• Log InLog In
  • Register
Liquid`
Team Liquid Liquipedia
EDT 09:15
CEST 15:15
KST 22:15
  • Home
  • Forum
  • Calendar
  • Streams
  • Liquipedia
  • Features
  • Store
  • EPT
  • TL+
  • StarCraft 2
  • Brood War
  • Smash
  • Heroes
  • Counter-Strike
  • Overwatch
  • Liquibet
  • Fantasy StarCraft
  • TLPD
  • StarCraft 2
  • Brood War
  • Blogs
Forum Sidebar
Events/Features
News
Featured News
Serral wins EWC 202521Tournament Spotlight: FEL Cracow 20259Power Rank - Esports World Cup 202580RSL Season 1 - Final Week9[ASL19] Finals Recap: Standing Tall15
Community News
[BSL 2025] H2 - Team Wars, Weeklies & SB Ladder2EWC 2025 - Replay Pack2Google Play ASL (Season 20) Announced35BSL Team Wars - Bonyth, Dewalt, Hawk & Sziky teams10Weekly Cups (July 14-20): Final Check-up0
StarCraft 2
General
Greatest Players of All Time: 2025 Update #1: Maru - Greatest Players of All Time Serral wins EWC 2025 Power Rank - Esports World Cup 2025 EWC 2025 - Replay Pack
Tourneys
Sea Duckling Open (Global, Bronze-Diamond) TaeJa vs Creator Bo7 SC Evo Showmatch Sparkling Tuna Cup - Weekly Open Tournament FEL Cracov 2025 (July 27) - $10,000 live event Esports World Cup 2025
Strategy
How did i lose this ZvP, whats the proper response
Custom Maps
External Content
Mutation # 484 Magnetic Pull Mutation #239 Bad Weather Mutation # 483 Kill Bot Wars Mutation # 482 Wheel of Misfortune
Brood War
General
Shield Battery Server New Patch Google Play ASL (Season 20) Announced Help: rep cant save [G] Progamer Settings StarCraft & BroodWar Campaign Speedrun Quest
Tourneys
[BSL] Non-Korean Championship - Final weekend [Megathread] Daily Proleagues [BSL20] Non-Korean Championship 4x BSL + 4x China CSL Xiamen International Invitational
Strategy
Does 1 second matter in StarCraft? Simple Questions, Simple Answers Muta micro map competition [G] Mineral Boosting
Other Games
General Games
Nintendo Switch Thread Stormgate/Frost Giant Megathread Total Annihilation Server - TAForever [MMORPG] Tree of Savior (Successor of Ragnarok) Path of Exile
Dota 2
Official 'what is Dota anymore' discussion
League of Legends
Heroes of the Storm
Simple Questions, Simple Answers Heroes of the Storm 2.0
Hearthstone
Heroes of StarCraft mini-set
TL Mafia
TL Mafia Community Thread Vanilla Mini Mafia
Community
General
US Politics Mega-thread Things Aren’t Peaceful in Palestine UK Politics Mega-thread Russo-Ukrainian War Thread Stop Killing Games - European Citizens Initiative
Fan Clubs
INnoVation Fan Club SKT1 Classic Fan Club!
Media & Entertainment
Anime Discussion Thread [\m/] Heavy Metal Thread Movie Discussion! [Manga] One Piece Korean Music Discussion
Sports
2024 - 2025 Football Thread Formula 1 Discussion TeamLiquid Health and Fitness Initiative For 2023 NBA General Discussion
World Cup 2022
Tech Support
Gtx660 graphics card replacement Installation of Windows 10 suck at "just a moment" Computer Build, Upgrade & Buying Resource Thread
TL Community
TeamLiquid Team Shirt On Sale The Automated Ban List
Blogs
Ping To Win? Pings And Their…
TrAiDoS
momentary artworks from des…
tankgirl
from making sc maps to makin…
Husyelt
StarCraft improvement
iopq
Socialism Anyone?
GreenHorizons
Eight Anniversary as a TL…
Mizenhauer
Flash @ Namkraft Laddernet …
namkraft
Customize Sidebar...

Website Feedback

Closed Threads



Active: 647 users

Dual Wielding Details

Forum Index > Diablo 3
Post a Reply
Normal
inReacH
Profile Blog Joined August 2008
Sweden1612 Posts
Last Edited: 2012-02-20 06:49:26
February 20 2012 02:43 GMT
#1
Hey all!

Starting a new video series today where I go beneath the surface of the systems being used in Diablo 3 and try to answer questions people have/will have about the less obvious mechanics in the game.

My goal with each video will be to get to questions that will actually allow for us to make better decisions when we play the game. You can see an example of a very "actionable" question in the third question of this episode.

The three questions I'll be answering today are:
1. How much does dual wielding increase attack speed?
2. Is your attack speed when dual wielding an average of each (weapon's) speed or independent?
3. Is it always better to be using two weapons?



Hope you guys enjoy and learn something from this.
I love making this style of video because I learn a lot myself, please ask questions you have and if the answer is something I think people want to know I'll include it in one of the videos!
=]
Nunc
Profile Joined January 2012
10 Posts
February 20 2012 02:48 GMT
#2
The most accurate numbers I've seen so far, nice job!
ShadowDrgn
Profile Blog Joined July 2007
United States2497 Posts
February 20 2012 03:16 GMT
#3
1. Don't spend 30 seconds on decimals to 6 places or fractions. The approximation is fine.
2. Your conclusion on the last question really sucked because you only provided one data point. How much of a DPS increase do you get by wielding 2 clubs versus 1, for instance? How big of a disparity in damage needs to exist before you do more damage with a single weapon?
3. No one cares where you're standing in the game.
Of course, you only live one life, and you make all your mistakes, and learn what not to do, and that’s the end of you.
LoliKuma
Profile Joined June 2010
United States237 Posts
February 20 2012 03:22 GMT
#4
Thanks a lot, great information in the video

Also ShadowDrgn quit being such a douche when someone's giving free info.
The End DOES Justifiy the Means
tWR
Profile Blog Joined March 2011
Canada138 Posts
February 20 2012 03:22 GMT
#5
Well done, I loved your math it was really detailed for a guide to dual wielding! Perhaps for another one of your videos check the correlation of how your attack speed relates to you spell casting speed and which weapons maybe better for which spells.

(i.e. a having a wizard with a fast wand and focus, compared to a slower but stronger staff)

Just some food for thought, keep up the great work!
dacthehork
Profile Blog Joined February 2008
United States2000 Posts
February 20 2012 03:43 GMT
#6
doesn't go into enough detail

1. Does damage of skills change based on what "weapon is ready" to be used? or is DPS averaged etc.
2. Does this mean we would want to use huge damage skills while a slow but high damage is "ready", and use a really high attack speed offhand to help with rage gaining etc.

Also can you swap from dual wield to two handed on the fly for more damage?
Warturtle - DOTA 2 is KING
inReacH
Profile Blog Joined August 2008
Sweden1612 Posts
Last Edited: 2012-02-20 03:59:21
February 20 2012 03:45 GMT
#7
On February 20 2012 12:43 dacthehork wrote:
doesn't go into enough detail

1. Does damage of skills change based on what "weapon is ready" to be used? or is DPS averaged etc.
2. Does this mean we would want to use huge damage skills while a slow but high damage is "ready", and use a really high attack speed offhand to help with rage gaining etc.

Also can you swap from dual wield to two handed on the fly for more damage?


Well the thing is that those questions you asked there that I agree are related to the stuff covered in this video also lead to other related questions of their own. I wanted to make the video clear and concise and I wouldn't have wanted it any longer that it is now.

Thanks for commenting and being the first one to ask more questions!
Assault_1
Profile Joined April 2009
Canada1950 Posts
February 20 2012 03:59 GMT
#8
so if left/right hand are independent then (for example) you can swing your left hand 3 times for every right-hand swing?
Nunc
Profile Joined January 2012
10 Posts
Last Edited: 2012-02-20 05:02:00
February 20 2012 04:17 GMT
#9
To answer some of the details you guys didn't take the hint on:

The game actually automatically assigns one of your hands to determine the weapon damage/DPS used for each attack. This assignment is always sequential and damage is strictly based on that weapon and not the other! It's in this sense that they are "independent", no averages apply, but no multiple attacks from the same hand in a row either (certain skills will always use main/right hand though)! The only benefit you get from dual-wielding is a passive ~14.6% attack speed bonus.

Whether it always starts with main-hand or off-hand after each break is a detail that I personally wanted to check and is up to further testing and videos.

Surely another question this brings up is how often you'd want to swap your weapons around. You can indeed swap weapons all you want, but right-clicking on a weapon in your inventory always replace the main-hand only. Not sure if there's another way to do it other than point-and-clicking both weapons when you want to go back to dual-wielding.

Beta patch 13 also introduces the notion that some skills will always use your main-hand weapon and never rotate weapons between those specific attacks (the example of Cyclone Strike was given, and it clearly is a skill that doesn't use a visible "weapon swing").

If you ask about averages, the video shows very well how it works since that was one of the actual questions, starting at 2:30. Please watch it carefully.

Finally, if you feel like telling everyone what to do, you're the one who needs to realize no one really has to care about what you say either. It's his video and he made legitimate choices.


Logo
Profile Blog Joined April 2010
United States7542 Posts
February 20 2012 04:30 GMT
#10
Good video one point though from patch 13...

When dual-wielding, certain skills will now always use your main-hand weapon to determine damageFor example, Cyclone Strike will use your main-hand weapon damage even if your off-hand weapon is set to swing next


It's not entirely clear what the full range of skills that apply are, but if any skills without a cooldown apply to this you could get some pretty big damage increase by having a fast attacking offhand and a heavy hitting mainhand.
Logo
Nunc
Profile Joined January 2012
10 Posts
Last Edited: 2012-02-20 04:38:52
February 20 2012 04:33 GMT
#11
Yep I think your highest DPS should always be kept on MH, they just gave you a reason to since if you don't do that you'd be screwing yourself over - I'm guessing this change was because if it was random then you'd be losing DPS for no reason, now you can place the weapon accordingly to get the maximum benefit.

I expect people who want fast attacks will usually pick similarly quick weapons though, this patch change mainly avoids missing out on DPS that you could be doing.

It does make it more clear how you can calculate the DPS you get from using different skills without swapping weapons, but the options remain the same as they were before.
Soft`Soap
Profile Joined March 2011
Canada865 Posts
February 20 2012 06:07 GMT
#12
hey inreach
you should start posting these on the blizzard forums and maybe you can become a community mvp
I really appreciate you doing all this stuff(especially the people without beta like me)
I know that if you posted these on the blizz forums as much as you do on here I would nominate you for MVP, you really deserve it

anyways, good videos as usual
thanks!
MiXyass DjLadyDana SoftSoap RightClick DigicidaL l)H[Zodiak] 58^^
inReacH
Profile Blog Joined August 2008
Sweden1612 Posts
February 20 2012 07:17 GMT
#13
On February 20 2012 13:30 Logo wrote:
Good video one point though from patch 13...

Show nested quote +
When dual-wielding, certain skills will now always use your main-hand weapon to determine damageFor example, Cyclone Strike will use your main-hand weapon damage even if your off-hand weapon is set to swing next


It's not entirely clear what the full range of skills that apply are, but if any skills without a cooldown apply to this you could get some pretty big damage increase by having a fast attacking offhand and a heavy hitting mainhand.


Great note, looks like I'll have to do a continuation of this topic... or two
I am glad I didn't try to stuff too much information into one video though and I think this is a good start.
Theorycrafting is great but there's something to be said for small doses of it unless it's for a specific purpose.
Meta
Profile Blog Joined June 2003
United States6225 Posts
February 20 2012 07:22 GMT
#14
When I played through the beta on a barbarian I was thoroughly unimpressed with dual wielding. It always seemed to lower my dps compared to using a single 2handed weapon. That's probably just the items at the start of act 1 though, late game items might be really high dps 1handers compared to 2handers.
good vibes only
inReacH
Profile Blog Joined August 2008
Sweden1612 Posts
Last Edited: 2012-02-20 22:35:01
February 20 2012 07:31 GMT
#15
Most of the highest damage weapons in the beta are single handed... and they are all better in pairs
papaz
Profile Joined December 2009
Sweden4149 Posts
Last Edited: 2012-02-20 08:05:23
February 20 2012 08:05 GMT
#16
Nice video.

I didn't quite understand why the dps decreased in the last case? What was the calculation/mechanics behind a dps decrease when dual-wielding?
kaige
Profile Joined January 2011
United States10 Posts
February 20 2012 09:11 GMT
#17
On February 20 2012 17:05 papaz wrote:
Nice video.

I didn't quite understand why the dps decreased in the last case? What was the calculation/mechanics behind a dps decrease when dual-wielding?


The dps increase from the attack speed wasn't enough to offset the decrease in dps from using a much lower quality offhand weapon.
papaz
Profile Joined December 2009
Sweden4149 Posts
February 20 2012 09:18 GMT
#18
On February 20 2012 18:11 kaige wrote:
Show nested quote +
On February 20 2012 17:05 papaz wrote:
Nice video.

I didn't quite understand why the dps decreased in the last case? What was the calculation/mechanics behind a dps decrease when dual-wielding?


The dps increase from the attack speed wasn't enough to offset the decrease in dps from using a much lower quality offhand weapon.


Ah, now I understand

thanks
ShadowDrgn
Profile Blog Joined July 2007
United States2497 Posts
February 20 2012 09:20 GMT
#19
On February 20 2012 17:05 papaz wrote:
I didn't quite understand why the dps decreased in the last case? What was the calculation/mechanics behind a dps decrease when dual-wielding?


