Define a relative position to objects

[Serpent 0]

Privjet

Is there a possibility in OFP to define a position relativly to an object?

Situation:

A player (let's call him "pl1") got by a script the option to build up a certain object (let's take an M2-MG named "m2mg"). This object should get placed _xy meters in front of the player no matter what's his angle of sight.

My script so far:

_r= (getdir player)+0

standort setpos [(getpos player select 0)+(sin _r),(getpos player select 1)+(sin _r),0];

_m2mg="M2StaticMG" createvehicle (getpos standort);

exit;

As far as I see the "_m2mg" gets placed in a relative position to the main grid but not to the player. Am I right there? What is the mathematic formula to place the object right in front of the player? Can someone make an example for a player's angle of sight of _deg degrees? The object _m2mg should be placed always in front of the player no matter how much _deg is.

[Prospero 1]

Let the player's direction be _dir, and let the distance of the object to be placed in front of the player be _dist.

_dx = cos(_dir) * _dist + sin(_dir) * _dist

_dy = -sin(_dir) * _dist + cos(_dir) * _dist

Now, add these onto the X/Y position of the player.

Prospero

[Serpent 0]

hm, it doesn't work. The gun always appears on the right side of the player, guess in an angle of 90°...

//edit

now it works but in a very strange way. Here's my script:

_dir= (getdir player)+315

_dx = cos(_dir) * 2 + sin(_dir) * 2

_dy = -sin(_dir) * 2 + cos(_dir) * 2

standort setpos [(getpos player select 0)+_dx,(getpos player select 1)+_dy,0];

_mg="M2StaticMG" createvehicle (getpos standort);

_mg setdir (getdir player);

exit;

As long as I add +315° to +325° the gun appears in front of the player. Does anyone find the mistake in the script?

[Doolittle 0]

I didn't look at your script, but be aware that orientation and angle is different for objects and sin/cos functions.  For objects, facing North is 0 degrees and we count clockwise.  East would be 90 degrees for example.  For the sine and cosine stuff, 0 degrees is East and we count counter-clockwise, so 90 degrees would be North, etc.

Here's how I change an "angle function" to an object's direction:

_x = _pos select 0

_z = _pos select 1

_dir = _z atan2 _x

_obj setDir (90 - _dir)

Doolittle

[seedhe 0]

I am doing much the same thing in a script and what I'm doing is along the lines of-

</span><table border="0" align="center" width="95%" cellpadding="3" cellspacing="1"><tr><td>Code Sample </td></tr><tr><td id="CODE">_dist = 2

_dir= (getdir player)

_dx = _dist * sin _dir

_dy = _dist * cos _dir

standort setpos [(getpos player select 0)+_dx,(getpos player select 1)+_dy,0];

_mg="M2StaticMG" createvehicle (getpos standort);

_mg setdir (getdir player);

exit;<span id='postcolor'>

[Dunkelheit II 0]

Hi. Could anyone suggest a simple script that would check a relative position between two points? The aim is just to check the angle between the vectors of two randomly generated points so that they are not situated on a more or less straight line vertically or horizontally. In the readme of the ECP mod, it is said:

 

"[<unit>,<position>] call (ECP_resources select 4)
ECP_getDirRelPos returns the relative direction in degrees from <unit> to <position>. 0 degrees
is considered to be to the direct front of <unit>, 90 is to the right etc."

 

But I can't find even the place where this array, ECP_recources, is announced, not to say the script itself.
 

[zwobot 22]

What about [OFPRes] calcrelpos.sqf by joltan or [OFPRes] anglefacingdiff.sqf by LCD ?

[Dunkelheit II 0]

7 hours ago, zwobot said:

What about [OFPRes] calcrelpos.sqf by joltan or [OFPRes] anglefacingdiff.sqf by LCD ?

 

The problem is that there are just two sets of coordinates, not objects. So the first script is irrelevant because it uses getdir, and the second – because it does not return angle and uses _northdirection that should somehow be known.

[zwobot 22]

So why don't you write a custom function yourself then? It seems all you need is a dot product of two vectors?

[Dunkelheit II 0]

18 hours ago, zwobot said:

It seems all you need is a dot product of two vectors? 

The direction matters. I cannot just take an arbitrary third point to define two vectors.