Particle rope and chain question

[BadEddy]

any way to make ropes and chain look better? By that I mean like on this image bellow:

http://imageshack.us/a/img845/126/particlerope001.jpg

as you can see on the image the rope is not a continuous rope and is cut at many places. I find that annoying. Any tricks to prevent that ?

 

[Freya Mokusei]

The short answer is no. The longer answer is.... not really.

Ropes and chains are done using particle physics, 2D sprites that 'stretch' between an emitter object and a target object. The emitter spawns particles and fires them at the target in quick succession. The 'rope' doesn't exist, it's only a series of 2D images being propelled across the 3D space. The rendering of this process is carried out entirely client-side, so the breakages you see are different to the breakages another person would see, and there's no way to make them attach 'end to end'. There is no way to standardise this appearance, as two peoples' viewers and computers can vary significantly.

You can improve the illusion by using more accurate values, or by not moving around, but you can't 'fix' the rope, because the particles are functioning in precisely the manner they were designed.

There IS a new trick for tying rope between two body parts on the same avatar, however: Rigged mesh.

[Madelaine McMasters]

---

BadEddy wrote:

any way to make ropes and chain look better? By that I mean like on this image bellow:

http://imageshack.us/a/img845/126/particlerope001.jpg

as you can see on the image the rope is not a continuous rope and is cut at many places. I find that annoying. Any tricks to prevent that ?

 

---

Hiya BadEddy:

It's possible you simply aren't emitting enough particles/second to avoid gaps in the stream. Try decreasing PSYS_SRC_BURST_RATE (I don't know why they call it rate, it's an interval). You can set it to zero to burst particles as fast as the viewer will allow. You might also try a PSYS_SRC_BURST_PART_COUNT greater than one, though I'm not sure that'll help. If all particles in the burst are rezzed in the same location and undergo the same forces, they'll appear as one particle.

Even if these changes improve things, you may still see gaps in the rope if your viewer is bogged down with too many particles, either from your own emitter or others within view. So make sure your viewer's particle count is set high enough and that  you aren't in a place filled with other particle emitters.

Good luck!

ETA: As Freya mentioned while I was composing this, motion at either the emitting or target end of the rope may also cause breaks, as the particles are like a stream of droplets emitted from a garden hose. If either end of the rope moves, the particles may not appear fast enough to cover the distance traveled by either end.

[Rolig Loon]

It looks to me as if the gaps are there because either the particles aren't the right size or they are just a little too far apart.  That's sort of a clumsy particle rope, though, since all the segments are long and straight.  It would be wiser to use much shorter segments.  That way, you could get a rope that moves like a real rope when either the emitter or the target moves.

Visit the Particle Lab in world and spend some time playing with the demos and doing experiments in their sandbox.    >>>>  http://maps.secondlife.com/secondlife/Teal/200/61/22

EDIT:  Ooooops.  I pasted the wrong SLURL in.  Just fixed it.  Sorry about that.  :smileyembarrassed:

[Duckie Dickins]

LL is developing a new particle type called ribbon which should solve this issue. We were shown an image of what ribbon would look like at the beta server user group held every thursday at 3pm in the morris region on the beta grid by Maestro  Linden. There wasn't an ETA for the new particle type but its being worked on and is on the "coming soon" list.

[Madelaine McMasters]

---

Duckie Dickins wrote:

LL is developing a new particle type called ribbon which should solve this issue. We were shown an image of what ribbon would look like at the beta server user group held every thursday at 3pm in the morris region on the beta grid by Maestro  Linden. There wasn't an ETA for the new particle type but its being worked on and is on the "coming soon" list.

---

I like the "ribbon" euphemism. I wonder how many will use it for making ribbons.

;-)

[BadEddy]

I'm still struggling a little with this. 

What I did so far is to use a new rope texture for the particle, one that match some prim rope included in the built. So that part works well. 

Now I'm back to how the particle cut and does not give a nice long continuous line. So I play with the length and width of the particle as well as the length and width of the texture I'm using. seems like this might get me somewhere. I'm thinking that maybe the key is to set the length of the particle the same as that of the texture? 

Lockguard use something like that for size: 

size 0.6 0.6

how does that translate in pixel?

