Jump to content


Some Artillery Dispersion Observations (Wall of text and math)

Artillery Accuracy Dispersion

  • This topic is locked This topic is locked
18 replies to this topic

chri105n #1 Posted 09 July 2019 - 06:09 PM

    Private

  • Player
  • 21958 battles
  • 39
  • Member since:
    07-05-2011

Hello,

 

After a particularly frustrating round as arty where it felt like all my shots were missing, I decided to do a few calculations. Using the in game range indicator i found that some of my shells were missing by as much as 10-11 meters at ranges of around 700 meters. Which really doesn't add up with a dispersion of 0.63 (Crusader SP). Now my first inclination was that due to the high arc the flight path might be long enough to cause this deviation. However, this would imply a flight path of around 1.5 kilometers, which is impossible since the impact angle is quite shallow. My next step was to assume that the aiming circle was a projection of a hypothetical circle which is placed at an angle equal to the impact angle, and then projected unto the ground. The dispersion given in the game is then based upon this hypothetical circle (Many of you might already know this). Using a bit of trigonometry I calculated what the actual dispersion is, i.e. what is the dispersion values of the projected aiming circle? And the numbers are quite shocking. First of all, a few approximations:


1. The angle at which the shell exits the gun barrel is equal to the angle of impact.

2. WG have no extra hidden stats affecting fully aimed shots from a still position.

3. There is no elevation difference between firing position and the target.

and finally the big one

4. The Shell travel distance doesn't impact the final dispersion values, i.e. the beeline distance to the target is the distance used to calculate final maximum deviation not distance of the shell travel arc. This would affect the results a bit, but not so much that we cannot use this approximation, just keep in mind that dispersion values for high arcing shells are a bit lower than what they otherwise would be.

 

Just in case the images fails to upload:

Here is a graph of the data: link

Here is the data used: link

 

Using the Crusader SP as a bench mark, at maximum range (1288) meters the gun is elevated around ca. 35 degrees. Now, my crusader has an dispersion value of 0.63 which means that at a range of 700 meters and assuming an impact angle of around 20 (based on in-game observations), if my calculations are at least slightly accurate, my Crusader SP's actual dispersion value is then 1.84. This is quite a bit WORSE than the apparent dispersion (1.08) of the Conqueror Gun Carriage. Now I don't have the CGC so I can't test this, but let us make a generous assumption an impact angle of 35 degrees at 700 meters, this gives the CGC an actual dispersion of 1.88... wait, so because of the higher arc, the CGC has basically the same actual dispersion as the Crusader SP? I would appear that this is indeed the case. Now what about the FV304 (Bert), at its maximum range of 499 meters, Bert fires at an angle of 45 degrees, which gives it an actual accuracy of 0.89. Which is only a bit worse than the base 0.6 of the KV-2.

 

Now what does all of this mean, basically, if you fired a shot in your Crusader SP which RNG gave a dispersion of 0.63m/100m too short, shooting at at flat ground 700 meters away, instead of hitting 4.41 meters from where you aimed it, you will hit 12.8 meters from where you aimed it.

 

TL;DR Dispersion of SPG's are very misleading and changes based upon impact angle making the CGC and the Crusader SP almost equally inaccurate.

 

What do you guys think of this? I'm not necessarily against the inaccuracy of artillery but i like to know how and why things work out the way they do. 

 

Sorry for the long post.

Attached Files

  • Attached File   Accuracy.png   34.03K
  • Attached File   Data.png   18.46K

Edited by chri105n, 09 July 2019 - 06:10 PM.


FrantisekBascovansky #2 Posted 09 July 2019 - 06:23 PM

    Warrant Officer

  • Player
  • 26360 battles
  • 633
  • Member since:
    03-26-2018

RNG. And yes, soft stats dispersion is very misleading.

But still, you can find persons yelling about need to nerf current SPG accuracy and raging about stunning friendly tank 30m away of fully aimed enemy at the same time.


Edited by FrantisekBascovansky, 09 July 2019 - 06:25 PM.


chri105n #3 Posted 09 July 2019 - 06:34 PM

    Private

  • Player
  • 21958 battles
  • 39
  • Member since:
    07-05-2011

View PostFrantisekBascovansky, on 09 July 2019 - 05:23 PM, said:

RNG. And yes, soft stats dispersion is very misleading.

But still, you can find persons yelling about need to nerf current SPG accuracy and raging about stunning friendly tank 30m away of fully aimed enemy at the same time.

 

I don't think this is due to soft stats, unless WG has some even more hidden stats, other than dispersion when turning hull/turret and dispersion when moving since all of these require movement and all the calculations are made assuming the SPG is stationary.

