 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Hi,
> That's some extreme detail in your caustics. What was the total photon
> count?
Sorry Nathan, as soon as the trace was finished, I upgraded to Megapov 0.4,
and traced all the demo scenes, so the stats have been cleared out of the
history. I can however tell you that calculating the photons took around 25
minutes on my P2 450 with 128mb of ram, so I saved the photons to a .ph
file, and the file's size is 30.6 megs.
Equiprawn
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Hi,
> Wow, I should take the comment about glass out of my reply to Andre's
"More
> Alcohol" posting and put it here instead. In case it's not seen I
basically
> said glass textures are both a good and bad thing, for obvious reasons.
I am begining to agree :)
> The B&W version is actually sepia and not all gray shades I see. Has a
good
> look to it.
I know, I thought it might help the photo look a little, plus I'm a fan of
sepia tinted grayscale images.
> That purple pitcher has too soft a highlight IMO, hate to say it because
of the
> tremendously long render time you endured.
Oh no worries. I was going for a frosted kind of plasticy (is that even a
real word?) look.
> Any plans to try a speeded up, less photon-intensive, version or has it
already
> been done?
Maybe. I really need to learn exactly what each of the settings do, so I can
balance quality with time. Though I am wondering what caused the long render
time for my image. The jug, water and glass meshes have thousands and
thousands of smooth triangles - could that be what took all the time, or was
it just that I went overboard on my settings? In global settings I used :
photons{
gather 20, 100
radius 0.1*phd, 2, 2//, 0.1*phd
autostop 0
jitter .4
expand_thresholds 0.2, 40
load_file "jugwaterglass.ph"
}
and then for the objects I used:
photons {
separation 0.02*phd
refraction on
ignore_photons
}
except for the two glasses which had reflective photons also.
Equiprawn
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Hi,
Thanks. That's basically what I was going for. I have some glass marbles
here that are made of frosted glass, and I though "I'm going to try that!".
I went for a frosted plastic look though in the end, to make a change from
the two glasses.
Equiprawn
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Hi,
Speaking of animation, before I realised how long the render was going to
take, I was considering doing an animation with photons. There is a physics
simulation program out there somewhere, can't remember the name right now,
that uses simple primitives in the simulation. So I was thinking of building
a simple jug, filling it with thousands of tiny spheres and then pouring
them into a simple glass. Then for the final animation, I would change the
spheres into spherical blob components, with the hope that they would blob
togethere to create realistic flowing water.
But I'm not going to do that now ;)
Equiprawn
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Hi,
> Wow! Both the coloured and black & white versions are
> fantastic, I love things like this.
Thanks. I was partially inspired by the still life rendering on the front
page of the Radiance site.
> This reminds me that I need to read about the changes
> to photons in 0.4, as I've tried a render and found
> they're totally different to 0.3.
I know, I've just noticed now. Hopefully this will make it easier though.
> If you rerender this try a slightly wider camera
> angle, just that glass on the right seems to be
> leaning over a bit.
I think the glass is leaning because I used too wide a camera angle. I think
it was somewhere near 70 degrees or something. The camera is fairly close to
the objects as well, which doesn;t help.
Equiprawn
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
No doubt the refraction and reflection eating into the render time. You didn't
say what 'phd' was equal to but I suppose 1.
I was thinking when I first saw the pitcher that it might be non-smooth glass or
plastic but the clear refractions don't prove that out. A good candidate for
Chris Huff's transparency blurring patch.
All in all it seems the makings of a year long project if everything you say
were to be done, either that or render farm comes to mind, heh-heh.
Bob
"Equiprawn" <equ### [at] tinet ie> wrote in message
news:38b93b08@news.povray.org...
|
| > That purple pitcher has too soft a highlight IMO, hate to say it because
| of the
| > tremendously long render time you endured.
|
| Oh no worries. I was going for a frosted kind of plasticy (is that even a
| real word?) look.
|
| > Any plans to try a speeded up, less photon-intensive, version or has it
| already
| > been done?
|
| Maybe. I really need to learn exactly what each of the settings do, so I can
| balance quality with time. Though I am wondering what caused the long render
| time for my image. The jug, water and glass meshes have thousands and
| thousands of smooth triangles - could that be what took all the time, or was
| it just that I went overboard on my settings? In global settings I used :
|
| photons{
| gather 20, 100
| radius 0.1*phd, 2, 2//, 0.1*phd
| autostop 0
| jitter .4
| expand_thresholds 0.2, 40
| load_file "jugwaterglass.ph"
| }
|
| and then for the objects I used:
|
| photons {
| separation 0.02*phd
| refraction on
| ignore_photons
| }
|
| except for the two glasses which had reflective photons also.
|
| Equiprawn
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
I'd have to say that the green glass lying down is one of my favorite
object i've ever seen rendered.
ross
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Equiprawn wrote:
>
> Hmmm... there's an idea - would there be any way to get Povray to
> trace more than one ray at a time?
On less than two processors, no. If a processor is designed a
particular way, it is possible for it to trace two rays at once, but it
would be far more trouble than it's worth; the rendering engive would
almost certainly have to be rewritten directly in assembler, and I doubt
the POV team is interested in doing that.
Regards,
John
--
ICQ: 46085459
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Quantuum-processors would reduce the rendering-time tremendously. Not only
would they be fast, but they calculate implicit equations instantly, just
like optimal angles between atoms in molecules are found instantly with
quantuum-porcesses. Ray-object intersections could be calculated instatly
for _any_ surface. Now _that_ would be something.
I've heard that they now have constructed a simple quantuum-ciruit that does
some simple calculations. Int the future ... blabla.
Simen.
John VanSickle skrev i meldingen <38B992D0.DF970A41@erols.com>...
>Equiprawn wrote:
>>
>> Hmmm... there's an idea - would there be any way to get Povray to
>> trace more than one ray at a time?
>
>On less than two processors, no. If a processor is designed a
>particular way, it is possible for it to trace two rays at once, but it
>would be far more trouble than it's worth; the rendering engive would
>almost certainly have to be rewritten directly in assembler, and I doubt
>the POV team is interested in doing that.
>
>Regards,
>John
>--
>ICQ: 46085459
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Jay Raney wrote:
> Isnt it amazing that POV takes 4.5 weeks to render this but nature could render
> it instantly. :)
> Very nice image, caustics are impressive.
No, nature's render time is however long it takes the light to get from the source
to the destination. Now if you factor in the miniscule, unnoticable effects of
distant stars, nature took billions of years to render it! :-)
--
___ ______________________________________________________
| \ |_ <dav### [at] faricynet> <ICQ 55354965>
|_/avid |ontaine http://www.faricy.net/~davidf/
"Sitting on a cornflake, waiting for the van to come" -Beatles
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
