• Log InLog In
  • Register
Liquid`
Team Liquid Liquipedia
EST 14:42
CET 20:42
KST 04:42
  • 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
RSL Revival - 2025 Season Finals Preview8RSL Season 3 - Playoffs Preview0RSL Season 3 - RO16 Groups C & D Preview0RSL Season 3 - RO16 Groups A & B Preview2TL.net Map Contest #21: Winners12
Community News
ComeBackTV's documentary on Byun's Career !10Weekly Cups (Dec 8-14): MaxPax, Clem, Cure win4Weekly Cups (Dec 1-7): Clem doubles, Solar gets over the hump1Weekly Cups (Nov 24-30): MaxPax, Clem, herO win2BGE Stara Zagora 2026 announced15
StarCraft 2
General
ComeBackTV's documentary on Byun's Career ! Micro Lags When Playing SC2? When will we find out if there are more tournament Weekly Cups (Dec 8-14): MaxPax, Clem, Cure win RSL Revival - 2025 Season Finals Preview
Tourneys
Sparkling Tuna Cup - Weekly Open Tournament $100 Prize Pool - Winter Warp Gate Masters Showdow $5,000+ WardiTV 2025 Championship Winter Warp Gate Amateur Showdown #1 RSL Offline Finals Info - Dec 13 and 14!
Strategy
Custom Maps
Map Editor closed ?
External Content
Mutation # 505 Rise From Ashes Mutation # 504 Retribution Mutation # 503 Fowl Play Mutation # 502 Negative Reinforcement
Brood War
General
Klaucher discontinued / in-game color settings Anyone remember me from 2000s Bnet EAST server? BGH Auto Balance -> http://bghmmr.eu/ How Rain Became ProGamer in Just 3 Months FlaSh on: Biggest Problem With SnOw's Playstyle
Tourneys
[BSL21] LB QuarterFinals - Sunday 21:00 CET Small VOD Thread 2.0 [Megathread] Daily Proleagues [BSL21] WB SEMIFINALS - Saturday 21:00 CET
Strategy
Simple Questions, Simple Answers Game Theory for Starcraft Current Meta Fighting Spirit mining rates
Other Games
General Games
Path of Exile Stormgate/Frost Giant Megathread General RTS Discussion Thread Nintendo Switch Thread Mechabellum
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
Deck construction bug Heroes of StarCraft mini-set
TL Mafia
Mafia Game Mode Feedback/Ideas Survivor II: The Amazon Sengoku Mafia TL Mafia Community Thread
Community
General
US Politics Mega-thread The Games Industry And ATVI Russo-Ukrainian War Thread Things Aren’t Peaceful in Palestine YouTube Thread
Fan Clubs
White-Ra Fan Club
Media & Entertainment
Anime Discussion Thread [Manga] One Piece Movie Discussion!
Sports
2024 - 2026 Football Thread Formula 1 Discussion
World Cup 2022
Tech Support
Computer Build, Upgrade & Buying Resource Thread
TL Community
TL+ Announced Where to ask questions and add stream?
Blogs
The (Hidden) Drug Problem in…
TrAiDoS
I decided to write a webnov…
DjKniteX
James Bond movies ranking - pa…
Topin
Thanks for the RSL
Hildegard
Customize Sidebar...

Website Feedback

Closed Threads



Active: 2112 users

DPS Items: A Brief Comparison (6.83b)

Blogs > aboxcar
Post a Reply
1 2 Next All
aboxcar
Profile Blog Joined May 2013
United States447 Posts
January 23 2015 04:32 GMT
#1
I'll try to keep this short and to the point.

Daedalus (5550)
+81 damage
25% chance to critical strike for 240% damage

In the basic case where there are no other sources of crit, we can call Daedalus +81 damage and +35% dps.

It should be noted that critical strike does not apply against buildings, and does benefit illusions. Illusions do not benefit from +81 added damage.

