Jump to content


Dispesion

Disperion Accuracy aiming

  • Please log in to reply
20 replies to this topic

shamelbazoqa #1 Posted 04 June 2019 - 02:13 PM

    Lance-corporal

  • Player
  • 7527 battles
  • 61
  • [JRTM] JRTM
  • Member since:
    01-28-2013

Did you know that your tank's gun can be accurate, but have high dispersion!
yes, it is only logical, check out the graph Attached:

Dispersion.png

So this simply tells you: you see the green outer circle? the size of it indicates dispersion, and it means how far can a shot go from the aim point, but it doesn't tell you how often it does!
As you can see, it might just once out of 10 hit at the max dispersion, or maybe hits at the max all the time, or maybe its just something in between.
I hope WOT include an indicator for accuracy in addition to dispersion, so people would stop banging rocks together.

This of course also means the other way around, your gun can have good dispersion but bad accuracy!

That of course is in the case where you have waited until the aim circle cannot go smaller. apparently many tanks has 2 accuracy modes:
1. While the circle is till zooming.
2. After it is fully zoomed.

Some tanks has extreme bad accuracy (and dispersion) while the circle is still zooming, and then suddenly jumps to better accuracy (and dispersion) after the circle fully zoomed, like the Jg.Pz.E100 for an example, if you wait for the circle to fully zoom, you get a somewhat acceptable accuracy, but if you shoot anytime before that... you should expect that your shot will most likely land at the outer part of the circle, at the lower half usually.

So i hope that WOT would include more data in the game about the gun accuracy of tanks, so we'd stop cursing the RNG, and realize the true capabilities of our guns.

ps. a historical note: in average WWII tank guns was twice as accurate as the Grille 15


Edited by shamelbazoqa, 04 June 2019 - 02:21 PM.


jack_timber #2 Posted 04 June 2019 - 02:19 PM

    Lieutenant Сolonel

  • Player
  • 40323 battles
  • 3,249
  • Member since:
    07-26-2014
Would like to see the graph....

fwhaatpiraat #3 Posted 04 June 2019 - 02:20 PM

    Lieutenant

  • Player
  • 58686 battles
  • 1,777
  • [RGT] RGT
  • Member since:
    05-04-2013
The term 'sigma' has been used related to this. Look it up.

shamelbazoqa #4 Posted 04 June 2019 - 02:26 PM

    Lance-corporal

  • Player
  • 7527 battles
  • 61
  • [JRTM] JRTM
  • Member since:
    01-28-2013

View Postjack_timber, on 04 June 2019 - 02:19 PM, said:

Would like to see the graph....

 

it didn't take the first one, i just attached it again, can you see it now?

 

13:27 Added after 1 minute

View Postfwhaatpiraat, on 04 June 2019 - 02:20 PM, said:

The term 'sigma' has been used related to this. Look it up.

 

i have studied statistics, but in this post i am trying to speak in a simple language.

Dava_117 #5 Posted 04 June 2019 - 02:30 PM

    Brigadier

  • Player
  • 22512 battles
  • 4,968
  • [T-D-U] T-D-U
  • Member since:
    12-17-2014
It's well known that in WoT, the gun dispersion is biased toward the center in a gaussian bell curve. Unfortunately the sigma parameter of that curve is not known. We just know is the same for all the tanks.

jack_timber #6 Posted 04 June 2019 - 02:31 PM

    Lieutenant Сolonel

  • Player
  • 40323 battles
  • 3,249
  • Member since:
    07-26-2014

View Postshamelbazoqa, on 04 June 2019 - 01:26 PM, said:

 

it didn't take the first one, i just attached it again, can you see it now?

 

13:27 Added after 1 minute

 

i have studied statistics, but in this post i am trying to speak in a simple language.

 

Yes working now, and yes keep it simple....

shamelbazoqa #7 Posted 04 June 2019 - 02:31 PM

    Lance-corporal

  • Player
  • 7527 battles
  • 61
  • [JRTM] JRTM
  • Member since:
    01-28-2013

View PostDava_117, on 04 June 2019 - 02:30 PM, said:

It's well known that in WoT, the gun dispersion is biased toward the center in a gaussian bell curve. Unfortunately the sigma parameter of that curve is not known. We just know is the same for all the tanks.

