 |
Range/Melee/Pet Haste bonuses stack question |
 |
07-12-2013, 12:42 AM
|
#1
|
Rare Collector
Atropos is offline
Join Date: Dec 2011
Location: World of Pain
Posts: 896
|
Range/Melee/Pet Haste bonuses stack question
Glitchy, how do attack speed bonuses such as Range/Melee/Pet Haste bonuses stack? Are they multiplicative or do they just summarize? I mean bonuses such as for example the haste bonus from Quiver and speed bonus from Swift arrow rune or Regulus/Aliangel BP ...
|
|
|
07-12-2013, 09:20 AM
|
#2
|
Administrator
Glitchless is offline
Join Date: May 2008
Posts: 8,569
|
The stacking is cumulative without any loss of effectiveness. It would best be described as multiplicative, and to find the factor by which you multiply you do this:
100 / (100 + Percent Bonus)
So a 100% haste bonus is 100 / (100+100) or a .5 multiplier of your original delay. You can repeat the process with all your haste bonuses on the resulting delay, order will not affect it.
__________________
Computing the probability that at least one of the following events will occur:
P(a or b ... or z) = 1 - P(!a and !b ... and !z)
Probability
|
|
|
07-12-2013, 12:45 PM
|
#3
|
Rare Collector
Atropos is offline
Join Date: Dec 2011
Location: World of Pain
Posts: 896
|
Clear to me now, thank You for explaning the mechanics  !
|
|
|
07-12-2013, 03:43 PM
|
#4
|
Gem Pouch Expert
Hooker is offline
Join Date: May 2011
Location: concerned white citizen
Posts: 469
|
i do not get this at all. i have melee haste r2 and get 6% bonus, sharpen steel r2 and get 5% bonus, and quicksteel r5+++ and get 10% bonus. do you just add the % all together? so i would get a 21% bonus to melee haste? or is each individual one taken and then added together? like for melee haste r2 100/(100+6) and then repeat for the others.
__________________
Quote:
Originally Posted by Hoodrat
Stand your ground, concerned white citizen !
|
Quote:
Originally Posted by Glitchless
Player mistakes have to be lived with.
|
Quote:
Originally Posted by Glitchless
I have it on good authority that the founding fathers were pro PKing.
|
|
|
|
07-12-2013, 05:09 PM
|
#5
|
Seer's BFF
Lurkee is offline
Join Date: Mar 2009
Posts: 707
|
Quote:
Originally Posted by Glitchless
You can repeat the process with all your haste bonuses on the resulting delay
|
hooker,
maybe do the math as described and it will make sense
__________________
If you make a sad face when you sing I hate you, die.
|
|
|
07-12-2013, 05:46 PM
|
#6
|
Gem Pouch Expert
Hooker is offline
Join Date: May 2011
Location: concerned white citizen
Posts: 469
|
clearly you dont get what "i do not get this at all" means.
for me nod math is confusing i cant do it myself. hence why i am asking. i did not read close enough to see that he said it repeats since i had my own questions with the beginning part of his post. go troll some where else
__________________
Quote:
Originally Posted by Hoodrat
Stand your ground, concerned white citizen !
|
Quote:
Originally Posted by Glitchless
Player mistakes have to be lived with.
|
Quote:
Originally Posted by Glitchless
I have it on good authority that the founding fathers were pro PKing.
|
Last edited by Hooker; 07-12-2013 at 05:49 PM..
|
|
|
07-12-2013, 05:48 PM
|
#7
|
Seer's BFF
Lurkee is offline
Join Date: Mar 2009
Posts: 707
|
no, i think i do
it means "help me do something that has already been explained to me, but im too lazy to do it."
you can't possibly be too stupid to take glitchless's formula and plug in the numbers
__________________
If you make a sad face when you sing I hate you, die.
|
|
|
07-12-2013, 05:51 PM
|
#8
|
Gem Pouch Expert
Hooker is offline
Join Date: May 2011
Location: concerned white citizen
Posts: 469
|
ya actually i am too stupid
__________________
Quote:
Originally Posted by Hoodrat
Stand your ground, concerned white citizen !
|
Quote:
Originally Posted by Glitchless
Player mistakes have to be lived with.
|
Quote:
Originally Posted by Glitchless
I have it on good authority that the founding fathers were pro PKing.
|
|
|
|
07-12-2013, 05:54 PM
|
#9
|
Gem Pouch Expert
Hooker is offline
Join Date: May 2011
Location: concerned white citizen
Posts: 469
|
ok i did the math and got 2.804 you seem to know it what do i do with that? or is that the 21%?
