I've been looking at the AR for a bit today.
The horizontal velocity of the grenades seems to be fine, from what I can see. Unreal uses Unreal Units, which correspond approximately to meters in a ratio of 52.5:1, so the horizontal speed of the grenade should be 787.5uu/s -- It's currently set at 800, which is fine.
However the falling behavior of the grenade ... From what I can see, gravity is handled natively in Unreal, so I can't simply adjust the strength of gravity. I know it'd be possible to adjust the projectile's Z velocity each tick and make it work that way, although that's not ideal... Not sure if there are better ways to approach the issue.
: I'm glad you like the feedback; it encourages me to offer
: my two cents more often.
: Regarding the grenades, I think it should be possible to
: calculate their behavior in Marathon. I made a new
: physics model in Anvil, and the grenade speed was set
: to 256. Judging from Mark Levin's physics model
: reference (he worked on the recent Marathon: Durandal
: if I recall), units of projectile speed in Anvil are
: likely in internal units per engine tick. There are
: 1024 internal units per world unit, a world unit is 2
: meters, and there are 30 ticks per second. So the
: horizontal speed of a grenade in Marathon is 15 meters
: per second.
: The vertical acceleration is harder to pinpoint. Overall
: gravity in Marathon is low (possibly to make the
: absence of fall damage less noticeable), so 9.8 m/s/s
: would likely be too high. I recall that projectiles
: start at a height of 0.6 WU above the ground, and I
: confirmed this experimentally in M1A1. I timed the
: time it takes for a grenade to hit the ground on a
: level plane. Using a stopwatch and averaging for a
: clip of 7 grenades, I come up with 1.02 seconds. I
: confirmed that by playing a replay in slow motion,
: where I counted off approximately 31 ticks from
: launching a grenade to the grenade hitting the ground,
: which is 1.03 seconds. In contrast, it takes half that
: time for a grenade to hit the ground in M:R, using the
: center view key.
: Doing a quadratic regression on these datapoints, I get
: an acceleration of -1.15 m/s/s. Startlingly low,
: really, but that's what it is in Marathon. If the time
: in the air is actually exactly 1 second, then the
: acceleration becomes exactly -1.2 m/s/s.
: So, long story short, I believe grenades will behave
: correctly if they're given a 'muzzle velocity' of 15
: m/s, and the effect of gravity on them is 1.15 m/s/s.
: I don't know how projectiles are defined in UT, but
: assuming things are already standardized to World
: Units, it should be implementable.