To explain a little further, you have to attack with your right hand and then your left hand before you can use your right hand weapon again. So even though you're getting a net 15% increase in attack speed, each individual weapon is attacking much slower than if it was wielded alone. If your offhand does significantly less damage than your mainhand, you're going to lose dps.

I think it's easier to conceptualize the numbers in terms of seconds per attack instead of attacks per second so let's look at the two weapons he used (ignoring strength and the magic damages).

axe: 12-21 damage per attack * 1.39 attacks per second (23.0dps)
club: 5-9 damage per attack * 1.2 attacks per second (8.4dps)

By themselves:
axe: 16.5 damage every .72 seconds
club: 7 damage every .833 seconds

With 15% dual wield bonus:
axe: 16.5 damage every .626 seconds
club: 7 damage every .725 seconds

Added together:
23.5 damage every 1.351 seconds = 17.39 dps (obviously lower than the 23.0 dps on the axe itself)

Disclaimer: might not be entirely accurate, but I think it's fundamentally correct.
Of course, you only live one life, and you make all your mistakes, and learn what not to do, and that’s the end of you.
Alakabon
Profile Joined November 2010
Canada111 Posts
February 20 2012 09:57 GMT
#20
Very insightful, thank you for doing this !
Na_Dann_Ma_GoGo
Profile Joined March 2010
Germany2959 Posts
February 20 2012 11:09 GMT
#21
A good thing something like this gets discussed. I remember that at least in the older videos of ForceStrategy he kept saying that 2 hand weapons are better, because they have higher damage, period.
Not questioning his first thought at all.

The first thing that should strike him is, why would Blizzard implement it that way, when there's obviously no advantage to 1hand weapons?

Well and then I saw PsyStarcraft do a walkthrough and he thought the same for a moment and then was like "wait a sec...".
He then tested out the attack speed and realised it gets increased when dual wielding. <- that's how you do it ;P

It's just annoying when someone who gets multiple hundret thousand views on his video keep explaining stuff wrong :/

Either way, as simple as the concept of dual wielding appears to be, there's actually quite a lot to take into consideration to make both, single and dual wielding appealing. As noted in the patch notes, you e.g. have to consider which attribute you're going to use for spells etc.
WrathBringerReturns said: No no no. Sarcasm is detected in the voice. When this forum is riddled with stupidity, you think I can tell every post apart? Fair enough it was intended sarcastically, was it obvious? Of course not.
Nunc
Profile Joined January 2012
10 Posts
February 20 2012 16:18 GMT
#22
Yeah, Force has a few videos where he simply equips a 2nd weapon, reduces his DPS and never stops to notice. Eventually he finds 2 comparatively good weapons but I also found it pretty annoying how his preferred class being the Barbarian who can dual-wield has never caused him to look into that properly. That means you should sub to Armada Gaming for all your juicy details in the future! :>
Logo
Profile Blog Joined April 2010
United States7542 Posts
Last Edited: 2012-02-20 16:56:27
February 20 2012 16:50 GMT
#23
If your offhand does significantly less damage than your mainhand, you're going to lose dp


Be careful because that only applies when doing the same attack again and again and may not be accurate if a big part of your DPS is coming from spending resources (dual wielding will always generate more resource than 1 weapon). Or for a class like the monk you get through your combos faster by dual wielding.

--------

One thing that would be good to look into is how the increased resource regeneration can effect your DPS.

For example what happens if you use multishot with your mainhand attack, then grenades (runed to generate additional hatred to cover the multi-shot's hatred expenditure) with your offhand. The multishot is the primary source of DPS so if the offhand attack (the grenades) are performed with a faster attacking weapon, even if it's less dps, then your overall DPS may go up.

Another example might be combining frenzy and cleave. With the former red rune 1 hit of frenzy would give +15% attack speed & +15% damage. You could use a fast offhand to gain one frenzy stack then hit with a cleave taking advantage of the +15% damage on the cleave with the +15% faster attack speed also factoring in to give you more attacks overall. Maybe the math doesn't work out to be more DPS than cleaving twice, idk, but it could for certain values.

I think overall the system is nice. It seems like..
1h - defensive or the best for stats (shield or orb-like offhand)
dual wielding - greatest resource generation and flexibility in applying effects or procing on hit effects (like immobilize for the DH)
2h - best at maximizing the damage/resource of abilities you spend resources on like multi-shot, impale, etc. and great for big burst damage overall.

Seems like each possible weapon configuration has a compelling reason to go it that's unique from the other configurations.
Logo
n0ave
Profile Joined January 2011
180 Posts
February 20 2012 17:06 GMT
#24
Good job, keep the videos coming.
Jermstuddog
Profile Blog Joined June 2010
United States2231 Posts
Last Edited: 2012-02-21 03:49:55
February 21 2012 03:10 GMT
#25
One of the big issues I have with the game right now comes from the Witch Doctor.

The Witch Doctor has no resource generating abilities, everything they do is mana-based and, at least in the beta, the lowest mana-cost spell can still run you OOM if you chain-cast it long enough.

At any rate, this seems to make 2H weapons significantly better specifically due to the fact that they are SLOWER, and therefore cause your damage-per-mana and damage-per-cast to be significantly higher than 1Hers.

Why would I ever use a 1Her as a Witch Doctor?

I also have a lesser problem with the Demon Hunter as 2H bows + quiver provides a ridiculous amount more stats than using 2 hand-crossbows. This is a lesser problem though as the threshold for surpassing the effectiveness by DW is nowhere near as large a gap compared to the Witch Doctor.
As it turns out, marines don't actually cost any money -Jinro
theaxis12
Profile Joined March 2011
United States489 Posts
February 21 2012 05:26 GMT
#26
Cool thanks
Shut your mouth and put your head back in the clouds.
Perseverance
Profile Joined February 2010
Japan2800 Posts
February 21 2012 07:02 GMT
#27
The first two questions were answered very well. Your last question answered was horrible. "No, but usually it is" with no example of anything that is better, and only one grossly exaggerated example of it not being better.

If basic, clear, and concise are your goals for these videos then you might want to work on examples for all possibilities (better, equal, and worse).

Another thing you could think about doing is figuring out the formula (I'm sure there is one that can be found out in the beta) for dual wield weapons to be better/worse/or equal versus just using a single weapon.

I mean, my guess is people aren't going to search youtube/forums for videos explaining stuff that they can figure out in 30 seconds on their own, they are going to look on youtube/forums for formulas and meaningful numbers.

Just my two cents, other people in this thread can feel free to point it out if they feel like I'm wrong.
<3 Moonbattles
bludragen88
Profile Joined August 2008
United States527 Posts
February 21 2012 16:57 GMT
#28
Out of curiosity, how do you get identical pairs of weapons? Are you using a game editor/does the beta allow you to do stuff like that? Or are there always crappy stock items you can buy infinitely at the store?
Malkavian183
Profile Joined February 2011
Turkey227 Posts
February 21 2012 18:36 GMT
#29
nice video, i really wondered how that will be thank you
Inject Bitch!
Logo
Profile Blog Joined April 2010
United States7542 Posts
February 21 2012 19:09 GMT
#30
On February 22 2012 01:57 bludragen88 wrote:
Out of curiosity, how do you get identical pairs of weapons? Are you using a game editor/does the beta allow you to do stuff like that? Or are there always crappy stock items you can buy infinitely at the store?


Stock items.
Logo
inReacH
Profile Blog Joined August 2008
Sweden1612 Posts
Last Edited: 2012-02-21 22:20:29
February 21 2012 22:20 GMT
#31
On February 21 2012 16:02 Perseverance wrote:
The first two questions were answered very well. Your last question answered was horrible. "No, but usually it is" with no example of anything that is better, and only one grossly exaggerated example of it not being better.

If basic, clear, and concise are your goals for these videos then you might want to work on examples for all possibilities (better, equal, and worse).

Another thing you could think about doing is figuring out the formula (I'm sure there is one that can be found out in the beta) for dual wield weapons to be better/worse/or equal versus just using a single weapon.

I mean, my guess is people aren't going to search youtube/forums for videos explaining stuff that they can figure out in 30 seconds on their own, they are going to look on youtube/forums for formulas and meaningful numbers.

Just my two cents, other people in this thread can feel free to point it out if they feel like I'm wrong.


Fair criticism. I guess when I decided to include this question I had one objective. To get people to check their DPS when adding a second weapon rather than assuming the second weapon would be additive.

The question of whether or not to use one or two weapons has a lot of factors to consider, many of which have been mentioned here such as resource generation being higher when you attack faster or wanting lower damage/faster attack speed to minimize wasted overkill damage. Needless to say I didn't want to get into all of that in this video as it was long enough I feel.

Anyways I appreciate the feedback Pers and everyone else
I am really happy with the response to this video and I'll try to improve the next one.
Perseverance
Profile Joined February 2010
Japan2800 Posts
February 22 2012 00:22 GMT
#32
On February 22 2012 07:20 inReacH wrote:
Show nested quote +
On February 21 2012 16:02 Perseverance wrote:
The first two questions were answered very well. Your last question answered was horrible. "No, but usually it is" with no example of anything that is better, and only one grossly exaggerated example of it not being better.

If basic, clear, and concise are your goals for these videos then you might want to work on examples for all possibilities (better, equal, and worse).

Another thing you could think about doing is figuring out the formula (I'm sure there is one that can be found out in the beta) for dual wield weapons to be better/worse/or equal versus just using a single weapon.

I mean, my guess is people aren't going to search youtube/forums for videos explaining stuff that they can figure out in 30 seconds on their own, they are going to look on youtube/forums for formulas and meaningful numbers.

Just my two cents, other people in this thread can feel free to point it out if they feel like I'm wrong.


Fair criticism. I guess when I decided to include this question I had one objective. To get people to check their DPS when adding a second weapon rather than assuming the second weapon would be additive.

The question of whether or not to use one or two weapons has a lot of factors to consider, many of which have been mentioned here such as resource generation being higher when you attack faster or wanting lower damage/faster attack speed to minimize wasted overkill damage. Needless to say I didn't want to get into all of that in this video as it was long enough I feel.

Anyways I appreciate the feedback Pers and everyone else
I am really happy with the response to this video and I'll try to improve the next one.


It's all good man, I think you have started something that can become great here. All you have to remember is more information is always better.

I'm looking forward to your next video ^^
<3 Moonbattles
VoirDire
Profile Joined February 2009
Sweden1923 Posts
February 22 2012 01:18 GMT
#33
Are you sure it's not just displayed damage that gets truncated? In-game damage could very well be un-rounded.

Check it with 2 players and make 100 1h swings and 115 dual wield swings with the same speed weapons.
R0YAL
Profile Blog Joined September 2009
United States1768 Posts
Last Edited: 2012-02-22 08:52:45
February 22 2012 08:29 GMT
#34
On February 20 2012 16:22 Meta wrote:
When I played through the beta on a barbarian I was thoroughly unimpressed with dual wielding. It always seemed to lower my dps compared to using a single 2handed weapon. That's probably just the items at the start of act 1 though, late game items might be really high dps 1handers compared to 2handers.

When you are dual wielding, if one weapon is at least 15% less damage than the other then your overall dps will be lower than if you used only the higher dps of the two. That is because of the 15% attack speed increase when dual wielding. Some factors to keep in mind when deciding to dual wield or use 2 handers on the classes that have skill resource generators (monk, barb, dh) is that dual wielding will cause you to generate your resource (spirit, fury, hatred) quicker since you will be attacking faster. You may ask why would anyone dual wield with these characters then? Well their resource spenders will be doing more damage with two handers (when compared to same dps output as dual wielding, ofc). So while dual wielding will assist in gaining resource faster, using a two hander will assist in spending your resource more efficiently. So it all balances out in the end right?! Not when you factor in support skills such as Breath of Heaven or War Cry. Dual wielding will benefit support skills over two handers because support skills are not based on weapon damage like the offensive spells are, therefore the faster you generate your resource, the more effective support skills are.

On February 21 2012 12:10 Jermstuddog wrote:
One of the big issues I have with the game right now comes from the Witch Doctor.

The Witch Doctor has no resource generating abilities, everything they do is mana-based and, at least in the beta, the lowest mana-cost spell can still run you OOM if you chain-cast it long enough.

At any rate, this seems to make 2H weapons significantly better specifically due to the fact that they are SLOWER, and therefore cause your damage-per-mana and damage-per-cast to be significantly higher than 1Hers.

Why would I ever use a 1Her as a Witch Doctor?

I also have a lesser problem with the Demon Hunter as 2H bows + quiver provides a ridiculous amount more stats than using 2 hand-crossbows. This is a lesser problem though as the threshold for surpassing the effectiveness by DW is nowhere near as large a gap compared to the Witch Doctor.

Well if you have a problem with the WD then you should also have a problem with the Wizard because they are in the same boat. Both of these classes do however have their own class specific off hand items that tend to focus on stats that benefit their respected class a lot (ie +damage) as well as have class specific stats (ie +mana). It is not very evident in the beta due to the underwhelming amount of content, but in the full version with the right combination of stats the WD and Wizard will most likely benefit from 1h + offhand more. You need to consider that their offhands can get +damage and +intellect which both increase damage output, therefore you can have a 1h + offhand each have 2 affixes that add damage combined with a naturally faster attack speed vs a two hander which can only have a max of 2 affixes that add damage. Looking at the end game an ideal 1h + offhand in terms of stats should beat out a two hander with ideal stats, but it will be much more arduous and expensive to aquire an ideal 1h + offhand.

I'm not sure what you are getting at with the DH, do you mean two 1 hand-crossbows, because it should come out to the same number of stats (maybe not early game but my point still stands).

