Anyone knows the mechanics of crushing blow?

[Ice]

After playing with 4-5 different crushing blow builds and doing a bit of testing in 1.10 beta, it does seem like the mechanics of crushing blow has changed quite a bit from 1.09:

- champs and bosses seem to take the same amount of damage as their normal counterparts,
- physical resistance affects CB but doesn't seem like a direct relation (ie. 50% res != 50% less damage),
- % damage done is lesser at higher player count,
- range attacks are still half the effectiveness of melee attacks though.

Does anyone know more precisely how crushing blow works in 1.10 beta?

[black_spy]

Hi,

i've looked a bit on the code calculating CB.

It seems [haven't checked all] to be calculated as follow

reduction_value = [HitPoints * 1 / ((x + [x * (#players - 1) * 0.5]) * 2 if ranged attack)]

x = 4 if normal Monster or an unique Version of it
x = 8 if act boss or SuperUnique (haven't checked this really - just a guess while looking at the code in the debugger)
x = 10 if player or merc

after this reduction_value is multiplied with the phys. resistence if there is any.


But there seems to be a bug in the calling code. First i did some tests with just one CB item, and everything worked as it should. Then i equiped a second CB item and the function was called twice.


HTH

[Ice]

Thanks a lot for your reply.

Does the last statement mean that CB percentages no longer stacks but each item is calculated separately? As in... having a guillaume's face (35%) + goblin toes (25%) gives you 48.75% of no CB, 42.5% of 1 CB and 8.75% of 2 CBs? Nope, just tried it with a 100% CB weapon and a 100% CB helm and it doesn't give 2 CBs per hit.

With the formula using melee and against normal monsters:
- players 1: 1/4_ or 25%
- players 2: 1/6_ or 17%
- players 3: 1/8_ or 13%
- players 4: 1/10 or 10%
- players 5: 1/12 or 8%
- players 6: 1/14 or 7%
- players 7: 1/16 or 6%
- players 8: 1/18 or 6%

Below are the results of a few tests I did using a socketd JahBerBerBerBerBer crystal sword and using phtoshop to measure the lengths of life bar (is that even accurate in the first place) on screenies taken after the first hit.

Zombie
(50% physical resistance)
Undamaged: 90 pixels
After first hit in players 1: 46 pixels (49%) - champion: 82/162 (49%)
After first hit in players 2: 58 pixels (35%)
After first hit in players 3: 64 pixels (29%) - boss: 154/214 (28%), 160/224 (28%)
After first hit in players 7: 72 pixels (20%)

Fallen
(15% physical resistance)
Undamaged: 90 pixels
After first hit in players 1: 28 pixels (69%)
After first hit in players 2: 42 pixels (53%)
After first hit in players 3: 52 pixels (42%)
After first hit in players 7: 68 pixels (24%)

Carrion Bird
(0% physical resistance)
Undamaged: 138 pixels
After first hit in players 1: 32-34 pixels (76%)
After first hit in players 2: 56 pixels (59%)
After first hit in players 3: 70 pixels (49%) - boss: 91/179 (49%)
After first hit in players 7: 98 pixels (29%) - champion: 150/212 (29%)

The damage definitely seems to be a lot more than what I would be expecting based on 1.09 or even the formula given. The effect of physical resistance on the % damage done also seems a bit strange based on the assumption that the monster's life bar are even accurate in the first place.

[Nicodemus Phaulkon]

Does the last statement mean that CB percentages no longer stacks but each item is calculated separately? As in... having a guillaume's face (35%) + goblin toes (25%) gives you 48.75% of no CB, 42.5% of 1 CB and 8.75% of 2 CBs? Nope, just tried it with a 100% CB weapon and a 100% CB helm and it doesn't give 2 CBs per hit.

You're confusing me, Icewraith. I recognized your post from the similar one from the Basin, but your explanation of what you believe is happening with multiple adders of Crushing Blow has me a tad confused.
From all appearances, it's seeming like multiple adders of CB actually decrease the effectiveness?

Explain it to me like I'm a three year old? ;)

[black_spy]

>Does the last statement mean that CB percentages no longer stacks but each item is calculated separately?

The chances stack as in 1.09, but the calculating Code seems to be called once for each equiped item with CB on it.


>As in... having a guillaume's face (35%) + goblin toes (25%) gives you 48.75% of no CB, 42.5% of 1 CB and 8.75% of 2 CBs?

No. 16% of no CB, ...


I've tried to reproduce your test, but my results are different to yours.

Zombie with 50% physical resistence and 100% magic resistence (i used berserk).

With just one CB item equiped:
Before first hit

After


Now with two CB items:
Before

After



Maybe there is another strange bug in the lifebar code :huh:


The code that calculates CB it at :6FD04560 in d2game.dll. You can take a look if you want.


Another guess for the differences: it seems like blizz changed the order of CB and normal dmg. CB is now applied before normal dmg.
Just take your Zombies, give them 100 life, take a weapon with 75-76 dmg and wonder why the zombies die after the first hit :huh:

[Nicodemus Phaulkon]

The chances stack as in 1.09, but the calculating Code seems to be called once for each equiped item with CB on it.