 

i assure you it is not the same for all tanks ;)

 



Dava_117 #8 Posted 04 June 2019 - 02:37 PM

    Brigadier

  • Player
  • 22512 battles
  • 4,968
  • [T-D-U] T-D-U
  • Member since:
    12-17-2014

View Postshamelbazoqa, on 04 June 2019 - 02:31 PM, said:

 

i assure you it is not the same for all tanks ;)

 

 

WG said so.

But

*Thin hat on*

It may be they did some errors while coding. In examole my soviet 122mm gun that have 440 alpha tend to roll sub 400 damage a lot. At the same time my 390 alpha 122mm gun roll a lot of 420+ damage.

*Thin hat off*

 

Uff. Sometime I need to vent too. ;)


Edited by Dava_117, 04 June 2019 - 02:38 PM.


Koriin #9 Posted 04 June 2019 - 02:41 PM

    Sergeant

  • Player
  • 14339 battles
  • 224
  • Member since:
    11-02-2012

View Postshamelbazoqa, on 04 June 2019 - 02:31 PM, said:

 

i assure you it is not the same for all tanks ;)

 

 

Always thought the same, but never found any numbers supporting this.

I am quite interrested in the curves used and how the dispersion value is used and especially how the value adapts to the given values for driving/turret rotation and tank rotation.

 

While im wearing my tin foil hat, i always think tanks seems more biased to the centre of the rectangle while driving/rotating the turret, but shells fly all over the place when you stop and aim for a little (so no fully aimed shots).

But those are just tin foil hat theories...



XxKuzkina_MatxX #10 Posted 04 June 2019 - 02:51 PM

    Major General

  • Player
  • 53231 battles
  • 5,350
  • Member since:
    04-02-2016

My problem with the final accuracy in this game is that at most engagement ranges, the difference between an accurate and inaccurate guns is very little.

 

This table represents the chances to hit different sized targets (1.5m radius, 1m radius and 0.3m radius) with different dispersion at multiple distances (100m~600m).

 

Spoiler

 

Short story: The developers don't like snipers so accuracy is of secondary importance and it means very little in this game.



stokerel #11 Posted 04 June 2019 - 03:00 PM

    Sergeant

  • Player
  • 14389 battles
  • 208
  • Member since:
    05-25-2012

View Postshamelbazoqa, on 04 June 2019 - 01:31 PM, said:

 

i assure you it is not the same for all tanks ;)

 

 

It's most probably not, regardless how much WG state otherwise.

What's also true is that the accuracy Bell curve got significantly 'flattened' a while back, can't remember which patch. This most affected the accurate German guns but also derp guns in general.



shamelbazoqa #12 Posted 04 June 2019 - 03:04 PM

    Lance-corporal

  • Player
  • 7527 battles
  • 61
  • [JRTM] JRTM
  • Member since:
    01-28-2013

View PostXxKuzkina_MatxX, on 04 June 2019 - 02:51 PM, said:

My problem with the final accuracy in this game is that at most engagement ranges, the difference between an accurate and inaccurate guns is very little.

 

This table represents the chances to hit different sized targets (1.5m radius, 1m radius and 0.3m radius) with different dispersion at multiple distances (100m~600m).

 

Spoiler

 

Short story: The developers don't like snipers so accuracy is of secondary importance and it means very little in this game.

 

0.3m radius = 60 cm target. that is a big target, like a flat plate, most targets that are in the "hard to hit" category in the game are (from the top fo my head) 25cm in diameter and below, that means .125 radius.. make your calculation again ;)

 



XxKuzkina_MatxX #13 Posted 04 June 2019 - 03:11 PM

    Major General

  • Player
  • 53231 battles
  • 5,350
  • Member since:
    04-02-2016

View Postshamelbazoqa, on 04 June 2019 - 06:04 PM, said:

0.3m radius = 60 cm target. that is a big target, like a flat plate, most targets that are in the "hard to hit" category in the game are (from the top fo my head) 25cm in diameter and below, that means .125 radius.. make your calculation again

 

I didn't make these calculations myself but i trust the author and i did verify them.

 

0.3m is an arbitrary number based on tank dimensions like a part of a lower plate or a cupola. The ratios are still correct, doesn't matter what size your target is!