But as far as the beta goes I think you are spot on
Absence of evidence is not evidence of absence.
Mataza
Profile Blog Joined August 2010
Germany5364 Posts
Last Edited: 2012-02-22 14:32:10
February 22 2012 12:58 GMT
#35
Ugh.
Please do actual math and calculate shit. It´s not hard and you very well do not need someone to tell you something when you can just put it on a piece of paper and have it in 10 minutes.

Dual wielding increases DPS as long as the worse weapon has around 3/4(73.91%) of the DPS of the better weapon.
1. Dual Wield increases Attack speed by 15%, that´s the factor 1.15.
2. The actual DPS is the mean of the DPS of both weapons, thats a+b divided by 2.

We want the point at which the dual wield dps equals the dps of the better(mainhand) weapon:
(Mainhand dps + Offhand dps)*1.15/2 = Mainhand dps
+ Show Spoiler [Let me calculate that for you] +
Mdps *1.15/2 + Odps *1.15/2 = Mdps

Mdps + Odps = 2* Mdps/1.15

We can simplify that as 2 / 1.15 equals 1.7391304 which gives us:

Mdps + Odps = 1.739 Mdps

Odps = 1.739 Mdps -1 Mdps

Odps = 0.739 Mdps


So the threshold is both weapons in 75% of each other, not 85%

For the given example with the 23.0 DPS axe, your DPS should rise as long as your second weapon has more than roughly 17 DPS.
I don´t have a betakey, but I would appreciate if someone confirms this.

[edit]
People commented on unneeded digits in the fraction. If you want accurate results you need those, dammit.
I recalculated with the daggers going to 1,72 from 1,5 changing the DW ias to 14,6%, changing the threshold to 74,4%
This is as good as it gets with only rounded numbers.
[/edit]
If nobody hates you, you´re doing something wrong. However someone hating you doesn´t make you right
Jermstuddog
Profile Blog Joined June 2010
United States2231 Posts
February 24 2012 14:06 GMT
#36
The major difference between the WD and the Wizard is that the Wizard has free spells that work as solid main attacks.

The WD only has spells that cost mana. Even Poison dart, the lowest mana cost nuke the WD has, is still a net loss. With a 1.4 speed 2h axe.

This makes attack speed still benefitial to the Wizard in some way as he can dump his mana pool faster and just proceed onto his free spells for recovery mode.

The WD does not have a recovery mode, and therefore you want your mana bar to last as long as possible. Making slower better in all situations.
As it turns out, marines don't actually cost any money -Jinro
thragar
Profile Blog Joined May 2010
Canada450 Posts
February 24 2012 17:13 GMT
#37
Can you please provide a text post? I prefer to read this kind of information.
inReacH
Profile Blog Joined August 2008
Sweden1612 Posts
Last Edited: 2012-02-24 22:01:14
February 24 2012 20:17 GMT
#38
On February 22 2012 21:58 Mataza wrote:
Ugh.
Please do actual math and calculate shit. It´s not hard and you very well do not need someone to tell you something when you can just put it on a piece of paper and have it in 10 minutes.

Dual wielding increases DPS as long as the worse weapon has around 3/4(73.91%) of the DPS of the better weapon.
1. Dual Wield increases Attack speed by 15%, that´s the factor 1.15.
2. The actual DPS is the mean of the DPS of both weapons, thats a+b divided by 2.

We want the point at which the dual wield dps equals the dps of the better(mainhand) weapon:
(Mainhand dps + Offhand dps)*1.15/2 = Mainhand dps
+ Show Spoiler [Let me calculate that for you] +
Mdps *1.15/2 + Odps *1.15/2 = Mdps

Mdps + Odps = 2* Mdps/1.15

We can simplify that as 2 / 1.15 equals 1.7391304 which gives us:

Mdps + Odps = 1.739 Mdps

Odps = 1.739 Mdps -1 Mdps

Odps = 0.739 Mdps


So the threshold is both weapons in 75% of each other, not 85%

For the given example with the 23.0 DPS axe, your DPS should rise as long as your second weapon has more than roughly 17 DPS.
I don´t have a betakey, but I would appreciate if someone confirms this.

[edit]
People commented on unneeded digits in the fraction. If you want accurate results you need those, dammit.
I recalculated with the daggers going to 1,72 from 1,5 changing the DW ias to 14,6%, changing the threshold to 74,4%
This is as good as it gets with only rounded numbers.
[/edit]



Bad attitude and bad math.

You didn't account for attack speed difference between weapons.
You also didn't account for the possibility of the second weapon having a increase to attack speed that transfers over to the first weapon.+ Show Spoiler +
If you're curious, attack speed increases on weapons that are given in decimal form (this is rare) DO transfer over do other weapons. This is one way you can actually substitute a lower DPS weapon but still have your DPS go up. Attack speed increases given in percentage form (common) DO NOT transfer over to other weapons.
+ Show Spoiler +
Even if that weren't the case, you are still proven wrong by the fact that you didn't account for the difference in attack speed. Consider the following

You have a 100-100Damage 1APS Axe in your right hand and a 5-5Damage 1APS(5DPS) Sword in your left hand. If you then remove the Sword for a 1-1Damage 4APS(4DPS Sword, your DPS will go up even though the left hand's DPS is 4 instead of 5 because the axe will be attacking more often.

You also didn't account for the second weapon increasing the damage of the first weapon by adding to primary attribute, critical hit chance or critical hit damage.

If you can see now how many factors that have to be accounted for to see if a weapon is good or bad before you put it on then you'll see that it's silly to bother when you can just try it on.

Which is exactly what that part of the video was trying to get people to do.

Please lose the know it all attitude if you're going to continue posting in the thead and if you do I will thank you for your contribution.
HypertonicHydroponic
Profile Blog Joined October 2010
437 Posts
February 25 2012 07:34 GMT
#39
On February 25 2012 05:17 inReacH wrote:
Show nested quote +
On February 22 2012 21:58 Mataza wrote:
Ugh.
Please do actual math and calculate shit. It´s not hard and you very well do not need someone to tell you something when you can just put it on a piece of paper and have it in 10 minutes.

Dual wielding increases DPS as long as the worse weapon has around 3/4(73.91%) of the DPS of the better weapon.
1. Dual Wield increases Attack speed by 15%, that´s the factor 1.15.
2. The actual DPS is the mean of the DPS of both weapons, thats a+b divided by 2.

We want the point at which the dual wield dps equals the dps of the better(mainhand) weapon:
(Mainhand dps + Offhand dps)*1.15/2 = Mainhand dps
+ Show Spoiler [Let me calculate that for you] +
Mdps *1.15/2 + Odps *1.15/2 = Mdps

Mdps + Odps = 2* Mdps/1.15

We can simplify that as 2 / 1.15 equals 1.7391304 which gives us:

Mdps + Odps = 1.739 Mdps

Odps = 1.739 Mdps -1 Mdps

Odps = 0.739 Mdps


So the threshold is both weapons in 75% of each other, not 85%

For the given example with the 23.0 DPS axe, your DPS should rise as long as your second weapon has more than roughly 17 DPS.
I don´t have a betakey, but I would appreciate if someone confirms this.

[edit]
People commented on unneeded digits in the fraction. If you want accurate results you need those, dammit.
I recalculated with the daggers going to 1,72 from 1,5 changing the DW ias to 14,6%, changing the threshold to 74,4%
This is as good as it gets with only rounded numbers.
[/edit]



Bad attitude and bad math.

You didn't account for attack speed difference between weapons.
You also didn't account for the possibility of the second weapon having a increase to attack speed that transfers over to the first weapon.+ Show Spoiler +
If you're curious, attack speed increases on weapons that are given in decimal form (this is rare) DO transfer over do other weapons. This is one way you can actually substitute a lower DPS weapon but still have your DPS go up. Attack speed increases given in percentage form (common) DO NOT transfer over to other weapons.
+ Show Spoiler +
Even if that weren't the case, you are still proven wrong by the fact that you didn't account for the difference in attack speed. Consider the following

You have a 100-100Damage 1APS Axe in your right hand and a 5-5Damage 1APS(5DPS) Sword in your left hand. If you then remove the Sword for a 1-1Damage 4APS(4DPS Sword, your DPS will go up even though the left hand's DPS is 4 instead of 5 because the axe will be attacking more often.

You also didn't account for the second weapon increasing the damage of the first weapon by adding to primary attribute, critical hit chance or critical hit damage.

If you can see now how many factors that have to be accounted for to see if a weapon is good or bad before you put it on then you'll see that it's silly to bother when you can just try it on.

Which is exactly what that part of the video was trying to get people to do.

Please lose the know it all attitude if you're going to continue posting in the thead and if you do I will thank you for your contribution.

Well, I have to say that this guys post (and bad math :D) got me interested in the real answer to this question, that is, how do you tell just by looking at a weapon if it is bette or worse for your DPS. Obviously, you need to calculate out all your bonuses and whatnot (and the new calculator you posted I'm sure is great for that), but you should be able to tell looking at the two fundemental properties (damage and attacks per second) and see whether that item should be in the range of worse, same, or better.

I made a little spreadsheet for myself (I'd post it once I get it finished and polished) to help visualize the refutation you gave and that I already thought was correct, that differing weapon speeds will change the way the DPS might change. Again, I haven't gotten it finished, but there are some interesting implications:

1) Mataza is correct about look for things with ~75% DPS averaging out to about the same DPS as the primary weapon, except that is ONLY the case when the weapon speeds are the same.

2) There is a curve where, given X% of the original dps, you need the weapon speed to be increased by Y% to equal the original DPS (I'm working on this).

3) The faster the offhand weapon, the closer to the original DPS plus the bonus 15% (from speed) you get, even as the offhand dps goes to zero.

4) The greater DPS weapon should always be in the main hand (barring any other special game considerations).

5) Two weapons of the same DPS, even though they have different DMG & APS, will always yield an average of the original DPS plus the 15% bonus from speed.

Anyway, I just thought I'd throw this out there -- tickled my math bone today.
[P] The Watery Archives -- http://www.teamliquid.net/forum/viewmessage.php?topic_id=279070
Mataza
Profile Blog Joined August 2010
Germany5364 Posts
Last Edited: 2012-02-26 14:34:15
February 25 2012 19:08 GMT
#40
On February 25 2012 05:17 inReacH wrote:
Show nested quote +
On February 22 2012 21:58 Mataza wrote:
Ugh.
Please do actual math and calculate shit. It´s not hard and you very well do not need someone to tell you something when you can just put it on a piece of paper and have it in 10 minutes.

Dual wielding increases DPS as long as the worse weapon has around 3/4(73.91%) of the DPS of the better weapon.
1. Dual Wield increases Attack speed by 15%, that´s the factor 1.15.
2. The actual DPS is the mean of the DPS of both weapons, thats a+b divided by 2.

We want the point at which the dual wield dps equals the dps of the better(mainhand) weapon:
(Mainhand dps + Offhand dps)*1.15/2 = Mainhand dps
+ Show Spoiler [Let me calculate that for you] +
Mdps *1.15/2 + Odps *1.15/2 = Mdps

Mdps + Odps = 2* Mdps/1.15

We can simplify that as 2 / 1.15 equals 1.7391304 which gives us:

Mdps + Odps = 1.739 Mdps

Odps = 1.739 Mdps -1 Mdps

Odps = 0.739 Mdps


So the threshold is both weapons in 75% of each other, not 85%

For the given example with the 23.0 DPS axe, your DPS should rise as long as your second weapon has more than roughly 17 DPS.
I don´t have a betakey, but I would appreciate if someone confirms this.

[edit]
People commented on unneeded digits in the fraction. If you want accurate results you need those, dammit.
I recalculated with the daggers going to 1,72 from 1,5 changing the DW ias to 14,6%, changing the threshold to 74,4%
This is as good as it gets with only rounded numbers.
[/edit]



Bad attitude and bad math.

You didn't account for attack speed difference between weapons.
You also didn't account for the possibility of the second weapon having a increase to attack speed that transfers over to the first weapon.+ Show Spoiler +
If you're curious, attack speed increases on weapons that are given in decimal form (this is rare) DO transfer over do other weapons. This is one way you can actually substitute a lower DPS weapon but still have your DPS go up. Attack speed increases given in percentage form (common) DO NOT transfer over to other weapons.
+ Show Spoiler +
Even if that weren't the case, you are still proven wrong by the fact that you didn't account for the difference in attack speed. Consider the following

You have a 100-100Damage 1APS Axe in your right hand and a 5-5Damage 1APS(5DPS) Sword in your left hand. If you then remove the Sword for a 1-1Damage 4APS(4DPS Sword, your DPS will go up even though the left hand's DPS is 4 instead of 5 because the axe will be attacking more often.

You also didn't account for the second weapon increasing the damage of the first weapon by adding to primary attribute, critical hit chance or critical hit damage.

If you can see now how many factors that have to be accounted for to see if a weapon is good or bad before you put it on then you'll see that it's silly to bother when you can just try it on.

Which is exactly what that part of the video was trying to get people to do.

Please lose the know it all attitude if you're going to continue posting in the thead and if you do I will thank you for your contribution.

I´m sorry to come off as rude and that rudeness was definitely not aimed at you.
I got annoyed at other posters demanding a formula, like this one:
+ Show Spoiler +
On February 21 2012 16:02 Perseverance wrote:
Another thing you could think about doing is figuring out the formula (I'm sure there is one that can be found out in the beta) for dual wield weapons to be better/worse/or equal versus just using a single weapon.