Desolator (4100)
+60 damage
-7 armor

Each point of armor is worth very close to 6% effective hit points (EHP). In the simple case that you remove all the armor from a hero with 7 armor (reducing from 7 armor to 0), this is basically +42% dps. If you reduced a level 16 unequipped Juggernaut from 10 armor to 3, this is like +36% dps.

There are some consequences to armor working like this. First, every point of additional armor is worth less and less. (Because you are getting less and less % increase in effective hitpoints. For example, going from 106 EHP to 112 EHP is not as good as going from 100 EHP to 106 EHP.) High armor agility heroes thus prefer evasion.

Inversely, each additional point of armor reduction is worth more and more! Consult the graphs here for further information: http://dota2.gamepedia.com/Armor

Thus desolator tends to have additional value on heroes that already have armor reduction, and against heroes that lack armor.

It should be noted that the desolator orb does apply to buildings, and fully stacks with other armor reduction. It also benefits other sources of physical damage. An exception is cleave, which ignores armor.

Monkey King Bar (5400)
+88 damage
+15 attack speed
35% chance to minibash for 100 damage.
True strike

Several things to note. First, attack speed (http://dota2.gamepedia.com/Attack_speed) in this game is not multiplicative. +15 attack speed is NOT the same as +15% dps, it is far inferior. Like armor, additional attack speed is worth less and less.

Minibash deals magical damage for melee carriers, and physical damage for ranged carriers (the bash is physical for melee and magical for ranged). Therefore gamepedia is incorrect and it is not the equivalent of adding 35 damage to every attack, as for melee carriers it is subject to magic resistance.

If a hero had no increased attack speed (IAS) (it would also have to have 0 agility, as each point of agility increases attack speed), then +15 attack speed would be equivalent to +15% dps. If this hero had only treads, this 15 IAS is now only worth 11.5% dps.

Now pretend a hero has 50% evasion and 1/2 incoming attacks miss. If you took that away, it's like a 100% dps increase! If a hero has 35% evasion from Butterly, the true strike is worth about +54% dps.

It should be noted that true strike benefits illusions, does not apply to attacks against buildings uphill. Attacks from cleave and flak cannon cannot be evaded.

Conclusion: MKB is a poor choice for a DPS item unless an opponent has evasion or a way to make you miss.
everything that rises must converge
ahswtini
Profile Blog Joined June 2008
Northern Ireland22208 Posts
January 23 2015 10:27 GMT
#2
On January 23 2015 13:32 aboxcar wrote:
Each point of armor is worth very close to 6% effective hit points (EHP). In the simple case that you remove all the armor from a hero with 7 armor (reducing from 7 armor to 0), this is basically +42% dps. If you reduced a level 16 unequipped Juggernaut from 10 armor to 3, this is like +36% dps.

There are some consequences to armor working like this. First, every point of additional armor is worth less and less. (Because you are getting less and less % increase in effective hitpoints. For example, going from 106 EHP to 112 EHP is not as good as going from 100 EHP to 106 EHP.) High armor agility heroes thus prefer evasion.

Armour does not stack with diminishing returns.

I understand your perspective, if you take a hero with 1000HP, the first 10 armour will give him 1600EHP (60% increase), but the next 10 armour will give him 2200EHP (120% increase from 0 armour, but only 37.5% increase from 10 armour). You're still getting +60% EHP for every 10 armour, because EHP should be calculated from your actual HP. You don't calculate your EHP by using EHP, if you understand what I mean.

You could say the same thing about stacking vit boosters. Each vit booster (like armour) gives you an absolute increase in the amount of HP, but they will give you decreasing relative increase.