Dorander #14 Posted 04 June 2019 - 03:13 PM

    Lieutenant General

  • Player
  • 21025 battles
  • 6,041
  • Member since:
    05-07-2012

Your actual accuracy of a shot is a function of dispersion, bloom and aim-time, as well as the distribution of shots within that circle. The dispersion value is a value per 100m and IIRC determines the minimum size of the circle, not the maximum size. Bloom and aim speed determine the maximum size of the circle which will expand upon movement, moreso on tank movement than turret movement. The Jageroo has some fairly large bloom values last I checked so moving the thing will increase the size of your aiming circle quite a bit.

 

Also keep in mind that client reticle and server reticle are not identical so if you fire too early, the shot will land anywhere in an even larger circle than you see, and the normal distribution points are further from the center than when you are fully aimed. It's not a matter of two formulas or two accuracies, it's just a matter of how the different mechanics interact with each other. Long story short: aim properly before you fire.



Nexuss9383 #15 Posted 04 June 2019 - 03:18 PM

    Private

  • Player
  • 17648 battles
  • 46
  • [USAGI] USAGI
  • Member since:
    03-29-2012
Who needs numbers and graphs when the hand of Stalin guides my shells when you play russia tank

Edited by Nexuss9383, 04 June 2019 - 03:19 PM.


fwhaatpiraat #16 Posted 04 June 2019 - 03:26 PM

    Lieutenant

  • Player
  • 58686 battles
  • 1,777
  • [RGT] RGT
  • Member since:
    05-04-2013

View Postshamelbazoqa, on 04 June 2019 - 02:26 PM, said:

 

it didn't take the first one, i just attached it again, can you see it now?

 

13:27 Added after 1 minute

 

i have studied statistics, but in this post i am trying to speak in a simple language.

Google: "wot dispersion sigma".



shamelbazoqa #17 Posted 04 June 2019 - 03:40 PM

    Lance-corporal

  • Player
  • 7527 battles
  • 61
  • [JRTM] JRTM
  • Member since:
    01-28-2013

View PostDorander, on 04 June 2019 - 03:13 PM, said:

Your actual accuracy of a shot is a function of dispersion, bloom and aim-time, as well as the distribution of shots within that circle. The dispersion value is a value per 100m and IIRC determines the minimum size of the circle, not the maximum size. Bloom and aim speed determine the maximum size of the circle which will expand upon movement, moreso on tank movement than turret movement. The Jageroo has some fairly large bloom values last I checked so moving the thing will increase the size of your aiming circle quite a bit.

 

Also keep in mind that client reticle and server reticle are not identical so if you fire too early, the shot will land anywhere in an even larger circle than you see, and the normal distribution points are further from the center than when you are fully aimed. It's not a matter of two formulas or two accuracies, it's just a matter of how the different mechanics interact with each other. Long story short: aim properly before you fire.

 

watch-out people here comes a WG employee XD



Dorander #18 Posted 04 June 2019 - 04:28 PM

    Lieutenant General

  • Player
  • 21025 battles
  • 6,041
  • Member since:
    05-07-2012

View Postshamelbazoqa, on 04 June 2019 - 02:40 PM, said:

 

watch-out people here comes a WG employee XD

 

Where?

 

I'm talking about mechanics as far as I know them, feel free to point out where I made a mistake and to back up the claim.



burbage1 #19 Posted 04 June 2019 - 04:40 PM

    Corporal

  • Player
  • 994 battles
  • 129
  • Member since:
    09-09-2017
It's an arcade game. With .37 dispersion I can hit a PzIC with only the cupola to aim at on the hill in Himmlesdorf from across the map, at pretty much the same rate I hit the PZIC when I have the whole tank to aim at at 100m. As far as I can see it rolls you either a hit or a miss, and then applies the dispersion to the hit (or concentration, as it should be called, as it is weighted to the centre of the aim point, not the edge). But in the end, if it was too hard to hit someone behind cover they'd either have to remove the cover, or lengthen the game.

arthurwellsley #20 Posted 04 June 2019 - 05:28 PM

    Brigadier

  • Player
  • 54060 battles
  • 4,018
  • [-B-C-] -B-C-
  • Member since:
    05-11-2011

View Postshamelbazoqa, on 04 June 2019 - 01:13 PM, said:

Did you know that your tank's gun can be accurate, but have high dispersion!
yes, it is only logical, check out the graph Attached:

Dispersion.png

So this simply tells you: you see the green outer circle? the size of it indicates dispersion, and it means how far can a shot go from the aim point, but it doesn't tell you how often it does!
As you can see, it might just once out of 10 hit at the max dispersion, or maybe hits at the max all the time, or maybe its just something in between.
I hope WOT include an indicator for accuracy in addition to dispersion, so people would stop banging rocks together.

