 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Here the particle system is used to simulate a trail of smoke behind a
missile-like thing.
Each particle is colored dependent on its time of birth, and its density is
controlled by its age. Also, the particles are automatically shot backwards
relative to the movement direction of the emitter.
Chris Huff, although I've read the particle_system patch documentation, you
still know it much better than me. Could you tell me which of the features
mentioned above are also possible with the particle_system patch?
Rune
--
\ Include files, tutorials, 3D images, raytracing jokes,
/ The POV Desktop Theme, and The POV-Ray Logo Contest can
\ all be found at http://rsj.mobilixnet.dk (updated January 28)
/ Also visit http://www.povrayusers.org
Post a reply to this message
Attachments:
Download 'smoketrail.mpg' (60 KB)
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <3a97b94c@news.povray.org>, "Rune" <run### [at] iname com>
wrote:
> Chris Huff, although I've read the particle_system patch
> documentation, you still know it much better than me. Could you tell
> me which of the features mentioned above are also possible with the
> particle_system patch?
All of them (sort of). In the current patch, you would have to do some
of it yourself, running the particle system for short intervals of time
and updating the position of the particle emitter between them, but this
will probably lead to artifacts from the emitter jerkily moving from
place (aliasing, like those "puffs" of smoke from your rocket), and will
be slowed down by the processing overhead of POV Script.
I'll be working on a dynamic_transform or transform_spline that will
interpolate between transformations, allowing things like emitters and
objects to move while the particle system runs. Hopefully, it will also
be useful for ordinary animations...
The age of each particle is available, so you could do your own
particles using that information. (the time of birth, of course, is just
current_time-particle_age)
--
Christopher James Huff
Personal: chr### [at] maccom, http://homepage.mac.com/chrishuff/
TAG: chr### [at] tagpovrayorg, http://tag.povray.org/
<><
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Chris Huff" wrote:
> In the current patch, you would have to do some of it yourself,
> running the particle system for short intervals of time and
> updating the position of the particle emitter between them,
Why? Does it not support moving emitter-points directly?
> but this will probably lead to artifacts from the emitter
> jerkily moving from place (aliasing, like those "puffs" of
> smoke from your rocket),
Those puffs are not caused by aliasing. Each puff is only one particle, not
a group of particles. They're caused by a low particle_frequency. If the
particle_frequency were high enough, there would be no visible puffs, even
if the particle_frequency were higher than the calculation frame-rate.
> I'll be working on a dynamic_transform or transform_spline that
> will interpolate between transformations, allowing things like
> emitters and objects to move while the particle system runs.
Moving particle-emitters are already supported in my system, but not moving
environment-objects.
> Hopefully, it will also be useful for ordinary animations...
I think it would be very useful to be able to interpolate transformations. I
have often done similar things myself (in POV-script).
> The age of each particle is available
Not in the current version, according to the documentation, but it does say
that it will be available in future versions...
Rune
--
\ Include files, tutorials, 3D images, raytracing jokes,
/ The POV Desktop Theme, and The POV-Ray Logo Contest can
\ all be found at http://rsj.mobilixnet.dk (updated January 28)
/ Also visit http://www.povrayusers.org
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <3a97d688@news.povray.org>, "Rune" <run### [at] inamecom>
wrote:
> Why? Does it not support moving emitter-points directly?
No. There isn't any way to specify that type of information yet, that's
why I'm working on the dynamic_transform patch. I wanted to use a method
that was at least as flexible as ordinary clock-based animation, so I
didn't want to use splines.
> Moving particle-emitters are already supported in my system, but not
> moving environment-objects.
One problem I have yet to figure out how to get around is the "trapping"
of particles inside the environment object...maybe transforming
particles within a certain distance of the object (close enough to be
considered "on" it) along with the object, and modifying their velocity
will work.
> > The age of each particle is available
> Not in the current version, according to the documentation, but it
> does say that it will be available in future versions...
Oops! Well, my version in progress supports radius, mass, and
temperature as well as age, velocity, and position.
--
Christopher James Huff
Personal: chr### [at] maccom, http://homepage.mac.com/chrishuff/
TAG: chr### [at] tagpovrayorg, http://tag.povray.org/
<><
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Chris Huff" wrote:
> "Rune" wrote:
>
> > Why? Does it not support moving emitter-points directly?
