Combat Mechanics? Attack Speed, Recovery

[Elric Galad]

Yes, it stacks.

[Boeroer]

You can also switch weapons right after attack and have no recovery at all. Also skips reload animation sometimes. Very powerful, horrible micro.

[Blunderboss]

You can also switch weapons right after attack and have no recovery at all. Also skips reload animation sometimes. Very powerful, horrible micro.

This glitch still there ? I remember this was a thing after release of PoE , cipher with quickswitch could spam powers no recovery if switching between weapons while casting spells , i havent tried to use it on current playtrough, micro is so intense for this that its better left for solo playtroughs

Edited March 28, 2016 by Blunderboss

[Stasis_Sword]

You can also switch weapons right after attack and have no recovery at all. Also skips reload animation sometimes. Very powerful, horrible micro.

 

It skips the reload not just the reload animation/recover right? Fire pistol -> Dual Wield -> pistol will still result in reloading when you get back to the pistol right?

[Ichthyic]

It skips your recovery phase, not reload.  yes, that's correct.

 

 

You can also switch weapons right after attack and have no recovery at all.

yes, but that requires that you switch to the new weapon immediately after using the previous.

what the stacking quickness does, is that if you decide to switch while your action bar is already on cooldown, it will not bump your action bar progress back.  this is very valuable for long fights I have found.  which makes it very valuable for pretty much all of POTD
 

Edited March 28, 2016 by Ichthyic

[Boeroer]

No. It also skips reloading from time to time if you have both items and switch right after the bang. Not every time, but often. Timing is critical. Don't know if it only happens when you switch from one gun to the other or if it glitches with every weapon set. But every time I try a quickswitcher (cipher, paladin, rogue, ranger) they can skip reloading with this setup. It's not reliable and a micro nightmare, but it works. The reloading icon will appear. Then there will be a shudder in the animation and then the gunner will use his shooting animation and fire.

Edited March 29, 2016 by Boeroer

[Blunderboss]

No. It also skips reloading from time to time if you have both items and switch right after the bang. Not every time, but often. Timing is critical. Don't know if it only happens when you switch from one gun to the other or if it glitches with every weapon set. But every time I try a quickswitcher (cipher, paladin, rogue, ranger) they can skip reloading with this setup. It's not reliable and a micro nightmare, but it works. The reloading icon will appear. Then there will be a shudder in the animation and then the gunner will use his shooting animation and fire.

Normally this never happens tho , with a 6 man party its just too micro heavy ( to go for the glitch , normal QS still best way to play guns char ) . So this bug is not so gamebreaking .

Anyway i had side question , any1 tried quick switch with grimoires ?

Edited March 29, 2016 by Blunderboss

[Boeroer]

Actually I nevr switch grimoires - so nope. But now I'm curious what happens.

[Blunderboss]

Wizard Rogue maybe, switching between grimoires one for self buffing other for offensive spells , also u could switch between Gun and Dual Wield for free as a bonus or switch gun sets one with spell procs like jolting touch other for dps , i want to play such character on my next playtrough melee dual wield wizard should be interesting :D

Edited March 29, 2016 by Blunderboss

[bebeto]

I read all this thread since im very interested in game mechanics and to sum what was important and obvious to me and aparently works correct:

 

animation speed(dex based)   =default attack speed animation /{1+[3x(dex-10)/100]}

default recovery speed(melee)=default attack speed animationx 1.2 in case of dual weilding 

                                                =default attack speed animationx 1.8 for 2 handed weapons,weapon+shield or 1hand style  

              

default recovery speed(range)=default attack speed animationx  1.5 in case of firearms,arbalest,crossbow

                                                =default attack speed animationx  2 in case of bows

recovery speed=default recovery speedx(1+recovery penalty-recovery bonus)

reload time=base reload time/(AxBxCxD) where AxBxCxD-gunner chant swif aim dex.mod 1.2x1.2x1.5x1.3

Also Str isnt as vital as i thought initially.Str bonus is another additional bonus into the equation and higher the bonus u get from different else buffs such as sneak,deatblows or focus weapons for ex. Str starts to loose from effectiveness.While dex bonus applies to all 3 meaning animation speed,recovery and reload time and is multiplicative while bonuses in game from attack speed are additive and matters only for recovery speed.Reload time bonuses are all multiplicative including one u get from dex.

