Hi folks,

I hesitated a while to give you this image, since it is far away from being
finished. It depicts a certain kind of volcanic eruption, a so called
pyroclastic flow. That is an Eruption of a mixture of gases and ashes first
described by Pliny the Younger who witnessed the downfall of Pompeji. I think
the flow itself is looking good enough now - may be some play with the grays is
needed for the "layers" of the media. But what to do with this empty areas? The
scale of the image is a region of some 20 km^2 viewed at an altitude of 5 km.
The mountain happens to be Mount Rainier near Seattle, since very detailed
hight_field data are available at Georgia Tech. I modelled the flow - and
especially his extend - after the hazard map given in the Wikipedia and pictures
of other pyroclastic flows. The problem is the shear scale of the image. To have
a tree recognizable I would have it 150 m high at least. I have no problem to
scale things a little bit higher as they are in reallity, but 10 times? I will
be thankful for every idea. My only one so far is, to look through the window of
a plane, just to reduce the problem a little bit.

If you understand German, the file name states some flaws with the ground flows
I'm repairing at the moment but cannot be seen with this prespective.

Best regards,
Michael

Post a reply to this message

Attachments:
Download 'untersteschichtenfalsch5kmhoch.png' (1243 KB)

Preview of image 'untersteschichtenfalsch5kmhoch.png'

On 17/07/2013 7:28 PM, MichaelJF wrote:
> Hi folks,

It looks impressive, Michael. I don't think that you can do anything 
about the trees unless you have an insert that shows a magnified view of 
an area. Screen.inc should help here.
How did you create the cloud?
I would have expected it to look as if the internals were rolling like a 
hydrocarbon fire. (I have spent many hours looking at then, when I 
worked on oil rigs, at night. They are most hypnotic.) And the images on 
google have lots of white in them. But that is, I think, being very 
critical.

-- 
Regards
     Stephen

Post a reply to this message

"MichaelJF" <mi-### [at] t-onlinede> wrote:
> Hi folks,
>
> I hesitated a while to give you this image, since it is far away from being
> finished.

This is a great start though, the smoke/flow looks pretty good could maybe be a
bit more dense and more samples added (but I guess this is just a test render so
you would be doing this).

It is a tough one with what to put in, my initial thought was a lake would be
good but tough to fit into that terrain. Maybe some low level clouds, I think
trees may work but you would need a lot of them.

Sean

Post a reply to this message

Stephen <mca### [at] aolcom> wrote:
> On 17/07/2013 7:28 PM, MichaelJF wrote:
> > Hi folks,
>
> It looks impressive, Michael. I don't think that you can do anything
> about the trees unless you have an insert that shows a magnified view of
> an area. Screen.inc should help here.

Thank you Stephen, yes a magnifying would show the trees but I think it would be
a bit disturbing to the scene.

> How did you create the cloud?

First I played with Rune's particle system to find a believable ground flow. If
one finds the correct position and size of the emitter object it is fine to find
the general direction of flows. But it has its limitations since the particles
interact only with the mountain in this case and not with each other. Then I
created ten layers of the cloud using the particles and for each cloud ist
predecessors as the ground object in a modified version of Gilles Tran's
makecloud2 macro (mainly allowing for different directions and using a gaussian
distribution than his uniform). For his tomography I grouped them into four
groups at the moment. The resulting df3-files were used as different media in
the same box. Rendering time is now some 5 hours.


> I would have expected it to look as if the internals were rolling like a
> hydrocarbon fire. (I have spent many hours looking at then, when I
> worked on oil rigs, at night. They are most hypnotic.) And the images on

Take a look at this:
http://www.maniacworld.com/erupting-volcano.htm

> google have lots of white in them. But that is, I think, being very
> critical.

I don't think it very critical. I plan to work a little bit on the colors and
the densities. May be I
change the grouping of the clouds.

