Because the glow fire works on a black background only I've had to
experiment with a media solution. This is the result so far. It's not
perfect but it's a start.

Media has both advantages and disadvantages compared to glows. Media works
on all backgrounds but it is also slower. With media one have greater
control over every aspect, but that also make it more difficult to control.
Media is more precise (for example the turbulence is really 3D), but because
of the complexity, errors are also more likely to occur. (There are a few
annoying artefacts in the media in this animation and I don't know what
causes it.)

In this animation I tried placing the fire in a simple scene, and I think it
works well. I also tried attaching a light_source to 20% of the particles
(there are about 30 particles alive at a time), but I don't think it works
particularly well. The fire doesn't flicker enough to the sides to make it
worth while.

Anyway, as always I'd really appreciate your feedback!

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 March 29)
/ Also visit http://www.povrayusers.org

Post a reply to this message

Attachments:
Download 'mediafire.mpg' (164 KB)

Rune wrote:
> 
> Because the glow fire works on a black background only I've had to
> experiment with a media solution. This is the result so far. It's not
> perfect but it's a start.
> 

Quite impressive, as already said some more fire like structures would be
good, meaning sharper border and somehow more splitting it up into
flames.  I don't know if that's practicable with the current particle
system approach.  

> 
> In this animation I tried placing the fire in a simple scene, and I think it
> works well. I also tried attaching a light_source to 20% of the particles
> (there are about 30 particles alive at a time), but I don't think it works
> particularly well. The fire doesn't flicker enough to the sides to make it
> worth while.
> 

The moving shadows are looking interesting, maybe radiosity (or even only
radiosity lighting) would be worth trying.  

Christoph

-- 
Christoph Hormann <chr### [at] gmxde>
IsoWood include, radiosity tutorial, TransSkin and other 
things on: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

This looks like a gigantic candle. Pretty, though.

Post a reply to this message

Maybe my imagination (it *is* late at night right now), but I think I see a
ghostly face in the top of the flames, just for an instant.Strange artifacts
though.

I wonder what the overall effect would be if you had two or three of these
flames side by side, but close enough to look as though they were part of
the same fire, each having different turbulence settings, and each
interacting. AFAIK, fires like the one in your animation don't have a single
flame 'source', instead having their flames distributed over a circular
area.

Just a thought. Good work.

Andy

"Rune" <run### [at] inamecom> wrote in message
news:3acb8527@news.povray.org...
> Because the glow fire works on a black background only I've had to
> experiment with a media solution. This is the result so far. It's not
> perfect but it's a start.
>
> Media has both advantages and disadvantages compared to glows. Media works
> on all backgrounds but it is also slower. With media one have greater
> control over every aspect, but that also make it more difficult to
control.
> Media is more precise (for example the turbulence is really 3D), but
because
> of the complexity, errors are also more likely to occur. (There are a few
> annoying artefacts in the media in this animation and I don't know what
> causes it.)
>
> In this animation I tried placing the fire in a simple scene, and I think
it
> works well. I also tried attaching a light_source to 20% of the particles
> (there are about 30 particles alive at a time), but I don't think it works
> particularly well. The fire doesn't flicker enough to the sides to make it
> worth while.
>
> Anyway, as always I'd really appreciate your feedback!
>
> 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 March 29)
> / Also visit http://www.povrayusers.org
>
>
>

Post a reply to this message

Good work there.

The lighting has quite a striking effect there. Nice.

Maybe if you add smoke the artefacts wouldn't be so obvious, but that
wouldn't really solve the problem...

Real flames are transparent at the centre. Do you think you could try that?

Nekar

"Rune" <run### [at] inamecom> wrote in message
news:3acb8527@news.povray.org...
> Because the glow fire works on a black background only I've had to
> experiment with a media solution. This is the result so far. It's not
> perfect but it's a start.
>
> Media has both advantages and disadvantages compared to glows. Media works
> on all backgrounds but it is also slower. With media one have greater
> control over every aspect, but that also make it more difficult to
control.
> Media is more precise (for example the turbulence is really 3D), but
because
> of the complexity, errors are also more likely to occur. (There are a few
> annoying artefacts in the media in this animation and I don't know what
> causes it.)
>
> In this animation I tried placing the fire in a simple scene, and I think
it
> works well. I also tried attaching a light_source to 20% of the particles
> (there are about 30 particles alive at a time), but I don't think it works
> particularly well. The fire doesn't flicker enough to the sides to make it
> worth while.
>
> Anyway, as always I'd really appreciate your feedback!
>
> 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 March 29)
> / Also visit http://www.povrayusers.org
>
>
>

Post a reply to this message

Rune <run### [at] inamecom> wrote:

> Because the glow fire works on a black background only I've had to
> experiment with a media solution. This is the result so far. It's not
> perfect but it's a start.