[Innula Zenovka]

I don't know how it translates into pixels, but those would be the x and y values for either (or more likely both, for a rope or chain) PSYS_PART_START_SCALE and PSYS_PART_END_SCALE and are measured in SL metres.  See the article on llParticleSystem for further details.

I think I would try decreasing the value of PSYS_SRC_BURST_RATE.

Do, though, keep in mind that particle effects aren't ever fully under your control -- how they appear to people depends, among other things, on how many particles they've set their viewer to be able to see, and how many particles there are on the sim.   There's all sorts of things can interfere with your system.

[BadEddy]

Yeah, I understand the sim and many other things play in this. I found out about that the day I was running that rain system all over the sim lol. 

All I'm trying to do is to get the best effect I can (without freaking out too much about it). when I increase the length of the particles (from the script) I noticed I get a lot less breaking. Thats good. Now the problem with that is the particle rope will not alway stop at the anchor point because of that. I guess I'll have to find the best between both problems. 

now I looked through the script for that PSYS_SRC_BURST_RATE and could not find anything like it in this lockguard script. 

 

thx

[Madelaine McMasters]

---

BadEddy wrote:

I'm still struggling a little with this. 

What I did so far is to use a new rope texture for the particle, one that match some prim rope included in the built. So that part works well. 

Now I'm back to how the particle cut and does not give a nice long continuous line. So I play with the length and width of the particle as well as the length and width of the texture I'm using. seems like this might get me somewhere. I'm thinking that maybe the key is to set the length of the particle the same as that of the texture? 

Lockguard use something like that for size: 

size 0.6 0.6

how does that translate in pixel?

---

The size you referenced is 0.6 x 0.6 meters, which seems huge for a li'l bit of rope. I'm also surprised to see that your particle is square when the rope is long and skinny. Is your texture mostly transparent, with just a thin length of rope visible in it? If so, you're making the graphics engine work harder, as it must think about a lot of pixels which are not visible. Given the shape of the rope pieces in your OP, I'd expect a particle size of something like 0.5 x 0.05 (or 0.05 x 0.5). The pixel dimension of a particle depends solely on the size of the applied texture. For the rope snippets you showed in your OP, I think a 256x16 texture would be sufficient.

If you still have gaps, you aren't emitting enough particles. As Innula suggested, decrease the burst rate. If you make the rope snippets longer, the illusion will break as you'll start to see the straight segments and the ends of the snippets won't match well. The right combination of burst rate and particle length is something you'll have to determine by experimentation.

[Innula Zenovka]

Actually, I was mistaken about the burst rate.. that's set at 0.0 in the basic LG script, I think (it's in the function lgLink.   That means "spit them out as fast as you can".   So, as well as playing with the size, as Madelaine suggests, try playing with the "life" parameter (if you're using the LG system, that is).   The shorter the particle's life, the faster it must move between the emitter and the target.

Is there a particular reason why you want to use the Lockguard system?

 

[Madelaine McMasters]

---

Innula Zenovka wrote:

The shorter the particle's life, the faster it must move between the emitter and the target.

 

---

Oooh, that makes perfect sense, but as I've never emitted particles at a target, this didn't occur to me. Neat!

llParticleSystem remains my favorite function, Innula. I must show you what I discovered with it some years back.

[Rolig Loon]

---

BadEddy wrote:

[ ...]

Lockguard use something like that for size: 

size 0.6 0.6

how does that translate in pixel?

---

I missed reading this earlier today because I'm away from home and have only spotty Internet access.  The answer, though, is that it doesn't translate into pixels at all.  Think about it.  How many pixels does it take to display a 1m x 1m x1m  cube on screen?  Well, how far away is it?  What's the resolution of your monitor?  Unless you know whether you are so close to the cube that it fills your screen or so far away that it's a flyspeck in the distance, asking how many pixels it takes is a meaningless question.

[BadEddy]

well, seems like I've done some good progress here. Thanks to you ladies.

Mostly, lowering  the life time of the particles and using a new texture of a much better size (and not just a thin rope surrounded by invisible) is what made the biggest differences. Its not totally perfect but at least I know what I get is the best I could. So I'm happy. 

 

And about LG, the only reason I use it is because its the one that came with the avsitter I use. Is there better options out there?