originally on the people who wanted a scaling defense: the defense scaler doesn't automatically say anything about whether tohit debuffs should be treated the same or differently: its not a precident of any value in and of itself.

But if you think I'm wrong, the best place to start is to look at how the slow in Ice works, and doesn't work, and what that says about the slow in Ice's overall mitigation. Then see if a similar situation can be constructed with tohit debuffs. And a simple "tohit debuffs are clearly better *this* way rather than *that* way" is insufficient because its obvious: the question is the justification behind the *why* one scenario makes more sense than the other.

Do that, and make a compelling case, and I'll reconsider. Of course, your real target is the devs.

[ QUOTE ]
[ QUOTE ]
Of course, in Korea, hero don't talk.

[/ QUOTE ] Where does that come from?

[/ QUOTE ]

It comes from watching this: Asian Launch Video

Although to be honest, I was vaguely hearing it in Joel Grey's voice ("In Korea, door handles do not break").

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
sometimes it gets the best of both worlds, and sometimes it gets the worst of both worlds

[/ QUOTE ] Correct me if I'm wrong, but the PP was introduced to the game. It wasn't part of the original design strategy as far as we know. That is why I don't see why Debuffs would be any less effective than +DEF if we ignore the context of the game.

[/ QUOTE ]

You're wrong. The idea that power effectiveness (accuracy, effect, duration, etc) would be lessened by level difference was in the game since launch. The purple patch simply narrowed the range of levels you could fight from something like +10 to +3.

The purple patch was eased a bit, to expand the level range to +5, if I recall correctly.

[/ QUOTE ]

I believe Mieux should have used the word "should" instead of "would." We all know it does: he's asking if it ought to.

Personally I believe it ought to, but the reasoning is lengthy.

[ QUOTE ]

Left alone, though, it would be a way "fixed" DEF would now work vs. mobs better than toHit debuffs.


[/ QUOTE ]

The fact that the purple patch reduces the effect of tohit debuffs is an intended effect. Its unclear to me if there are any specific instances where tohit debuffs are explicitly meant to be as strong as pure defense.

However, the issue of purple patch fall off for debuffs relative to defense and resistance did come up with regard to Ice tankers: by giving Ice tanks foe debuffs to help compensate for defense not scaling, the devs (possibly inadvertantly) gave them yet another thing that didn't scale with increasing level of foes. That was clearly a case where the purple patch scaling was doing something almost certainly unintended in that specific instance.

I'm not sure, though, that things like radiation infection are meant to sidestep the purple patch: I believe the purple patch reducing the tohit debuff in RI is probably working as intended.


Although I've pushed for this change at least as hard as anyone, I'm *not* an advocate of making everything work exactly, precisely the same way. That's why even though I've harped on tohit buffs for a long time, I *dont* want a solution that completely nullifies their strong effect on defense. I *want* the devs to have a tool in their toolbox that allows them to make enemies that are hard on defense to create challenges above the norm: I just wanted that effect not be used *pervasively* (i.e. all higher ranks, all higher levels). I don't want defense to be immune to tohit buffs, I want the devs to use tohit buffs to make things special, ala villains with psi damage or regeneration debuffs.

I'm also not in favor of making tohit debuffs work precisely the same way as defense either. Defense, tohit buffs, tohit debuffs, and resistance should work differently for variety sake: what we need to make sure about is that each AT gets the appropriate one (or ones) necessarily to function as they should.

You can get too crazy trying to make *everything* "equal." For example, blasters are said to use offense as their primary defense: they are presumably supposed to kill faster than something like a tank, which has higher mitigation and lower damage. But tank mitigation is stable relative to higher level foes, while the purple patch reduces blaster damage substantially. It reduces tanker damage similarly, but the tank isn't relying on high offense for its survivability: in effect the tank gets high mitigation and low damage, while the blaster gets high damage and low mitigation, and one of those two is hit harder by the purple patch. Does this mean blasters need a damage boost against higher foes? I don't personally think thats a good idea myself, at least not directly.