__________________
Quote:
Originally Posted by Hoodrat
Stand your ground, concerned white citizen !
|
Quote:
Originally Posted by Glitchless
Player mistakes have to be lived with.
|
Quote:
Originally Posted by Glitchless
I have it on good authority that the founding fathers were pro PKing.
|
Last edited by Hooker; 07-12-2013 at 06:32 PM..
|
|
|
07-12-2013, 06:03 PM
|
#10
|
Seer's BFF
Lurkee is offline
Join Date: Mar 2009
Posts: 707
|
show your work
__________________
If you make a sad face when you sing I hate you, die.
|
|
|
07-12-2013, 06:32 PM
|
#11
|
Gem Pouch Expert
Hooker is offline
Join Date: May 2011
Location: concerned white citizen
Posts: 469
|
(100/(100+6))+(100/(100+5))+(100/(100+10)=2.804
__________________
Quote:
Originally Posted by Hoodrat
Stand your ground, concerned white citizen !
|
Quote:
Originally Posted by Glitchless
Player mistakes have to be lived with.
|
Quote:
Originally Posted by Glitchless
I have it on good authority that the founding fathers were pro PKing.
|
|
|
|
07-12-2013, 08:08 PM
|
#12
|
Seer's BFF
Lurkee is offline
Join Date: Mar 2009
Posts: 707
|
don't add them, multiply them
(100/(100+6))*(100/(100+5))*(100/(100+10) = .8168
then multiply that number by your original delay to get your new delay
ie original delay of 30, 30 * .8168 = 24.5
if you wanted to figure out what that ends up being if it were one single haste buff you do
100/(100+x) = .8168 and solve for x
100 = 81.68 + .8168x
x = 22.4
so 6%, 5%, and 10% haste actually end up being more than the 21% haste that one might approximate if they just added them. you could also get the number by doing 1.05 * 1.06 * 1.10
__________________
If you make a sad face when you sing I hate you, die.
Last edited by Lurkee; 07-12-2013 at 08:24 PM..
|
|
|
07-12-2013, 08:48 PM
|
#13
|
Gem Pouch Expert
Hooker is offline
Join Date: May 2011
Location: concerned white citizen
Posts: 469
|
thank you for the help
__________________
Quote:
Originally Posted by Hoodrat
Stand your ground, concerned white citizen !
|
Quote:
Originally Posted by Glitchless
Player mistakes have to be lived with.
|
Quote:
Originally Posted by Glitchless
I have it on good authority that the founding fathers were pro PKing.
|
|
|
|
07-12-2013, 09:05 PM
|
#14
|
Seer's BFF
Lurkee is offline
Join Date: Mar 2009
Posts: 707
|
we can be frenz nao
__________________
If you make a sad face when you sing I hate you, die.
|
|
|
07-13-2013, 08:50 AM
|
#15
|
Gem Pouch Expert
Hooker is offline
Join Date: May 2011
Location: concerned white citizen
Posts: 469
|
sure
__________________
Quote:
Originally Posted by Hoodrat
Stand your ground, concerned white citizen !
|
Quote:
Originally Posted by Glitchless
Player mistakes have to be lived with.
|
Quote:
Originally Posted by Glitchless
I have it on good authority that the founding fathers were pro PKing.
|
|
|
|
07-13-2013, 11:33 AM
|
#16
|
Epic Scholar
flipynifty is offline
Join Date: Aug 2008
Posts: 4,529
|
Lmfao
__________________
omg i wanna sammich nao
|
|
|
12-28-2013, 03:57 PM
|
#17
|
Gem Pouch Expert
Zan is offline
Join Date: Feb 2011
Posts: 269
|
Bump for questioning:
With penalties such as 25% per whip, is the multiplier given to mod the delay (100 + Penalty)/100?
And with two whips or whip+shield do I compute each extra delay separately or add them together and then apply the mod?
EDIT: well that cant be the formula.. or a 50% penalty would mean x1.5 the delay.. but I understand the game sees a 50% penalty as double delay.. would love if someone could point out the direction.
