Login

russian armor

Tank Destroyer Moving Penalty When Turning to Aim

7 May 2020, 03:09 AM
#1
avatar of Cyanara

Posts: 769 | Subs: 1

This has always felt like a distinct problem with the Jagdpanzer IV, and after seeing it a few times in streams recently it has been on my mind again.

When tank destroyers (TDs) rotate to target an enemy tank they have a tendency to shoot before they have finished turning. This seems to coincide with a high miss rate that makes it look like they must have a moving penalty being applied that a stationary turreted tank would not have.

I did some tests in CheatMod with an SU85 and JPIV. I always had the bias that the JPIV was far more heavily affected by this issue, but surprisingly it seemed pretty even from an anecdotal perspective. Both tanks distinctly shoot before they finish rotating and often miss.

I'm curious as to whether increasing general TD aim time by 300 milliseconds or so (and reducing cooldown/reload time accordingly) would help to resolve this issue and allow TDs to perform more consistently in their role.

If the enemy tank was close enough to make the TD keep rotating, the moving penalty would probably still appropriately apply.

Anyone with modding knowledge able to comment on whether or not this makes any sense?

Edit


And the results are in:



Recordings of the experiments in case you want to see it yourself or recreate methodology (I recommend 2x speed :p )




7 May 2020, 05:20 AM
#2
avatar of WhySooSerious

Posts: 1248

I think it's also possible to not have the penalty apply when ONLY rotating as another option.
7 May 2020, 09:28 AM
#3
avatar of Hannibal
Senior Moderator Badge

Posts: 3104 | Subs: 2

This has always felt like a distinct problem with the Jagdpanzer IV, and after seeing it a few times in streams recently it has been on my mind again.

When tank destroyers (TDs) rotate to target an enemy tank they have a tendency to shoot before they have finished turning. This seems to coincide with a high miss rate that makes it look like they must have a moving penalty being applied that a stationary turreted tank would not have.

I did some tests in CheatMod with an SU85 and JPIV. I always had the bias that the JPIV was far more heavily affected by this issue, but surprisingly it seemed pretty even from an anecdotal perspective. Both tanks distinctly shoot before they finish rotating and often miss.

I'm curious as to whether increasing general TD aim time by 300 milliseconds or so (and reducing cooldown/reload time accordingly) would help to resolve this issue and allow TDs to perform more consistently in their role.

If the enemy tank was close enough to make the TD keep rotating, the moving penalty would probably still appropriately apply.

Anyone with modding knowledge able to comment on whether or not this makes any sense?

Do you have any counted numbers for yout test?
But what I could imagine hapoening:
The tank shoots as soon as the first edge of the target reaches the tolerated angle. It might be that the tank then targets the edge instead of the center. If the shot scatters, then one half of the scatter area is not covered by the enemy target which leads to a higher chance to miss overall. But I don't know how target aquisition is checked. Maybe it is always the center of the unit that needs to be in range. But I remember a discussion in the forums about Elefant and the British AT emplacement being weird, because although both have 70 range, the Elefant could target the emplacement but not the pther way round due to hit boxes and the projectile spawn point or something.

But it might just be that CoH bugs out and applies some modifiers during rotation.

I am not sure if fire aim time would fix this. It depends on where the tank actually aims at. If the original target would still be the edge and not be updated after the aim time, then nothing would change.
7 May 2020, 10:04 AM
#4
avatar of Fire and Terror

Posts: 306

thanks for the info! In order to properly test this one must compare the turning numbers with the numbers while chasing a tank. So one can actually do moving acc vs turning acc. I always feel like the moving acc can be quite bad, so i often halt before the shot.
7 May 2020, 10:22 AM
#5
avatar of Hannibal
Senior Moderator Badge

Posts: 3104 | Subs: 2

thanks for the info! In order to properly test this one must compare the turning numbers with the numbers while chasing a tank. So one can actually do moving acc vs turning acc. I always feel like the moving acc can be quite bad, so i often halt before the shot.

Moving accuracy is usually halved. Additionally the scatter area gets larger which further reduces hit chance.
7 May 2020, 10:35 AM
#6
avatar of ullumulu

Posts: 2243


Do you have any counted numbers for yout test?
But what I could imagine hapoening:
The tank shoots as soon as the first edge of the target reaches the tolerated angle. It might be that the tank then targets the edge instead of the center. If the shot scatters, then one half of the scatter area is not covered by the enemy target which leads to a higher chance to miss overall. But I don't know how target aquisition is checked. Maybe it is always the center of the unit that needs to be in range. But I remember a discussion in the forums about Elefant and the British AT emplacement being weird, because although both have 70 range, the Elefant could target the emplacement but not the pther way round due to hit boxes and the projectile spawn point or something.

