Molotov Area of Effect

(DB Genome editor) #1

I’m not sure if it’s a bug or something intentional that’s simply not obvious, but on a couple of occasions, I’ve been killed by a Molotov while a was clearly out of the flames VFX.

  • First occasion, I threw a Molotov in a corridor on Waterloo during a fight and when my opponent went down I was standing just outside the flame area and then I died, killed by my own Molotov according to the kill feed. The guy I had killed went “LOL” on the chat so I pointed out I wasn’t in the flames but he said from his POV I was standing right in them.

  • More recently, similar situation except the opponent had thrown the Molotov. Again, my last few HPs get drained by the fire even though I don’t seem to be touching it and I go down. Once in incap 3rd person view, I can clearly see that I’m outside the flame area, but a second later that same fire finished me off.

Is it possible that the area affected by the Molotov is not necessarily what you see? Either larger than the flames VFX (I could live with that if there is proper feedback that you’re taking heat damage) or worse positioned slightly different on each client and/or on the server? We know that mismatches exist with the ragdolls (getting gibbed or revived beside the body you see in incap state), and since the Molotov area of effect is supposed to be defined by the physics engine (as shown in an earlier dev video), is it possible it’s not totally WYSIWYG?

The only other explanation I can see would be that there is some sort of afterburn effect causing fire damage to continue for a short while after you’ve left the flame area. But if it is the case, some sort of obvious feedback that your are still on fire would be welcome. And that wouldn’t explain my first case where we both saw something different…

(1-800-NOTHING) #2

can confirm “afterburn”/lag damage.

(KoP) #3

Thanks for reporting. I just had a look and found some problems, many of your observations were spot on.

There is indeed a bug causing the damage inflicting checks to be slightly longer than they should be, enough to extend incorrectly just beyond the extent of the fire.

As for fire being positioned differently on each client it is synced but there was a very small chance of a very small difference that shouldn’t really be noticeable - I’ve dealt it just to be safe.

This is all fixed and should be included as part of any future patch.
EDIT: Sorry, it doesn’t look like the fix is going to be merged into the next patch, but it will be in the one after.