Last edited by Zan; 12-28-2013 at 04:08 PM..
|
|
|
12-28-2013, 05:16 PM
|
#18
|
Administrator
Glitchless is offline
Join Date: May 2008
Posts: 8,569
|
Quote:
Originally Posted by Glitchless
The stacking is cumulative without any loss of effectiveness. It would best be described as multiplicative, and to find the factor by which you multiply you do this:
100 / (100 + Percent Bonus)
So a 100% haste bonus is 100 / (100+100) or a .5 multiplier of your original delay. You can repeat the process with all your haste bonuses on the resulting delay, order will not affect it.
|
It's the same formula as in the quote above, but you're subtracting.
100 / (100 - 50) is 2. ie a 50% penalty to speed means your attacks take twice as long.
__________________
Computing the probability that at least one of the following events will occur:
P(a or b ... or z) = 1 - P(!a and !b ... and !z)
Probability
|
|
|
12-28-2013, 11:21 PM
|
#19
|
Gem Pouch Expert
Zan is offline
Join Date: Feb 2011
Posts: 269
|
Thanks for answering.
I do believe, however, that this brings a problem with "stalking" skill. I'll try to be clear in explaining:
I assume the damage multiplier is linear with the skill lvl, so
xDAMAGE = 1 + SS/100, capping at 2 with skill 100
And the delay, by the formula you gave me:
xDELAY = 1/(1-0.4*SS/100), capping at 40% delay at skill 100 (xDELAY = 1,6667)
So the final formula to apply to the previous DPS would be
DPS * (xDAMAGE/xDELAY), which caps at 20% more damage.
However,
(xDAMAGE/xDELAY) = (1+SS/100) / ( 1/(1-0.4*SS/100) )
is non-linear and has its peak at skill 75, with 22,5% extra damage.
Where "SS" stands for Stalking Skill
Either the xDAMAGE linear assumption is wrong or there is an issue with the way its calculated.
Last edited by Zan; 12-29-2013 at 06:48 AM..
|
|
|
 |
|
 |
12-29-2013, 10:08 AM
|
#20
|
Administrator
Glitchless is offline
Join Date: May 2008
Posts: 8,569
|
That's an interesting find, worthy of a TC reward.
Though this doesn't necessarily need a "fix", since PvPers may want the higher burst, we decided to make the delay curve logarithmic. The result is an initial minor loss in DPS when you first train the skill that is regained as you surpass level 25. Here is the resulting DPS mods for whip users:
0.9696
0.962300087
0.958639507
0.9568
0.956085145
0.956141635
0.956761844
0.957811948
0.9592
0.960859783
0.962741859
0.964808248
0.967029082
0.969380423
0.971842766
0.9744
0.977038657
0.97974736
0.98251641
0.985337474
0.988203336
0.991107711
0.994045089
0.997010618
1
1.003009417
1.006035457
1.009075066
1.012125496
1.01518427
1.018249147
1.021318096
1.024389267
1.027460978
1.030531692
1.0336
1.036664613
1.039724347
1.042778111
1.045824902
1.048863793
1.051893928
1.054914516
1.057924824
1.060924172
1.063911929
1.066887509
1.069850369
1.0728
1.075735931
1.078657723
1.081564965
1.084457275
1.087334296
1.090195694
1.093041158
1.095870397
1.09868314
1.10147913
1.104258132
1.107019921
1.10976429
1.112491044
1.1152
1.117890989
1.12056385
1.123218435
1.125854604
1.128472227
1.131071182
1.133651356
1.136212641
1.138754941
1.141278161
1.143782217
1.146267029
1.148732521
1.151178626
1.15360528
1.156012422
1.1584
1.160767962
1.163116262
1.165444858
1.16775371
1.170042784
1.172312047
1.17456147
1.176791026
1.179000693
1.18119045
1.183360278
1.185510161
1.187640086
1.189750041
1.191840017
1.193910005
1.195960001
1.19799
1.2
__________________
Computing the probability that at least one of the following events will occur:
P(a or b ... or z) = 1 - P(!a and !b ... and !z)
Probability
|
|
|
All times are GMT -5. The time now is 10:10 PM Boards live since 05-21-2008 |
|
|
|