If we take a situation where we have a medallion and we are against a Puck with 1000HP/7 armour, and a centaur with 2700HP/17 armour. If you medallion the puck you reduce her EHP by about 42%. If you medallion the centaur, you reduce his EHP by about 21%. However, in terms of actual damage, you're essentially nuking the centaur for over 1000 damage, whereas you're only nuking the puck for 400 damage. Naturally this only holds true if you outright kill both of them, and you only do so with physical damage (which is unlikely in a real game). Most people are still going to medallion the puck just to kill her faster, but if you have a team of expert mathematicians who can calculate the the most efficient way to kill, medallioning the centaur does more "effective damage"
"As I've said, balance isn't about strategies or counters, it's about probability and statistics." - paralleluniverse
FFGenerations
Profile Blog Joined April 2011
7088 Posts
Last Edited: 2015-01-23 11:57:03
January 23 2015 11:56 GMT
#3
do heroes have or attain evasion in some way other than buying a butterfly or pa's passive?

like, does a lv25 antimage or mirana with no items have 12% evasion?

ie is it only worth getting an mkb when opponent makes butterfly or pa? (ignoring high ground miss) fuck macs holy fucking shit
Cool BW Music Vid - youtube.com/watch?v=W54nlqJ-Nx8 ~~~~~ ᕤ OYSTERS ᕤ CLAMS ᕤ AND ᕤ CUCKOLDS ᕤ ~~~~~~ ༼ ᕤ◕◡◕ ༽ᕤ PUNCH HIM ༼ ᕤ◕◡◕ ༽ᕤ
Saechiis
Profile Blog Joined May 2010
Netherlands4989 Posts
Last Edited: 2015-01-23 12:08:04
January 23 2015 12:06 GMT
#4
http://dota2.gamepedia.com/Evasion

The page doesn't seem to include Troll's whirling axes (60% evasion), or KotL's blinding light (80% evasion).

Edit: I guess they're considered blinds: http://dota2.gamepedia.com/Blind
I think esports is pretty nice.
ahswtini
Profile Blog Joined June 2008
Northern Ireland22208 Posts
January 23 2015 13:19 GMT
#5
On January 23 2015 20:56 FFGenerations wrote:
do heroes have or attain evasion in some way other than buying a butterfly or pa's passive?

like, does a lv25 antimage or mirana with no items have 12% evasion?

ie is it only worth getting an mkb when opponent makes butterfly or pa? (ignoring high ground miss) fuck macs holy fucking shit

No you don't get evasion as part of any of the attributes.

You not only get mkb to deal with evasion, but when you're dealing with miss chances caused by blinds.

And yeah troll axes and blinding light are blinds and not evasion. There's a big difference between the two - evasion is a property of a hero, whereas blinds are debuffs placed on enemy heroes. If you have evasion, you can evade attacks from all enemy heroes, whereas if you apply a blind, it only affects that enemy hero.

Brewmaster being especially annoying and worth buying mkb for, since he has both evasion AND can apply blind.
"As I've said, balance isn't about strategies or counters, it's about probability and statistics." - paralleluniverse
aboxcar
Profile Blog Joined May 2013
United States447 Posts
Last Edited: 2015-01-23 14:04:25
January 23 2015 14:02 GMT
#6
On January 23 2015 20:56 FFGenerations wrote:
do heroes have or attain evasion in some way other than buying a butterfly or pa's passive?

like, does a lv25 antimage or mirana with no items have 12% evasion?

ie is it only worth getting an mkb when opponent makes butterfly or pa? (ignoring high ground miss) fuck macs holy fucking shit

The posts above mentioned a few sources of miss chance. There are more like Tinker's laser and windrun (and halberd). It's usually worth getting MKB against these heroes. BKB won't even remove the haze from panda after he's already made you drunk!

I didn't explicitly point it out b/c I didn't want OP to get long and math-y, but if you have super low damage and low attack speed, like a support CM, MKB will give you more damage than Daedalus.

It's the same with phase boots vs treads. For most heroes, at the early levels, phase boots actually gives you more damage.
As you level and farm up, treads becomes the better item from a DPS point of view.
everything that rises must converge
Mafe
Profile Joined February 2011
Germany5966 Posts
Last Edited: 2015-01-23 16:05:11
January 23 2015 16:04 GMT
#7
Well if you try to put most stats into dps-terms, you might as well do so with any flat damage bonus.

Thus the 81 damage on daedalus is +81%dps if you are have 100 damage, but it's just +40,5% dps if you have 200 damage.