I mean, my guess is people aren't going to search youtube/forums for videos explaining stuff that they can figure out in 30 seconds on their own, they are going to look on youtube/forums for formulas and meaningful numbers.

or giving out simply bad information, like this one:
+ Show Spoiler +
On February 22 2012 17:29 R0YAL wrote:
Show nested quote +
On February 20 2012 16:22 Meta wrote:
When I played through the beta on a barbarian I was thoroughly unimpressed with dual wielding. It always seemed to lower my dps compared to using a single 2handed weapon. That's probably just the items at the start of act 1 though, late game items might be really high dps 1handers compared to 2handers.

When you are dual wielding, if one weapon is at least 15% less damage than the other then your overall dps will be lower than if you used only the higher dps of the two. That is because...



My math isn´t bad btw, it´s just simplified. I just had the feeling if I go into a really indepth formula with lots of variables for attacks per second etc. that some people would demand something easier to make decisions on the fly.

If one takes up an offhand weapon for the stats it provides, like ias, crit% or flat out strength/agility/vitality, then you should consider it a stat stick rather than a weapon.
Comparing a vanilla weapon with high dps to another weapon with low dps but lots of stats on it is not an easy comparison. If someone figures that out for single weapons it will be rather easy to apply to dual wielding.
The point of my bad math though was to work with the facts presented in the video.

As for differing attack speeds, yes my blunder. If one had a weapon with 0 dps but that attacks incredibly fast, like 1 hit every 0.1 seconds, that would almost be 15% ias without any drawback. I´m saying almost. Note that things like this probably won´t be in the game.

I might look further into that. What are the most extreme(slowest & fastest) attack speeds weapon have in the beta so far? 1 APS or less, 2 APS or more?

On another topic, I somehow get the feeling 2h weapons will be largely inferior again, just like in D2.


[edit]
I checked the battlenet item guide for these infos.
The weapons range from 1.2 attacks per seconds to 1.69 attacks per second(only 1 legendary).
With these in mind I got 68% if you have 1.2 on the main weapon and 1.7 on the offhand. Or if you change them around 77%
If there was a weapon with 2 attacks per second these numbers would be 65% and 80%. Nothing outrageous really.
If nobody hates you, you´re doing something wrong. However someone hating you doesn´t make you right
HypertonicHydroponic
Profile Blog Joined October 2010
437 Posts
February 28 2012 06:18 GMT
#41
Ok kiddies, I've got the magic formula. Now, I haven't run it with Mataza's range of speeds to get the "relevant" details but I'm sure you won't mind the practice, and tbh, I'm done being interested in the math now.

So basically, let's just lay the ground work.

DPS = Dmg / Time

Total DPS = Total Dmg / Total Time

Total Dmg = Dmg1 + Dmg2

Total Time = Time1 + Time2


Okay, so the numbers we are given to work with on a weapon are:

Dmg = X
Atk/Sec = Y
DPS = Z


How do we get time:

Time = 1 Atk / Y Atk/Sec = 1/Y Sec

So plugging into our formula above:

Total DPS = Total Dmg / Total Time = (X + X') / (T + T') = (X + X') / [(1 / Y) + (1 / Y')]

Are we good so far? Now, this is the equation we would be using if we were just going to do the work and plug stuff in, but this really doesn't do anything for us on the fly. What is it we really want to know? What we want to know is at what point does adding second weapon EQUAL the DPS of just using the primary weapon. When we know this, we know something better will make the total DPS better, and something worse will make the total DPS worse. So to set up the equation for what we want to know:

Z = (X + X') / [{1 / (Y * 1.15)} + {1 / (Y * 1.15)}]

Doesn't look much better, but let's see what happens if we switch things around a bit:

*****
C = 1.15 --- note: obviously this could have generic use, as such I'll put the generic formula and then plug back in at the end
*****

Z = (X + X') / [{1 / (Y * C)} + {1 / (Y * C)}]

Z = X + X' / [({1 / Y} * {1 / C}) + ({1 / Y'} * {1 / C})]
Z = X + X' / [({1 / Y} + {1 / Y'}) * {1 / C}]
Z = C(X + X') / [(1 / Y) + (1 / Y')]
Z = C(X + X') / [(Y + Y') / YY']
Z = C(X + X')YY' / (Y + Y')

Subsitution: Z = X / (1 / Y) = X * Y

XY = C(X + X')YY' / (Y + Y')
XY(Y + Y') = C(X + X')YY'
XY^2 + XYY' = CXYY' + CX'YY'
XY + XY' = CXY' + CX'Y'
XY = (C - 1)XY' + CX'Y'

Z = (C - 1)XY' + CZ' <-- generic formula

Z = (1.15 - 1)XY' + 1.15Z'
Z = .15XY' + 1.15Z' <-- specific formula


I hope I don't need to explain any of that, all just standard algebra: substitution, distributive property, etc. Now this is nice when we want to solve for the original DPS, but that's not what we are after (we started with the original DPS after all), what we are after is an equation that tells us something about the offhand weapon. There are two equations that we can draw from for this based on the above:

Z = (C - 1)XY' + CZ'

Z - CZ' = (C - 1)XY'
(Z - CZ') / [(C - 1)X] = Y'

Y' = (Z - CZ') / [(C - 1)X] <-- generic formula

Y' = (Z - 1.15Z') / [(1.15 - 1)X]
Y' = (Z - 1.15Z') / .15X <-- specific formula


to find the break-even offhand speed for a given offhand DPS, and:

Z = (C - 1)XY' + CZ'

Z - (C - 1)XY' = CZ'
[Z - (C - 1)XY'] / C = Z'

Z' = [Z - (C - 1)XY'] / C <-- generic formula

Z' = [Z - (1.15 - 1)XY'] / 1.15
Z' = (Z - .15XY') / 1.15 <-- specific formula


to find the break-even offhand DPS for a given offhand speed.

Personally I would use the former because I think it is easier to think in those terms, but the latter is probably has a more broad application since it seems the range of speed is more limited than the range of DPS -- but there they both are for you.

Now please note that these formulas work for:

Z > Z' > 0

When Z = Z', you have reached a limit and you will get some strange results like 1 = -1. Basically, you cannot have two weapons with the same DPS such that whatever the ratio of Dmg to Atk/Sec makes the total DPS go down to that of the original DPS. In fact, when Z = Z', your total DPS is simply going to be 1.15Z (or 1.15Z', or {1.15Z + 1.15Z'} / 2 ... doesn't really matter, it's all the same in this special case).

I should probably mention here in case it isn't obvious that once you find your break-even offhand speed or DPS, higher numbers than that are better, and lower are worse (just to cover the bases of someone not understanding why we are finding the point of breaking even).

Anywho, I hope that helps simply some things for people. I think there are a few cool things to take from this:

1) Two "quick and easy" formulas to choose from for helping to determine at a glance whether a weapon is better for dual-wielding (obviously, just at the core, not with extra bonuses and considerations).

2) That the damage of the bigger DPS weapon is what matters, in conjunction with the speed of the lower DPS weapon (I think we knew this but it is interesting to note).

3) That the generic formula(s) can be used for further breakpoint finding (like when a weapon gives a decimal IAS bonus).

4) It's just that cool! Hope this helps someone!
[P] The Watery Archives -- http://www.teamliquid.net/forum/viewmessage.php?topic_id=279070
inReacH
Profile Blog Joined August 2008
Sweden1612 Posts
Last Edited: 2012-02-28 12:45:23
February 28 2012 09:18 GMT
#42
On February 28 2012 15:18 HypertonicHydroponic wrote:

Total DPS = Total Dmg / Total Time = (X + X') / (T + T') = (X + X') / [(1 / Y) + (1 / Y')]



This is oversimplifed and unneededly complicated at the same time.
You don't need to create the custom variable TIME when you can get the same answer you have there by simply doing this

DPS = DMG * APS

This will get you the same result as

DPS = DMG / (1/APS)

Proof by example:
1.4 APS
4.5 DAMAGE
6.3 DPS

Your method:

6.3 = 4.5 / (1/1.4)

My method:

6.3 = 4.5 * 1.4

So this is how it's overly complicated.

It is overly simplified because we don't care about a weapon's DPS off character, only on character, and for on character you neglected to factor in

Crit chance
Crit bonus
Primary attribute
Bonus to damage
Decimal attack speed modifier from other weapon
Percentage attack speed modifier from other items

The correct formula for a single weapon on character is

DPS = (DAMAGE + DAMAGE BONUS) * (WEAPON ATTACK SPEED) * (1+ATTACK SPEED BONUS[N])* (1+ PRIMARY ATTRIBUTE*.01) * (1 + CRIT CHANCE * CRIT BONUS)

The [N] is there to show that there could be many of these, one for each percentage bonus to attack speed.

This is the formula I use for my calculator here: http://armadagaming.com/dpscalculator.php
So far it hasn't shown any incorrect figures when compared to ingame.
I'll be adding support for two weapons in the next few days hopefully



Have you viewed this yet?


It shows conclusively that any threshhold formula is pretty useless because of the other factors that you have to consider
+ Show Spoiler +
I show that my DPS can go up even though I equip a lower DPS weapon



Anyways I don't think your math is wrong per se and respect for taking the time to do this.
I'm actually tackling a D3 math problem now that I'm having trouble with.

PM me if you're interested in giving it a look
HypertonicHydroponic
Profile Blog Joined October 2010
437 Posts
February 29 2012 06:37 GMT
#43
I would like to reply to your last post, but first I want to give a quick example of how my formula(s) works to give people an idea of how it could be relevant since I neglected to do so in my last post:

X = 20 dmg
Y = 1.5 A/S --> T = 1 / 1.5 = .667
Z = 30 DPS

X' = 10 dmg
Y' = 1.2 A/S --> T' = 1/1.2 = .833
Z' = 12 DPS


Let's do the actual calculation first, then I'll show how the shortcut works:

DPS = (X + X') / [(1 / {Y * 1.15}) + (1 / {Y' * 1.15})]
DPS = (10 + 12) / [(1 / 1.5 * 1.15) + (1 / 1.2 * 1.15)]
DPS = 30 / [(1 / 1.725) + (1 / 1.38)]
DPS = 30 / (0.57971014492753623188405797101449 + 0.72463768115942028985507246376812)
DPS = 30 / 1.3043478260869565217391304347826
DPS = 23


Clearly we've wound up with a worse DPS, but couldn't we have just averaged the DPS's?

(30 + 12) / 2 = 42 / 2 = 21 <-- NO

So how would it work using my shortcut formula? "Find speed" version:

Y' = (Z - 1.15Z') / .15X


We plug in the primary DPS, offhand DPS, and primary damage...

Y' = (30 - 1.15(12)) / .15(20)
Y' = (30 - 13.8) / 3
Y' = 16.2 / 3
Y' = 5.4


So an offhand weapon with 12DPS need to have an attack speed of 5.4 A/S to make the original DPS (30) happen overall. Clearly, its 1.2 is much less than this, so we know this weapon will decrease the DPS.

Now using the "Find DPS" version:

Z' = [Z - .15XY'] / 1.15

We plug in the primary DPS, offhand speed, and primary damage...

Z' = [30 - .15(20)(1.2)] / 1.15
Z' = [30 - 3.6] / 1.15
Z' = 26.4 / 1.15
Z' = 22.956521739130434782608695652174


So with an offhand attack speed of 1.2, the offhand DPS needs to be roughly 23 to make the overall DPS equal the original DPS (30).

Can we check this? Sure:

#1
X’ = Z’ / Y’
X’ = 12 / 5.4
X’ = 2.2222222222222222222222222222222
DPS = (X + X') / [(1 / {Y * 1.15}) + (1 / {Y' * 1.15})]
DPS = (20 + 2.2222222222222222222222222222222) / [(1 / {1.5 * 1.15}) + (1 / {5.4 * 1.15})]
DPS = 22.2222222222222222222222222222222 / [(1 / 1.725) + (1 / 6.21)]
DPS = 22.2222222222222222222222222222222 / [0.57971014492753623188405797101449 + 0.16103059581320450885668276972625]
DPS = 22.2222222222222222222222222222222 / 0.74074074074074074074074074074074
DPS = 30 = Z

#2
X’ = Z’ / Y’
X’ = 22.956521739130434782608695652174 / 1.2
X’ = 19.130434782608695652173913043478
DPS = (X + X') / [(1 / {Y * 1.15}) + (1 / {Y' * 1.15})]
DPS = (20 + 19.130434782608695652173913043478) / [(1 / {1.5 * 1.15}) + (1 / {1.2 * 1.15})]
DPS = 39.130434782608695652173913043478 / [0.57971014492753623188405797101449 + 0.72463768115942028985507246376812]
DPS = 39.130434782608695652173913043478 / 1.3043478260869565217391304347826
DPS = 30 = Z



Okay, onto your post...



Show nested quote +
On February 28 2012 15:18 HypertonicHydroponic wrote:

Total DPS = Total Dmg / Total Time = (X + X') / (T + T') = (X + X') / [(1 / Y) + (1 / Y')]


This is oversimplifed and unneededly complicated at the same time.
You don't need to create the custom variable TIME when you can get the same answer you have there by simply doing this

DPS = DMG * APS

This will get you the same result as

DPS = DMG / (1/APS)

Proof by example:
1.4 APS
4.5 DAMAGE
6.3 DPS

Your method:

6.3 = 4.5 / (1/1.4)

My method:

6.3 = 4.5 * 1.4

So this is how it's overly complicated.


How do we get time:

Time = 1 Atk / Y Atk/Sec = 1/Y Sec

...