> Media has both advantages and disadvantages compared to glows. Media works
> on all backgrounds but it is also slower. With media one have greater
> control over every aspect, but that also make it more difficult to control.
> Media is more precise (for example the turbulence is really 3D), but because
> of the complexity, errors are also more likely to occur. (There are a few
> annoying artefacts in the media in this animation and I don't know what
> causes it.)

Yeah, those artifacts are really strange.  You're using a blob, right?  Have
you tried looking at the shape of the blob in those frames?  Is it really
the media being clear, or is it the media being thin in those spots?  For
some reason (shape of the spots?), I think it might be the former, and if
so, that can be fixed.  If the latter, maybe different turbulence would
help.

> In this animation I tried placing the fire in a simple scene, and I think it
> works well. I also tried attaching a light_source to 20% of the particles
> (there are about 30 particles alive at a time), but I don't think it works
> particularly well. The fire doesn't flicker enough to the sides to make it
> worth while.

I thought the flickers were really nice, personally, especially on the back
stones.

I agree with others about the multiple sources.  That might also help the
flicker.  However, when doing multiple sources, overlapping might be a
problem (unless they're a single media object, which might not be too bad)
Also, the flames to the outside get drawn in towards the center somewhat,
which might be tricky to model, although an attraction force between the
particles may be sufficient.  In fact, that might be something you want to
do anyway, as the current flame is a little too broad at the top.  The point
effect is due to the hottest parts rising faster and the surrounding area
getting drawn in, after all.

Geoff

Post a reply to this message

"Christoph Hormann" wrote:
> Quite impressive, as already said some more fire like
> structures would be good, meaning sharper border and
> somehow more splitting it up into flames.

I suppose I came closer in the "Glow Fire take 3" approach?
This media is difficult to control! :(

> The moving shadows are looking interesting, maybe
> radiosity (or even only radiosity lighting) would be
> worth trying.

No thanks, not right now.
Radiosity in stills is slow enough for me! ;)

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 March 29)
/ Also visit http://www.povrayusers.org

Post a reply to this message

"Tony[B]" wrote:
> This looks like a gigantic candle. Pretty, though.

I'll remember when I need to make a gigantic candle! ;)

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 March 29)
/ Also visit http://www.povrayusers.org

Post a reply to this message

"Andy Cocker" wrote:
> I wonder what the overall effect would be if you had two or
> three of these flames side by side, but close enough to look
> as though they were part of the same fire, each having
> different turbulence settings, and each interacting. AFAIK,
> fires like the one in your animation don't have a single
> flame 'source', instead having their flames distributed over
> a circular area. Just a thought.

I'll try some of your ideas.
It should all be possible with just a single emitter though.

>  Good work.

Thanks!

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 March 29)
/ Also visit http://www.povrayusers.org

Post a reply to this message

"Nekar Xenos" wrote:
> Good work there.

Thanks!

> The lighting has quite a striking effect there. Nice.

I'm not quite satisfied with it myself.

> Real flames are transparent at the centre.

Are you sure? I mean, a candle light is, but not a camp fire I believe.
Reference images would be great.

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 March 29)
/ Also visit http://www.povrayusers.org

Post a reply to this message

"Geoff Wedig" wrote:
> Yeah, those artifacts are really strange.

Yeah.

> You're using a blob, right?

No. I considered using a blob with a blob_pattern density but had to reject
the idea, as it doesn't allow me to assign separate turbulence to the
individual particles, which I need to.

I'm using a merge of spheres. Inside the merge is a single interior with
multiple media statements, one for each particle.

Using a union of spheres forces a whole lot of samples to be taken because
of the many intersections; that's why I use merge.

When using merge I can't place the media inside the individual spheres (that
doesn't work correctly). So I place all the media statements in the parent
object (the merge).

Using this method is fairly fast.

> I thought the flickers were really nice, personally,
> especially on the back stones.

I think it would be both faster and prettier to use just one light_source
(possible area_light) and manually moving it around a little and change the
brightness.

> I agree with others about the multiple sources.

