Weapon Spread

[Doctor J]

The spread values given for firemodes in the .ufo files - are these in degrees, radians, 'gosh units' or what?  I tried grepping the sources but didn't find anything like a useful comment, only it being used in G_ShootSingle().

[BTAxis]

They're multipliers to a gaussian random value that's generated in the code.

[blondandy]

I think the gaussian spread is dimensionless and the spread values are in degrees.

http://ufoai.ninex.info/wiki/index.php/Damage_rate_vs_distance

angle = angles[YAW] * (M_PI * 2 / 360);

it converts from degrees to radians before calling cos and sin.

[Doctor J]

I think the gaussian spread is dimensionless and the spread values are in degrees.

That's what i was looking for, thanks!

[BhOg]

What do the different formats in the weapons_xx.ufo for "spread" mean?

Most entries read e.g. "spread 1.0 1.0", but flamer has "spread 1.5 0.5", machinegun has "spread 3 20" for full auto sweep, and grenade launcher has "spread 0 1.5" for aimed single shot?!

[H-Hour]

From the wiki:

width of gaussian applied to attempted line of fire, in degrees (TODO: check it really is degrees). First element affects pitch, second element affects yaw. Lower values mean greater precision.

[BhOg]

Oh, thanks, but let me rephrase ... why have those 3 given entries those "strange" values?

Is the "3 20" used to achieve the sweep effect?
Why do flamer and GL then have different deviations in vertical/horizontal direction?

[H-Hour]

Is the "3 20" used to achieve the sweep effect?
Why do flamer and GL then have different deviations in vertical/horizontal direction?

First number is pitch, second number is yaw. So yes, "3 20" will mean wildly inaccurate horizontally.