Archived

This topic is now archived and is closed to further replies.

-Define-

Inaccurate Damage Numbers? Let's find the Solution!

Recommended Posts

Hey there,

i started playing Warframe 3 weeks ago and finally decided to take a deep look into damage numbers etc so i am able to do calculations and theorycrafting myself.

Now upon reading up on damage 2.0 in the Wiki, I started doing calculations in an Google Docs Sheet and am kinda confused since the numbers dont add up.

Example A:

Spoiler

A Vectis Prime (unmodded):

qPfNAjd.png

_______________________________________________________________________________________

 

The Enemy is an Infested Charger lvl 85:

uQSK7V9.png

His Health type is "Infested", which means "Impact" and "Puncture" are unaffected and "Slash" damage is buffed by 25%

The Charger doesn't have Armor thus there should be no damage reduction.

The expected Damage (Actual Dmg) is:

yRGP45f.png

_______________________________________________________________________________________

 

 

Now I tested it in the Simulacrum and got the following result consistently:

uYmo0cP.png

 

I did more complex calculations regarding armor first and the outcome was even more scewed so I figured to do it the simplest way possible.

 

Example B:

Spoiler

Despair(unmodded):

lketWp2.png

_______________________________________________________________________________________

The Enemy is a Corrupted Heavy Gunner, who has:

  • Cloned Flesh (-25% Impact, 0% Puncture, +25% Slash)
  • Ferrite Armor (0% Impact, +50% Puncture, -15% Slash)

The Scaled Armor to LvL 85:

uUviOUy.png

_______________________________________________________________________________________

 

Now we apply the formula of the Damage Modifiers from the Wiki latest?cb=20151018085648

to every Damage Type with the respective Health and Armor Modifiers:

 

Impact:

H5oVRC3.png

 

Puncture:

8bDh75l.png

 

Slash:

Cqy6OOD.png

 

this results in an Expected Damage of:

EsPM0zv.png

__________________________________________________________________________________

 

Now i went to the Simulacrum and tested it and the result was off again, this time higher than the expected Damage:

UbikBzh.png

_______________________________________________________________________________________

 

The tested and expected Damage on Armored Targets changes with levels, since the Armor Rating (AR) scales with enemy level and is used to determine the damage modifiers.

Now since im new i might be missing something, but the numbers dont add up and i dont know where i could have made a mistake in this simple equasion.

It isn't Falloff Damage as well, since im literally point blank on the target.

This is the Link to the Sheet, in case someone wants to look into the Formulas:

https://docs.google.com/spreadsheets/d/1kH4LG0utLG0_bNayMFceI9crZRGBE4vXtnyoGqx6NAw/edit?usp=sharing

 

People might say isnt a big difference.

The thing is, this is unmodded Damage. So since im trying to ultimatively do more complex calculations with killtimers and comparing Weapons,

the basics need to be accurate, otherwise once scaling hits, the gap between Expected and Actual Damage widens exponentionally.

________________________________________________________________________________________

 

The Results and Solutions found in this Thread which are accurate and reproducable:

1. Submitted by (Xbox One)R3d P01nt : Ingame Physical Damage Values are Calculated as follows:

 

Ingame Value: ROUNDDOWN((I+P+S)*(1+Modded Multiplier))*Portion of Damage

 

Example:

Soma Prime (1,2 I | 4,8 P | 6 S) = 12 Total Physical DMG      Modded Multiplier: R0 Serration = 15%

Portion of     (10% | 40%  | 50%)= 100%

for Impact: ROUNDDOWN((1,2+4,8+6)*(1+0,15))*0,1 = ROUNDDOWN(12*1,15)*0,1 = ROUNDDOWN(13,8)*0,1 = 13*0,1 = 1,3

 

________________________________________________________________________________________

I'd appreciate any form of help despite as insignificant it may seem, it could bring us a step closer to figuring this out.

The more people participate the better, since different point of views will result in us eventually finding the answers faster together!

Share this post


Link to post
Share on other sites

Sorry for the late reply, i just got back home.

The Damage against unarmored enemys isn't affected by level, since they dont have any form of mitigation. That being said i tested it and the result was what I expected.

Thank you for the suggestion tho, every form of help is appreciated!

