It seem that flamer is effected by elevation and does little damage.
This can be seen in the trenches in rails and metal.
elevation and weapons
25 Jan 2021, 16:29 PM
#1
Posts: 13477 | Subs: 1
11 Feb 2021, 22:34 PM
#2
Posts: 13477 | Subs: 1
For whatever reason, most flame bursts fired do nothing to the conscript squad targeted in the trenches on Rails and Metal, though there is a lick of flame that consistently appears some distance behind the targeted squad, as can be seen in the screenshot. This lick of flame does no damage, and is apparently just a visual bug. I had thought at first that the flame AOE was "Deflected" behind the squad somehow, but that doesn't appear to be the case. The damaged squad in the image is from a couple of mosin shots fired by a conscript squad I spawned, not from the flames, I merely forgot to set all squads HP to 100% before beginning the tests.
Interestingly, flame bursts do rarely "connect" with the squad in the trench, and does 20 damage (To the lowest three models in the trench) for some reason, despite them being in Green Cover. The flame burst also did 16 damage to one model from the closest Conscript squad to the Pioneer (The model closest to the aformentioned three Conscripts).
Is this a known bug? I wonder what causes this. This doesn't seem to just be a case of Cover not working.
12 Feb 2021, 07:19 AM
#3
Posts: 4928
Aren't flames technically projectiles? Is it possible the flames are actually colliding with the raised ground infront of the trench and not actually reaching the squad inside?
all you filthy hotchkiss lovers need to l2p #TeamPanzerIII
> Kriegsbeute Doctrine|Partisan Rework|British Infantry Rework|Soviet Assault Squad <
12 Feb 2021, 07:26 AM
#4
Posts: 13477 | Subs: 1
Aren't flames technically projectiles? Is it possible the flames are actually colliding with the raised ground infront of the trench and not actually reaching the squad inside?
I doubt that.
Even if it was flamer has a accuracy of 1 and should be able to hit the conscripts squad with 100% chance.
12 Feb 2021, 08:19 AM
#5
Posts: 3111 | Subs: 2
Could Pip elaborate if he tested at different ranges?
I don't think this is a bug but an animation and some kind of calculation issue.
From my experience flamer squads sometimes stop early (they probably are at the very edge of their range) and start attacking. The animation also looks like it should hit but there's barely any damage if at all.
It is usually fixed if you order them to walk a couple of steps closer and attack then.
I stopped issuing attack commands because of that and rather let them walk a bit too close because sometimes it is buggy.
I highly doubt anything of this has to do with elevation or terrain in general.
I don't think this is a bug but an animation and some kind of calculation issue.
From my experience flamer squads sometimes stop early (they probably are at the very edge of their range) and start attacking. The animation also looks like it should hit but there's barely any damage if at all.
It is usually fixed if you order them to walk a couple of steps closer and attack then.
I stopped issuing attack commands because of that and rather let them walk a bit too close because sometimes it is buggy.
I highly doubt anything of this has to do with elevation or terrain in general.
12 Feb 2021, 08:26 AM
#6
Posts: 13477 | Subs: 1
Could Pip elaborate if he tested at different ranges?
I don't think this is a bug but an animation and some kind of calculation issue.
From my experience flamer squads sometimes stop early (they probably are at the very edge of their range) and start attacking. The animation also looks like it should hit but there's barely any damage if at all.
It is usually fixed if you order them to walk a couple of steps closer and attack then.
I stopped issuing attack commands because of that and rather let them walk a bit too close because sometimes it is buggy.
I highly doubt anything of this has to do with elevation or terrain in general.
It could be that flamer stop at right distance in 2 D but the elevation increases the distance although the weapon has an AOE of 2...
Or it could be that the AOE only work for target at same height level.
12 Feb 2021, 08:44 AM
#7
Posts: 3111 | Subs: 2
It could be that flamer stop at right distance in 2 D but the elevation increases the distance although the weapon has an AOE of 2...
Or it could be that the AOE only work for target at same height level.
