Particle flame - how to control the starting point?

Can anyone give me advice on how to control the starting point of a particle flame?

When I put the particle script into the prim the flame goes from the center of the prim rather from the top...

When I change the RADIUS from 0 to some positive number it lifts the flame and brings it to the top of the prim which is great but I see some few particles on the bottom of the prim too on the same distance from the center of the prim as the top particle..Its like its mirroring the top flame...

Does anyone know how to fix it or know any other way to bring the flame particle to the top of the prim without changing RADIUS from 0? Thanks a lot in advance!

And the other 20-odd parameters besides radius?

I don't know the answer to your question, but I have seen products with invisible prims containing the particle scripts, for the ostensible purpose of doing what you're asking.

For LOTS of information about particles, go to the Particle Lab! Search ingame. It's a blast. Kudos to its owner/creator, Jopsy Pendragon.

Ditto.. can't say enough about Jopsy's AWESOME Particle Lab. I scripted for a long time, and avoided learning the ins and outs of the particle function for an equally long time. Then, I went to the Particle Lab in Teal, and took about an hour and a half or so, and went from kiosk to kiosk, reading the short, but super easy to understand explanations of the various particle system controls and flags, played with/watched the examples shown, etc. Now, whenever I want to do something with particles, it's no longer a hit and miss with editing a particle script, saving, seeing the changes, repeat. But rather now, I get the idea in my head what I want, go right to the flags, make the changes, and it works as I want it to...

another great way is a way i use for 1 prim candless. do this:

Rez a Cylinder, make it the size you want, except make it twice as tall.
Change it to a sphere, dimple the end to 0.500
Change it back to a cylinder.

It should be the size you want, and it shows the center point at the top, That's where the particles will start. You can also try dimpling the sphere to 0.475 and then changing it to a cylinder, this will make the particles start slightly above the candle. (so it doesn't look like the flame is slightly imbedded in the the candle)

**blah i can't type lol, edited to fix that

Cool, Ruthven -- thanks

Post #7 of /54/ae/255682/1.html gives an example of how to use a parameter to raise the starting point of a candle flame without having the "few particles on the bottom of the prim"

Hard to know for sure without seeing the actual script you are using, but my hunch is that the problem you are having is being caused by the pattern you are using. If you are not already doing so, try using pattern PSYS_SRC_PATTERN_ANGLE

Regardless, the above post gives a particle system scripting solution; you'll just need to change the colour, alpha, texture, speed etc parameters on it to match those which your current script is using.

right, but the problem with that is, if you later resize the prim, you'd have to re-adjust the radius. this is why i use the dimple method. also, just a thought, although i'm not in world to try it, is there some way to get the script to detect the prim size and readjust the radius accordingly?
IE. using some form of code like this for the burst radius: Get Prim Size.Z* 0.5?

True; there is alot to be said for the dimple method you suggest. I'd never previously considered that as a solution but it one more option in the toolbox. Thanks

The script I posted, however, does answer both OP's original questions and, althou' I didnt post the section of script involved, I do have particle scripts that adjust the radius based on the size of prim.

Depending on the circumstances, I would use llGetPrimitiveParams or llGetScale to determine the prim size, and multiple by an adjustment factor

Sure.. The llGetScale function returns a vector equal to the size of the prim.

http://wiki.secondlife.com/wiki/LlGetScale

ok, i used that and i got the scale vector, but of course the radius isn't a vector, not sure how to make the script tell the scale.z to the radius to use it for the calculation

The scale x and y are the diameter of a cylinder and z is the height. You probably want to do something like this to get the radius:

vector size = llGetScale();
float radius;

if (size.x > size.y)
radius = size.y / 2.0;
else
radius = size.x / 2.0;

ah ok, that looks like what i was looking for, i'll try it when i go in world later, thanks sindy

PS: why does it say size.x, and size.y, when the #i'm looking for is from size.z? i'm still a very newbie scripter, my script skills is limited to editing other scripts to fine tune it to my preferences, so comments in scripts are my friend lol. and thanks again for the help

I think Sindy is just giving the direct answer to the "radius of a cylinder?" question. For a cylinder, size.z is the length and x & y are the diameter widths - her code snippet does 1/2 the diameter of the smallest width as the radius.

so would i want it to look more like the following?

vector size = llGetScale();
float radius;
radius = size.z / 2.0;

alright, so i did as i wrote in the previous post, works perfectly! thanks for the help, and i guess i'm learning more than i thought i was lol

edit to add this, hmm, it works for explosions at least, tried it on a cone particle system, didn't seem to do anything

The various different particle system parameters are not necessaily transferable between the different patterns, as you have just discovered. An absolutely fabulous effect can quickly degrade into complete junk, just be changing the system pattern. Of course the reverse can be true too, leading to some pleasant surprises.

The explode pattern is not the natural choice for a candle flame. Thats not to say you can not produce a candle flame from an explode pattern...you most definately can. But to do so requires setting other parameters to "contort" and "force" the particles to behave in a somewhat angle-cone-type way. It is little wonder, then, that when you changed from explode to angle-cone that the results were disappointing.

The example below, based on Post #7 of /54/ae/255682/1.html, uses an angle cone pattern and calculates the particle system radius based on the prim size. Note, however, that the multiplier factor is likely to have changed from 2, and this is another possible reason why you are not getting the results you expected

ParticleStart(Radius)
{
llParticleSystem([
PSYS_PART_FLAGS, 0x103,
PSYS_SRC_PATTERN, 0x004,
PSYS_PART_START_ALPHA, 0.50,
PSYS_PART_END_ALPHA, 0.00,
PSYS_PART_START_COLOR, <1.00,1.00,0.00>,
PSYS_PART_END_COLOR, <1.00,0.00,0.00>,
PSYS_PART_START_SCALE, <0.06,0.12,0.00>,
PSYS_PART_END_SCALE, <0.01,0.01,0.00>,
PSYS_PART_MAX_AGE, 0.70,
PSYS_SRC_MAX_AGE, 0.00,
PSYS_SRC_ACCEL, <0.00,0.00,0.04>,
PSYS_SRC_ANGLE_BEGIN, 0.00,
PSYS_SRC_ANGLE_END, 0.09,
PSYS_SRC_BURST_PART_COUNT, 1,
PSYS_SRC_BURST_RADIUS, Radius,
PSYS_SRC_BURST_RATE, 0.05,
PSYS_SRC_BURST_SPEED_MIN, 0.09,
PSYS_SRC_BURST_SPEED_MAX, 0.13,
PSYS_SRC_OMEGA, <0.00,0.00,0.00>,
PSYS_SRC_TEXTURE, "b708e484-aeef-8cc9-3896-dd661d4e6807"
]);

}
default
{
state_entry()
{
vector PrimSize = llGetScale();
float DerivedRadius = size.z / 2.0;
ParticleStart(DerivedRadius);
}
}

[/php]

actually, i used a similar code like this but i put it in the particle state, (particle uses on off switch, and automatically updates if you've changed the prim size when you re-activate the particles.) i know you can also write in somewhere so that the script detects if the prim has been changed and resets the script or whatever setting needs to be changed for you, but not sure about how to go about doing that. that would be usefull so you don't have to manually update the script each time you want to change the prim size, or if you're putting it in a mod product but have the script no-mod.