Best regards,
Michael

Post a reply to this message

"s.day" <s.d### [at] uelacuk> wrote:
> "MichaelJF" <mi-### [at] t-onlinede> wrote:
> > Hi folks,
> >
> > I hesitated a while to give you this image, since it is far away from being
> > finished.
>
> This is a great start though, the smoke/flow looks pretty good could maybe be a
> bit more dense and more samples added (but I guess this is just a test render so
> you would be doing this).
>

Thank you Sean,

yes indeed I will work a little bit about the densities. First I had the lower
media completely invisible.
Then I found the following trick with quiet insane settings, but worked:
http://news.povray.org/povray.binaries.images/thread/%3Cweb.4ad45c1d4c643aec1524ae1d0@news.povray.org%3E/


> It is a tough one with what to put in, my initial thought was a lake would be
> good but tough to fit into that terrain.

That's the problem. But I will first try to find another location for the camera
having my ground flows repaired.

Best regards,
Michael

Post a reply to this message

Good work Michael. The cloud mass needs more density and a more 
billowing effect, but you were certainly aware of that.

I am not too troubled by any empty space you mention. After all, it is a 
wide panorama and small details are invisible. Try a plane's wing maybe...

Thomas

Post a reply to this message

"MichaelJF" <mi-### [at] t-onlinede> wrote:
> It depicts a certain kind of volcanic eruption, a so called
> pyroclastic flow. That is an Eruption of a mixture of gases and ashes first
> described by Pliny the Younger who witnessed the downfall of Pompeji.


Caribbean island of Martinique erupted in 1902.  At that time, there were plenty
of descriptions from witnesses on ships offshore, and they got photographs of
the pyroclastic flows.  Their descriptions were in remarkable accord with
Pliny's.  Today, this type of eruption is known to be common.

During the early days of the eruption, the attitude of the authorities in
Martinique was remarkably cavalier.  As it happened, one of the visiting sailors
lived in the shadow of Mt. Vesuvius, the very volcano that destroyed Pompei.  He
told the locals, "I don't know very much about your volcano, but if Mt. Vesuvius
looked the way yours does, I'd get out of Naples."  When the mountain finally
blew, the city of St.-Pierre was totally anihilated; there were only two
survivors in the blast zone.  According to Wikipedia, the pyroclastic flows were
estimated at 670 kph.  Many victims never knew what hit them.

>  The problem is the shear scale of the image. To have
> a tree recognizable I would have it 150 m high at least. I have no problem to
> scale things a little bit higher as they are in reallity, but 10 times?

If that's the case, then is it necessary to model individual trees at all?
Would a suitably clever texture (pigment and normal) be enough?

Post a reply to this message

Stephen <mca### [at] aolcom> wrote:
> I would have expected it to look as if the internals were rolling like a
> hydrocarbon fire.

A pyroclastic flow is not a hydrocarbon fire.


ardentes.  The glow is difficult to see in daylight, but I have seen it in video

local news for me.)  The edge of the flow, where it contacts the ground, looks
like glowing embers.

Post a reply to this message

Thomas de Groot <tho### [at] degrootorg> wrote:
> Good work Michael. The cloud mass needs more density and a more
> billowing effect, but you were certainly aware of that.

Thanks, and yes I'm aware of this as you may have guessed from my answers to
Stephen and Sean.

> I am not too troubled by any empty space you mention. After all, it is a
> wide panorama and small details are invisible. Try a plane's wing maybe...

Together with Stephens proposal of some magnifying object this gave me a very
silly idea. But I will solve the problems at hand first. Then I can go to silly
things. Only one: not only planes can reach this altitude...

Thanks again,
Michael

Post a reply to this message

Hi Ricky,

yes you are right completely. But as I learned so far Pliny's report is accepted
nowadays. May be I'm wrong with that. I'm no geologist but a statistician and I
learned it from the Wikipedia too. And yes with this image a procedural texture
would really suffice. I still try to have a better ground view of the flow, but
failed so far. So I think I will stick to the air view given.

Best regards,
Michael

Post a reply to this message