Subsitution: Z = X / (1 / Y) = X * Y

As you can see, I did address this, though just briefly. The problem with the simplified form is that when you have to take into account two different weapons of possibly differing damages and speeds, you cannot use the simplified form because:

Y + Y' != 1 / (T + T')

DPS is just that, Damage Per Second -- it has nothing to do with the number of attacks. But we do need to account for the number of attacks being averaged together. In a lot of ways it is like the following:

A car travels at 40mph for 15 minutes, 30mph for 20 minutes, and 60mph for 25 minutes. The total distance travelled is:

[40 mph * (15 minutes * {1 hour / 60 minutes})] + [30 mph * (20 minutes * {1 hour / 60 minutes})] + [60 mph * (25 minutes * {1 hour / 60 minutes})]
= [40mph * (1/4)h] + [30mph * (1/3)h] + [60mph * (5/12)h]
= 10m + 10m + 25m
= 45 miles


But that's just the distance ("damage"), what is the average speed ("DPS")? You cannot just:

40mph + 30mph + 60mph / 3 = 130mph / 3 = 43.333mph

because they are weighted by the amount of time for each speed. So to find the average speed we need the Total Distance divided by the Total Time.

45 miles / [(1/4)h + (1/3)h + (5/12)h]
= 45 miles / [(15/60)h + (20/60)h + (25/60)h]
= 45 miles / (60/60)h
= 45 miles / 1h
= 45mph


So average speed is clearly different than the average of the speeds. This is quite similar to the conversions needed to find the "Total DPS" which is really an average. But in the case of DPS, the miles are damage, the minutes are attacks, and the hours are seconds.

We make 15 attacks with our 40dps weapon, 20 attacks with 30dps weapon, and 25 attacks with the 60dps weapon: what was the average (or "Total") DPS? The math is the same (but our conversion rate is different since there will always be 60 minutes in 1 hour, but there will be a varying amount of attacks in 1 second).

But what we are dealing with is just a simple two attacks, which brings us back to:

(X + X') / [(1 / Y) + (1 / Y')]
OR
(Dmg1 + Dmg2) / [(1 atk / (Y atk/sec)) + (1 atk / (Y' atk/sec))]


There is no needless complication -- it is what it is. Just because the inverse of the inverse equals the starting value, does not mean the inverse of the addition of inverses equals the addition of the starting values:

Y = 1 / (1 / Y)
5 = 1 / (1 / 5)
5 = 1 / .2
5 = 5

Y + Y' != 1 / [(1 / Y) + (1 / Y')]
2 + 5 != 1 / [(1 / 2) + (1 / 5)]
7 != 1 / [(5 / 10) + (2 / 10)]
7 != 1 / (7 / 10)
7 != 10 / 7


it does not even equal the average:

7/2 != 10/7

The only time you can do:

(Dmg1 * Speed1 + Dmg2 * Speed2) / 2 = Avg DPS is when Speed1 = Speed2 because:

Y = Y'

(X + X') / [(1 / Y) + (1 / Y')]
= (X + X') / [(1 / Y) + (1 / Y)]
= (X + X') / (2 / Y)
= (X + X')Y / 2
= (XY + X'Y) / 2


another way to simplify which I think demonstrates the point more closely is:

(X + X') / [(1 / Y) + (1 / Y')]
(X + X') / [(1 / Y) + (1 / Y)]
= (X + X') / [(Y / Y^2) + (Y / Y^2)]
= (X + X') / [(Y + Y) / Y^2]
= (X + X')Y^2 / 2Y
= (XY^2 + X'Y^2) / 2Y <-- this is...
= (XY + X'Y) / 2


When Y != Y'

(X + X') / [(1 / Y) + (1 / Y')]
= (X + X') / [(Y' / YY') + (Y / YY')]
= (X + X') / [(Y + Y') / YY']
= (X + X')YY' / (Y + Y')
= (XYY' + X'YY') / (Y + Y') <-- ... this


It should be obvious that the following is not always the case, and in fact only the case when Y = Y':

Y / 2 = YY' / (Y + Y')

And this is why you must do

(X + X') / [(1 / Y) + (1 / Y')]

and not

(XY + X'Y') / 2

when you are dealing with the Average or Total DPS. So no, it is not over complicated, it is correct. Your "My method" is a nice-to-have that is only doable when you are dealing with a single rate of attack.

It is overly simplified because we don't care about a weapon's DPS off character, only on character, and for on character you neglected to factor in

Crit chance
Crit bonus
Primary attribute
Bonus to damage
Decimal attack speed modifier from other weapon
Percentage attack speed modifier from other items

The correct formula for a single weapon on character is

DPS = (DAMAGE + DAMAGE BONUS) * (WEAPON ATTACK SPEED) * (1+ATTACK SPEED BONUS[N])* (1+ PRIMARY ATTRIBUTE*.01) * (1 + CRIT CHANCE * CRIT BONUS)

The [N] is there to show that there could be many of these, one for each percentage bonus to attack speed.

This is the formula I use for my calculator here: http://armadagaming.com/dpscalculator.php
So far it hasn't shown any incorrect figures when compared to ingame.
I'll be adding support for two weapons in the next few days hopefully



Have you viewed this yet?


It shows conclusively that any threshhold formula is pretty useless because of the other factors that you have to consider
+ Show Spoiler +





Anyways I don't think your math is wrong per se and respect for taking the time to do this.
I'm actually tackling a D3 math problem now that I'm having trouble with.

PM me if you're interested in giving it a look

Even this is not quite an accurate assessment. While my formula does not take into account a lot of the extras in your formula, there is a way you can use my formula which does not even need to take those things into consideration. That is not to say we won't take those things into consideration, but that is to say we don't need to take them into consideration *in the context of* the average DPS.

My formula(s) works with four things:

Primary DPS
Primary Damage
Offhand DPS
Offhand Speed

If you have the first two, and can get one of the last two, you can use my formula.

Even if primary DPS is:

DPS = (DAMAGE + DAMAGE BONUS) * (WEAPON ATTACK SPEED) * (1+ATTACK SPEED BONUS[N])* (1+ PRIMARY ATTRIBUTE*.01) * (1 + CRIT CHANCE * CRIT BONUS)

You can just get this from your formula or from equiping the weapon in a 1hand configuration and reading the number of the menu. Repeat for the offhand weapon. Now we have the two DPS values.

So all we need now is the Primary Damage. Obviously, this might need a little bit of calculation, but it can be boiled down to the following elements:

(DAMAGE + DAMAGE BONUS) * (1+ PRIMARY ATTRIBUTE*.01) * (1 + CRIT CHANCE * CRIT BONUS)

because we don't need the primary speed for our formula (it is already taken into account in the form of the Primary DPS). This shouldn't be too hard to calculate on the fly.

Then you plug in and compare the result to the current Offhand speed. If the offhand speed equals or is better than the result, you can dual wield without hurting DPS.

The only exception to this I can see is when the offhand weapon adds decimal IAS, in which case you would need to recalculate the Primary DPS manually (e.g., via your calculator [or ***]) since it gets added directly to the WEAPON ATTACK SPEED before percentage modifiers as shown in your video.

Of course, if the game is going to be showing what your overall DPS is anyway, you might as well just stick a weapon in your offhand and see what it says -- all of this really is then just understanding what is going on and maybe helping to know some special breakpoints. My intent was never to generate DPS given a set of theoretical conditions, but to find how offhand speed factors into the resulting DPS, which I did. If you want a more complicated version of my formula given your DPS formula (which does not look like it takes decimal speed increase into account, btw) that would take into account all of the different things that can affect the Primary or Offhand DPS, I can do that.



*** Ok, one thing just occurred to me about the decimal IAS that might not make it all that difficult to recalculate manually. Since my formula works based on individual weapon DPS, you can recalculate the "but what if there is a bonus decimal ias" for the single weapon in the following manner:

First I'm going to reformat your equation to read a bit shorter, and also add in the decimal IAS:

DPS = (Dmg + BDmg) * (1 + {Atrb * .01}) * ({1 + CritC} * CritB) * (ApS + DApS) * (1 + PApS) [* (1 + PApS') ...]

Now, just for the sake of making things really readable, lets make:

Q = (Dmg + BDmg) * (1 + {Atrb * .01}) * ({1 + CritC} * CritB) * (1 + PApS) [* (1 + PApS') ...]

Or everything but the " * (ApS + DApS) "

Let's assume that we know what our (ApS + DApS) is. Well, if we know that, and we know our DPS, then we can get Q:

DPS = (Dmg + BDmg) * (1 + {Atrb * .01}) * ({1 + CritC} * CritB) * (ApS + DApS) * (1 + PApS) [* (1 + PApS') ...]
DPS = (ApS + DApS) * Q
DPS / (ApS + DApS) = Q


Why is this relevant? Because

DPS = (ApS + DApS) * Q
DPS = (ApS * Q) + (DApS * Q)


So what? Well, now we want to find DPS' which has its own (ApS' + DApS'), and:

ApS' = (ApS + DApS)

So if we apply Q, we get:

DPS' = (ApS' + DApS') * Q
DPS' = (ApS' * Q) + (DApS' * Q)
DPS' = [(ApS + DApS) * Q] + (DApS' * Q)
DPS' = DPS + (DApS' * Q)


So then all we need to do is multiply the DApS' by the Q we've determined, and slap it on our original DPS.

We then plug in the DPS' and solve for our threshold Speed and we are off and running again. Anyway, I haven't tested it yet, so I could be overlooking something with this part, but still, if your calculator is accurate for single weapon DPS, you can still use the threshold formulas without this trickery.
[P] The Watery Archives -- http://www.teamliquid.net/forum/viewmessage.php?topic_id=279070
inReacH
Profile Blog Joined August 2008
Sweden1612 Posts
Last Edited: 2012-02-29 07:12:51
February 29 2012 07:08 GMT
#44
On February 29 2012 15:37 HypertonicHydroponic wrote:
Y + Y' != 1 / (T + T')

You are right, sorry for my shortsightedness

I'm 99% sure the problem I'm now having is the same problem you have but you don't know it yethehe
Just replied to your PM, looking forward to showing you what I've got =]
gunr
Profile Joined April 2011
Australia1 Post
March 01 2012 13:26 GMT
#45
Is the "dps" value of any use to casters? By that I mean the dps number on the stats page of your character. Since all the skills are based off "weapon damage". Or is this the same thing and just worded differently.
KiLL_ORdeR
Profile Blog Joined June 2009
United States1518 Posts
Last Edited: 2012-04-27 20:30:57
April 27 2012 20:17 GMT
#46
I have 2 questions about dual wielding.

The first: when the game is calculating you dual wielding attacks, does each hand get one attack once every so many seconds, or is it 1 hand attacking, and then the other.

If that's not clear, let's say you have you have a weapon in your right hand that attacks once every second (1.00) and another in your left that attacks twice every second (.5), with both values assuming that the dual wield speed bonus has already been applied. would you attack with the .5 weapon first, then both simultaneously, then the .5, simultaneous, etc. or would you attack first after 1 second, then half a second later, then a second later, then half a second later, and so on.

The second question, are barbarians allowed to dual wield any 2 handed weapon, or are there weapon classes that can be dual wielded while others cannot. The reason why I am asking this is because I thought that Barbarians could dual wield any 2 hander. However, their class specific Mighty Weapons have a 1 hand and 2 hand category. It seems stupid to have 2 handed class specific items for a class that can hold any 2 handed weapon in 1 hand.
In order to move forward, we must rid ourselves of that which holds us back. Check out my stream and give me tips! twitch.tv/intotheskyy
Belial-
Profile Joined April 2007
United States132 Posts
April 27 2012 21:37 GMT
#47
On April 28 2012 05:17 KiLL_ORdeR wrote:
I have 2 questions about dual wielding.

The first: when the game is calculating you dual wielding attacks, does each hand get one attack once every so many seconds, or is it 1 hand attacking, and then the other.

If that's not clear, let's say you have you have a weapon in your right hand that attacks once every second (1.00) and another in your left that attacks twice every second (.5), with both values assuming that the dual wield speed bonus has already been applied. would you attack with the .5 weapon first, then both simultaneously, then the .5, simultaneous, etc. or would you attack first after 1 second, then half a second later, then a second later, then half a second later, and so on.

The second question, are barbarians allowed to dual wield any 2 handed weapon, or are there weapon classes that can be dual wielded while others cannot. The reason why I am asking this is because I thought that Barbarians could dual wield any 2 hander. However, their class specific Mighty Weapons have a 1 hand and 2 hand category. It seems stupid to have 2 handed class specific items for a class that can hold any 2 handed weapon in 1 hand.



As for your first question, the attack speed is calculated independently, but the hands attack alternatingly ( fairly certain that isn't a word but whatever). This is why having a low DPS offhand can actually lower your dps, if they just attacked at their independently calculated dual-wield speed than having two weapons would ALWAYS be a dps increase over a single one-handed weapon. Basically, your second example is correct.

Unfortunately I can't help you on the second one :/
Skorpion
Profile Joined March 2011
United States15 Posts
April 27 2012 21:58 GMT
#48
On March 01 2012 22:26 gunr wrote:
Is the "dps" value of any use to casters? By that I mean the dps number on the stats page of your character. Since all the skills are based off "weapon damage". Or is this the same thing and just worded differently.


The DPS number on its own doe not necessarily mean anything, but the weapon speed also factors into your casting speed if I'm not mistaken.
Mataza
Profile Blog Joined August 2010
Germany5364 Posts
April 27 2012 23:40 GMT
#49
On April 28 2012 06:58 Skorpion wrote:
Show nested quote +
On March 01 2012 22:26 gunr wrote:
Is the "dps" value of any use to casters? By that I mean the dps number on the stats page of your character. Since all the skills are based off "weapon damage". Or is this the same thing and just worded differently.