Gpyblwo.png

Share this post


Link to post
Share on other sites
7 hours ago, -Define- said:

Hey there,

i started playing Warframe 3 weeks ago and finally decided to take a deep look into damage numbers etc so i am able to do calculations and theorycrafting myself.

Now upon reading up on damage 2.0 in the Wiki, I started doing calculations in an Google Docs Sheet and am kinda confused since the numbers dont add up.

Example A:

  Hide contents

A Vectis Prime (unmodded):

qPfNAjd.png

_______________________________________________________________________________________

 

The Enemy is an Infested Charger lvl 85:

uQSK7V9.png

His Health type is "Infested", which means "Impact" and "Puncture" are unaffected and "Slash" damage is buffed by 25%

The Charger doesn't have Armor thus there should be no damage reduction.

The expected Damage (Actual Dmg) is:

yRGP45f.png

_______________________________________________________________________________________

 

 

Now I tested it in the Simulacrum and got the following result consistently:

uYmo0cP.png

 

I did more complex calculations regarding armor first and the outcome was even more scewed so I figured to do it the simplest way possible.

 

Example B:

  Reveal hidden contents

Again the Vectis Prime(unmodded):

qPfNAjd.png

_______________________________________________________________________________________

The Enemy is a Corrupted Heavy Gunner, who has:

  • Cloned Flesh (-25% Impact, 0% Puncture, +25% Slash)
  • Ferrite Armor (0% Impact, +50% Puncture, -15% Slash)

The Scaled Armor to LvL 85:

mf6SYxb.png

_______________________________________________________________________________________

 

Now we apply the formula of the Damage Modifiers from the Wiki latest?cb=20151018085648

to every Damage Type with the respective Health and Armor Modifiers:

 

Impact:

W178xTv.png

 

Puncture:

7VwZO7Q.png

 

Slash:

BbGZA84.png

 

this results in an Expected Damage of:

03RgzCQ.png

__________________________________________________________________________________

 

Now i went to the Simulacrum and tested it and the result was off again, this time higher than the expected Damage:

WPnXvkR.png

_______________________________________________________________________________________

 

The tested and expected Damage changes on armored Targets with  levels, since the Armor Rating (AR) Scales with enemy level and is used to determine the damage modifiers.

Now since im new i might be missing something, but the numbers dont add up and i dont know where i could have made a mistake in this simple equasion.

It isn't Falloff Damage as well, since im literally point blank on the target.

 

In Example B it is 28 instead of 25,47 Damage, which people might say isnt a big difference.

 

The thing is, this is unmodded Damage. So since im trying to ultimatively do more complex calculations with killtimers and comparing Weapons, the basics need to be accurate,

otherwise once scaling hits, the gap between Expected and Actual Damage widens exponentionally.

 

I'd appreciate any form of help, since I would love to do further calculations, but can't really continue since im stuck on this very essencial and basic part.

I was so confused cause you where writing the decimal points as commas. Please fix that, it's very annoying and very confusing. Decimal point should always be a dot   . 

 

As for the topic, it could be a rounding error on DEs side? Like their damage isn't being rounded correctly or maybe it's designed to round up or down depending on the circumstances

Share this post


Link to post
Share on other sites

Since im from Germany my google docs doesnt accept the Decimal Point as anything but a comma sadly, so I stuck to it to be consistent throughout the post to avoid confusion.

Sorry if that confused you :/

PS: Do you have any thoughts on the difference between the Calculation and the Tests?

Share this post


Link to post
Share on other sites

Im not an expert on damage calculations so im not sure if this even affects the overall calculations, but the base level of a corrupted heavy gunner isnt 1, its 8. Maybe that screws with the armor calculation?

Share this post


Link to post
Share on other sites

Will look into it tonight when i finish work.

At a glance, your unarmoured calculations seem legit but will take a closer look tonight. As for the armoured calc, i think you might have a couple of errors.

I also suspect that part of the discrepancy might have to do with you being a "client" in the simulacrum.

But as i've said, will investigate tonight. Talk later.

 

ps.

3 weeks into the game and already looking into dmg numbers... I like that.

A tenno like you, we'd like to have in our clan.

