Thanks for the clarification. I am still very much in the copy and paste fingers crossed stage of qc. I believe c0burn wants to release a bug smashed version of the code eventually. That's a great project IMO. 

Result is that a player in fight with an Axe_ogre will get pissed. If player is shot from distance, the ogre won't walk to the player's body. 

the point of doing it in ai_run is that:
a) self is the ogre that should do the widdle. there is absolutely no need to call find (because you can just use self). there is no need to loop etc (each monster will call the function at some point anyway).
however, you really ought to only do that logic if self.classname is actually what you expect. you don't really want scrags trying to widdle.
b) the player that its trying to urinate on is already stored inside self.enemy. This means that you don't need a seperate leakentity field - just use self.enemy! (do note that you'll want to avoid the self.enemy=world line if the ogre is meant to start widdling.) 

Thanks for your answer. Sorry I'm such a noob at coding.
If I use the self argument, how do I make it fit in?

As it is an experimental src, so maybe it is better to show an example
of what I am trying to do.
MF_Qtest00 

here's a demo of Qtest

the main problem, the "new" monsters seem too much overpowered (especially vomitus)

noticed a couple of issues, check the demo

http://quaketastic.com/files/demos/spy_MadfoxQtest_demo.rar 

Just finished a nice trick to change the serpent into a quaddam.
But don't touch it, cause it lives.., a bit husling with the skin file.

Yes, the src has a MISSILE weaponcode for each monster,
wizard is also much too strong. Number code for weapons won't work. Still beta.

Thanks for the demo, I'll look at it! 

You made your way quiet advanced on hard skill. Spy.
The missing sounds are qc depended.
I turned a bit strayed with the qtest.qc. adding the monsters,
especially the dragons, caused a lot of changes.

I assume the forced demo end had something to do with crashing,
it is still beta version, already glad it runs.

Thanks for taking your time. 

I am just getting started "cutting and pasting" QuakeC. I followed Preach's revised tutorial on monster spawning here:

https://tomeofpreach.wordpress.com/2017/10/08/teleporting-monsters-flag/

I'd like to add a small random delay automatically to each teleporting monster as you can do in Quoth. I am using custents multiple triggers to trigger these delays spawns. Similar to Quoth, it would be nice to have a little variation on the timing if the monsters all share a targetname.

I've tried a few things already, being exceedingly cautious not to mess up this code I can barely comprehend. I'm wondering if anyone would like to give me a hint of how to approach this little addition. 

Add a new function:
void monster_teleport_delay () {
self.think = monster_teleport_go;
self.nextthink =time + self.wait; // or self.delay if you prefer that
};

Then in your monster_teleport() function change self.use to be monster_teleport_delay; instead of monster_teleport_go.

DISCLAIMER: Untested. 

thank you for the help... I will try this after work this evening! 

Newb question: would this mean I need to add a wait or delay key value to the entities? I don't mind setting it manually but wanted to know if this is how this method works. 

If you don't want to set a wait value manually, instead of using self.wait in that code you can generate a random number between 0 and 1 using random() instead.

Can you see how to adapt the code to use random to create a delay of between 0 and 1 seconds?

How about a delay between 0 and 2 seconds?

What about between 0.5 and 1 seconds? 

Obviously I posted that before reading the other thread. 

No worries thanks for the great tutorials. I am enjoying messing around with QC finally. Next, I am going to try and figure out a silent spawn without the tfog and audio next. I think I can do it. 

 

After trying to implement random timings I felt it better to go with self.delay. More work but more control for the mapper. Thanks everyone! 

/*==========
te_lightning
==========*/
void(vector v1, vector v2) te_lightning =
{
WriteByte (MSG_BROADCAST, SVC_TEMPENTITY);
WriteByte (MSG_BROADCAST, TE_LIGHTNING1);
WriteEntity (MSG_BROADCAST, world);
WriteCoord (MSG_BROADCAST, v1_x);
WriteCoord (MSG_BROADCAST, v1_y);
WriteCoord (MSG_BROADCAST, v1_z);
WriteCoord (MSG_BROADCAST, v2_x);
WriteCoord (MSG_BROADCAST, v2_y);
WriteCoord (MSG_BROADCAST, v2_z);
};

Writing some TE_ helper functions has gotten me curious. WHY is there an entity parameter needed for the lightning, beam, etc? Is it safe to just write world? 

I figured it out. It uses the entity as the start point regardless of the co-ordinates passed. And this is why the lightning beam starts at the origin of the player, regardless of anything else. 

beams last 0.2 seconds. QC respawns them every 0.1 seconds.

the entity argument is just there to avoid the resulting double-beams issue (trying to depend upon precise timing would be too unreliable with network latency etc, hence why they last longer).
Using world would break when you have two players firing beams, although some engines do understand world to mean 'don't replace any beams'.
tbh 99% of the time that entity arg will be self.

ignoring the startpoint and using the player's position is just to smooth things out so that the player doesn't leave the beam behind when moving sideways etc.

note that there's already an engine extension to provide those as builtins (with that name), so beware of potential conflicts if a modder uses *extensions.qc. 

original code has

LightningDamage (self.origin, trace_endpos + v_forward*4, self, 30);

(even URQP has it that way)

But i had long changed it to

LightningDamage (org, (trace_endpos + (v_forward * 4)), self, 30);

this is especially obvious when shooting from RL window on DM3 and aiming at someone on the lower bridge.
this is because:

org = self.origin + '0 0 16';
traceline (org, org + v_forward*600, TRUE, self); 

The original is "waist lightning". Darkplaces has specific settings to let you pick which style you prefer, waist or gunpoint. 

That only fixes the position of the bolt VISUALLY, which is what Spike was explaining. The above fixes the tracelines/damage as well. 

ya the original deals damage from the gun but visially it came from the hip
i keep confusing myself that sel.origin is the bottom of the model
where the feet touch the floor 

Still taking baby steps into QC with very little coding background. Wanted to check with experienced coders. This bit of code makes a trigger_push silent with a spawnflag. It appears to be working fine in game and no errors in FTEQCC but I feel like the syntax is still wrong.

Next steps will be to learn how to toggle it and add custom sounds.

//============================================================================

float PUSH_ONCE = 1;
float PUSH_SHHH = 2; //DMS push silently

void() trigger_push_touch =
{
if (other.classname == "grenade")
other.velocity = self.speed * self.movedir * 10;
else if (other.health > 0)
{
other.velocity = self.speed * self.movedir * 10;
if (other.classname == "player")
if (!(self.spawnflags & PUSH_SHHH)) //DMS
{
if (other.fly_sound < time)
{
other.fly_sound = time + 1.5;
sound (other, CHAN_AUTO, "ambience/windfly.wav", 1, ATTN_NORM);
}
}
}
if (self.spawnflags & PUSH_ONCE)
remove(self);
};


/*QUAKED trigger_push (.5 .5 .5) ? PUSH_ONCE
Pushes the player
*/
void() trigger_push =
{
InitTrigger ();
precache_sound ("ambience/windfly.wav");
self.touch = trigger_push_touch;
if (!self.speed)
self.speed = 1000;
};

//============================================================================ 

You could also check the spawnflags in the spawn function and only precache the sound if needed.