 

The reedit of post was because of some minor errors 1 example being with bow recovery wich i saw later is different than fireguns.Such observations i deducted from this default values i saw here:https://docs.google.com/spreadsheets/d/1vDm5MOrCK6S95h5u0EzZbGTv_u_rYqjuO0-zupqk0_A/edit#gid=1264774515.

Anyway i hope all is correct or close to it.

Edited February 20, 2017 by bebeto

[bebeto]

After some experimenting i found out recovery formula that i posted is not quite acurate.Its very weird since from my results atk speed and recovery penalty from armor is neither exclusively addive or multiplicative and pretty much sure is a combination of both.Seems its more complex but what im sure about is that atk speed especially when u have multiple bonuses become more important than penalty.

For example my archer with alacrity that give 50% atk speed bonus while wearing armor with 50% recovery penalty shots without penalty compared to same archer while naked but without alacrity wich have a slightly small recovery.The other bonuses in the experiment were atk speed chant(20%) +glove of swift(15%)+the rain of godagh field durganized wich all stack(20%,15%)

If it was multiplicative than recovery would be 0 or close for the naked experiment and its neither aditive since my chanter with 100% recovery total but with armor -50% penalty still has a recovery time.

This recovery mechanic is the only thing i dont understand how it works,pretty hard to figure this out.

[Khagmas]

 Help calculate the attack speed!

First option:

1. 20 Agility,
2. Breastplate Armor (-40% reduced), Durgan-Reinforced Armor with - 15% Armor Speed Penalty
3. Pilferer's Grip (gloves) with -10% armor Speed Penalty (Stuck with durgan-reinforced or not?)
4. Dagger with Durgan-Refined Weapon +15% Attack Speed
5. Sabre with Durgan-Refined Weapon +15% Attack Speed
6. Dagger and Sabre are dual wielding (increases attack speed by 20%)

Second option:

1. 20 Agility,
2. Breastplate Armor (-40% reduced), Durgan-Reinforced Armor with - 15% Armor Speed Penalty.
3. Dagger with Durgan-Refined Weapon +15% Attack Speed and 'speed' enchant + 20% Attack Speed (Total +35% or not? )

4. Sabre with Durgan-Refined Weapon +15% Attack Speed
5. Dagger and Sabre are dual wielding (increases attack speed by 20%)

 

Third option:

1. 20 Agility,
2. Padded armor (-20% reduced), Durgan-Reinforced Armor with - 15% Armor Speed Penalty.
3. Dagger with Durgan-Refined Weapon +15% Attack Speed

4. Sabre with Durgan-Refined Weapon +15% Attack Speed
5. Dagger and Sabre are dual wielding (increases attack speed by 20%)


I would like to see a visual formula to understand how the calculation is going on.
 

[MaxQuest]

I dunno how visual is that, but here you go:
 
Base values:
Sabre: 30 frames attack phase -> 50 frames recovery phase
Dagger: 20 frames attack phase -> 33.3 frames recovery phase
recovery_factor = 1.2
 

---

First option:
dex_coef = 1.3
speed_coef = 0.15 - (0.40 - 0.15) + 0.2 = 0.1
note: pilferer gloves bonus is surpressed by durganized armor enchant
 
Sabre:
attack_duration = base attack_duration / dex_coef = 30 / 1.3 = 23.07 frames
recovery_duration = attack_duration * max(0, 1 - 2 * speed_coef) / recovery_factor = 23.07 * max(0, 1 - 2 * 0.1) / 1.2 = 23.07 * 0.8 / 1.2 = 15.38 frames
 
Dagger:
attack_duration = 20 / 1.3 = 15.38 frames
recovery_duration = 15.38  * max(0, 1 - 2 * 0.1) / 1.2 = 15.38 * 0.8 / 1.2 = 10.25 frames
 