I mean the way CoH2 works you can't exclude anything, but I'd still go for the easiest solution.
I have seen this "no damage" issue regularly, I think I remember seeing it even in tournament casts. From what I know it does not depend on elevation, nor on cover, formation or whatnot. It usually got fixed once the flaming guy got two steps closer.
I think it has to do with some weird range calculation and flame damage, that the game thinks the flamer is in range so it stops walking while it actually is only at the border.
I don't know how exactly flamer damage is calculated, but there might be some kind of random chance to it. But this is just speculation.
As I said, walking just a little closer fixes the issue for me.
12 Feb 2021, 08:49 AM
#8
Posts: 13477 | Subs: 1
I mean the way CoH2 works you can't exclude anything, but I'd still go for the easiest solution.
I have seen this "no damage" issue regularly, I think I remember seeing it even in tournament casts. From what I know it does not depend on elevation, nor on cover, formation or whatnot. It usually got fixed once the flaming guy got two steps closer.
I think it has to do with some weird range calculation and flame damage, that the game thinks the flamer is in range so it stops walking while it actually is only at the border.
I don't know how exactly flamer damage is calculated, but there might be some kind of random chance to it. But this is just speculation.
As I said, walking just a little closer fixes the issue for me.
As far as I know flamer work like most small arm weapon and in order to fire the entity has to be in range and it goes through an accuracy check although the weapon has higher accuracy than most weapons.
The weapon simply has an AOE damage in addition.
12 Feb 2021, 09:19 AM
#9
Posts: 1273
Flamers appear to ray cast their animation to the position of the 3D target perpendicularly to the ground normal vector cell that they are standing on, or somewhere in between from itself to target. With elevations it most likely results in the raycast hitting the ground locally due to the slope, but the engine still thinks it is hitting the target successfully due to the game's hit/miss rules and it is then broadcast over network. I'm generalising. It leads to some funky stuff in some places of maps, especially on places with negative slopes adjacent to high rise. There's likely some weird network client reproduction/synch stuff because one tells it is hitting the target, whereas the other says it is not hitting and the engine does not know which one to prioritise (but locally it works fine?). I would suggest it is more of a map bug than unit or network bug, but the mistake is probably on all sides. /source: game programmer assumption of how the engine works.
12 Feb 2021, 09:37 AM
#10
Posts: 3111 | Subs: 2
I stand corrected. Vipper and Pip seem to be on to something here. What JohnSmith said could make sense, but god knows how they programmed it.
I tested on Over the Rhine since it has the largest elevations I can remember. Don't give much about the unit's health since I regularly moved the flamer to test some positions where damage is applied. Most of the positions I tried with both ground attack as well as auto fire.
First, THIS is working:
-> Flamer has unobstructed view of the (now dead) unit.
As soon as I move it behind the rock, everything bugs out. Funnily enough a ground attack triggers different animations than the auto attack. Visually, more flames got stuck in the rock.
Also not working with a full squad, so AoE damage is not applied either
To my shame I can't fully remember if this was working or not. I think it was, but super, super slowly:
However, for some reason this did not work:
Maybe some rocks of the cliffs created obstruction? Visually there was none, but well who knows what invisible objects there are.
Burning a squad below the cliff from next to the right bunker (see burning bush to the right edges) worked although slower than I would have expected. Burning a squad at similar elevation (see the burned ground at the concrete cover, flamer was near the trees) worked as intended.
I tested on Over the Rhine since it has the largest elevations I can remember. Don't give much about the unit's health since I regularly moved the flamer to test some positions where damage is applied. Most of the positions I tried with both ground attack as well as auto fire.
First, THIS is working:
-> Flamer has unobstructed view of the (now dead) unit.
As soon as I move it behind the rock, everything bugs out. Funnily enough a ground attack triggers different animations than the auto attack. Visually, more flames got stuck in the rock.