Share this post


Link to post
Share on other sites
8 hours ago, TheRealArmadillo said:

Im not an expert on damage calculations so im not sure if this even affects the overall calculations, but the base level of a corrupted heavy gunner isnt 1, its 8. Maybe that screws with the armor calculation?

Thank you alot, i can't believe i forgot to change the base Level back to 8 after the charger calculations.

 

The updated Heavy Gunner Calculation for the Vectis Prime:

chOjJIH.png

Looking at this makes me wonder if the game is rounding or not, since it outputs 28 ingame, and the expected numer is ~28,95.

I will do some testing with other weapons to see if the same ruleset applies.

 

EDIT: I tested it with "Despair" (Unmodded) and the results are sadly off again despite having corrected the Base Level in the calculations. The result is in the main Post.

 

The Charger Calculation (Example A) still is a mystery for me tho ... any thoughts on it? 

Share this post


Link to post
Share on other sites

Sry, for taking a long time to reply.

First of all, as you may have already observed, warframe rounds off your damage by simply striking off the decimals. So completely ignore any number after the decimal.

As far as your calculations on armour they were indeed wrong because of not taking into account the base level of the h.gunner.

But when it comes to the charger calculation, I'm baffled too. Your calculations are correct. I really don't know why it's showing you 316 instead of 337. I went into the simulacrum and in a mission and reproduced the 316 myself, so i rule out any simulacrum bugs or auras etc. There is a 6% discrepancy give or take. I'll test a few more weapons and post again.

 

 

EDIT:

Braton Test

It gives dmg of 20 on a charger, so the display does not include the +25% on slash. It should give 22 dmg if we dont include any decimal.

Braton stats:

6.6 Impact

6.6 Puncture

8.5 Slash

Total: 20

 

Zenith Test

It gives dmg of 33 on a charger which is what we get if we do not include any decimal in our calculation.

Zenith stats:

4.5 Impact

6 Puncture

19.5 Slash

Total: 30

 

Soma Prime Test

It gives 14 dmg on a charger which is what we get if we include decimals in our calculation.

1.2 Impact

4.8 Puncture

6 Slash           x1.25 =7.5

Total: 12         Total = 13.5 = 14(roundoff)

Share this post


Link to post
Share on other sites
Spoiler
2 hours ago, disco_inferno6 said:

Sry, for taking a long time to reply.

First of all, as you may have already observed, warframe rounds off your damage by simply striking off the decimals. So completely ignore any number after the decimal.

As far as your calculations on armour they were indeed wrong because of not taking into account the base level of the h.gunner.

But when it comes to the charger calculation, I'm baffled too. Your calculations are correct. I really don't know why it's showing you 316 instead of 337. I went into the simulacrum and in a mission and reproduced the 316 myself, so i rule out any simulacrum bugs or auras etc. There is a 6% discrepancy give or take. I'll test a few more weapons and post again.

 

 

EDIT:

Braton Test

It gives dmg of 20 on a charger, so the display does not include the +25% on slash. It should give 22 dmg if we dont include any decimal.

Braton stats:

6.6 Impact

6.6 Puncture

8.5 Slash

Total: 20

 

Zenith Test

It gives dmg of 33 on a charger which is what we get if we do not include any decimal in our calculation.

Zenith stats:

4.5 Impact

6 Puncture

19.5 Slash

Total: 30

 

Soma Prime Test

It gives 14 dmg on a charger which is what we get if we include decimals in our calculation.

1.2 Impact

4.8 Puncture

6 Slash           x1.25 =7.5

Total: 12         Total = 13.5 = 14(roundoff)

 

Ty so much for your contribution!

 

Thanks to you being able to reproduce the numbers i can rule out buggs with character etc.

The damage Calculation seems totally random:

for 1 Weapon not taking into account the Damage multiplier from Flesh,

for another weapon not counting in decimals while counting in decimals for another?

This makes an actual calculation for a comparison impossible or extremely complicated.

I wonder if its actually that random or if we are missing another seperate element of the equasion, which isn't yet displayed on the wiki.

 

I wish there was a bit more transparency from DE's side to give us some formulas or at least a general idea how the damage is calculated differently between weapons.

