Mark_GGG

Neither. It means the exact opposite of the player half - if their action speed without that modifier would be above 92% of the base value then instead it is 92% of the base value.

This is not the same as negating all bonuses because those can still apply in combination with penalties to action speed to partially or fully counter them, just the final result can't be above 92% of the base value.

That would allow reducing the critical strike multiplier lower than 100%, which is undersirable for this stat, and is more expensive for performance.

The amount of extra damage can't be negative. In general, reducing things by values greater than 100% in PoE is prevented from negating the value.

That would mean it did nothing, because there isn't a critical strike damage multiplier applied to the reflected damage for it to modify.

And that wouldn't change any of what's complained about in the rest of this thread because increases to that stat are still additive with reductions to it.

Nah, you're fine, I just dislike the misleading description on that specific stat because it implies it stacks with other things that it really doesn't.

For physical damage specifically, we store the pre-mitigation value in the damage package so it can be referenced. The amount of damage "prevented" is the amount by which the actual physical damage taken differs from that pre-mitigation value of physical damage to deal.

So basically anything which lowers the amount of physical damage taken counts as preventing, with the specific exception of "damage taken as" modifiers, which apply before mitigation, and intentionally also before the pre-mitigation value is stored (because you didn't prevent that damage, you just took it as a different type).

And how does Untiring interact with Petrified Blood or Steelskin?

It doesn't, neither of those affect the amount of damage taken - they change what and how much you lose as a result of taking the damage, but they do not affect how much damage you take.

This is one of several potential changes to that system which have been discussed in the past as options for the long term. Personally I am in favour of something similar to this change, although it would certainly have consequecnes to other things that would need worked out if we ever do it.

Completely off topic but why is the term 'recently' used to describe different values of time.

It isn't (or if it is somewhere then that's a bug).

I.e. the Warcry Mastery that gives 20% increased Damage for each time you've Warcried Recently is 8 seconds, however almost every other example of 'Recently' I can find is 4 seconds.

That is definitely set to be 4 seconds in the code. Are you seeing this count a longer amount of time in-game?

It's a horrible pile of lies that technically doesn't actually modify damage, but the damage multiplier applied to damage from being a critical strike, in a way unlike anything else in the game.

What it actually does is look at the enemy's total critical strike multiplier (default 130% for monsters), subtract 100% from that to leave only the part which makes the damage higher (30% for monster with no crit modifiers), apply a percentage reduction to that value (e.g. 10% reduced from this stat would reduce the 30 to 27), then give the original 100% back and treat that as the critical strike multiplier to use for this hit.

The design reason it works that way makes sense - it shouldn't be able to make critical strikes damage you less than non-critical strikes, and 100% reduced extra damage taken from crits should just make them the same as non-crits - but I don't like how it's inconsistent with everything else.

I dearly hope to one day be able to burn it to the ...

If these modifiers used "less", then the 50% less and the 60% less would combine to an effective total of 80% less (you'd still take 50% of 40% of the damage, for 20%), not 110%, and no amount of stacking such modifiers would remove all of it completely.

I think puerility meant that it's not obvious that "reduced reflected damage taken" and "increased damage taken" stack additively because they could be in different "categories"

This is true, but neither is obvious that they would not stack additively - I am not arguing at all that the exact result is obvious to a new player, only that it is not unreasonably outside expectations - having already learned or assumed that the two reflected damage modifeirs stack additively, it would be unreasonable to assume nothing else could possibly stack additively with them, so the possibily that something can do so should not be a complete surprise to the player.

Just like "increased damage" and "increased damage taken (on the enemy)" stack multiplicatively (correct me if I'm wrong)

Only because you specifically asked, because this is extremely nitpicky: Those do not stack multiplicatively, th...

But if a player doesn't know these modifiers stack additively (which is described in the in-game help system), they wouldn't think they had 100% (or higher) reduced in the first place, because that didn't come from a single modifier. In the case in the OP, there's a 60% reduced and a 50% reduced, and those can only be seen as combining to 110% reduced if you either know or assume these modifiers stack additively, which at the very least implies the possibility that other things also stack additively with them.

if the question is "you take 100% reduced reflected damage; how much reflected damage do you take?" and the answer isn't "none", that's kind of a trick question

The game never poses that question to the player. In this case it's asking "If you have 50% reduced reflected damage taken and 60% reduced reflected damage taken, how much do you take?", and then "what about if you also have 30% increased damage taken?"

If they do, great as they will have the stack and therefore get the item rarity buff when they die.

No, that only means they benefit from that IR bonus if they die from the corrupted blood. If they die from the explosion damage, then that's what killed them, and the IRR support does nothing, regardless of whether they have Corrupted Blood on them from Corrupting Fever (I don't think they would from a single hit, but I can't be bothered checking since the answer doesn't actually matter).

Lots of stats on Jewellery either can't actually be allowed, or don't actually work, with negative values. In all those cases, we had to set up stats that were a reasonably "reversal" of the core effect to replace the stat if the league mechanic had to reverse those mods.

Regeneration is one of those cases. Regeneration is fundamentally is a form of recovery, and regeneration as a mechanic cannot be given a negative value to cause over-time loss - recovery and loss have to be kept separate internally for a bunch of reasons, many of which relate to keeping characters alive. So keeping the same stats and just setting the values negative would not have worked, and instead we replaced them with the closest equivalent the game can handle, which is over-time loss.

Similarly, added damage can't have negative values because it leads to a bunch of weird edge cases like having maximum damage lower than minimum which intentionally are not allowed and cause the game to spit out...

Yes. Only one poison can be assigned the kill. In practice, it will be the oldest one, because that will be earlier in the list and thus found first (may not be the case for some debuffs with more complicated behaviours, but poison is nice and simple, new ones are just chucked at the end of the list).

Only the currently running debuffs can even be considered for this - the ones that are ticking down in the background but not currently applying because of a stronger one of the same type aren't even looked at here.

Only the ones currently dealing damage, of a type that technically could have removed the last life, are looked at, and the first one found that fits the above is assigned the kill.

That was just an example. I was pointing out that the system for assigning damage-over-time kills takes into account immunities. That is not in any way tied to PVP.

I have no idea why me saying "I don't know when this will go live" is being intrpreted to mean "this will definitely go live before the next league", but that's not what I said. I was directly asked if this bug was intented, which is isn't, and responded to that. My job is to fix the bug, not chose when the fix is deployed.

I'd be 100% down with introducing negative charges as an actual mechanic, but that would mean actually following through with the concept, so e.g. the player's per-charge bonuses should also go negative if their current number of charges is negative. That wouldn't change the fact that Discharge specifically says it counts charges removed, so until it can actually remove charges below zero, getting bonuses as though it did so is clearly wrong.

No, it won't

Generally yes, excepting any that can be eliminated as not being able to contribute (such as poison against a CI character)