---

Second option:
same as #1
 

---

Third option:
dex_coef = 1.3
speed_coef = 0.15 - (0.20 - 0.15) + 0.2 = 0.3
 
Sabre:
attack_duration = 30 / 1.3 = 23.07 frames
recovery_duration = 23.07 * max(0, 1 - 2 * 0.3) / 1.2 = 23.07 * 0.4 / 1.2 = 7.69 frames
 
Dagger:
attack_duration = 20 / 1.3 = 15.38 frames
recovery_duration = 15.38  * max(0, 1 - 2 * 0.3) / 1.2 = 15.38 * 0.4 / 1.2 = 5.12 frames
 

---

Fourth (bonus) option:
If you had Gauntlets of Swift Action in your #3 scenario:
 
dex_coef = 1.3
speed_coef = (1.15*1.15 - 1) - (0.20 - 0.15) + 0.2 = 0.47
 
Sabre:
attack_duration = 30 / 1.3 = 23.07 frames
recovery_duration = 23.07 * max(0, 1 - 2 * 0.47) / 1.2 = 23.07 * 0.06 / 1.2 = 1.15 frames
 
Dagger:
attack_duration = 20 / 1.3 = 15.38 frames
recovery_duration = 15.38  * max(0, 1 - 2 * 0.47) / 1.2 = 15.38 * 0.06 / 1.2 = 0.77 frames

So for example in case of sabre, making one swing would go down from [5 + 30 + 50] frames to only [5 + 23.07 + 1.15]. Or ~x2.9 faster.

---

 

This recovery mechanic is the only thing i dont understand how it works,pretty hard to figure this out.

Hope this will help you out

Edited March 7, 2017 by MaxQuest

[bebeto]

Attack speed is what matters most,its more important than dex and aparently it has more impact than penalty from armor.In other words alacrity+50% armor penalty >naked without alacrity.My chanter with war bow and with 3 DEX has 0 recovery if its stacked with range atk speed chant+alacrity+the rain of godagh field durganized+gloves of swift.Wich is very weird because total bonus if it was additive would be 120% while penalty its -50%.If it was multiplicative total bonus would be 185% and would explain 0 recovery even with 50% recovery penalty.

But a naked char with multiplicative 100% still has a small recovery.So thats why im confused.

Dex i know for sure has no impact in making recovery to 0.Dex only give the values of attack speed/cast animation wich means the higher dex the less default values.What cuts recovery to 0 its attack speed only.So attack speed its the king here,because can make chars even with 2hand/bows with 3 dex having 0 recovery.Thats why i dislike mods like cautious attack,it does more harm than good.Even for tank if u take it ,than u can forget about dps and his role will be only as bait to soak damage nothing else.

Since i cant figure the exact formula and the formula i posted its not far from the truth but its not exactly correct either since it seems there might be some other variable involved with recovery bonus given by atk speed.

[Livegood118]

Helpful Info

 

 

Thanks for this. I was looking at the equations and I can't seem to find where the Dexterity co-efficient is applied to the recovery bonus though 

[JerekKruger]

Thanks for this. I was looking at the equations and I can't seem to find where the Dexterity co-efficient is applied to the recovery bonus though 

 

It's applied to the attack phase, and then the recovery phase is based off this modified attack phase, so it's there but a little hidden.

[Livegood118]

 

Thanks for this. I was looking at the equations and I can't seem to find where the Dexterity co-efficient is applied to the recovery bonus though

It's applied to the attack phase, and then the recovery phase is based off this modified attack phase, so it's there but a little hidden.

Ahh gotcha - thanks

[MaxQuest]

In other words alacrity+50% armor penalty >naked without alacrity

Without any other buffs or stuff that influences recovery, [DAoM + plate] and [no DAoM + naked] would net the same recovery duration.

But yes, if there are any recovery reducing effects, the first variant would be much more favourable, as majority of coefficients do stack with DAoM multiplicatively (unless suppressed).

 

My chanter with war bow and with 3 DEX has 0 recovery if its stacked with range atk speed chant+alacrity+the rain of godagh field durganized+gloves of swift.

Wich is very weird because total bonus if it was additive would be 120% while penalty its -50%.If it was multiplicative total bonus would be 185% and would explain 0 recovery even with 50% recovery penalty.

But a naked char with multiplicative 100% still has a small recovery.So thats why im confused.