Comparing dps is fun if you like theorycraft, but for practical relevance in the game, one should always keep in mind that "+20% dps" rarely translates into "20% better" or "20% more useful", in dota even more so than in many other video games.
FFGenerations
Profile Blog Joined April 2011
7088 Posts
January 23 2015 16:24 GMT
#8
ah yeah, fuck windrun!
Cool BW Music Vid - youtube.com/watch?v=W54nlqJ-Nx8 ~~~~~ ᕤ OYSTERS ᕤ CLAMS ᕤ AND ᕤ CUCKOLDS ᕤ ~~~~~~ ༼ ᕤ◕◡◕ ༽ᕤ PUNCH HIM ༼ ᕤ◕◡◕ ༽ᕤ
TomatoBisque
Profile Joined March 2013
United States6290 Posts
Last Edited: 2015-01-24 09:25:13
January 24 2015 09:16 GMT
#9
"Armor has diminishing returns" arguments were more interesting when it was people going "omg 10 armor gives 40% reduction but 20 only increases that to 60% diminishing returns!!!11!" now I just feel like it's people who didn't want to admit they were wrong about the former argument so they're making up their own definition for diminishing returns to still be right.

Diminishing returns has a pretty simple definition: If you increase your input by a factor of n, and the output is increased by a factor less than n, you have diminishing returns. This is clearly not true for armor, since, as pointed out above, 10 armor increases EHP by 60%, while 20 increases by 120%. When the input and output have the same factor, we have a constant return.

The fact that HP becomes more valuable as our armor increases does not change the fact that armor has a constant return.
rip
Saechiis
Profile Blog Joined May 2010
Netherlands4989 Posts
Last Edited: 2015-01-24 10:16:45
January 24 2015 10:16 GMT
#10
But the question is: what is the most cost-efficient way to increase my EHP? Armor's value does change considering the amount of HP and armor you already have. Armor has diminishing returns from a cost-efficiency stance.
I think esports is pretty nice.
vndestiny
Profile Blog Joined September 2011
Singapore3441 Posts
January 24 2015 10:37 GMT
#11
On January 24 2015 19:16 Saechiis wrote:
But the question is: what is the most cost-efficient way to increase my EHP? Armor's value does change considering the amount of HP and armor you already have. Armor has diminishing returns from a cost-efficiency stance.

Well hp also have a diminishing return if you wanna look at it that way.
Saechiis
Profile Blog Joined May 2010
Netherlands4989 Posts
January 24 2015 11:12 GMT
#12
On January 24 2015 19:37 vndestiny wrote:
Show nested quote +
On January 24 2015 19:16 Saechiis wrote:
But the question is: what is the most cost-efficient way to increase my EHP? Armor's value does change considering the amount of HP and armor you already have. Armor has diminishing returns from a cost-efficiency stance.

Well hp also have a diminishing return if you wanna look at it that way.


Yes, that makes it intersting to figure out what's the best buy.
I think esports is pretty nice.
aboxcar
Profile Blog Joined May 2013
United States447 Posts
January 24 2015 12:11 GMT
#13
I never used the term "diminishing returns" because it's confusing. You forget what the point you are talking about is. What is the return? and How is it diminishing?

The armor system is from WC3: http://classic.battle.net/war3/basics/armorandweapontypes.shtml


Damage Reduction or Increases for Armor
For positive Armor, damage reduction =((armor)*0.06)/(1+0.06*(armor))
For negative Armor, it is damage increase = 2-0.94^(-armor) since you take more damage for negative armor scores.

Consult this chart* for more information. A Positive Armor of 1 reduces damage by about 5.7%. A negative armor of 10 increases damage by 46.1%. A unit with 20 armor basically has 55% extra hitpoints -- 100 would become effectively 155.


Is EHP a "return" of armor, or is it just a tool to understand damage reduction? Is damage reduction the "return"?
If you gain 20 EHP, then 20 EHP, and then 20 more, are you returning less? What if equivalently your damage reduction increases less and less?

But what was the point again? To win an argument about "diminishing returns"?

