Arma3 random seed functions

[JeyR 2]

Helo. I have a seeding problem with the random function that I just can't figure out. 
 

while { surfaceIsWater _suitable || count _list > 0} do {
	while {_spot distance officer_jeff < 1000 || surfaceIsWater _spot} do {

		  _prevStep = _center getPos[ (random 2000), (random 350)];
		  for "_i" from 0 to 2 do {
		  	_step = _prevStep getPos[ (random 2001), (random 351)];
		  	_prevStep = _step;
		  };
	  	_spot = [_step select 0,_step select 1,0];
	};
	_suitable = [_spot, 0, 300, 20, 0, 0.7, 0] call BIS_fnc_findSafePos;
	_list = nearestTerrainObjects [_suitable,["TREE","BUILDING","RUIN","ROCK","HOUSE"], _howBig,false];
};

If the script can't find a position that is not a water surface, it restarts BUT the RANDOM functions just keeps giving me the same position every time the cycle ends. I don't understand, is it not supposed to be random? 
I also tried this with seed using the alternate syntax of random: 

while { surfaceIsWater _suitable || count _list > 0} do {
	while {_spot distance officer_jeff < 1000 || surfaceIsWater _spot} do {

		  _prevStep = _center getPos[ (time random 2000), (time random 350)];
		  for "_i" from 0 to 2 do {
		  	_step = _prevStep getPos[ (time random 2001), (time random 351)];
		  	_prevStep = _step;
		  };
	  	_spot = [_step select 0,_step select 1,0];
	};
	_suitable = [_spot, 0, 300, 20, 0, 0.7, 0] call BIS_fnc_findSafePos;
	_list = nearestTerrainObjects [_suitable,["TREE","BUILDING","RUIN","ROCK","HOUSE"], _howBig,false];
};

Here is an example output: 

15:12:48 "[823.795,10712.5] as _suitable, 56 as count _list, [816.133,10551.5,0] as _spot "
15:12:49 "[820.423,10745.8] as _suitable, 46 as count _list, [816.133,10551.5,0] as _spot "
15:12:50 "[828.004,10684.4] as _suitable, 76 as count _list, [816.133,10551.5,0] as _spot "

Any help would be much appreciated.
 

[gc8 977]

Can you please also post the code that prints the output?

[JeyR 2]

diag_log format ["%1 %2 %3  ", _suitable,count _list, _spot];

I don't see how is this important. 

[gc8 977]

Honestly I don't see anything wrong with the output. _spot is only printing the last value you assigned to it

[JeyR 2]

I know that the output is correct. Did you read my question? I want to know why the "[ (random 2000), (random 350)]" keeps giving back the same numbers in each cycle ?

[gc8 977]

Have you tried putting:

 

systemchat format["_prevStep = %1",_prevStep];

 

Because I doubt it really gives back same numbers...

[JeyR 2]

Yes it does, and if you don't believe it, try it. I can upload the whole file if need be. 

[gc8 977]

if you put the whole script somewhere I would like to test it

[JeyR 2]

https://pastebin.com/dvaUQzZu

[gc8 977]

I get only one print in the log but the numbers are different every time

[JeyR 2]

How do you get only 1 print? 

Here is my output, and I just raw copied my script into pastebin : 
https://i.gyazo.com/0a4263a9d70009a2ebbe9b1418d6e196.png

[gc8 977]

This is the code I tested with, had to make some modifications because I run the script from init.sqf and there was no _this variable there:

 

 officer_jeff = player;
 
_howBig = 11; // _this select 0;
missionNamespace setVariable["task_underway",true];

_center = missionNamespace getVariable ["World_center",[5840,5700,0]];
_suitable = [0,0,0];
private ["_step","_prevStep"];
_list = _suitable;
_spot = getPos officer_jeff;
while { surfaceIsWater _suitable || count _list > 0} do {
	while {_spot distance (officer_jeff) < 1000 || surfaceIsWater _spot} do {