The DPS number on its own doe not necessarily mean anything, but the weapon speed also factors into your casting speed if I'm not mistaken.

Don´t your own words contradict you?
Logically weapon damage increases your damage.
Weapon speed increases cast speed.
Put those together and you have Damage Per Second.

So yes, dps is generally good too for casters. Unless they completely remove the energy gaining feats from wiz, it should make next to no difference at all.
If nobody hates you, you´re doing something wrong. However someone hating you doesn´t make you right
HardlyNever
Profile Blog Joined July 2011
United States1258 Posts
April 27 2012 23:42 GMT
#50
I'm having trouble conceptualizing why damage can go DOWN because of dual wielding (I'm not denying that it can, I just can't figure out why).

I understand that the weaker weapon is dragging the dps down, but I can't understand WHY, and it is probably because I'm not understanding the dual-wield mechanic right.

The way I am understanding it is that EACH weapon swings 15% faster when you are dual wielding(this is the way it is presented in the video). So even if your offhand weapon did literally 0 dps, shouldn't your main-hand weapon be swinging 15% faster(making you do more damange)? I'm not understanding where the trade-off is that can bring the dps down. Something is being averaged here, it seems like, but I can't figure out what.

Again, I'm not refuting that dps can go down, I'm asking if anyone can explain more clearly WHY for me. Thanks.
Out there, the Kid learned to fend for himself. Learned to build. Learned to break.
m00nchile
Profile Joined July 2010
Slovenia240 Posts
Last Edited: 2012-04-27 23:55:37
April 27 2012 23:51 GMT
#51
On April 28 2012 08:42 HardlyNever wrote:
I'm having trouble conceptualizing why damage can go DOWN because of dual wielding (I'm not denying that it can, I just can't figure out why).

I understand that the weaker weapon is dragging the dps down, but I can't understand WHY, and it is probably because I'm not understanding the dual-wield mechanic right.

The way I am understanding it is that EACH weapon swings 15% faster when you are dual wielding(this is the way it is presented in the video). So even if your offhand weapon did literally 0 dps, shouldn't your main-hand weapon be swinging 15% faster(making you do more damange)? I'm not understanding where the trade-off is that can bring the dps down. Something is being averaged here, it seems like, but I can't figure out what.

Again, I'm not refuting that dps can go down, I'm asking if anyone can explain more clearly WHY for me. Thanks.

It's like this, you have a 30 damage weapon equipped, so your hits (excluding skill and stat bonuses) look like this: 30, 30, 30 etc. Then you equip a 5 damage weapon in the offhand slot, now you hit like this: 30, 5, 30, 5. Even though you attack faster, it doesn't make up for the fact you're causing very little damage every other hit, hence the dps drop. You just have to remember you're swinging the weapons one after the other, not simultaneously, just because you have two weapons equipped doesn't mean you add the damages together and apply them to every hit.
The above post was made by a noob. Take it as such.
HardlyNever
Profile Blog Joined July 2011
United States1258 Posts
Last Edited: 2012-04-28 00:19:02
April 28 2012 00:13 GMT
#52
On April 28 2012 08:51 m00nchile wrote:
Show nested quote +
On April 28 2012 08:42 HardlyNever wrote:
I'm having trouble conceptualizing why damage can go DOWN because of dual wielding (I'm not denying that it can, I just can't figure out why).

I understand that the weaker weapon is dragging the dps down, but I can't understand WHY, and it is probably because I'm not understanding the dual-wield mechanic right.

The way I am understanding it is that EACH weapon swings 15% faster when you are dual wielding(this is the way it is presented in the video). So even if your offhand weapon did literally 0 dps, shouldn't your main-hand weapon be swinging 15% faster(making you do more damange)? I'm not understanding where the trade-off is that can bring the dps down. Something is being averaged here, it seems like, but I can't figure out what.

Again, I'm not refuting that dps can go down, I'm asking if anyone can explain more clearly WHY for me. Thanks.

It's like this, you have a 30 damage weapon equipped, so your hits (excluding skill and stat bonuses) look like this: 30, 30, 30 etc. Then you equip a 5 damage weapon in the offhand slot, now you hit like this: 30, 5, 30, 5. Even though you attack faster, it doesn't make up for the fact you're causing very little damage every other hit, hence the dps drop. You just have to remember you're swinging the weapons one after the other, not simultaneously, just because you have two weapons equipped doesn't mean you add the damages together and apply them to every hit.


I get this, conceptually, but that doesn't explain what is going on with the numbers. The way it is displayed in the video, that shouldn't matter. The way attack speed was displayed in the video BOTH weapons were swinging 15% faster EACH. To use your example:

Lets say the axe alone swings for 30 damage 1.2 times a second. So we would have 30,30,30 1.2 times a second.

Then lets say you put a dagger in the off-hand doing that 5 damage 1.5 times a second . Your swings would look like 30, 5, 30, 5, 30, but your axe would be doing that 30 damage 1.38 times a second, just on its own (again, that is how the video presents it). That alone would net you a dps increase, regardless of what the off hand swings looked like.

Your explanation gives no consideration for speed in the example. Again, I'm not saying it isn't happening, I'm just not understanding how to model it based on what was presented in the video (that each weapon gains 15% IAS).
Out there, the Kid learned to fend for himself. Learned to build. Learned to break.
Belial-
Profile Joined April 2007
United States132 Posts
Last Edited: 2012-04-28 00:26:28
April 28 2012 00:20 GMT
#53
On April 28 2012 08:42 HardlyNever wrote:
I'm having trouble conceptualizing why damage can go DOWN because of dual wielding (I'm not denying that it can, I just can't figure out why).

I understand that the weaker weapon is dragging the dps down, but I can't understand WHY, and it is probably because I'm not understanding the dual-wield mechanic right.

The way I am understanding it is that EACH weapon swings 15% faster when you are dual wielding(this is the way it is presented in the video). So even if your offhand weapon did literally 0 dps, shouldn't your main-hand weapon be swinging 15% faster(making you do more damange)? I'm not understanding where the trade-off is that can bring the dps down. Something is being averaged here, it seems like, but I can't figure out what.

Again, I'm not refuting that dps can go down, I'm asking if anyone can explain more clearly WHY for me. Thanks.



It's because despite the 15% increased attack speed, your mainhand is actually attacking more slowly because it is alternating attacks with your offhand. For example you have Weapon A that does 20 damage per swing and attacks 1 time per second, for a total of 20 DPS, then you equip weapon B that does 5 damage a swing and attacks 1 time per second for 5 DPS. With the 15% attack speed buff you get 1.15 attacks per second from either weapon. So in 10 seconds you would get a total of 11.5 attacks if you were dual wielding, 5.75 of those doing 20 damage per hit (115 damage total) and 5.75 doing 5 damage per hit ( rounded up to 29 damage for simplicity's sake). Now in the same 10 seconds, if you were wielding only Weapon A you would hit 10 times for 200 damage. 115 + 29 is 144 damage, or 14.4 DPS , vs 200 (20 dps) of only wielding A means that equipping weapon B is a net DPS loss of about 23%.

The reason it's a loss is because both weapons don't attack at 1.15 speed EACH, they take turns attacking.

Edit:
"Your explanation gives no consideration for speed in the example. Again, I'm not saying it isn't happening, I'm just not understanding how to model it based on what was presented in the video (that each weapon gains 15% IAS)."

The reason is because the weapons don't work independently of each other, they are only treated as independent for purposes of calculating their attack speed.

If you had Weapon A with 2 attacks per second, and weapon B with 1 attack per second your attacks would look like this.

Weapon A attacks after ~.45 seconds, then weapon B attacks ~.83 seconds later, then .45 seconds after that Weapon A attacks again. (Obviously these times were completely geurilla math, but I think they illustrate the picture)

The weapons alternate and attack at the delay that is calculated independently for each weapon. They always attack A,B,A,B etc. no matter how much faster weapon A is than weapon B.
HardlyNever
Profile Blog Joined July 2011
United States1258 Posts
Last Edited: 2012-04-28 00:43:18
April 28 2012 00:36 GMT
#54
On April 28 2012 09:20 Belial- wrote:
Show nested quote +
On April 28 2012 08:42 HardlyNever wrote:
I'm having trouble conceptualizing why damage can go DOWN because of dual wielding (I'm not denying that it can, I just can't figure out why).

I understand that the weaker weapon is dragging the dps down, but I can't understand WHY, and it is probably because I'm not understanding the dual-wield mechanic right.

The way I am understanding it is that EACH weapon swings 15% faster when you are dual wielding(this is the way it is presented in the video). So even if your offhand weapon did literally 0 dps, shouldn't your main-hand weapon be swinging 15% faster(making you do more damange)? I'm not understanding where the trade-off is that can bring the dps down. Something is being averaged here, it seems like, but I can't figure out what.

Again, I'm not refuting that dps can go down, I'm asking if anyone can explain more clearly WHY for me. Thanks.



It's because despite the 15% increased attack speed, your mainhand is actually attacking more slowly because it is alternating attacks with your offhand.


This is contrary to what is being displayed in the video, (thus my hang up with this entire thing). The rest of your explanation is fine, except that this statement does not reflect what is being shown in the video. Skip to around 2:45 in the video, where he is equipping the club in one hand, and the dagger in the other.

If you watch, two things are happening. First and foremost, the attack speed of the CLUB goes UP from 1.2 to 1.38. So the main hand IS attacking faster. The second thing is that the video is saying that the attack speed is independent on BOTH weapons. In other words, the attack speed displayed isn't some average of the two, but BOTH weapons get to swing 15% faster.

My hang up is I'm not understanding how the other hand causes any sort of delay. I'm taking the displayed attack speed of each weapon literally, i.e. the club is going to swing 1.38 times a second, no matter what. What you are suggesting is that that isn't actually true, and that both weapons get a set attack speed (an average of the two, plus the 15% increase), and they take turns sharing that attack slot (in other words, your character will attack at that average speed, but alternate between the two weapons).

That is fine, and I would like it to work that way, but the way the information is presented in the video suggests otherwise. Perhaps the character sheet displayed attack speed isn't really accurate (i.e it doesn't truly alternate per weapon)?
Out there, the Kid learned to fend for himself. Learned to build. Learned to break.
m00nchile
Profile Joined July 2010
Slovenia240 Posts
April 28 2012 01:10 GMT
#55
On April 28 2012 09:36 HardlyNever wrote:
Show nested quote +
On April 28 2012 09:20 Belial- wrote:
On April 28 2012 08:42 HardlyNever wrote:
I'm having trouble conceptualizing why damage can go DOWN because of dual wielding (I'm not denying that it can, I just can't figure out why).

I understand that the weaker weapon is dragging the dps down, but I can't understand WHY, and it is probably because I'm not understanding the dual-wield mechanic right.

The way I am understanding it is that EACH weapon swings 15% faster when you are dual wielding(this is the way it is presented in the video). So even if your offhand weapon did literally 0 dps, shouldn't your main-hand weapon be swinging 15% faster(making you do more damange)? I'm not understanding where the trade-off is that can bring the dps down. Something is being averaged here, it seems like, but I can't figure out what.

Again, I'm not refuting that dps can go down, I'm asking if anyone can explain more clearly WHY for me. Thanks.



It's because despite the 15% increased attack speed, your mainhand is actually attacking more slowly because it is alternating attacks with your offhand.


This is contrary to what is being displayed in the video, (thus my hang up with this entire thing). The rest of your explanation is fine, except that this statement does not reflect what is being shown in the video. Skip to around 2:45 in the video, where he is equipping the club in one hand, and the dagger in the other.

If you watch, two things are happening. First and foremost, the attack speed of the CLUB goes UP from 1.2 to 1.38. So the main hand IS attacking faster. The second thing is that the video is saying that the attack speed is independent on BOTH weapons. In other words, the attack speed displayed isn't some average of the two, but BOTH weapons get to swing 15% faster.

My hang up is I'm not understanding how the other hand causes any sort of delay. I'm taking the displayed attack speed of each weapon literally, i.e. the club is going to swing 1.38 times a second, no matter what. What you are suggesting is that that isn't actually true, and that both weapons get a set attack speed (an average of the two, plus the 15% increase), and they take turns sharing that attack slot (in other words, your character will attack at that average speed, but alternate between the two weapons).

That is fine, and I would like it to work that way, but the way the information is presented in the video suggests otherwise. Perhaps the character sheet displayed attack speed isn't really accurate (i.e it doesn't truly alternate per weapon)?

But that's the point, the axe doesn't attack 1.38 times a second. It swings once, then the offhand, then the axe again. While the offhand is swinging, the axe isn't doing damage.
The above post was made by a noob. Take it as such.
Belial-
Profile Joined April 2007
United States132 Posts
Last Edited: 2012-04-28 01:14:38
April 28 2012 01:10 GMT
#56
The overall time it takes an individual weapon to attack is reduced by 15%, but because the hands are forced to take turns, your mainhand ends up attacking less often than it would if you were dual-wielding. I really don't know how to explain it any more plainly

Edit: Basically the MH's attack timer isn't on cooldown while the offhand is attacking. Whenever it is the offhand weapons turn to attack, the MH is basically just sitting there doing nothing.

Another way to say this is that the attacks per second that is displayed isn't a true representation of how many attacks per second that weapon makes while dual-wielding, but is instead a representation of the delay between when your offhand finishes attacking and your mainhand begins attacking. Basically blizzard chose to represent it in a counterintuitive way when considering how dual-wielding in general works.
KiLL_ORdeR
Profile Blog Joined June 2009
United States1518 Posts
April 28 2012 01:14 GMT
#57
On April 28 2012 06:37 Belial- wrote:
Show nested quote +
On April 28 2012 05:17 KiLL_ORdeR wrote:
I have 2 questions about dual wielding.