>
> No. There isn't any way to specify that type of information yet,
> that's why I'm working on the dynamic_transform patch. I wanted
> to use a method that was at least as flexible as ordinary
> clock-based animation, so I didn't want to use splines.
I completely agree. That's why the particle_emitter point and direction can
be animated exactly as flexible as regular clock animation in my system. I
have not used splines in any of my posted particle system animations.
> > Moving particle-emitters are already supported in my system,
> > but not moving environment-objects.
>
> One problem I have yet to figure out how to get around is the
> "trapping" of particles inside the environment object...
Yes, that problem we have in common...
> maybe transforming particles within a certain distance of the
> object (close enough to be considered "on" it) along with the
> object, and modifying their velocity will work.
But what if the object moves a greater distance between two calculation
frames than that certain distance you mentioned? And what if the surface
movement is not caused by a transformation but by - say an isosurface with a
varying function?
> Oops! Well, my version in progress supports radius, mass, and
> temperature as well as age, velocity, and position.
I have never understood what use the temperature feature is. What can it be
used to simulate?
From the documentation:
> As the simulation goes on, particles will tend toward the
> temperature of their environment. Particles hotter than
> their environment will tend to go against gravity and float,
> colder particles will sink.
I don't see the logic in this. It has noting to do with reality. Hot water
doesn't float just because it's hotter than the air around it. The
temperature at which it begins to float is much higher than the temperature
of the environment. In fact, the temperature of the environment seems to be
completely irrelevant. And wouldn't your system end up in a state where the
particles neither sink nor float?
Please explain the concept.
Rune
--
\ Include files, tutorials, 3D images, raytracing jokes,
/ The POV Desktop Theme, and The POV-Ray Logo Contest can
\ all be found at http://rsj.mobilixnet.dk (updated January 28)
/ Also visit http://www.povrayusers.org
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <3a98066b@news.povray.org>, "Rune" <run### [at] inamecom>
wrote:
> But what if the object moves a greater distance between two
> calculation frames than that certain distance you mentioned? And what
> if the surface movement is not caused by a transformation but by -
> say an isosurface with a varying function?
That would still be a problem. I'll probably add a feature to kill
particles that end up inside an environment object, so at least time
won't be wasted processing them.
> I have never understood what use the temperature feature is. What can
> it be used to simulate?
Mainly convection and the effects of smoke rising and cooling.
> From the documentation:
> > As the simulation goes on, particles will tend toward the
> > temperature of their environment. Particles hotter than
> > their environment will tend to go against gravity and float,
> > colder particles will sink.
>
> I don't see the logic in this. It has noting to do with reality. Hot
> water doesn't float just because it's hotter than the air around it.
Huh? Water normally doesn't float in air...
> The temperature at which it begins to float is much higher than the
> temperature of the environment. In fact, the temperature of the
> environment seems to be completely irrelevant.
The reason convection happens is the density difference between one mass
of air, and the surrounding air. The density is affected by the
temperature, which will cause a warmer mass of air to rise. And there is
no threshold of temperature difference below which it won't happen. It
can be on the order of a couple degrees, which is why we have weather.
> And wouldn't your system end up in a state where the particles
> neither sink nor float?
It will, eventually, if you have no heat sources or emitters of
unusually hot or code particles. In the same way that smoke will
eventually disperse throughout a room.
--
Christopher James Huff
Personal: chr### [at] maccom, http://homepage.mac.com/chrishuff/
TAG: chr### [at] tagpovrayorg, http://tag.povray.org/
<><
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Rune" <run### [at] inamecom> wrote in message
news:3a97b94c@news.povray.org...
> Here the particle system is used to simulate a trail of smoke behind a
> missile-like thing.
That is a very realistic (for POV) animation. I mean discounting the
missile and the colored smoke that fades a little jerkily and all the other
things one should discount...
What kind of render time did you get for that.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Chris Huff" wrote:
> "Rune" wrote:
> > I have never understood what use the temperature feature is.
> > What can it be used to simulate?
>
> Mainly convection and the effects of smoke rising and cooling.