This of course also means the other way around, your gun can have good dispersion but bad accuracy!

That of course is in the case where you have waited until the aim circle cannot go smaller. apparently many tanks has 2 accuracy modes:
1. While the circle is till zooming.
2. After it is fully zoomed.

Some tanks has extreme bad accuracy (and dispersion) while the circle is still zooming, and then suddenly jumps to better accuracy (and dispersion) after the circle fully zoomed, like the Jg.Pz.E100 for an example, if you wait for the circle to fully zoom, you get a somewhat acceptable accuracy, but if you shoot anytime before that... you should expect that your shot will most likely land at the outer part of the circle, at the lower half usually.

So i hope that WOT would include more data in the game about the gun accuracy of tanks, so we'd stop cursing the RNG, and realize the true capabilities of our guns.

ps. a historical note: in average WWII tank guns was twice as accurate as the Grille 15

 

Author - brumbarr (sadly now no longer playing - so none of this is mine) did the following work and posted it back in November 2017;

 

After a thread on the forums about  aimtime and how it works. I decided I wanted to figure it out exactly. All we know for now is that aim time is the time it take to reduce the aimcircle to 40% its size, dispersion is something that  say show much the circle gets bigger and accuracy is the size of the aimingcircle when fully aimed. However, we do not know  the exact relations between these 2 and how exactly they  influence the size of the aiming circle at all times.

So thats what I set out to do, finding a mathmatical description of the size of the aimingcircle. The method is simple: measure the size of the aimingcircle for different tanks and speeds in a trainingroom.   Thanks for uglycousin (still playing I think) for giving me a second person to set up the trainingroom.

To measure the size of the circle, first I  did the test driving in the room. Then I watched the replays and paused at certain moments. I then took a screenshot of  my whole screen, making sure I was always in 8x zoom. Then I took those screenshots into paint and measured the circle diameter in pixels.

I will now describe the process and results of my investigation. But if you dont want to read that, scroll down to the conclusion at the bottom.

Disclaimer: the following formulas are NOT what WG uses, I made a linear  model that describes the size of the aimingcircle as close as possible. 

Aiming circle bloom

 

I assumed there where 3 variables that had an influence on bloom: speed, dispersion and accuray. I tried to do test in which I held 2 variables constant to  see the influence of 1.

 

 I started  with gathering data of 2 different dispersion numbers for which I picked 4 tanks with different accuracy and measured the size at each speedincrease of 10 untill 50 kph.

 

These are the raw results:

 

 

lucGfBq.png

 

 

 

From that I made a graph of the dispersion in function of speed , and calculated the gradient  of the graph assuming linear increase.  Then obviously  the  aiming circle size = C*v+accuracy.

 

With V=speed and C being the gradient, which consist of unknown  factors. To check the linear approach was decent I plotted the model and experiment:

 

ry3mSBe.png

 

As you can see the linear approach to the speed factor isnt perfect but not massivly different, only in the middle it differs.  I am happy enough with this.

 

Now we need to determine what the C factor consist off. Since there are only 2 variables left, it has to consist dispersion or/and accuracy components.

 

 

 

As you can see in the data, with the same dispersion numbers, the aiming circle fort he same speed is bigger when the accuracy is bigger.  So there has to be an accuracy factor in C,  which is proportional to accuracy.   

 

Here you can see accuracy vs circle size:

 

5ZITQUm.png

 

As you can see, the increase isnt marginal. We can now rewrite our formula as:

 

Size=Acc(D*v+1)

 

With D an unknown factor containing dispersion in some form.  As we can see, size of aiming circle is directly proportional to accuracy. So an increase in accuracy of 25% will results in 25% better gun handling.  This is why the E50/E50M have such amazing gun handling , their dispersion isn't great , but good, but due to the very good accuracy their gun handling is much better than at first glance.  The WZ-132-1 has the exact same dispersion values, so you would think the gun handling would bet he same, but no, since it has 33% worse accuracy is will have 33% worse gun handling, which is massive! Thats more than a vstab! 