Share this post


Link to post
Share on other sites

Ty so much for your contribution!

 

Thanks to you being able to reproduce the numbers i can rule out buggs with character etc.

The damage Calculation seems totally random:

for 1 Weapon not taking into account the Damage multiplier from Flesh,

for another weapon not counting in decimals while counting in decimals for another?

This makes an actual calculation for a comparison impossible or extremely complicated.

I wonder if its actually that random or if we are missing another seperate element of the equasion, which isn't yet displayed on the wiki.

 

I wish there was a bit more transparency from DE's side to give us some formulas or at least a general idea how the damage is calculated differently between weapons.

 

No worries. We'll get to the bottom of this.

Share this post


Link to post
Share on other sites
On 11/2/2017 at 6:41 PM, Arniox said:

I was so confused cause you where writing the decimal points as commas. Please fix that, it's very annoying and very confusing. Decimal point should always be a dot   . 

Yeah, this is a European (and South American).  I don't know how decimal points became commas and commas became decimal points over there (or vice versa over here), but they did, and it's intensely frustrating.  Something as exacting as mathematics should at least have a standard notation for something as basic as the decimal point.  I mean, it's like some countries using + for multiplication and * for addition.
 

https://en.wikipedia.org/wiki/Decimal_mark#Hindu.E2.80.93Arabic_numeral_system

Nice map on the wikipedia page.  While the comma seems more widespread by land mass (friggin' Russia), India and China are incredibly populous, it wouldn't at all surprise me if the period were more common by usage.

Share this post


Link to post
Share on other sites
1 minute ago, PhotriusPyrelus said:

Yeah, this is a European (and South American).  I don't know how decimal points became commas and commas became decimal points over there (or vice versa over here), but they did, and it's intensely frustrating.  Something as exacting as mathematics should at least have a standard notation for something as basic as the decimal point.  I mean, it's like some countries using + for multiplication and * for addition.
 

https://en.wikipedia.org/wiki/Decimal_mark#Hindu.E2.80.93Arabic_numeral_system

Nice map on the wikipedia page.  While the comma seems more widespread by land mass (friggin' Russia), India and China are incredibly populous, it wouldn't at all surprise me if the period were more common by usage.

To me, commas have always separated two sets of numbers. Like thousands from hundreds, millions from billions and so on. 

So If I say 190,456,590,289,139,495,000.90

I know that, that is one hundred and ninety quintillion, four hundred and fifty six quadrillion, five hundred and ninety trillion, two hundred and eighty nine billion, one hundred and thirty nine million, four hundred and ninety five thousand point nine.

But how would you easily see the difference when using commas for the decimal point? Do you just have the two combined and you have to guess where the full numbers are? Or do you have a separation? Or do you just no have the comma separators and it's just a long string of numbers that's hard to read? 

Share this post


Link to post
Share on other sites
Just now, Arniox said:

To me, commas have always separated two sets of numbers. Like thousands from hundreds, millions from billions and so on. 

So If I say 190,456,590,289,139,495,000.90

I know that, that is one hundred and ninety quintillion, four hundred and fifty six quadrillion, five hundred and ninety trillion, two hundred and eighty nine billion, one hundred and thirty nine million, four hundred and ninety five thousand point nine.

But how would you easily see the difference when using commas for the decimal point? Do you just have the two combined and you have to guess where the full numbers are? Or do you have a separation? Or do you just no have the comma separators and it's just a long string of numbers that's hard to read? 

Fun fact: that's not a mathematically accepted form (at least in professional mathematics).  You would use spaces instead of commas (190 456 590 289 139 495 000.90) or (190 456 590 289 139 495 000,90).  I agree commas make at least a little it easier to read than spaces, and I suspect the reason they are standardized to be spaces instead of commas has to do with the commas-as-decimal-point people.

Again, something as precise and well-defined as mathematics should be standardized.  One way or the other, none of this "Oh, just use whatever you want, comma or decimal point".  And while I would grumble if it was decided to use the comma (China, India, and the US alone make up 3 billion of the 7.5 billion people on the planet), at least it would be standard.

Share this post


Link to post
Share on other sites

All right, this thread wasn't created to discuss which decimal separation is the correct one.

jjpdnArnioxPhotriusPyrelus stay on topic pls or create your own Thread if you need to.