The first: when the game is calculating you dual wielding attacks, does each hand get one attack once every so many seconds, or is it 1 hand attacking, and then the other.

If that's not clear, let's say you have you have a weapon in your right hand that attacks once every second (1.00) and another in your left that attacks twice every second (.5), with both values assuming that the dual wield speed bonus has already been applied. would you attack with the .5 weapon first, then both simultaneously, then the .5, simultaneous, etc. or would you attack first after 1 second, then half a second later, then a second later, then half a second later, and so on.

The second question, are barbarians allowed to dual wield any 2 handed weapon, or are there weapon classes that can be dual wielded while others cannot. The reason why I am asking this is because I thought that Barbarians could dual wield any 2 hander. However, their class specific Mighty Weapons have a 1 hand and 2 hand category. It seems stupid to have 2 handed class specific items for a class that can hold any 2 handed weapon in 1 hand.



As for your first question, the attack speed is calculated independently, but the hands attack alternatingly ( fairly certain that isn't a word but whatever). This is why having a low DPS offhand can actually lower your dps, if they just attacked at their independently calculated dual-wield speed than having two weapons would ALWAYS be a dps increase over a single one-handed weapon. Basically, your second example is correct.

Unfortunately I can't help you on the second one :/


Thanks!

That's how I thought it would work, but I looked around a bunch and couldn't find a definitive answer, but you're reasoning for why it isn't always a dps increase to dual wield makes perfect sense.

As for my second question, in case anyone is wondering, I did a search on the forums and it turns out that Barbarians simply cannot equip 2 handed weapons in 1 hand. I assumed that they could, but now that I think about it, I never actually tried it in the beta.
In order to move forward, we must rid ourselves of that which holds us back. Check out my stream and give me tips! twitch.tv/intotheskyy
HardlyNever
Profile Blog Joined July 2011
United States1258 Posts
April 28 2012 01:19 GMT
#58
I feel like you guys aren't watching the video. Are you watching the video? Watch the video (at like the 2:44 mark). The club speed goes up to 1.38(1.2 +15%). He swings it. Then the character sheet goes to the offhand speed, which is like 1.72 (15% faster than 1.5). That is what is displayed.

Again, I'm willing to accept that what is displayed isn't what is actually going on, but that isn't how the information is presented. It is presented as BOTH weapons have an INDEPENDENT attack speed, and that INDEPENDENT attack speed is INCREASED by 15% EACH. He basically says as much (please, watch that part of the video).

What I'm getting at is something isn't working quite how it is displayed, but I'm not sure what it is.
Out there, the Kid learned to fend for himself. Learned to build. Learned to break.
eluv
Profile Joined August 2010
United States1251 Posts
Last Edited: 2012-04-28 01:26:57
April 28 2012 01:21 GMT
#59
Watch the video again, and note that the attack speed changes after one attack. This is what makes it clear that the weapons take turns sharing the attack slot. However, when they are swung, they use their own attack rate (as reduced by DW) independent of the attack rate of the other weapon. Maybe a timeline would make things clearer. Assuming that A attacks once a second and B attacks twice a second when taking into account the DW speed increase, your attacks will look like:

[Weapon A Hits] - [Wait 1s] - [Weapon B Hits] - [Wait .5s] - [Weapon A Hits] - [Wait 1s] - ....

Note that I've simplified this a bit, in that depending on how attack animations are coded and such, the "hit" should actually occur somewhere in the middle of the "Wait" time. What is displayed is exactly what happens, in the sense that while "active" each weapon attacks at it's own rate, it's just that while the other weapon is swinging, the unused weapon is not "active", and so has no effect on the current attack.

"Yes I fucked my way to the GSL partnership" - Sundance
Belial-
Profile Joined April 2007
United States132 Posts
April 28 2012 01:23 GMT
#60
On April 28 2012 10:19 HardlyNever wrote:
I feel like you guys aren't watching the video. Are you watching the video? Watch the video (at like the 2:44 mark). The club speed goes up to 1.38(1.2 +15%). He swings it. Then the character sheet goes to the offhand speed, which is like 1.72 (15% faster than 1.5). That is what is displayed.

Again, I'm willing to accept that what is displayed isn't what is actually going on, but that isn't how the information is presented. It is presented as BOTH weapons have an INDEPENDENT attack speed, and that INDEPENDENT attack speed is INCREASED by 15% EACH. He basically says as much (please, watch that part of the video).

What I'm getting at is something isn't working quite how it is displayed, but I'm not sure what it is.




It's working fine, I've watched the video and while he doesn't do a great job explaining it, it's there. You just don't seem to be understanding, not sure what more I can say
HardlyNever
Profile Blog Joined July 2011
United States1258 Posts
Last Edited: 2012-04-28 01:36:27
April 28 2012 01:30 GMT
#61
On April 28 2012 10:10 Belial- wrote:
Another way to say this is that the attacks per second that is displayed isn't a true representation of how many attacks per second that weapon makes while dual-wielding, but is instead a representation of the delay between when your offhand finishes attacking and your mainhand begins attacking. Basically blizzard chose to represent it in a counterintuitive way when considering how dual-wielding in general works.


This is basically what I'm getting at, except that the attacks per second displayed isn't really indicative of anything at all. When I see attack speed of 1.38 attacks per second, I assume that weapon will attack 1.38 times per second. Period. The end. Full stop. If I see the other weapon has an attack speed of 1.72 attacks per second, I assume the same thing for that. That weapon will attack 1.72 times a second.

What you're saying is that it just isn't true. I don't see how it represents a delay in anyway, and if it represents a delay, how is that delay represented mathematically? Are you saying the true attack speed of your character is really an average of the 2 (i.e. (1.38 +1.72)/2)?
Out there, the Kid learned to fend for himself. Learned to build. Learned to break.
Belial-
Profile Joined April 2007
United States132 Posts
April 28 2012 01:38 GMT
#62
On April 28 2012 10:30 HardlyNever wrote:
Show nested quote +
On April 28 2012 10:10 Belial- wrote:
Another way to say this is that the attacks per second that is displayed isn't a true representation of how many attacks per second that weapon makes while dual-wielding, but is instead a representation of the delay between when your offhand finishes attacking and your mainhand begins attacking. Basically blizzard chose to represent it in a counterintuitive way when considering how dual-wielding in general works.


This is basically what I'm getting at, except that the attacks per second displayed isn't really indicative of anything at all. When I see attack speed of 1.38 attacks per second, I assume that weapon will attack 1.38 times per second. Period. The end. Full stop. If I see the other weapon has an attack speed of 1.72 attacks per second, I assume the same thing for that. That weapon will attack 1.72 times a second.

What you're saying is that it just isn't true. I don't see how it represents a delay in anyway, and if it represents a delay, how is that delay represented mathematically? Are you saying the true attack speed of your character is really an average of the 2 (i.e. (1.38 +1.72)/2)?



Essentially yes, your attack speed ends up being an average of the dual-wield modified attack speeds of both of your weapons. The reason I say it is representative of the delay between attacks is because you can calculate the delay between attacks by looking at that number. Why blizzard decided not to just display the time between attacks as opposed to attacks per second is beyond me, it would really clear this all up.
HardlyNever
Profile Blog Joined July 2011
United States1258 Posts
Last Edited: 2012-04-28 01:47:42
April 28 2012 01:45 GMT
#63
On April 28 2012 10:38 Belial- wrote:
Show nested quote +
On April 28 2012 10:30 HardlyNever wrote:
On April 28 2012 10:10 Belial- wrote:
Another way to say this is that the attacks per second that is displayed isn't a true representation of how many attacks per second that weapon makes while dual-wielding, but is instead a representation of the delay between when your offhand finishes attacking and your mainhand begins attacking. Basically blizzard chose to represent it in a counterintuitive way when considering how dual-wielding in general works.


This is basically what I'm getting at, except that the attacks per second displayed isn't really indicative of anything at all. When I see attack speed of 1.38 attacks per second, I assume that weapon will attack 1.38 times per second. Period. The end. Full stop. If I see the other weapon has an attack speed of 1.72 attacks per second, I assume the same thing for that. That weapon will attack 1.72 times a second.

What you're saying is that it just isn't true. I don't see how it represents a delay in anyway, and if it represents a delay, how is that delay represented mathematically? Are you saying the true attack speed of your character is really an average of the 2 (i.e. (1.38 +1.72)/2)?



Essentially yes, your attack speed ends up being an average of the dual-wield modified attack speeds of both of your weapons. The reason I say it is representative of the delay between attacks is because you can calculate the delay between attacks by looking at that number. Why blizzard decided not to just display the time between attacks as opposed to attacks per second is beyond me, it would really clear this all up.


Ok, I think we are on the same page. He might want to update the video, because the 2nd question (to my mind) asks if the character's attack speed is the average of the two weapon's attack speed (I know it is modified attack speed), and his answer seems to refute that, and says they don't affect each other at all (while technically true, because one weapon doesn't affect the speed of the other, it is misleading because your character's true attack speed is the average of the two modified speeds).

Edit: As for why they chose this way, I believe WoW displayed weapon speed as time between attacks, and people felt it was counter-intuitive that higher "attack speed" weapons actually swung slower than lower "attack speed" weapons. Both systems have their drawbacks.
Out there, the Kid learned to fend for himself. Learned to build. Learned to break.
caradoc
Profile Blog Joined January 2011
Canada3022 Posts
Last Edited: 2012-04-28 01:55:49
April 28 2012 01:52 GMT
#64
On April 28 2012 10:45 HardlyNever wrote:
Show nested quote +
On April 28 2012 10:38 Belial- wrote:
On April 28 2012 10:30 HardlyNever wrote:
On April 28 2012 10:10 Belial- wrote:
Another way to say this is that the attacks per second that is displayed isn't a true representation of how many attacks per second that weapon makes while dual-wielding, but is instead a representation of the delay between when your offhand finishes attacking and your mainhand begins attacking. Basically blizzard chose to represent it in a counterintuitive way when considering how dual-wielding in general works.


This is basically what I'm getting at, except that the attacks per second displayed isn't really indicative of anything at all. When I see attack speed of 1.38 attacks per second, I assume that weapon will attack 1.38 times per second. Period. The end. Full stop. If I see the other weapon has an attack speed of 1.72 attacks per second, I assume the same thing for that. That weapon will attack 1.72 times a second.

What you're saying is that it just isn't true. I don't see how it represents a delay in anyway, and if it represents a delay, how is that delay represented mathematically? Are you saying the true attack speed of your character is really an average of the 2 (i.e. (1.38 +1.72)/2)?



Essentially yes, your attack speed ends up being an average of the dual-wield modified attack speeds of both of your weapons. The reason I say it is representative of the delay between attacks is because you can calculate the delay between attacks by looking at that number. Why blizzard decided not to just display the time between attacks as opposed to attacks per second is beyond me, it would really clear this all up.


Ok, I think we are on the same page. He might want to update the video, because the 2nd question (to my mind) asks if the character's attack speed is the average of the two weapon's attack speed (I know it is modified attack speed), and his answer seems to refute that, and says they don't affect each other at all (while technically true, because one weapon doesn't affect the speed of the other, it is misleading because your character's true attack speed is the average of the two modified speeds).

Edit: As for why they chose this way, I believe WoW displayed weapon speed as time between attacks, and people felt it was counter-intuitive that higher "attack speed" weapons actually swung slower than lower "attack speed" weapons. Both systems have their drawbacks.


Your true attack speed is the sum of the reciprocal of each individual weapon's attacks per second, multiplied by the 15% bonus. It's not the average.

I actually just made a post about this on the b.net forums, though it deals with dual wielding dps, it's essentially the same thing:

dual wield dps = ([dmg1 + dmg2] / [ 1/aps1 + 1/aps2 ] ) x 1.15
Salvation a la mode and a cup of tea...
m00nchile
Profile Joined July 2010
Slovenia240 Posts
April 28 2012 02:23 GMT
#65
On April 28 2012 10:30 HardlyNever wrote:
Show nested quote +
On April 28 2012 10:10 Belial- wrote:
Another way to say this is that the attacks per second that is displayed isn't a true representation of how many attacks per second that weapon makes while dual-wielding, but is instead a representation of the delay between when your offhand finishes attacking and your mainhand begins attacking. Basically blizzard chose to represent it in a counterintuitive way when considering how dual-wielding in general works.


This is basically what I'm getting at, except that the attacks per second displayed isn't really indicative of anything at all. When I see attack speed of 1.38 attacks per second, I assume that weapon will attack 1.38 times per second. Period. The end. Full stop. If I see the other weapon has an attack speed of 1.72 attacks per second, I assume the same thing for that. That weapon will attack 1.72 times a second.

What you're saying is that it just isn't true. I don't see how it represents a delay in anyway, and if it represents a delay, how is that delay represented mathematically? Are you saying the true attack speed of your character is really an average of the 2 (i.e. (1.38 +1.72)/2)?

Look, 1.38 attacks per second is a measurement and a representation, and looking st single wielding, it works, but for dual wielding, it can be counterintuitive. If you invert the value (I think that's the expression, I mean 1/x in any case) you get the duration of a swing, which seems to be what you're after. That is basically what the dual wield bonus does, reduces the swing time. And saying a 1.38 aps weapon always attacks 1.38 times a second is like saying a car doing 30 MPH always drives for an hour and always travels 30 miles.
The above post was made by a noob. Take it as such.
HardlyNever
Profile Blog Joined July 2011
United States1258 Posts
April 28 2012 03:09 GMT
#66
On April 28 2012 11:23 m00nchile wrote:
Show nested quote +
On April 28 2012 10:30 HardlyNever wrote:
On April 28 2012 10:10 Belial- wrote:
Another way to say this is that the attacks per second that is displayed isn't a true representation of how many attacks per second that weapon makes while dual-wielding, but is instead a representation of the delay between when your offhand finishes attacking and your mainhand begins attacking. Basically blizzard chose to represent it in a counterintuitive way when considering how dual-wielding in general works.