That's because the stacking is neither pure additive, nor pure multiplicative.

- alacrity (daom), speed (rain of goddah), gloves and durganized weapon enchant stack multiplicatively between themselves [1.5 * 1.2 * 1.15 * 1.15 = 2.38]

- while sure-handed chant makes part of other 'speed category' (called 'Rate of Fire'), and it stacks additively with the above value [(2.38 - 1) + (1.2 - 1) = 1.58]

 

Now substract from that 1.58 a 0.5 penalty for making an attack with single weapon. Plus substract extra 0.5 due to 50% recovery penalty, and you end up with 0.58 coefficient, which is enough to achieve zero recovery, because: recovery_duration = attack_duration / recovery_factor * max(0, 1 - 2 * speed_coef)

 

---

Thanks for this. I was looking at the equations and I can't seem to find where the Dexterity co-efficient is applied to the recovery bonus though 

Jerek explained it right.

 

We can write the same thing in two ways:

attack_duration = base_attack_duration / dex_coef
recovery_duration = attack_duration / recovery_factor * max(0, 1 - 2 * speed_coef)

or:

attack_duration = (base_attack_duration / dex_coef)
recovery_duration = (base_recovery_duration / dex_coef) * max(0, 1 - 2 * speed_coef)
where:
base_recovery_duration = base_attack_duration / recovery_factor

Same thing. I've chosen the first, because that's how it is presented in code. The game doesn't stock base recovery values; recovery is calculated every swing based on total animation duration of last attack phase.

[bebeto]

Thanks for clarifications Max Quest.DAOM is indeed op when stacked with others bonuses such durganized and speed bonus from weapons or with passive such atk speed from dual weilding or from Sure-Handed Ila Nocked Her Arrows with Speed.This multiplication factor allows u still have a low recovery while wearing a heavy armor.

Btw Sure-Handed Ila Nocked her arrows with speed can reduce recovery from some spells of priest(buff ones).For my cipher it didnt Its very weird this chant since can reduce some and on others has no effect.In theory it shouldnt at all but for sure priest can buff faster the party with this chant on.

[Livegood118]

Same thing. I've chosen the first, because that's how it is presented in code. The game doesn't stock base recovery values; recovery is calculated every swing based on total animation duration of last attack phase.

 

 

Thanks again.

 

I was just wondering – are you working towards creating some kind of calculator utility/spreadsheet to work all this out? 

[MaxQuest]

Btw Sure-Handed Ila Nocked her arrows with speed can reduce recovery from some spells of priest(buff ones).For my cipher it didnt Its very weird this chant since can reduce some and on others has no effect.In theory it shouldnt at all but for sure priest can buff faster the party with this chant on.

Yeap. Sure-Handed makes part of Rate of Fire category. Besides ranged weapons recovery, it also affects recovery of spells that are internally considered as AttackRanged (basically majority of targeted spells)

 

I was just wondering – are you working towards creating some kind of calculator utility/spreadsheet to work all this out?

Not working, but I was thinking of that. If I'll have enough time, I actually could do such calculator. We'll see.

[bebeto]

 

Btw Sure-Handed Ila Nocked her arrows with speed can reduce recovery from some spells of priest(buff ones).For my cipher it didnt Its very weird this chant since can reduce some and on others has no effect.In theory it shouldnt at all but for sure priest can buff faster the party with this chant on.

Yeap. Sure-Handed makes part of Rate of Fire category. Besides ranged weapons recovery, it also affects recovery of spells that are internally considered as AttackRanged (basically majority of targeted spells)

 

I was just wondering – are you working towards creating some kind of calculator utility/spreadsheet to work all this out?

Not working, but I was thinking of that. If I'll have enough time, I actually could do such calculator. We'll see.

 

MaxQuest i analyze your formula but something is unclear to me.Your formula for recovery is:

Recovery phase duration=Attack phase duration * max (0,(1-2*speed coef_))/recovery factor

where recovery factor is 1.2

But acording to this tests from this list wich were performed by naked chars meaning speed coef_ being 0

https://docs.google.com/spreadsheets/d/1vDm5MOrCK6S95h5u0EzZbGTv_u_rYqjuO0-zupqk0_A/edit#gid=1264774515

Lets take dual weilding for example.A naked char with dual weilding stilettos with 10 dex has an attack animation of 20 frames while recovery is 24.Acording to your formula recovery should be 20/1.2  meaning ~16.Shouldnt recovery factor be a multiplier wich would explain the numbers on the list meaning 20*1.2=24?