If you have nothing productive to say about the actual topic, please refrain from posting in this thread.

Share this post


Link to post
Share on other sites
On 11/2/2017 at 6:13 PM, -Define- said:

The Damage against unarmored enemys isn't affected by level, since they dont have any form of mitigation. That being said i tested it and the result was what I expected.

Yeah, I was wondering if DE snuck in some sort of mitigation by level (or bugged it in).

In example A you are using a Vectis, correct?  Since you are not zooming, you get an accuracy penalty.  I wonder if DE also put a damage penalty on it?  If you zoom in, does it correct the discrepancy?

Share this post


Link to post
Share on other sites
7 minutes ago, (Xbox One)R3d P01nt said:

Yeah, I was wondering if DE snuck in some sort of mitigation by level (or bugged it in).

In example A you are using a Vectis, correct?  Since you are not zooming, you get an accuracy penalty.  I wonder if DE also put a damage penalty on it?  If you zoom in, does it correct the discrepancy?

 

Thats a good idea, i went into the simulacrum and tested it:

u1mRyqa.png

 

"Sadly" wasnt the case.

Share this post


Link to post
Share on other sites
1 minute ago, -Define- said:

 

Thats a good idea, i went into the simulacrum and tested it:

u1mRyqa.png

 

"Sadly" wasnt the case.

That was fast!

Maybe it's a Vectis problem, like a typo in the damage numbers displayed in the UI?  Other weapons seem to give the correct numbers, minus the bonus from slash damage, correct?  (As shown by @disco_inferno6.)

Share this post


Link to post
Share on other sites
39 minutes ago, (Xbox One)R3d P01nt said:

That was fast!

Maybe it's a Vectis problem, like a typo in the damage numbers displayed in the UI?  Other weapons seem to give the correct numbers, minus the bonus from slash damage, correct?  (As shown by @disco_inferno6.)

At the moment I don't see a consistent method of how the damage is calculated.

I just tested another Weapon to see if the statement would be correct:

Braton Prime (Unmodded):

Spoiler

______________________________________________________________________

Unmodded Stats:

1wI3yRw.png

 

______________________________________________________________________

Expected DMG:

jcjTdRf.png

 

______________________________________________________________________

Actual DMG:

jAF8J9D.png

 

So if we would ignore the Slash Multiplier, the difference would be even bigger.

 

Now an interesting Fact here is, if we took the Total Raw Damage of 35,1 and multiplied it with the Slash Multiplier of 1.25 the result is 43,875.

Disregarding Decimals completely would net us an expected Damage Value of 43.

 

Logically this doesn't make sence tho, why would Impact and Puncture Damage be affected by the Slash damage Multiplier.

Share this post


Link to post
Share on other sites
1 hour ago, -Define- said:

Now an interesting Fact here is, if we took the Total Raw Damage of 35,1 and multiplied it with the Slash Multiplier of 1.25 the result is 43,875.

Disregarding Decimals completely would net us an expected Damage Value of 43.

 

Logically this doesn't make sence tho, why would Impact and Puncture Damage be affected by the Slash damage Multiplier.

Yeah, I'm totally confused now.

Share this post


Link to post
Share on other sites

There might be different Calculations between weapon types like "Sniper" and "Assault Rifle".

Im looking deeper into thee Braton Prime right now and try to determine the exact way this weapon is calculated. I approach the rounding in multiple ways and look for changes between different  elements and their multipliers.

I will go into much more detail once i figured out an approach which can reliably reproduce a correct outcome.

 

Share this post


Link to post
Share on other sites
20 hours ago, -Define- said:

wish there was a bit more transparency from DE's side

Why bother with rigorousness and consistency when one can get away with absolute lack of transparency? I took YEARS of players requests for DE to add a few additional basic weapon stats in the arsenal UI. Some weapons, like the mutalist quanta, still have a completely erratic bevior, despite time consuming reviews and feedback from players. Most have gave up out of frustration. In fact, i'm pretty convinced DE themsleves aren't quite sure how their own game works XD

Your efforts are commendable, though.

Share this post


Link to post
Share on other sites