Also not working with a full squad, so AoE damage is not applied either
To my shame I can't fully remember if this was working or not. I think it was, but super, super slowly:
However, for some reason this did not work:
Maybe some rocks of the cliffs created obstruction? Visually there was none, but well who knows what invisible objects there are.
Burning a squad below the cliff from next to the right bunker (see burning bush to the right edges) worked although slower than I would have expected. Burning a squad at similar elevation (see the burned ground at the concrete cover, flamer was near the trees) worked as intended.
12 Feb 2021, 09:46 AM
#11
Posts: 1273
Great tests. I like your methodological approach. You're right, there is something deeper.
Aren't these rocks 3d models superseded on top of the map? This might make matters even weirder for the engine, as you suggest, they create obstructions.
I'm curious if it is the same in a map with a height difference but no models in between units.
Aren't these rocks 3d models superseded on top of the map? This might make matters even weirder for the engine, as you suggest, they create obstructions.
I'm curious if it is the same in a map with a height difference but no models in between units.
12 Feb 2021, 09:57 AM
#12
Posts: 999 | Subs: 1
interesting stuff... i always found flamers to act a bit weird on auto-attack; sometimes they seemingly do no damage at all even though the animation indicates the poor guys at the receiving end take a bath in liquid death and everything around them gets burned to a crisp.
usually what hannibal described solves this for me:
but apparently there seems to be more to this than i thought. also, if elevation is really responsible for the odd behavior, does this extend to other AoE weapons as well?
in any case, great detective work so far guys, hope this mystery can (and will) be solved!
usually what hannibal described solves this for me:
As I said, walking just a little closer fixes the issue for me.
but apparently there seems to be more to this than i thought. also, if elevation is really responsible for the odd behavior, does this extend to other AoE weapons as well?
in any case, great detective work so far guys, hope this mystery can (and will) be solved!
||| ScatterThis! - Tank AI calculation made easy ||| CoH2 Pen Calc - Tank vs tank... who wins against who and by what margin? |||
12 Feb 2021, 11:09 AM
#13
Posts: 13477 | Subs: 1
I had noticed a similar issues years ago that affected other weapons also and I pointed it out to Relic at the time who according to patch notes had fix it.
"Directional cover is now analyzed on a 2D basis. This prevents scenarios from occurring where entities would not receive their cover bonus."
Maybe what they tried then might be helpful
"Directional cover is now analyzed on a 2D basis. This prevents scenarios from occurring where entities would not receive their cover bonus."
Maybe what they tried then might be helpful
12 Feb 2021, 11:12 AM
#14
Posts: 208
From what I recall, flamers actually DO use a projectile. Their collision_type is tp_throw if I recall correctly, which can travel through most objects in the game with some exceptions. Perhaps those cliffs are one of them? Any map makers know the exact name of those in the editor?
12 Feb 2021, 11:59 AM
#15
Posts: 13477 | Subs: 1
From what I recall, flamers actually DO use a projectile. Their collision_type is tp_throw if I recall correctly, which can travel through most objects in the game with some exceptions. Perhaps those cliffs are one of them? Any map makers know the exact name of those in the editor?
But if the weapon has an accuracy of 1 why would it not hit in the accuracy check?
Even if there is a projectile why would it create an issue unless is somehow connected with the faust bug?
There is also this that might relevant:
"Projectile Tuning
Modified the behavior of tp_homing_inf projectiles to not collide with a number of world objects such as bushes or light fences. In effect, this change will increase the strength of any weapon using this projectile type. This will impact the reliability of hand held AT like the Bazooka and flak weaponry like the Ostwind or 251 Flak Half-track. These weapon types will no longer continuously hit world objects, thereby missing their targets."
12 Feb 2021, 12:17 PM
#16
Posts: 3111 | Subs: 2
But if the weapon has an accuracy of 1 why would it not hit in the accuracy check?
Even if there is a projectile why would it create an issue unless is somehow connected with the faust bug?
