So grenades actually have drag and deceleration depending on their projectile model

Vel0city

func_fish
aa
Dec 6, 2014
1,947
1,589
As found on reddit:https://www.youtube.com/watch?v=ACfafLuLmy8

Basically, despite all projectile speed bonuses being disabled in that video (Loch and LC), the Loose Cannon has the least range, and the Loch-N-Load travels the farthest. This is despite the fact that all projectiles in that video have the same projectile speed when exiting their respective launcher. So since the Loose Cannon is the largest projectile and has the largest area, it falls behind in range due to drag caused by a bigger physics model.

Loch-N-Load projectile is the same size as stock, but because it doesn't tumble it is more aerodynamic and therefore travels faster and farther despite not having its projectile speed.

I personally was not aware of this, and it's odd Valve would balance around it by having projectile speed stats. This means Loose Cannon balls travel slower than stock despite having a projectile speed bonus being stated in the stats.

I mean, we know Source is pretty advanced with physics simulating all kinds of things like mass and drag and angle-to-travel-direction calculations but to think actual weapons are balanced around it is kinda strange. For example, the Loose Cannon states it has a higher projectile speed making me think the projectiles travels further, yet in reality it's being held back by its bigger VPhysics model giving it more drag compared to stock. It's also how the Smissmass 2014 LnL change came to be: as seen in the video Valve removed the projectile spin giving it less drag making it travel faster and thus further.

I think I've never seen weapons in any game being balanced by physics properties, and certainly not in a way that makes a weapon not really do what its stats say it does, as the Iron Bomber bombs travel faster and further due to their small physics model, even though this isn't a listed stat on the weapon itself.

Interesting nevertheless.
 
Jul 6, 2015
1,425
820
giphy.gif
 

Dr. Spud

Grossly Incandescent
aa
Mar 23, 2009
880
855
The more you don't know. This isn't true.

There are parameters you can set in a model's qc file to dictate its physical properties. The physics mesh is used for collision and does not effect trajectory.

If the Source engine were simulating aerodynamics with a mesh in real time Valve would probably have told somebody.
 
Last edited:
Sep 10, 2015
142
39
I don't think this is completely true. I think that valve just has hardcoded projectile physics properties, Because in tf2classic it doesn't function the exact same. Not to mention that they didn't compensate for where projectiles come out, I'm pretty sure that the loose canons projectiles come out much closer to the body than the loch-n-load does. But I know for certain there is a difference between the projectile origins when they calculate the vectors for their movement.
 

Lain

lobotomy success story
aa
Jan 8, 2015
723
757
Yeah this is not true, sad that so many people take this at face value. The skeptic in me feels like it's just hard coded values randomized.