[ QUOTE ]

On a serious note, maybe Unyielding's defense debuff should be derived directly from the enhanced resistance value.


[/ QUOTE ]

If the debuff was meant to be balanced against the resistances in UNY, it should have been lowered when those resistances were lowered.

If the debuff was meant to be balanced against the other defenses in the set, specifically invincibility, then it should have been lowered when those defenses were lowered.

If the debuff was meant to be balanced specifically against the overall utility of invincibility (my theory) then it should have been lowered when invincibility's utility and strength were lowered.

If the debuff was meant to be a static penalty that the invuln could offset by slotting UNY, then it should have been lowered when ED cut the number of resistance slots you could put in UNY.

If the debuff was meant to scale upward with higher foes, in the same manner as resistances, then the Defense Scaling changes mentioned in this thread will do that, but that has no bearing on whether the actual base magnitude of the debuff is set right. What it does is amplify the problem if (as many people feel) the debuff is set wrong, because the defense scaler is a magnifying glass on the issue: if its balanced now, the Defense Scaler will magnify a zero into a zero; if its not, it will magnify the discrepancy to a stronger net effect.

The Defense Scaler didn't create the problem, nor does it need to be changed to solve the problem. The sole issue is whether the debuff in UNY is actually set correctly, but the DefScaler is the perfect opportunity to reexamine the debuff in light of the much simpler way that defense (and defense debuffs) affects mitigation overall (we don't have to look at 25 different tohit cases to see what's happening).

[ QUOTE ]
[ QUOTE ]

To me the social aspect is no more a part of the game


[/ QUOTE ]

The social aspect is an integral part of the game, it's part of the immersions of being a super hero and teaming with other super heroes.


[/ QUOTE ]

It also significantly reduces my damage mitigation and offensive output at times, which others can attest to. I factor it out of my calculations because its a constant factor for all ATs, and because I'm embarrassed to tell you how high that constant factor can be on some days.

Maybe typing in the chat window should grant a +35% defense bonus and +60% resistance bonus while you are typing, because in the comic books, heroes almost never gets hit in the middle of a snappy comeback.

Of course, in Korea, hero don't talk.

[ QUOTE ]
[ QUOTE ]

its possible that Geko is wrong


[/ QUOTE ]

It's more likely we have misinterpreted what was said.


[/ QUOTE ]

Although that is always your assumption, in this case extremely unlikely, since he didn't post verbage, but Excel-like formulas showing the MIN/MAX representation of collaring boundaries.

[ QUOTE ]
I appreciate that you have corroborated my explanation of the to-hit system as I was fairly confidant in my assertion, but always try to remain somewhat reserved considering I do not actually have the code sitting in front of me.

I will say one thing though... if they are checking for floors and ceilings twice then they are wasting processing power as they really only need to check for a floor when it comes to the "to-hit" factor as this is the only region where we can skew into the negative... and they only need to check for a ceiling after accuracy is taken into account.

You end up with the exact same result, but with two less checks on every to-hit calculation... unless of course I have missed something, but I do not believe I have in this case.

[/ QUOTE ]

So long as no such thing as negative accuracy ever comes into the game, or true accuracy (not tohit) debuffs, then you're basically right. However, you should never code that way: a good optimizing compiler is likely to detect the double boundary check and optimize it away anyway, so its better to code the algorithm cleanly. Otherwise, its a hidden landmine on anyone altering the code one day to introduce tohit changes.

[ QUOTE ]
[ QUOTE ]

Furthermore, we have a new consideration as the to-hit floor is demonstrably different for the various villain classes (minions 5%, lieutenants 6.25% and bosses 7.5%)


[/ QUOTE ]

I don't think so. The to-hit floor is still 5% for everyone. The coding should look to see if you have a value below 5% and if you do, simply raise it to that level. It should not then apply the +Acc for the same reason it doesn't apply our +Acc to our 95% ceiling. If we can be floored at 5% with our own +Acc enhancments, then so can mobs.