The point is, if you have lots of armor, more armor does not mean a lot to you. If you have no armor, a little armor goes a long way.
everything that rises must converge
Saechiis
Profile Blog Joined May 2010
Netherlands4989 Posts
Last Edited: 2015-01-24 12:54:02
January 24 2015 12:52 GMT
#14
It's not like we haven't already dereailed the thread, so yes, I'm on the internet and I'd like to argue about diminishing returns.
We all agree, but let's argue about whether the definition is strictly correct.
I think esports is pretty nice.
ahswtini
Profile Blog Joined June 2008
Northern Ireland22208 Posts
Last Edited: 2015-01-24 17:05:58
January 24 2015 17:01 GMT
#15
But that each point in armour gives you less and less is wrong, each point still gives you +6% of your HP as EHP. The question is, at what point is it a better use of money (and more importantly, inventory slots) to increase your HP.

This may help visualise it
[image loading]

That's purely physical EHP. There's no optimal hp to armour ratio, more you need to factor in gold and slots (and magical/pure sources of damage). Basically, you could spend 1400 gold to buy a platemail and get, say +300 more EHP, or you could spend 1100 on a vit booster and end up with +500 EHP

To put it another way, buying more armour doesn't make you gain less EHP, but you could be getting more EHP if you get HP items. Therefore, armour is not less effective the more you buy it, it's just that you could be getting more effective items.
"As I've said, balance isn't about strategies or counters, it's about probability and statistics." - paralleluniverse
TomatoBisque
Profile Joined March 2013
United States6290 Posts
January 24 2015 23:50 GMT
#16
On January 24 2015 19:16 Saechiis wrote:
But the question is: what is the most cost-efficient way to increase my EHP? Armor's value does change considering the amount of HP and armor you already have. Armor has diminishing returns from a cost-efficiency stance.

That's not what diminishing returns mean, you're saying something that is true (having more armor increases the effectiveness of buying HP), and then trying to change the definition of a term because you don't understand it.
rip
aboxcar
Profile Blog Joined May 2013
United States447 Posts
Last Edited: 2015-01-25 01:16:07
January 25 2015 01:13 GMT
#17
HP and armor is not the whole story though. Evasion is another way to look at it.

0 to 9 armor (giving 35.1% damage reduction) is, all other things being equal, worth almost the same as evasion from butterfly. They will both give about 54% more effective hit points.

but 9 to 18 armor is not worth as much as 35% evasion from butterfly. Can you see that?


Or try thinking about it backwards. If you reduce someone from 25 armor to 20 armor, this is not as potent as reducing them from 5 armor to 0 armor. You can do the math to show it, but maybe this is intuitive based on game experience.. If you can accept this, then the inverse must be true too.

It's not just about the raw number of EHP, and it's certainly not about the definition of whatever terms you want to pull out. It just confuses people...
everything that rises must converge
SKC
Profile Joined October 2010
Brazil18828 Posts
Last Edited: 2015-01-25 01:49:48
January 25 2015 01:39 GMT
#18
That's because evasion gives more EHP the more armor you have. But armor remains equally eficient no matter how much armor you have. Diminishing returs implies that it drops in value, when it doesnt. The fact that other stats increase in value is meaningless to how armor works, armor doesn't get worse.

It's like saying damage has diminishing returns because at some point it's better to buy attack speed. It's dumb to treat armor that way because that's how evey single stat works. It's almost always better to not stack only a single stat, be it armor, evasion, damage, attack speed, magic reduction or whathever.

It's you who explained what you wanted to say poorly and keeps trying to defend it when it makes no sense.
aboxcar
Profile Blog Joined May 2013
United States447 Posts
Last Edited: 2015-01-25 02:23:14
January 25 2015 02:21 GMT
#19
Evasion will give the same % increase no matter what the other variables are. Armor will not.

And actually, not all attributes work this way.

Magic resistance and, recently after icefrog added it, evasion, stack simply multiplicatively.