Also want to ask u if atk speed bonus that u get from dual weapon passive is multiplicative?Or is additive like in case with sure ila-handed for range?

[JerekKruger]

@bebeto I can't speak to your question, though MaxQuest has his own table of attack speeds for weapons here which I believe he found through in game tests. Given the amount of work Max put into finding out the formula for attack speed, I'd be inclined to believe this table over the one you link, though of course it might be a mistake on Max's part. Hopefully he'll let us know.

 

I can answer your second question though: if you look at Max's formula carefully you'll notice that the RED_coef component of his speed_coef covers this bonus. However it isn't actually a bonus, it's a -0.5 penalty applied to characters who aren't dual wielding (which can be reduced to a -0.35 with a Durgan reinforced shield). However, to answer your question, this penalty is additive with other recovery speed modifiers.

 

However I will note that Sure-Handed Ila is not really additive. As a BLUE_coef you multiply it with all other Blue_coefs (which are the majority of speed bonuses), then add the product to GREEN_coefs and RED_coefs. 

[bebeto]

@bebeto I can't speak to your question, though MaxQuest has his own table of attack speeds for weapons here which I believe he found through in game tests. Given the amount of work Max put into finding out the formula for attack speed, I'd be inclined to believe this table over the one you link, though of course it might be a mistake on Max's part. Hopefully he'll let us know.

 

I can answer your second question though: if you look at Max's formula carefully you'll notice that the RED_coef component of his speed_coef covers this bonus. However it isn't actually a bonus, it's a -0.5 penalty applied to characters who aren't dual wielding (which can be reduced to a -0.35 with a Durgan reinforced shield). However, to answer your question, this penalty is additive with other recovery speed modifiers.

 

However I will note that Sure-Handed Ila is not really additive. As a BLUE_coef you multiply it with all other Blue_coefs (which are the majority of speed bonuses), then add the product to GREEN_coefs and RED_coefs. 

Sure-Handed Ila its not multiplicative with general atk speed buffs such frenzy,alacrity etc but its rather additive.But if it would be to quess its multiplicative with same type of buffs if exists(cant think other than swift mode from ranger and i think one gets supressed because dont stack....wich im not really sure didnt test it)..As MaxQuest pointed its different cattegory such as range atk speed.I just wonder if this principle applies to melee atk speed skills such as dual weild talent wich applies only for melees.

I have no doubt that MaxQuest knows his math.That list i found its in same thread and its abit contradictory with MaxQuest formula or maybe i skiped something and its my fault.What is important its to figure this straight doesnt matter who is wrong or right.

Edited March 14, 2017 by bebeto

[JerekKruger]

Sure-Handed Ila its not multiplicative with general atk speed buffs such frenzy,alacrity etc but its rather additive.

 

If MaxQuest's formula is correct then Sure-Handed Ila is multiplicative with Frenzy or Deleterious Alacrity of Motion (not both, since they don't stack). Look at his formula here.

 

As MaxQuest pointed its different cattegory such as range atk speed.

 

It's listed under a different tab in his table, but it ultimately interacts in the formula the same way i.e. it's a BLUE_coef and is multiplied with all other BLUE_coefs (including Frenzy and DAOM) to obtain a total which is then added to the totals for GREEN_coefs and RED_coefs.

 

What is important its to figure this straight doesnt matter who is wrong or right.

 

Agreed, what matters is finding out the correct formula.

 

I have no doubt that MaxQuest knows his numbers.That list i found its in same thread and its abit contradictory with MaxQuest formula or maybe i skiped something and its my fault.

 

Well, when Max posted his findings they didn't agree with the current thinking at the time. My reason for believing Max rather than the traditional model is the fact that he not only tested things in game, he also looked at the decompiled code. I might be wrong, but I don't think anyone had done that before.

Edited March 14, 2017 by JerekKruger