Page MenuHomeFeedback Tracker

Remote explosive indirectHit falloff inconsistent/reversed
Feedback, NormalPublic

Description

Whilst developing content for SPE a strange behavior was discovered with certain remote explosives and the damage value they apply depending on their placement proximity to the model center of the object they are intended to damage.

This behavior was originally found with SPE assets but has found to behave the same with vanilla Satchel Charge and Demolition Block. The satchel charge is the worst offender, when placed via script at model center of a building and detonated it applies zero damage. Depending on the building it can take up to a 4 meter offset before damage is applied correctly.

Conversely the vanilla demolition block appears to correctly apply damage at model center, however in testing it has been found that small offsets E.G 1 meter. Can cause the demo block to briefly not apply any damage before resuming correct falloff behavior again.

Some initial investigation has led to the speculation that the explosive models bounding sphere may be a contributing factor. The original explosives from SPE that the bug was found with are very similar in size to the vanilla satchel and demolition block. When shrinking down the service lods on the SPE satchel charge equivalent the falloff behavior started to behave more as expected.

Video below briefly demonstrates the issue comparing vanilla satchel and demo block placed at model center. The height offset given is to more closely approximate how the explosive would be positioned if placed by a player through actual gameplay.

Details

Severity
Minor
Resolution
Open
Reproducibility
Always
Operating System
Windows 10 x64
Category
Explosives
Steps To Reproduce

Mission attached to demonstrate.

  • Play in editor.
  • Scroll actions will spawn either satchel charge or demolition block at the indicated offset from buildings model center and report the damage taken by the building as a hint.
  • Graph option runs each explosive type through offsets up to 35 meters and plots them on a graph. (Warning loud noises to follow)

Alternatively to execute manual offsets as shown in the video:

  • Load VR in editor.
  • Place down building of your choice. We have been using Land_i_Barracks_V1_F in testing.
  • Look at building and execute below code
target = cursorObject; 
target setDamage 0;
bomb = "SatchelCharge_Remote_Ammo" createVehicle [0,0,0]; 
bomb setPosASL ((getPosASL target)vectorAdd [0,0,0.6]);
bomb setDamage 1;
  • Omit the last line if you just want to look at where your offset is positioned inside the building, before detonating.

Ammo classes:

SatchelCharge_Remote_Ammo
DemoCharge_Remote_Ammo

Additional Information

I think this has largely gone unnoticed because in practical gameplay it is difficult to actually place explosives close to model centers. For assets such as vehicles it probably has little effect, for buildings which are probably less commonly desired to be intentionally destroyed, it becomes more of a factor.

We originally found this whilst balance testing a new destructible building asset in SPE, which it is quite easy to place the explosives on or close to the model center.

Event Timeline

-Ben- created this task.Jul 25 2024, 12:46 PM
dedmen added a subscriber: dedmen.Aug 5 2024, 11:31 AM

Objects, shield other objects behind them.
The game checks, how visible the CenterOfMass of the building is, when viewing from the position of the explosion.

CenterOfMass = 64.0000000, 5.50000000, 64.0000000
ExplosionPos = 64.0000000, 6.18880224, 64.0000000
Visibility = 0

It thinks the CenterOfMass, is completely shielded.
And because the explosion is completely obscured by another object that's in the way, it does no damage.

And the object that is in the way is...... The satchel charge!

dedmen changed the task status from New to Feedback.Aug 5 2024, 11:37 AM
-Ben- added a comment.Aug 5 2024, 11:52 AM

Excellent. That falloff look a lot more sensible now.

-Ben- added a comment.Aug 6 2024, 2:36 PM

My graph seems a little different to your test, but still looks appropriate.

Also tested our original case in SPE and it is working much better now too, so seems all good from my end.

This comment was removed by Geez.
Geez added a subscriber: Geez.Aug 7 2024, 11:02 AM