		  _prevStep = _center getPos[ (random 2000), (random 350)];
		  for "_i" from 0 to 2 do {
		  	_step = _prevStep getPos[ (random 2001), (random 351)];
		  	_prevStep = _step;
		  };
	  	_spot = [_step select 0,_step select 1,0];
	};
	_suitable = [_spot, 0, 300, 20, 0, 0.7, 0] call BIS_fnc_findSafePos;
	_list = nearestTerrainObjects [_suitable,["TREE","BUILDING","RUIN","ROCK","HOUSE"], _howBig,false];
	diag_log format [">>>>>>>>>>>>>>>>>>>>>>>>>> %1 %2 %3 ", _suitable,count _list, _spot];
};
_marker1 = createMarker ["Marker"+ str _suitable,_suitable];
_marker1 setMarkerType "hd_objective";

 

[JeyR 2]

_howbig is always more than 100 meters, of course it works with 11. I might have forgot to mention that. 

[gc8 977]

Ok I got it printing more with 50 in the _howBig

 

and this is what it printed:

 

18:25:52 "_spot = [5424.32,2814.91,0]"
18:25:52 "_spot = [6579.24,1790.43,0]"
18:25:52 "_spot = [6060.41,6817.24,0]"
18:25:52 "_spot = [6682.16,6311.07,0]"
18:25:52 "_spot = [7446.01,7323.48,0]"
18:25:52 ">>> [7185.68,7461.17] 0 [7446.01,7323.48,0] "

second run:

 

18:26:13 "_spot = [6449.77,3067.1,0]"
18:26:13 "_spot = [6000.58,4245.9,0]"
18:26:13 ">>> [6037.21,4216.71] 3 [6000.58,4245.9,0] "
18:26:13 ">>> [6245.18,4352.97] 7 [6000.58,4245.9,0] "
18:26:13 ">>> [6033.74,4244.55] 7 [6000.58,4245.9,0] "
18:26:13 ">>> [6024.26,4214.05] 5 [6000.58,4245.9,0] "
18:26:13 ">>> [5954.42,4244.34] 3 [6000.58,4245.9,0] "
18:26:13 ">>> [5902.36,4290.74] 2 [6000.58,4245.9,0] "
18:26:13 ">>> [5932.08,4268.41] 2 [6000.58,4245.9,0] "
18:26:13 ">>> [5923.96,4278.38] 2 [6000.58,4245.9,0] "
18:26:13 ">>> [5869,4270.3] 4 [6000.58,4245.9,0] "
18:26:13 ">>> [6234.19,4349.69] 10 [6000.58,4245.9,0] "
18:26:13 ">>> [6269.02,4344.69] 2 [6000.58,4245.9,0] "
18:26:13 ">>> [5863.95,4260.2] 7 [6000.58,4245.9,0] "
18:26:13 ">>> [5924.03,4258.01] 1 [6000.58,4245.9,0] "
18:26:13 ">>> [5974.62,4262.46] 3 [6000.58,4245.9,0] "
18:26:13 ">>> [5983.3,4359.46] 6 [6000.58,4245.9,0] "
18:26:13 ">>> [6007.38,4256.17] 5 [6000.58,4245.9,0] "
18:26:13 ">>> [5915.53,4280.58] 2 [6000.58,4245.9,0] "
18:26:13 ">>> [6084.11,4323.63] 2 [6000.58,4245.9,0] "
18:26:13 ">>> [5910.75,4262.52] 0 [6000.58,4245.9,0] "

 

Seems to work to me (The _spot is changing)

[JeyR 2]

No it doesn't, I can see it clearly. The _spot didn't change in the last 20 cycles [6000.58,4245.9,0]. Don't tell me you don't see it. _spot is supposed to be random, not the same number in each cycle. Get it ?

[gc8 977]

Just now, JeyR said:

No it doesn't, I can see it clearly. The _spot didn't change in the last 20 cycles [6000.58,4245.9,0]. Don't tell me you don't see it. _spot is supposed to be random, not the same number in each cycle. Get it ?

 

Of course it's the same in the second print because it prints the last value set for _spot. But in the "_spot = " line all the values assigned to _spot are printed and those vary

[JeyR 2]

I can't see it, the 2nd run just printed _spot as [6000.58,4245.9,0] 20 times in a row. What do you mean it varies ?