But it might just be that CoH bugs out and applies some modifiers during rotation.

I am not sure if fire aim time would fix this. It depends on where the tank actually aims at. If the original target would still be the edge and not be updated after the aim time, then nothing would change.



this dont need tests, since everybody which played TD with no turret know this issue...the first shoot misses often while the vehicle stay but must turn the chasemate.


jackson and firefly has a big advantage here
7 May 2020, 11:03 AM
#7
avatar of Hannibal
Senior Moderator Badge

Posts: 3104 | Subs: 2




this dont need tests, since everybody which played TD with no turret know this issue...the first shoot misses often while the vehicle stay but must turn the chasemate.


jackson and firefly has a big advantage here

Of course it needs testing. Subjective bias is strong and otherwise fixing it would just be guess work.
MMX
7 May 2020, 11:21 AM
#8
avatar of MMX

Posts: 999 | Subs: 1




this dont need tests, since everybody which played TD with no turret know this issue...the first shoot misses often while the vehicle stay but must turn the chasemate.


jackson and firefly has a big advantage here


while it is indeed known that rotation counts as moving and, hence, rotation tends to make shots miss, this is not what the OP wants to test. the question is if a longer aim time could be used to delay the shooting animation so that the moving accuracy penalty no longer applies.

i did some quick testing on that and it seems that is indeed possible.

setting the fire aim and ready aim time to 2 seconds did the trick in my setup, which used a base acc of 1 and moving acc of 0, together with absurdly high scatter angle to quickly distinguish scatter shots from natural (accuracy based) hits.

in the tests the jp4 took a brief moment to aim after rotating and thus never missed a shot. so it'd be merely about testing how much delay is needed to force the rotation to fully stop before shooting
7 May 2020, 11:40 AM
#9
avatar of Cyanara

Posts: 769 | Subs: 1

jump backJump back to quoted post7 May 2020, 11:21 AMMMX


setting the fire aim and ready aim time to 2 seconds did the trick in my setup, which used a base acc of 1 and moving acc of 0, together with absurdly high scatter angle to quickly distinguish scatter shots from natural (accuracy based) hits.


Oh, very cool. Thanks for doing that. The scatter change is nice idea for testing. I'll record some testing of vanilla rules to help provide a baseline.
7 May 2020, 12:25 PM
#10
avatar of grammar

Posts: 28

Does changing the aim time make it easier to circle the TD?
7 May 2020, 12:28 PM
#11
avatar of Katitof

Posts: 17875 | Subs: 8

Does changing the aim time make it easier to circle the TD?

It means you need to have target in sight for longer to shoot. It should make it easier to run away from firing arc.
7 May 2020, 18:47 PM
#12
avatar of Stug life

Posts: 4474

i thought no-turret TD CAN'T shoot on the move, they don't have the line of code right ?
7 May 2020, 18:56 PM
#13
avatar of ShadowLinkX37
Director of Moderation Badge

Posts: 4183 | Subs: 4

i thought no-turret TD CAN'T shoot on the move, they don't have the line of code right ?


Pretty sure all of them can. I think the JT and ISU are the only ones that cannot. Maybe the elephant as well?
7 May 2020, 19:44 PM
#14
avatar of Katitof

Posts: 17875 | Subs: 8

i thought no-turret TD CAN'T shoot on the move, they don't have the line of code right ?

Only super heavies can't.
8 May 2020, 09:02 AM
#15
avatar of Cyanara

Posts: 769 | Subs: 1


Do you have any counted numbers for your test?


I do now! Turns out it was pretty much as I suspected.


And the results are in:



Recordings of the experiments in case you want to see it yourself or recreate methodology (I recommend 2x speed :p )



1 user is browsing this thread: 1 guest

Ladders Top 10

  • #
    Steam Alias
    W
    L
    %
    Streak
Data provided by Relic Relic Entertainment

Replay highlight

VS
  • U.S. Forces flag cblanco ★
  • The British Forces flag 보드카 중대
  • Oberkommando West flag VonManteuffel
  • Ostheer flag Heartless Jäger
uploaded by XXxxHeartlessxxXX

Board Info

470 users are online: 470 guests
17 posts in the last 24h
43 posts in the last week
97 posts in the last month
Registered members: 44644
Welcome our newest member, felayo364
Most online: 2043 users on 29 Oct 2023, 01:04 AM