Aha.
> > From the documentation:
> > > As the simulation goes on, particles will tend toward the
> > > temperature of their environment. Particles hotter than
> > > their environment will tend to go against gravity and float,
> > > colder particles will sink.
> >
> > I don't see the logic in this. It has noting to do with reality. Hot
> > water doesn't float just because it's hotter than the air around it.
>
> Huh? Water normally doesn't float in air...
When water is below 100 degrees Celsius / 212 degrees Fahrenheit it falls.
When it is above, it rises. The "threshold" is at 100 degrees Celsius, not
at the temperature of the environment. So the "rule" does not apply to
water, not always anyway.
> The reason convection happens is the density difference between
> one mass of air, and the surrounding air. The density is affected
> by the temperature, which will cause a warmer mass of air to rise.
But the particles themselves are not air, right? They just drift around with
the air surrounding them. So if smoke is surrounded by hot air, it will
drift upwards along with the air. However, all this only applies to
materials that have densities very close to the density of air (like smoke).
It doesn't apply to many other materials such as cool water which does not
normally drift along with air, even if its temperature is higher than the
environment temperature.
Since the temperature feature does not apply to all uses of the system, can
it be turned off?
Rune
--
\ Include files, tutorials, 3D images, raytracing jokes,
/ The POV Desktop Theme, and The POV-Ray Logo Contest can
\ all be found at http://rsj.mobilixnet.dk (updated January 28)
/ Also visit http://www.povrayusers.org
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
In article <3a983257@news.povray.org>, "Rune" <run### [at] inamecom>
wrote:
> > Huh? Water normally doesn't float in air...
>
> When water is below 100 degrees Celsius / 212 degrees Fahrenheit it
> falls. When it is above, it rises. The "threshold" is at 100 degrees
> Celsius, not at the temperature of the environment. So the "rule"
> does not apply to water, not always anyway.
Uh, water boils above 100C. This is phase change, not convection. My
patch doesn't simulate phase changes. (well, it might, if you use
inter-particle forces and simulate temperature change by changing the
velocity of the particles, but it has nothing to do with the temperature
feature of atmospheric particles)
> > The reason convection happens is the density difference between one
> > mass of air, and the surrounding air. The density is affected by
> > the temperature, which will cause a warmer mass of air to rise.
>
> But the particles themselves are not air, right? They just drift
> around with the air surrounding them. So if smoke is surrounded by
> hot air, it will drift upwards along with the air. However, all this
> only applies to materials that have densities very close to the
> density of air (like smoke). It doesn't apply to many other materials
> such as cool water which does not normally drift along with air, even
> if its temperature is higher than the environment temperature.
The particle system doesn't simulate the interactions of different kinds
of fluids yet. The atmospheric particles simply assume the environment
has a very similar density when at the same temperature. And as I said
before, it doesn't do phase change, so you can't simulate the boiling
off of water into steam which then rises into the air with a single
group of particles, you will have to use separate particles for the
liquid water and for the steam, maybe separate systems.
I plan on eventually giving each particle a density value, along with
support of multiple fluid types.
> Since the temperature feature does not apply to all uses of the
> system, can it be turned off?
It only applies to atmospheric particles.
--
Christopher James Huff
Personal: chr### [at] maccom, http://homepage.mac.com/chrishuff/
TAG: chr### [at] tagpovrayorg, http://tag.povray.org/
<><
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Bill DeWitt" wrote:
> That is a very realistic (for POV) animation. I mean
> discounting the missile and the colored smoke that fades
> a little jerkily and all the other things one should
> discount...
I didn't go for realism... :)
> What kind of render time did you get for that.
I used emission and absorption media, method 3, with only one sample. I
think though that more samples are created internally, since the spheres
overlap so much.
The animation is 40 frames. It took perhaps 50-55 sec. a frame. 48 of those
are tracing.
And note that my computer is a slow P150 w. 16 MB RAM.
Rune
--
\ Include files, tutorials, 3D images, raytracing jokes,
/ The POV Desktop Theme, and The POV-Ray Logo Contest can
\ all be found at http://rsj.mobilixnet.dk (updated January 28)
/ Also visit http://www.povrayusers.org
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
|
|