So Gorerider, Guillaume's Helm and a Ribcracker would give 100% Crushing blow, with the possibility that Crushing blow could be effected three separate times? Interesting.

Another guess for the differences: it seems like blizz changed the order of CB and normal dmg. CB is now applied before normal dmg.

Which makes it even more effective, as it would be applying its percentage damage against the larger sum of life before the normal physical damage.

While this looks insane at first glance, I'm wondering at this theory in the light of the fact that there seems to be diminishing returns when the player index rises. Crushing Blow's greatest aspect was the ability to whittle down the HUGE amounts of life that a P8 monster could possess... if that aspect has been changed to become a much lesser percentage, the multiple calls to Crushing Blow could be a purposeful feature, instead of a bug?

Now... how is Deadly Strike doing? ;)

[Occhidiangela]

Jarulf spelled out very succinctly that CB is a tag that is checked once per attack, that the chances are added and one check is made. If they changed that, I wonder at why.

Note:

LIghtning Fury Seems to be taking the CB bonus with it. I have level 3 Lightning Fury on a non optimal Versatile Zon, and she simply ate some Act V NM monsters alive. It also appears that the mana leach is also re attached to CB and LF, but I may not have been seeing it clearly.

[Ice]

I've tried to reproduce your test, but my results are different to yours.

Your tests definitely looks more logical and corresponds to what I would expect based on the formula (1/4 for a players 1 game and halved the effectiveness from 50% pr).

But I'm still lost as to why I'm still getting around damage around the 50% mark pretty consistently over probably 100+ different games and 1000+ hits on different zombies. Using two items with CB also didn't gave me extra damage from each crushing blow hit :(.

[Occhidiangela]

My observation on my level 42 Amazon in players 4 NM diff Act II was that CB seemed to be doing 50% versus 25% damage with only Goblin Toe equipped for CB.

I will have to look at it again.

[black_spy]

Thx for giving me a copy of your testing chars.

Now i have an idea about your observations.

I tried your 5 BER sword on a 50% phys. res. zombie with 100 life. The function was called 5 times:
life reduction
1: 6400h C80h
2: 5780h AF0h
3: 4C90h 992h
4: 42FEh 860h
5: 3A9Eh 754h

This values are 256ths. So after the 5th call the life of the zombie is 51.

After this normal damage is applied (but i think you used a 1-2 dmg sword).

This is a reduction of 49%.


/edit: Another crazy observation. The calling code must be very buggy :(
Having equiped the 5 BER sword while game creation leads to the 5 calls (once for each rune).
Equiping the sword before attacking a monster leads two only one call for the sword (+ calls from another items).

[Brista]

Is it summing the chances and calling 5 times

BER is, what?, 20% chance

So 5 BERs would sum to 100% chance

And then it checks for each BER

I wonder if 5 BERs + Guillaumes, Rattlecage and Goblin Toe would give 8 chances?

One warning: Blizzard have warned people that editors are producing strange (buggy) results and that they are not accepting data on bugs produced by editors

[black_spy]

>One warning: Blizzard have warned people that editors are producing strange (buggy)
>results and that they are not accepting data on bugs produced by editors

Hmm ... i didn't use a character editor. I only edited some txt files and started the game with -direct -txt parameters. Then i created a new character and did my tests.

I'd wonder if this is enough to produce strange results.

[HunkyDory]

One warning: Blizzard have warned people that editors are producing strange (buggy) results and that they are not accepting data on bugs produced by editors

And I can underwrite my fellow Amazon Basin and Lurkerlounge member Brista's warning.

DO NOT USE EDITED 1.10 CHARACTERS OR ITEMS TO TEST BUGS!

I did a lot of testing and just threw away a big stack of notes because they are worthless.

Let me just give one example .

A 40% / +10 Minimum Damage jewel correctly applying the 40% to Maximum Damage but not to the Minimum Damage.

You can use edited chars and items to get a "feelie" of the patch but that is all.

Regards, Hunky

edit: edited means imported too

[RedFin]

Jarulf spelled out very succinctly that CB is a tag that is checked once per attack, that the chances are added and one check is made.  If they changed that, I wonder at why.

Note:

LIghtning Fury Seems to be taking the CB bonus with it.  I have level 3 Lightning Fury on a non optimal Versatile Zon, and she simply ate some Act V NM monsters alive.  It also appears that the mana leach is also re attached to CB and LF, but I may not have been seeing it clearly.

Now, what do you mean on that CB/LF and mana leech?

Physical part of javelins were leeching in 1.09, and i am not 100% sure but i do think that physical javelins carried CB too. If you are talking that bolts carry CB, which they did not in 1.09 then this is very big change.

Then it should be checked also for other lightning skills.

RF

[Occhidiangela]

IN the LoD beta, and early patch, crushing blow and leach were carried on the Lightning fury's bolts, not just on the spear you threw.

That got fixed.

It looked to me as though that "feature" had returned.

I can say for certain that the same is NOT happening on Lightning Strike. :)