Hi all

I spend most of last night playing with fire (not literally)
I've got a fairly decent candle flame, so I thought I'd post it.

I'm not totally happy with the shape (difference between
streched sphere and unscaled sphere)

Main media consists of a gradient y density and a spherical
density. There is a second object at the bottom to create the blue
glow.

Comments? suggestions??

Gail
********************************************************************
* gsh### [at] monotixcoza              * Reality.dat not found         *
* http://www.rucus.ru.ac.za/~gail/ * Attempting to reboot universe *
********************************************************************
* The best way to accelerate Windows NT is at 9.8 m/s^2      *
********************************************************************

Post a reply to this message

Attachments:
Download 'candle.jpg' (3 KB)

Preview of image 'candle.jpg'

How about showing this on a white background?  Sorry, I don't really mean that
:-}
Candle flames are always much more of a yellow-white than just yellow, I know
everyone seems to not believe that though.  Also varies a lot depending on the
surrounding light and background, but I still always try for a near white
yellow which goes to yellow-orange at the edges and top.
Turbulence is another thing, sometimes there's none sometimes a lot.  As far
as the general shape goes I think you've come close enough but I'd say there's
not enough blue (or rather invisible) part going up inside the yellow part.
The amount of the merging parts I'm talking about, maybe more diffused boudary
line there would be better?
I found this photo: http://www.hd.org/Damon/photos/_I/cat/lw/eooz2q1mh6.HTM
and it has no blue base to the flame!

Bob

Post a reply to this message

Very nice, but IMHO looks more like a match-flame.

Post a reply to this message

Bob Hughes wrote in message <396c532b@news.povray.org>...
>How about showing this on a white background?  Sorry, I don't really mean
that
>:-}

It disappears completely on a white background :-<

>Candle flames are always much more of a yellow-white than just yellow, I
know
>everyone seems to not believe that though.  Also varies a lot depending on
the
>surrounding light and background, but I still always try for a near white
>yellow which goes to yellow-orange at the edges and top.

I had a candle burning on my bookshelf while modelling this. I'll agree that
the tip
was more white (tried but couln't get that effect) but most of the candle
flame
was yellow/orange

>Turbulence is another thing, sometimes there's none sometimes a lot.  As
far
>as the general shape goes I think you've come close enough but I'd say
there's
>not enough blue (or rather invisible) part going up inside the yellow part.
>The amount of the merging parts I'm talking about, maybe more diffused
boudary
>line there would be better?


Between yellow and blue? I agree, at the moment they're different objects.

Gail
********************************************************************
* gsh### [at] monotixcoza              * Reality.dat not found         *
* http://www.rucus.ru.ac.za/~gail/ * Attempting to reboot universe *
********************************************************************
* The best way to accelerate Windows NT is at 9.8 m/s^2      *
********************************************************************

Post a reply to this message

The blue should have a shape that looks like a rounded / \ or something :P
but otherwise it's really good : )

Post a reply to this message

If it was mostly blue it would look very much like a gas flame. I like it.

Alan

"Gail Shaw" <gsh### [at] monotixcoza> wrote in message
news:396c4381@news.povray.org...
> Hi all
>
> I spend most of last night playing with fire (not literally)
> I've got a fairly decent candle flame, so I thought I'd post it.
>
> I'm not totally happy with the shape (difference between
> streched sphere and unscaled sphere)
>
> Main media consists of a gradient y density and a spherical
> density. There is a second object at the bottom to create the blue
> glow.
>
> Comments? suggestions??
>
> Gail
> ********************************************************************
> * gsh### [at] monotixcoza              * Reality.dat not found         *
> * http://www.rucus.ru.ac.za/~gail/ * Attempting to reboot universe *
> ********************************************************************
> * The best way to accelerate Windows NT is at 9.8 m/s^2      *
> ********************************************************************
>
>
>
>
>

Post a reply to this message

if you use a color map in your spherical density, you can include all of
the necessary colors.  and are you using multiple densities, or multiple
medias?  there is quite a difference between them.

Gail Shaw wrote:
> 
> Hi all
> 
> I spend most of last night playing with fire (not literally)
> I've got a fairly decent candle flame, so I thought I'd post it.
> 
> I'm not totally happy with the shape (difference between
> streched sphere and unscaled sphere)
> 
> Main media consists of a gradient y density and a spherical
> density. There is a second object at the bottom to create the blue
> glow.
> 
> Comments? suggestions??
> 
> Gail
> ********************************************************************
> * gsh### [at] monotixcoza              * Reality.dat not found         *
> * http://www.rucus.ru.ac.za/~gail/ * Attempting to reboot universe *
> ********************************************************************
> * The best way to accelerate Windows NT is at 9.8 m/s^2      *
> ********************************************************************
> 
>  [Image]

Post a reply to this message

Bob Hughes wrote:

> How about showing this on a white background?  Sorry, I don't really mean that
> :-}
> Candle flames are always much more of a yellow-white than just yellow, I know
> everyone seems to not believe that though.

It also depends on what is burning

> Also varies a lot depending on the
> surrounding light and background, but I still always try for a near white
> yellow which goes to yellow-orange at the edges and top.
> Turbulence is another thing, sometimes there's none sometimes a lot.  As far
> as the general shape goes I think you've come close enough but I'd say there's
> not enough blue (or rather invisible) part going up inside the yellow part.
> The amount of the merging parts I'm talking about, maybe more diffused boudary
> line there would be better?
> I found this photo: http://www.hd.org/Damon/photos/_I/cat/lw/eooz2q1mh6.HTM
> and it has no blue base to the flame!
>
> Bob

Post a reply to this message

ryan constantine wrote in message <396### [at] yahoocom>...
>if you use a color map in your spherical density, you can include all of
>the necessary colors.  and are you using multiple densities, or multiple
>medias?  there is quite a difference between them.
>


Yup, I know. Multiple densities. I needed another object for the blue as it
was outside of the container object of the rest of the flame.

The spherical density creates the darker patch at the bottom. The gradient
density creates the colours.

Gail
********************************************************************
* gsh### [at] monotixcoza              * Reality.dat not found         *
* http://www.rucus.ru.ac.za/~gail/ * Attempting to reboot universe *
********************************************************************
* The best way to accelerate Windows NT is at 9.8 m/s^2      *
********************************************************************

Post a reply to this message

Alan Nickerson wrote in message <396c66ff$1@news.povray.org>...
>If it was mostly blue it would look very much like a gas flame. I like it.
>


didn't think about that, but you're right. I tried a gas flame (with no
source of reference for the colour. It's been 3 years since I last
used a gas flame)

I also played a bit more with the original flame. Not much better


Gail
********************************************************************
* gsh### [at] monotixcoza              * Reality.dat not found         *
* http://www.rucus.ru.ac.za/~gail/ * Attempting to reboot universe *
********************************************************************
* The best way to accelerate Windows NT is at 9.8 m/s^2      *
********************************************************************

Post a reply to this message

Attachments:
Download 'candle.jpg' (4 KB)

Preview of image 'candle.jpg'