Bordhaw #4 Posted 09 July 2019 - 07:10 PM

    Brigadier

  • Player
  • 14499 battles
  • 4,507
  • Member since:
    01-29-2017

View Postchri105n, on 09 July 2019 - 05:09 PM, said:

 

TL;DR Dispersion of SPG's are very misleading and changes based upon impact angle making the CGC and the Crusader SP almost equally inaccurate.

 

What do you guys think of this? I'm not necessarily against the inaccuracy of artillery but i like to know how and why things work out the way they do. 

 

 

 

Did you have a crew in it ?

 



Gruff_ #5 Posted 09 July 2019 - 08:06 PM

    Warrant Officer

  • Player
  • 20069 battles
  • 816
  • [T-D-U] T-D-U
  • Member since:
    09-22-2011
A gust of wind must of blown them off course, try a less windy map.. :unsure:

Nishi_Kinuyo #6 Posted 09 July 2019 - 10:49 PM

    Lieutenant General

  • Player
  • 8821 battles
  • 6,055
  • [GUP] GUP
  • Member since:
    05-28-2011
Might be a perfect post for that SPG discussion thread we got pinned in gameplay. :girl:

chri105n #7 Posted 11 July 2019 - 04:24 PM

    Private

  • Player
  • 21958 battles
  • 39
  • Member since:
    07-05-2011

View PostNishi_Kinuyo, on 09 July 2019 - 09:49 PM, said:

Might be a perfect post for that SPG discussion thread we got pinned in gameplay. :girl:

 

Should I re-post it over there?

LordMuffin #8 Posted 11 July 2019 - 05:22 PM

    Field Marshal

  • Player
  • 51622 battles
  • 12,923
  • [-GLO-] -GLO-
  • Member since:
    06-21-2011

At 700m with a 0.62 dispersion gun.
You should be looking at a reticule with a radius of 2.17m

 

But if you miss with 10m something is weird.

 

May I suggest you try to measure the actual in-game reticule at 700m aswell. When fully aimed.

 

It should be doable by having a target at 700m and then put a tank on the edge of the reticule and then you have it.


Edited by LordMuffin, 11 July 2019 - 05:25 PM.


Telham #9 Posted 11 July 2019 - 07:18 PM

    Lance-corporal

  • Player
  • 11884 battles
  • 56
  • Member since:
    07-22-2015

OP, you stumbled at the first hurdle.

It's hard for me to put this into writing but the best I can come up with is firing into hilly terrain. Fring into an area which is sloping away from you shows you an enormously elongated dispersion circle ( or ovoid)  and you can miss center target by a mile. On the other hand firing into a hill, you can sometimes get a dispersion "circle" that almost actually IS a circle and it's hard to miss the target.

For arti the target is rarely square to the flight path of the shell, thus the dispersion circle ( or the dispersion cone if going from close to far away) doesn't show as a round aiming circle. The dispersion cone  is rarely bisected at a right angle thus you get aiming "circles" that are ovoid in shape and misses that are much farther away then the dispersion for that given distance warrants.

 



Balc0ra #10 Posted 11 July 2019 - 07:30 PM

    Field Marshal

  • Player
  • 71547 battles
  • 20,020
  • [WALL] WALL
  • Member since:
    07-10-2012
One thing to keep in mind is that the aim, don't take into account uneven ground. So if you aim downhill. Or up hill, vs flat in an arty. The shell might go past the circle, or land at the far edge or before it. 

Homer_J #11 Posted 11 July 2019 - 08:20 PM

    Field Marshal

  • Beta Tester
  • 31782 battles
  • 34,938
  • [WJDE] WJDE
  • Member since:
    09-03-2010

View PostLordMuffin, on 11 July 2019 - 05:22 PM, said:

At 700m with a 0.62 dispersion gun.
You should be looking at a reticule with a radius of 2.17m

 

Only looking straight down the shell flight path.

 

You can think of the point where it intersects the ground like a sausage sliced diagonally.

 

One of the things you learn with arty is when to aim short and when to aim long.



G01ngToxicCommand0 #12 Posted 11 July 2019 - 08:51 PM

    Warrant Officer

  • Player
  • 37465 battles
  • 800
  • Member since:
    11-03-2011

View Postchri105n, on 11 July 2019 - 04:24 PM, said:

 

Should I re-post it over there?

 

No one reads it, which is the purpose of that pinned topic.

LordMuffin #13 Posted 11 July 2019 - 10:22 PM

    Field Marshal

  • Player
  • 51622 battles
  • 12,923
  • [-GLO-] -GLO-
  • Member since:
    06-21-2011

View PostHomer_J, on 11 July 2019 - 08:20 PM, said:

 

Only looking straight down the shell flight path.

 

You can think of the point where it intersects the ground like a sausage sliced diagonally.