This is basically what I'm getting at, except that the attacks per second displayed isn't really indicative of anything at all. When I see attack speed of 1.38 attacks per second, I assume that weapon will attack 1.38 times per second. Period. The end. Full stop. If I see the other weapon has an attack speed of 1.72 attacks per second, I assume the same thing for that. That weapon will attack 1.72 times a second.

What you're saying is that it just isn't true. I don't see how it represents a delay in anyway, and if it represents a delay, how is that delay represented mathematically? Are you saying the true attack speed of your character is really an average of the 2 (i.e. (1.38 +1.72)/2)?

Look, 1.38 attacks per second is a measurement and a representation, and looking st single wielding, it works, but for dual wielding, it can be counterintuitive. If you invert the value (I think that's the expression, I mean 1/x in any case) you get the duration of a swing, which seems to be what you're after. That is basically what the dual wield bonus does, reduces the swing time. And saying a 1.38 aps weapon always attacks 1.38 times a second is like saying a car doing 30 MPH always drives for an hour and always travels 30 miles.


Umm... not really.

I get having the weapon speed for all weapons in your inventory and in shops not be adjusted for gear, dual wielding etc., but when a character sheet has "attack speed" that is adjusted to reflect my gear and stuff, I would expect it to show my actual attack speed.

It is a closed game world, it can display whatever it wants, and better, what it is really needs to tell the players (i.e. it isn't affected by numerous outside factors like a car would be). I would expect this to change within a few months after release.
Out there, the Kid learned to fend for himself. Learned to build. Learned to break.
m00nchile
Profile Joined July 2010
Slovenia240 Posts
April 28 2012 03:27 GMT
#67
On April 28 2012 12:09 HardlyNever wrote:
Show nested quote +
On April 28 2012 11:23 m00nchile wrote:
On April 28 2012 10:30 HardlyNever wrote:
On April 28 2012 10:10 Belial- wrote:
Another way to say this is that the attacks per second that is displayed isn't a true representation of how many attacks per second that weapon makes while dual-wielding, but is instead a representation of the delay between when your offhand finishes attacking and your mainhand begins attacking. Basically blizzard chose to represent it in a counterintuitive way when considering how dual-wielding in general works.


This is basically what I'm getting at, except that the attacks per second displayed isn't really indicative of anything at all. When I see attack speed of 1.38 attacks per second, I assume that weapon will attack 1.38 times per second. Period. The end. Full stop. If I see the other weapon has an attack speed of 1.72 attacks per second, I assume the same thing for that. That weapon will attack 1.72 times a second.

What you're saying is that it just isn't true. I don't see how it represents a delay in anyway, and if it represents a delay, how is that delay represented mathematically? Are you saying the true attack speed of your character is really an average of the 2 (i.e. (1.38 +1.72)/2)?

Look, 1.38 attacks per second is a measurement and a representation, and looking st single wielding, it works, but for dual wielding, it can be counterintuitive. If you invert the value (I think that's the expression, I mean 1/x in any case) you get the duration of a swing, which seems to be what you're after. That is basically what the dual wield bonus does, reduces the swing time. And saying a 1.38 aps weapon always attacks 1.38 times a second is like saying a car doing 30 MPH always drives for an hour and always travels 30 miles.


Umm... not really.

I get having the weapon speed for all weapons in your inventory and in shops not be adjusted for gear, dual wielding etc., but when a character sheet has "attack speed" that is adjusted to reflect my gear and stuff, I would expect it to show my actual attack speed.

It is a closed game world, it can display whatever it wants, and better, what it is really needs to tell the players (i.e. it isn't affected by numerous outside factors like a car would be). I would expect this to change within a few months after release.

Ok, this is the last time I reply to you, as I feel I'm banging my head against a wall. 1.38 attacks per second is the exact same thing as saying 725ms per weapon swing. What you're saying is I drove 15 miles in half an hour, therefore I didn't average 30 MPH but 15 MP1/2H. Again, the representation Blizzard chose for weapon speed can be counterintuitive, but it's still valid. The old example, a 1.2 speed weapon goes to 1.38 when dual wielded. You just went from a 833ms swing to a 725ms swing. The weapon swings faster. If you equipped a pos in the offhand, that weapon attacks with lousy damage when it could have been the high damage weapon swinging. Less damage per time = less DPS.
The above post was made by a noob. Take it as such.
HardlyNever
Profile Blog Joined July 2011
United States1258 Posts
April 28 2012 03:43 GMT
#68
On April 28 2012 12:27 m00nchile wrote:
Show nested quote +
On April 28 2012 12:09 HardlyNever wrote:
On April 28 2012 11:23 m00nchile wrote:
On April 28 2012 10:30 HardlyNever wrote:
On April 28 2012 10:10 Belial- wrote:
Another way to say this is that the attacks per second that is displayed isn't a true representation of how many attacks per second that weapon makes while dual-wielding, but is instead a representation of the delay between when your offhand finishes attacking and your mainhand begins attacking. Basically blizzard chose to represent it in a counterintuitive way when considering how dual-wielding in general works.


This is basically what I'm getting at, except that the attacks per second displayed isn't really indicative of anything at all. When I see attack speed of 1.38 attacks per second, I assume that weapon will attack 1.38 times per second. Period. The end. Full stop. If I see the other weapon has an attack speed of 1.72 attacks per second, I assume the same thing for that. That weapon will attack 1.72 times a second.

What you're saying is that it just isn't true. I don't see how it represents a delay in anyway, and if it represents a delay, how is that delay represented mathematically? Are you saying the true attack speed of your character is really an average of the 2 (i.e. (1.38 +1.72)/2)?

Look, 1.38 attacks per second is a measurement and a representation, and looking st single wielding, it works, but for dual wielding, it can be counterintuitive. If you invert the value (I think that's the expression, I mean 1/x in any case) you get the duration of a swing, which seems to be what you're after. That is basically what the dual wield bonus does, reduces the swing time. And saying a 1.38 aps weapon always attacks 1.38 times a second is like saying a car doing 30 MPH always drives for an hour and always travels 30 miles.


Umm... not really.

I get having the weapon speed for all weapons in your inventory and in shops not be adjusted for gear, dual wielding etc., but when a character sheet has "attack speed" that is adjusted to reflect my gear and stuff, I would expect it to show my actual attack speed.

It is a closed game world, it can display whatever it wants, and better, what it is really needs to tell the players (i.e. it isn't affected by numerous outside factors like a car would be). I would expect this to change within a few months after release.

Ok, this is the last time I reply to you, as I feel I'm banging my head against a wall. 1.38 attacks per second is the exact same thing as saying 725ms per weapon swing. What you're saying is I drove 15 miles in half an hour, therefore I didn't average 30 MPH but 15 MP1/2H. Again, the representation Blizzard chose for weapon speed can be counterintuitive, but it's still valid. The old example, a 1.2 speed weapon goes to 1.38 when dual wielded. You just went from a 833ms swing to a 725ms swing. The weapon swings faster. If you equipped a pos in the offhand, that weapon attacks with lousy damage when it could have been the high damage weapon swinging. Less damage per time = less DPS.


We're way passed that point... I was just saying your car analogy sucked lol.
Out there, the Kid learned to fend for himself. Learned to build. Learned to break.
Belial-
Profile Joined April 2007
United States132 Posts
Last Edited: 2012-04-28 09:08:19
April 28 2012 08:04 GMT
#69
On April 28 2012 10:52 caradoc wrote:

Your true attack speed is the sum of the reciprocal of each individual weapon's attacks per second, multiplied by the 15% bonus. It's not the average.

I actually just made a post about this on the b.net forums, though it deals with dual wielding dps, it's essentially the same thing:

dual wield dps = ([dmg1 + dmg2] / [ 1/aps1 + 1/aps2 ] ) x 1.15



Alright, I guess I didn't think it through completely, but wouldn't your attack speed total end up at the average of the two speeds anyway?

First hand attacks at 2 attacks per second, the other at 1 per second, for simplicity's sake let's say those are the speeds post modifier. Weapon 1 attacks after .5 seconds, then weapon two attacks 1 second later which ends that attack cycle in 1.5 seconds which is also the average of 2 and 1.

Or maybe I'm simplifying it too far...

Edit: Yeah, turns out I'm an idiot. After thinking about it I realized that that equates to 1.33 attacks per second >.>
Lavit2099
Profile Joined November 2011
United States390 Posts
June 05 2012 08:28 GMT
#70
Assuming all things being equal (in terms of dps/speed) on weapons. What is the real gain of duel wielding over using a two hander, or even a one hander and a shield/offhand? My main's a DH, just started Act 3 on Nightmare (the story drove me away from the game once I beat it but I found a friend to play with) and I use a 2h crossbow and quiver. Assuming I found "like" dps/speed hand crossbows, would my net gain in "damage" just be that my attacks would be slightly faster?

Ignoring "those items aren't in the game!" outrage, going to use this as an example:

2 handed crossbow, 100dps (1.3aps), quiver with 10% attack speed.
vs.
Two one handed crossbows, 100dps (1.3aps).

If I'm reading everything right, I'd attack a teeny tiny bit faster while wielding two than when I'm not?
Normal
Please log in or register to reply.
Live Events Refresh
OSC
12:30
King of the Hill Weekly #220
CranKy Ducklings116
Liquipedia
[ Submit Event ]
Live Streams
Refresh
StarCraft 2
Harstem 520
Hui .211
StarCraft: Brood War
Britney 46702
Horang2 3871
Bisu 2251
ggaemo 1937
Flash 1485
Jaedong 1153
Hyun 1071
Barracks 871
EffOrt 724
Mini 483
[ Show more ]
Larva 423
Soulkey 327
actioN 323
Last 234
ZerO 165
Snow 164
Killer 153
ToSsGirL 112
Zeus 112
Rush 74
JYJ67
Leta 65
Sharp 51
Backho 47
sSak 45
Movie 43
sorry 39
sas.Sziky 30
zelot 27
yabsab 26
Sacsri 25
[sc1f]eonzerg 21
Shinee 15
Noble 14
Hm[arnc] 10
ajuk12(nOOB) 10
Terrorterran 5
IntoTheRainbow 4
Aegong 4
Shine 0
Stormgate
RushiSC2
Dota 2
Gorgc5687
qojqva2400
XcaliburYe280
420jenkins114
Counter-Strike
markeloff83
kRYSTAL_38
Other Games
singsing2456
B2W.Neo1179
hiko739
Fuzer 352
DeMusliM340
Happy184
ToD114
QueenE29
ZerO(Twitch)11
Organizations
StarCraft 2
Blizzard YouTube
StarCraft: Brood War
BSLTrovo
sctven
[ Show 15 non-featured ]
StarCraft 2
• StrangeGG 72
• davetesta41
• AfreecaTV YouTube
• intothetv
• Kozan
• IndyKCrew
• LaughNgamezSOOP
• Migwel
• sooper7s
StarCraft: Brood War
• BSLYoutube
• STPLYoutube
• ZZZeroYoutube
League of Legends
• Nemesis2236
• Jankos886
Other Games
• WagamamaTV309
Upcoming Events
uThermal 2v2 Circuit
2h 45m
The PondCast
20h 45m
Online Event
1d 2h
Korean StarCraft League
2 days
CranKy Ducklings
2 days
BSL20 Non-Korean Champi…
2 days
Mihu vs QiaoGege
Zhanhun vs Dewalt
Fengzi vs TBD
Online Event
3 days
Sparkling Tuna Cup
3 days
BSL20 Non-Korean Champi…
4 days
Bonyth vs TBD
OSC
5 days
Liquipedia Results

Completed

BSL 20 Non-Korean Championship
FEL Cracow 2025
Underdog Cup #2

Ongoing

Copa Latinoamericana 4
Jiahua Invitational
BSL 20 Team Wars
CC Div. A S7
IEM Cologne 2025
FISSURE Playground #1
BLAST.tv Austin Major 2025
ESL Impact League Season 7
IEM Dallas 2025
PGL Astana 2025

Upcoming

BSL 21 Qualifiers
ASL Season 20: Qualifier #1
ASL Season 20: Qualifier #2
ASL Season 20
CSLPRO Chat StarLAN 3
BSL Season 21
RSL Revival: Season 2
Maestros of the Game
SEL Season 2 Championship
WardiTV Summer 2025
uThermal 2v2 Main Event
HCC Europe
Yuqilin POB S2
ESL Pro League S22
StarSeries Fall 2025
FISSURE Playground #2
BLAST Open Fall 2025
BLAST Open Fall Qual
Esports World Cup 2025
BLAST Bounty Fall 2025
BLAST Bounty Fall Qual
TLPD

1. ByuN
2. TY
3. Dark
4. Solar
5. Stats
6. Nerchio
7. sOs
8. soO
9. INnoVation
10. Elazer
1. Rain
2. Flash
3. EffOrt
4. Last
5. Bisu
6. Soulkey
7. Mini
8. Sharp
Sidebar Settings...

Advertising | Privacy Policy | Terms Of Use | Contact Us

Original banner artwork: Jim Warren
The contents of this webpage are copyright © 2025 TLnet. All Rights Reserved.