For example, if you have two instances of 20% evasion, your total evasion is 1 - .8 * .8 = 36%

I also agree that buying more attack speed at certain points is not worth it.
everything that rises must converge
SKC
Profile Joined October 2010
Brazil18828 Posts
Last Edited: 2015-01-25 02:42:40
January 25 2015 02:42 GMT
#20
If you have a lot of evasion, more evasion is worse than other stats.
If you have a lot of MR, more MR is worse than other stats.
If you have a lot of armor, armor is worse than other stats.
If you have a lot of HP, HP is worse than other stats.
If you have a lot of damage, damage is worse than other stats.
If you have a lot of attack speed, attack speed is worse than other stats.

Every single stat can be argued to have "diminishing returns" in the same way you argued armor does.

Armor gives the exact same amount of EHP for every point of armor. It's not about %s. Increasing the output by the same % is broken and that's why almost no stat actually does it. Evasion doesn't do it because it's multiplicative. If lets, say, every 100 damage you got always doubled your DPS damage would be a broken stat.
1 2 Next All
Please log in or register to reply.
Live Events Refresh
Ladder Legends
17:00
WWG Masters Showdown
SteadfastSC232
Liquipedia
[ Submit Event ]
Live Streams
Refresh
StarCraft 2
White-Ra 585
SteadfastSC 232
ProTech142
BRAT_OK 116
IndyStarCraft 21
StarCraft: Brood War
Calm 1950
Shuttle 445
Dewaltoss 121
Mini 103
Hyun 90
ggaemo 55
yabsab 48
910 24
soO 19
Killer 15
[ Show more ]
HiyA 12
NaDa 4
Counter-Strike
fl0m1069
Heroes of the Storm
Grubby2118
Khaldor807
Liquid`Hasu503
Other Games
FrodaN3023
Beastyqt789
B2W.Neo693
mouzStarbuck264
ToD172
ArmadaUGS170
QueenE84
Mew2King57
Chillindude13
Organizations
Other Games
PGL1725
gamesdonequick838
StarCraft 2
Blizzard YouTube
StarCraft: Brood War
BSLTrovo
sctven
[ Show 18 non-featured ]
StarCraft 2
• StrangeGG 49
• Reevou 20
• HeavenSC 19
• Adnapsc2 11
• Kozan
• Migwel
• AfreecaTV YouTube
• intothetv
• sooper7s
• IndyKCrew
• LaughNgamezSOOP
StarCraft: Brood War
• STPLYoutube
• ZZZeroYoutube
• BSLYoutube
Dota 2
• WagamamaTV851
• HappyZerGling73
Other Games
• imaqtpie1307
• Shiphtur185
Upcoming Events
BSL 21
18m
StRyKeR vs eOnzErG
Bonyth vs Sziky
Replay Cast
13h 18m
Wardi Open
16h 18m
Monday Night Weeklies
21h 18m
WardiTV Invitational
2 days
Replay Cast
3 days
WardiTV Invitational
3 days
ByuN vs Solar
Clem vs Classic
Cure vs herO
Reynor vs MaxPax
Replay Cast
5 days
Sparkling Tuna Cup
6 days
Liquipedia Results

Completed

Acropolis #4 - TS3
RSL Offline Finals
Kuram Kup

Ongoing

C-Race Season 1
IPSL Winter 2025-26
KCM Race Survival 2025 Season 4
BSL Season 21
Slon Tour Season 2
CSL Season 19: Qualifier 1
META Madness #9
eXTREMESLAND 2025
SL Budapest Major 2025
ESL Impact League Season 8
BLAST Rivals Fall 2025
IEM Chengdu 2025
PGL Masters Bucharest 2025
Thunderpick World Champ.
CS Asia Championships 2025
ESL Pro League S22

Upcoming

CSL Season 19: Qualifier 2
CSL 2025 WINTER (S19)
BSL 21 Non-Korean Championship
Acropolis #4
IPSL Spring 2026
Bellum Gens Elite Stara Zagora 2026
HSC XXVIII
Big Gabe Cup #3
OSC Championship Season 13
Nations Cup 2026
ESL Pro League Season 23
PGL Cluj-Napoca 2026
IEM Kraków 2026
BLAST Bounty Winter 2026
BLAST Bounty Winter 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.