 

One of the things you learn with arty is when to aim short and when to aim long.

Indeed, which is,why you should measure the width.



ValkyrionX #14 Posted 11 July 2019 - 11:30 PM

    Captain

  • Player
  • 50816 battles
  • 2,269
  • [NOPAS] NOPAS
  • Member since:
    02-07-2015
remove them , problem solved :)

Jauhesammutin #15 Posted 12 July 2019 - 09:38 AM

    Warrant Officer

  • Player
  • 23536 battles
  • 716
  • [KANKI] KANKI
  • Member since:
    11-05-2013

View Postchri105n, on 09 July 2019 - 05:09 PM, said:

Hello,

 

After a particularly frustrating round as arty where it felt like all my shots were missing, I decided to do a few calculations. Using the in game range indicator i found that some of my shells were missing by as much as 10-11 meters at ranges of around 700 meters. Which really doesn't add up with a dispersion of 0.63 (Crusader SP). 

 

View PostLordMuffin, on 11 July 2019 - 04:22 PM, said:

At 700m with a 0.62 dispersion gun.
You should be looking at a reticule with a radius of 2.17m

 

But if you miss with 10m something is weird.

 

May I suggest you try to measure the actual in-game reticule at 700m aswell. When fully aimed.

 

It should be doable by having a target at 700m and then put a tank on the edge of the reticule and then you have it.

 

At 700m with a 0.63 dispersion gun your radius would be 700m*0,62=4,41m. The 4,41m only includes 95% of the shots, so the rest 5% can go anywhere. This has been changed a long time ago and now the 5% get a new roll.

 

When you shoot from an angle your aiming reticle looks like an ellipse rather than a circle. Was that 10-11m miss to the side or behind the vehicle?

The aiming circle should be 8,82m wide but much longer if you aim from an angle.


Edited by Jauhesammutin, 12 July 2019 - 09:40 AM.


chri105n #16 Posted 21 July 2019 - 11:04 PM

    Private

  • Player
  • 21958 battles
  • 39
  • Member since:
    07-05-2011

View PostTelham, on 11 July 2019 - 06:18 PM, said:

OP, you stumbled at the first hurdle.

It's hard for me to put this into writing but the best I can come up with is firing into hilly terrain. Fring into an area which is sloping away from you shows you an enormously elongated dispersion circle ( or ovoid)  and you can miss center target by a mile. On the other hand firing into a hill, you can sometimes get a dispersion "circle" that almost actually IS a circle and it's hard to miss the target.

For arti the target is rarely square to the flight path of the shell, thus the dispersion circle ( or the dispersion cone if going from close to far away) doesn't show as a round aiming circle. The dispersion cone  is rarely bisected at a right angle thus you get aiming "circles" that are ovoid in shape and misses that are much farther away then the dispersion for that given distance warrants.

 

 

Yes. This was what I was trying to explaing with this part:

View Postchri105n, on 09 July 2019 - 05:09 PM, said:

My next step was to assume that the aiming circle was a projection of a hypothetical circle which is placed at an angle equal to the impact angle, and then projected unto the ground. The dispersion given in the game is then based upon this hypothetical circle (Many of you might already know this).

 

I'm sorry if I failed to communicate what my thoughts were. I've made a crude illustrations that should show what I am trying to communicate, link

 

In addition, all the calculations are made assuming the target is on level ground. I'm well aware that shooting a tank that is on angled terrain will affect the numbers.


Edited by chri105n, 21 July 2019 - 11:06 PM.


PhooBar #17 Posted 22 July 2019 - 03:21 AM

    Captain

  • Player
  • 20216 battles
  • 2,459
  • [T-D-U] T-D-U
  • Member since:
    01-22-2013

Just a non-math related note:

If you get closer to the target, the impact circle gets smaller, therefore RNG has less chance to tuck with you.  (see what I did there?)

Mount a camo net, paint, skill up your crew and use whatever terrain you can.

Getting closer gets more hits.

 



BeefCrtinBandit #18 Posted 22 July 2019 - 05:22 AM

    Warrant Officer

  • Player
  • 6951 battles
  • 643
  • [-SJA-] -SJA-
  • Member since:
    05-03-2016
spgs have been quitely nerfed in accuracy this year.  also its like the aim time scam. it's not the actual aim time, its aim time plus the time it takes to get your giant bloom down to fully aimed. same scam, different soft stat.  

Asklepi0s #19 Posted 22 July 2019 - 11:40 AM

    Moderator

  • Moderator
  • 10469 battles
  • 1,106
  • Member since:
    01-23-2017
Please use the pinned thread.

Thread is closed.






Also tagged with Artillery, Accuracy, Dispersion

1 user(s) are reading this topic

0 members, 1 guests, 0 anonymous users