 

Next task is determining the factor D. The only  variable left is dispersion, so I tested different tanks with differnt dispersion at the same speed, their accuracy was different, but thats fine, since we can normalise for that.    These numbers showed that the factor D was proportional to the dispersion values, so D=c*dispersion, with c an unknown constant.

 

Now the formula looks like this:

 

S=Acc(c*d*v+1)

 

Determining c was done by plotting the experimental result and  trying some numbers until the model best fits the experiment. I took c=0.68.

 

The influence of dispersion can be see in this graph:

 

xK8nmDx.png

Now we have a formula that gives a perfect description of aiming circle size in function of all variables.  Next up is determing  the time it takes for the circle to shrink, or the actuall aiming time for the tank.

 

 

Aiming time:

 

We know aiming time is the time it takes fort he circle to shrink by 60% its starting size.

 

 So we can write:

 

S1=S2*(4/10)^(t/T) with T=aiming time, S1 size after time t, S2= starting size.

 

Solving this for t we get: t=T*(log(S1/S2)/log(4/10)).

 

 

We can now determine the time it take from any speed to reach any size we want.

 

To determe the time it takes to fully aim, jsut replace S1 by the accuracy of the gun. Note this time is independant of accuracy! ( which is logical, since it needs to go to  a smaller circle but also does it faster, these 2 cancel out)

 

Plotting this for 3 different tank in fucntion of time comming to a stop from a speed of 50 (40 for conway) we get:

 

kJEdGx6.png

 

Influence of equipment/skills etc.

 

Now that we have every formula we need we can quantify the influence of equipment/skills/directives/modules. To do so simply multiply the variable that gets influenced by (1-0,01*improvement in %). Dispersion values only get influenced by vstabs and the smooth ride skill.Other equipment only influences the accuracy value.

 

 Note that the same improvement to acc or dispersion results in a bigger improvement in size for what improves acc than what improves dispersion. Vstabs for example do not make the size of the circle shrink by 20%, they make the increase in size  less by 20%.

 

Lets take a look at a common dillema:vstabs vs gun laying drive, lets try this on 2 different tanks:

 

 

Uz6R9iQ.png

zfC2SPp.png

 

We can clearly see what the difference in vstab and gld is, vstab makes the circle smaller, so you start smaller but the decrease is still the same, gld starts at a bigger size but then starts to decrease faster, catching up to the vstabs. In the BCs case, the time to fully aim is actually lower when equiping gld than when equiping vstabs.

 

Mathematicaly, gld decreases the total time to aim by 10%,  whereas vstabs decreases the total time to aim by subtracting 20% *initial size. To know whether vstabs or gld is better depends on the tank and how much you want to aim, you can determine this by pluggin in the numbers and plotting it for each vehicle, since it will be different for each.

 

As general rules however, these apply:

 

·      - Bad dispersion + bad aimtime:

 

 Vstab better, unless you  fully aim from full speed.

 

·     -  Bad dispersion + good aimtime

 

:Vstab better, unless at high speed when fully aiming.

 

·       -Good dispersion + bad aimtime

 

:Vstabs always superior

 

·       -Good aimtime + good dispersion

 

:Vstabs always superior

 

 

 

 

 

Conclusion and TLDR:

 

·      - Accuracy has a massive  influence on aiming circle size on the move, they are proportional.

 

·      - Aiming circle size is proportional to speed/dispersion.

 

·       -To determine what gun has better actual gun handling: multply accuracy with dispersion, the lower the numbers the better the gun handling.

 

·     -  Size of aiming circle= Acc(0.68*d*v+1)

 

·      - Time to fully aim = Aiming time*(-log(0.68*d*v+1)/log(4/10))

 

·      - Vstabs is superior to gld in most situations.

 

·     -  Influencing accuracy gives a better boost than influencing dispersion values.


Edited by arthurwellsley, 04 June 2019 - 05:35 PM.






Also tagged with Disperion, Accuracy, aiming

1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users