Instead of using multiple sources I could use one source that moves around
in a circle really fast (currently possible) or a source that is an object
(not yet implemented).

> when doing multiple sources, overlapping might be a problem

I'm already using overlapping media if that's what you mean, and I've used
it often before. Could that be causing the artefacts?

> Also, the flames to the outside get drawn in towards the
> center somewhat, which might be tricky to model, although
> an attraction force between the particles may be sufficient.

Sorry, I've decided not to implement any kind of inter-particle forces in my
system.

> In fact, that might be something you want to do anyway,
> as the current flame is a little too broad at the top.

Real camp fire isn't necessarily cone-shaped is it?

Thanks for the feedback!

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 March 29)
/ Also visit http://www.povrayusers.org

Post a reply to this message

Rune <run### [at] inamecom> wrote:

> "Geoff Wedig" wrote:
>> Yeah, those artifacts are really strange.

> Yeah.

>> You're using a blob, right?

> No. I considered using a blob with a blob_pattern density but had to reject
> the idea, as it doesn't allow me to assign separate turbulence to the
> individual particles, which I need to.

> I'm using a merge of spheres. Inside the merge is a single interior with
> multiple media statements, one for each particle.

Ok, I thought those artifacts looked like spherical shapes.  I really think
if you do a fast anim of the spheres themselves (opaque, in other words)
that you'll find the same holes in the surface.

> Using this method is fairly fast.

Yeah, and it should work, with some fine tuning.

>> I thought the flickers were really nice, personally,
>> especially on the back stones.

> I think it would be both faster and prettier to use just one light_source
> (possible area_light) and manually moving it around a little and change the
> brightness.

Hmm, maybe take a weighted average of the particles' locations would work. 
Ie, weight based upon the strengths (amount of emission) of the individual
bits.

Or maybe use two lights (areas), one large one near the bottom as above, but
one up higher that only uses the highest points.  That is where most of the
flicker comes from, anyway.

> Instead of using multiple sources I could use one source that moves around
> in a circle really fast (currently possible) or a source that is an object
> (not yet implemented).

Same effect, really.

> I'm already using overlapping media if that's what you mean, and I've used
> it often before. Could that be causing the artefacts?

No, the artefacts are caused (well, assuming I'm right) by places outside
the container.

> Sorry, I've decided not to implement any kind of inter-particle forces in my
> system.

Drat. :/

>> In fact, that might be something you want to do anyway,
>> as the current flame is a little too broad at the top.

> Real camp fire isn't necessarily cone-shaped is it?

Not really, no.  More of a teardrop.

Geoff

Post a reply to this message

i like it.  except that i think the hotter areas (yellow and white)
would be located 'on' the burning object, not above it.  looks great
though!

Rune wrote:
> 
> Because the glow fire works on a black background only I've had to
> experiment with a media solution. This is the result so far. It's not
> perfect but it's a start.
> 
> Media has both advantages and disadvantages compared to glows. Media works
> on all backgrounds but it is also slower. With media one have greater
> control over every aspect, but that also make it more difficult to control.
> Media is more precise (for example the turbulence is really 3D), but because
> of the complexity, errors are also more likely to occur. (There are a few
> annoying artefacts in the media in this animation and I don't know what
> causes it.)
> 
> In this animation I tried placing the fire in a simple scene, and I think it
> works well. I also tried attaching a light_source to 20% of the particles
> (there are about 30 particles alive at a time), but I don't think it works
> particularly well. The fire doesn't flicker enough to the sides to make it
> worth while.
> 
> Anyway, as always I'd really appreciate your feedback!
> 
> 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 March 29)
> / Also visit http://www.povrayusers.org
> 
>                     Name: mediafire.mpg
>    mediafire.mpg    Type: WMPLAYER File (video/mpg)
>                 Encoding: x-uuencode

Post a reply to this message

On Thu, 5 Apr 2001 17:24:08 +0200, "Rune" <run### [at] inamecom> wrote:

>
>> Real flames are transparent at the centre.
>
>Are you sure? I mean, a candle light is, but not a camp fire I believe.
>Reference images would be great.
>
Two here Rune.  A street bonfire in Guingamp some years ago.
Hope these are some help.
----------------------
dav### [at] hamiltonitecom
http://hamiltonite.com/

Post a reply to this message

Attachments:
Download 'GuingampFire1.jpg' (78 KB) Download 'GuingampFire3.jpg' (41 KB)

Preview of image 'GuingampFire1.jpg'