[/ QUOTE ]

Geko's explicit description and subsequent other red name additions to the description of how tohit is calculated all but state that (tohit + tohitbuffs - tohitdebuffs - defense +defensedebuffs) is floor/ceiling checked once, and then accuracy is factored in, and then the entire thing is floor/ceiling checked again. Unless they made significant changes to how floors and ceilings are calculated, the pseudo-effective floor for higher ranked things is going up: its an automatic consequence of attempting to satisfy the design constraint "defense should work just as effectively against higher ranked and higher level foes" by setting mitigation average constant, which is exactly, precisely what the devs have effectively stated they are doing.

You should keep in mind that the absolute tohit floor has *never* been 5%. The intermediate terms have 5% floors, but in every case where accuracy has come into play, the effective floor - according to the current dev-acknowledged formula - has always been higher.

We ignore that because we assume that foes don't have accuracy enhancements, only tohit buffs, but back when I5 was on test I proved that was not the case: certain mobs *right now* have heightened accuracy (in the sense of the formula) and not tohit buffs - and this was basically all but confirmed by a red name to me.

However, having said all of that, its possible that Geko is wrong, at least about the tohit calculator checking for floors and ceilings twice. If it did, then in the arena anyone that was packing 6 dmg/acc HOs should have never seen a tohit on a perma-elude or MoG scrapper less than 20% - because that would be the limit when defense floors to 5%, then the +300% accuracy boost increased that number to 20%. But that is *not* what was happening. The intermediate floor might not be functioning correctly, and if that is the case, then in fact the global floor is really still 5% - but it would almost certainly be a bug that the devs would fix, because tohit would then not be working as they specifically described to us.

The short answer is that I've in fact played SR, invuln, regen, and DA to significant levels now. I'm very familiar with how they play. The main reason why DA is not in my SR I5 comparisons (listed in my sig) is due to not having direct experience with DA at that time. That's been sufficiently rectified to allow me to extend that into a four-way I6 comparison.

I can play the "this build outperforms that build" game; its part of what I do when I analyze the various sets. But I won't do it as part of a "who should get nerfed" discussion. I'll simply say that the performance you suggest SR can achieve by wrapping SR in a ton of power pools can be exceeded by invuln and regen builds that take comparably insane power pool selections. The absolute best mitigation build I can construct right now is not an SR build. In fact, the absolute best SR build I can make isn't in the top three (I wouldn't want to play any of them, btw).

[ QUOTE ]

Before people get too happy for defense based sets, I think one
has to take a look at some of the numbers. First realize
resistance is capped at 75% for most archtypes, this system will
cap defense-based mitigation for everyone at 90%. Now that the
performance of the two is brought in line, I can't expect that to
stay.


[/ QUOTE ]

Invuln with 75% capped resistances and running dull pain (something SR does not have) can buy 84% direct mitigation not counting the 60% heal that dull pain offers, while DP is up. The theoretical maximum mitigation of non-defensive scrappers is not stuck at 75%.

If the *absolute maximum* credible defense an SR scrapper can buy is 80% net mitigation, that's not bad: invuln can theoretically do better.

Of course, invuln cannot achieve 75% resists for any length of time, but then again an SR scrapper running FF, FS, evasion, CJ, maneuvers, and weave all simultaneously is either going to have some END consumption issues, or is going to have serious slot consumption issues.

More to the point: whatever SR will be able to do in I7, it can already do now. The only difference is that it *if* it chooses to expend all that effort to absolutely maximize defense, that defense will work across all levels and ranks in the same way, just like resistance does now.

Do you want to take a guess as to what the maximum damage mitigation is of regen? I'll give you a hint: the number is in the triple digits.


And that theoretical maximum defense build? I've tested it at length. You're better off dropping tough/weave and picking up aid self. One less power selection, significantly more survivability.