There is also this that might relevant:
"Projectile Tuning
Modified the behavior of tp_homing_inf projectiles to not collide with a number of world objects such as bushes or light fences. In effect, this change will increase the strength of any weapon using this projectile type. This will impact the reliability of hand held AT like the Bazooka and flak weaponry like the Ostwind or 251 Flak Half-track. These weapon types will no longer continuously hit world objects, thereby missing their targets."
Because it apparently collides with terrain or something else. Just like a tank shot that would hit can collide with terrain or another tank if front of it.
12 Feb 2021, 12:58 PM
#17
Posts: 13477 | Subs: 1
Because it apparently collides with terrain or something else. Just like a tank shot that would hit can collide with terrain or another tank if front of it.
I don't know if ballistic shots that have passed accuracy check and have score a "hit" can collide with unit that stand between gun and target or if collision hit checks are only for shot that have "missed".
I guess I can test with one of the high accuracy TDs at some point.
12 Feb 2021, 13:00 PM
#18
Posts: 8154 | Subs: 2
Because it apparently collides with terrain or something else. Just like a tank shot that would hit can collide with terrain or another tank if front of it.
I tried this on the test range map with cheatcommands and a 100% accuracy shot will go through other vehicles and terrain*.
I basically put a JPIV to shoot a JT, while there were a bunch of other JTs in the middle.
In regards to terrain, you know how if you delete a base building the ground below sunk? Well i just parked a JP4 so it was basically below ground.
EDIT:
Terrain as the ground, not any other objects or different entities.
My controls weren't working
And if they were, you were playing dishonorably,
And if you weren't, you were playing without skill,
And if you were, it's not fun to play that way,
And if it is, you only care about winning.
Bible of Scrublord, Psalt 3:16.
And if they were, you were playing dishonorably,
And if you weren't, you were playing without skill,
And if you were, it's not fun to play that way,
And if it is, you only care about winning.
Bible of Scrublord, Psalt 3:16.
12 Feb 2021, 13:01 PM
#19
Posts: 1594
I don't know if ballistic shots that have passed accuracy check and have score a "hit" can collide with unit that stand between gun and target or if collision hit checks are only for shot that have "missed".
I guess I can test with one of the high accuracy TDs at some point.
As far as I'm aware "hit" shots cannot miss/hit intervening obstacles, only missed shots that otherwise would have struck the opposing tank's collision box through favourable scatter.
Though I believe there used to be an issue with Snares striking obstacles en-route to a target, so it may be dependent on collision tables.
12 Feb 2021, 13:04 PM
#20
Posts: 13477 | Subs: 1
I tried this on the test range map with cheatcommands and a 100% accuracy shot will go through other vehicles and terrain*.
I basically put a JPIV to shoot a JT, while there were a bunch of other JTs in the middle.
In regards to terrain, you know how if you delete a base building the ground below sunk? Well i just parked a JP4 so it was basically below ground.
Seem someone beat me to the test and things where as I suspected.
Great work, well done.
The only other similar case I know is faust bug where the projectile should hit the target and collides in process and even then I have not checked it it actually uses accuracy check.
(maybe this is something worth looking into as a fix to faust bug)
1 user is browsing this thread:
1 guest
Livestreams
 |
|
|
57 | ||
 |
|
|
30 | ||
|
|
|
|
23 | ||
|
|
|
|
4 | ||
 |
|
|
1 | ||
 |
|
|
132 | ||
 |
|
|
38 | ||
|
|
|
|
1 | ||
|
|
|
|
1 |
Ladders Top 10
-
#Steam AliasWL%Streak
- 1.278110.716+16
- 2.270144.652+2
- 3.29446.865-1
- 4.558230.708+2
- 5.253150.628+14
- 6.968566.631-4
- 7.402.952+10
- 8.22491.711+11
- 9.8120.802-1
- 10.19078.709+4
Data provided by
Relic Entertainment
Relic Entertainment
Replay highlight
VS
-
cblanco ★ -
보드카 중대
-
VonManteuffel -
Heartless Jäger
Einhoven Country
Honor it
7
Download
781