[gc8 977]

Try with this new test code:

 

officer_jeff = player;
 
_howBig = 50; // _this select 0;
missionNamespace setVariable["task_underway",true];

_center = missionNamespace getVariable ["World_center",[5840,5700,0]];
_suitable = [0,0,0];
private ["_step","_prevStep"];
_list = _suitable;
_spot = getPos officer_jeff;
while { surfaceIsWater _suitable || count _list > 0} do {
	while {_spot distance (officer_jeff) < 1000 || surfaceIsWater _spot} do {

		  _prevStep = _center getPos[ (random 2000), (random 350)];
		  for "_i" from 0 to 2 do {
		  	_step = _prevStep getPos[ (random 2001), (random 351)];
		  	_prevStep = _step;
		  };
	  	_spot = [_step select 0,_step select 1,0];
		
		diag_log format["_spot = %1", _spot];
	};
	_suitable = [_spot, 0, 300, 20, 0, 0.7, 0] call BIS_fnc_findSafePos;
	_list = nearestTerrainObjects [_suitable,["TREE","BUILDING","RUIN","ROCK","HOUSE"], _howBig,false];
	
	diag_log format [">>> %1 %2 %3 ", _suitable,count _list, _spot];
};
_marker1 = createMarker ["Marker"+ str _suitable,_suitable];
_marker1 setMarkerType "hd_objective";

 

Then check the "_spot =" prints in the log

 

[JeyR 2]

17:49:03 "_spot = [9666.13,3028.81,0]"
17:49:03 ">>> [9531.69,2972.07] 167 [9666.13,3028.81,0] "
17:49:03 ">>> [9455.88,3236.07] 38 [9666.13,3028.81,0] "
17:49:03 ">>> [9409.98,3141.88] 44 [9666.13,3028.81,0] "
17:49:03 ">>> [9784.95,2756.75] 103 [9666.13,3028.81,0] "
17:49:03 ">>> [9478.4,3219.99] 66 [9666.13,3028.81,0] "
17:49:03 ">>> [9432.54,3081.66] 112 [9666.13,3028.81,0] "
17:49:03 ">>> [9464.37,3226.52] 45 [9666.13,3028.81,0] "
17:49:03 ">>> [9672.66,2894.3] 131 [9666.13,3028.81,0] "
17:49:03 ">>> [9427.07,3079.41] 109 [9666.13,3028.81,0] "
17:49:03 ">>> [9411.67,3120.7] 75 [9666.13,3028.81,0] "
17:49:03 ">>> [9499.26,3231.2] 92 [9666.13,3028.81,0] "
17:49:03 ">>> [9650.26,2901.7] 129 [9666.13,3028.81,0] "
17:49:03 ">>> [9455.83,3235.39] 38 [9666.13,3028.81,0] "
17:49:04 ">>> [9454.94,3095.55] 126 [9666.13,3028.81,0] "
17:49:04 ">>> [9778.71,2757.8] 105 [9666.13,3028.81,0] "
17:49:04 ">>> [9486.06,3108.22] 126 [9666.13,3028.81,0] "

 

[gc8 977]

This is what it printed for me:

 

18:52:20 "_spot = [7819.88,3367.59,0]"
18:52:20 "_spot = [7012.37,4690.59,0]"
18:52:20 "_spot = [7960.8,4514.42,0]"
18:52:20 "_spot = [4829.26,7487.83,0]"
18:52:20 "_spot = [6527.62,6185.35,0]"
18:52:20 ">>> [7228,6987,0] 0 [6527.62,6185.35,0] "

 

[JeyR 2]

Okey, I have to be missing something then. I am calling this script via debug console. like this : [100] spawn CHAB_fnc_findSpot; Each of my functions are listed in CfgFunctions.hpp. What do I do wrong? 

[gc8 977]

try placing your target guy in forest or something.

[pierremgi 4862]

What is the aim of this code?  In plain language.

[JeyR 2]

Find a safe spot for a composition to spawn , randomly. 

[pierremgi 4862]

Just now, JeyR said:

Find a safe spot for a composition to spawn , randomly. 

That's perhaps the problem. You need a too big safe pos (good luck on Tanoa!) and random is not efficient. Try a spiral, or some filter to avoid already checked areas.