Preview of image 'GuingampFire3.jpg'

Rune wrote:
> 
> No thanks, not right now.
> Radiosity in stills is slow enough for me! ;)
> 

With glows it can be reasonably fast.  I remember i made a short
animations with glows and radiosity:

Subject: New Glow (296k)
Date: Tue, 12 Sep 2000 16:47:54 +0200
Newsgroups: povray.binaries.animations

Christoph

-- 
Christoph Hormann <chr### [at] gmxde>
IsoWood include, radiosity tutorial, TransSkin and other 
things on: http://www.schunter.etc.tu-bs.de/~chris/

Post a reply to this message

> >Are you sure? I mean, a candle light is, but not a camp fire I believe.
> >Reference images would be great.
> >
> Two here Rune.  A street bonfire in Guingamp some years ago.
> Hope these are some help.

I have some bonfire photos on my website too.
Check out the last 5 images here...

http://www.xlcus.co.uk/photos/1999-11-06-fireworks/


TTFN,

--
Jonathan Hunt
http://www.xlcus.co.uk/

Post a reply to this message

"David Wilkinson" wrote:
> >Reference images would be great.
>
> Two here Rune.  A street bonfire in Guingamp some years ago.
> Hope these are some help.

Thanks, they're great reference images!

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 March 29)
/ Also visit http://www.povrayusers.org

Post a reply to this message

"Geoff Wedig" wrote:
> Ok, I thought those artifacts looked like spherical shapes.
> I really think if you do a fast anim of the spheres themselves
> (opaque, in other words) that you'll find the same holes in
> the surface.

No, I think they're caused by a low number of samples being used.
More specifically, one sample...

> > I think it would be both faster and prettier to use
> > just one light_source
>
> Hmm, maybe take a weighted average of the particles'
> locations would work.

Good idea! I'll try that.

> > Instead of using multiple sources I could use one source
>
> Same effect, really.

Exactly.

> Not really, no.  More of a teardrop.

Well, I think it's often more irregular in the shape.

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 March 29)
/ Also visit http://www.povrayusers.org

Post a reply to this message

"Ryan Constantine" wrote:
> i like it.

Thanks!

> except that i think the hotter areas (yellow and white)
> would be located 'on' the burning object, not above it.

I'll try moving the fire a bit downwards.

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 March 29)
/ Also visit http://www.povrayusers.org

Post a reply to this message

"Christoph Hormann" wrote:
> Rune wrote:
> >
> > No thanks, not right now.
> > Radiosity in stills is slow enough for me! ;)
>
> With glows it can be reasonably fast.

Well, I'm not using glows...

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 March 29)
/ Also visit http://www.povrayusers.org

Post a reply to this message

Rune wrote in message <3acceeac$1@news.povray.org>...
>"Geoff Wedig" wrote:
>> Ok, I thought those artifacts looked like spherical shapes.
>> I really think if you do a fast anim of the spheres themselves
>> (opaque, in other words) that you'll find the same holes in
>> the surface.
>
>No, I think they're caused by a low number of samples being used.
>More specifically, one sample...

When I first saw the artifacts, it immediately reminded me of a tute on
halos in the 3.0 docs: declaring a halo in a unit sphere with turbulence of
.5 caused artifacts like this. Declaring the container as 1.5 and the sphere
pattern at 1 prevented the clipping effect. IOW maybe your
turb/particle-placement values are extending outside the boundaries of your
merge object. It would make sense since they mostly occur near the top where
you are likely increasing the amount of chaos.

P.S.
The flickering light looked just fine to me as is. <shrug> And yes, it's
still cool!

Batronyx™ ^"^
bat### [at] cadronhsacom

Post a reply to this message

"Batronyx" wrote:
> When I first saw the artifacts, it immediately reminded
> me of a tute on halos in the 3.0 docs: declaring a halo
> in a unit sphere with turbulence of .5 caused artifacts
> like this.

I should have been more specific. The problem you and others are referring
to is not the same artefacts I'm talking about. What I'm talking about is
some very abrupt transparent cracks and holes in the media.

> Declaring the container as 1.5 and the sphere pattern at
> 1 prevented the clipping effect.

There's probably a bit of the clipping problem too, but that should be easy
to fix, like you describe.

> P.S.
> The flickering light looked just fine to me as is.

Well, not to me. I'll try a different approach in the next anim.

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 March 29)
/ Also visit http://www.povrayusers.org

Post a reply to this message