 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"JWV" <jwv|at|planet.nl> wrote in message news:3ecdcc30@news.povray.org...
> These are the only things which can be improved according to me:
>
> 1: I think you should use less focal blur, it seems to be to much for me.
It depends whether you think it's a full size car or a small model. The blur
makes it look small, so I think I agree with you and I'll reduce it.
> 2: Try to find a real TVR model.
Couldn't find any nice ones so I decided to build a vaguely TVR shaped car
in Wings3D. It's not ready yet.
> 3: use radiosity (if possible) this might make it look more realistic.
Yes, I've been thinking about this. The trouble is the effect relies heavily
on specular highlights, which aren't compatible with radiosity. But if I use
a really high specular and really dim light sources I can get the highlights
working without interfering with radiosity's diffuse illumination. So I'll
give it a try :)
> 3: Change the color (black perhaps :-P)
:-P
I chose the colour 'cause it's popular on TVRs, not because it actually
looks nice :) I think I might do some more pearlescent shades.
> JWV
>
> BTW: how did you make the "table" which it stands on? is it just a simple
> image map? I like it.
I don't have the source to hand, but it's basically a granite pigment for
the veins in the marble, applied to a standard checker pattern. The
reflection settings are similar to the paint but a bit more reflective. I
can post the source if you want it.
--
Tek
http://www.evilsuperbrain.com
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> > BTW: how did you make the "table" which it stands on? is it just a
simple
> > image map? I like it.
>
> I don't have the source to hand, but it's basically a granite pigment for
> the veins in the marble, applied to a standard checker pattern. The
> reflection settings are similar to the paint but a bit more reflective. I
> can post the source if you want it.
please post the source of the table, i think i can make it by myself, but
i'm sort of lazy today...
"Tek" <tek### [at] evilsuperbrain com> wrote in message
news:3ecdd5da@news.povray.org...
> "JWV" <jwv|at|planet.nl> wrote in message news:3ecdcc30@news.povray.org...
> > These are the only things which can be improved according to me:
> >
> > 1: I think you should use less focal blur, it seems to be to much for
me.
>
> It depends whether you think it's a full size car or a small model. The
blur
> makes it look small, so I think I agree with you and I'll reduce it.
>
> > 2: Try to find a real TVR model.
>
> Couldn't find any nice ones so I decided to build a vaguely TVR shaped car
> in Wings3D. It's not ready yet.
>
> > 3: use radiosity (if possible) this might make it look more realistic.
>
> Yes, I've been thinking about this. The trouble is the effect relies
heavily
> on specular highlights, which aren't compatible with radiosity. But if I
use
> a really high specular and really dim light sources I can get the
highlights
> working without interfering with radiosity's diffuse illumination. So I'll
> give it a try :)
>
> > 3: Change the color (black perhaps :-P)
>
> :-P
> I chose the colour 'cause it's popular on TVRs, not because it actually
> looks nice :) I think I might do some more pearlescent shades.
>
> > JWV
> >
> > BTW: how did you make the "table" which it stands on? is it just a
simple
> > image map? I like it.
>
> I don't have the source to hand, but it's basically a granite pigment for
> the veins in the marble, applied to a standard checker pattern. The
> reflection settings are similar to the paint but a bit more reflective. I
> can post the source if you want it.
>
> --
> Tek
> http://www.evilsuperbrain.com
>
>
>
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Here ya go. It's pretty simple. I used a pigment pattern so the scaling and
turbulence on the granite wouldn't affect the checkers.
union {
cylinder { -y/2, -y*2/3, 4 }
torus { 4, 1/2 translate -y }
no_shadow
pigment {
pigment_pattern {
granite poly_wave .3 scale <1,1,.3>*3 rotate <45,45,0> scale 1/10 warp {
turbulence .1 octaves 6 } scale 10
}
warp { repeat x offset y } warp { repeat z offset x } //break up the pattern
at the edge of the tiles
pigment_map {
[0 rgb .2]
[.9 checker rgb 1, rgb 0]
}
}
finish {
diffuse 0.5 ambient 0
reflection { 0.1, 0.5 falloff 5 }
}
}
--
Tek
http://www.evilsuperbrain.com
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Looks great, but you are lucky that the sky_sphere image is blurred,
otherwise you would need to add blurry reflections on the metal.
Since you need the light_sources I wouldn't recommend that you rely too much
on radiosity (arnold renders / almost pure rad). It needs a high quality to
animate well, so I would use it subtle. Besides it probably won't benefit
this particular scene, because you are using highly reflective materials.
But it looks impressive.
Good luck in your further work!
Hugo
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"Hugo Asm" <hua### [at] post3teledk> wrote in message
news:3ece8d05@news.povray.org...
> Looks great, but you are lucky that the sky_sphere image is blurred,
> otherwise you would need to add blurry reflections on the metal.
Well actually the reflection is from a smooth gloss layer over the metallic
effect. Just like real cars the reflection would be perfectly sharp.
> Since you need the light_sources I wouldn't recommend that you rely too much
> on radiosity (arnold renders / almost pure rad). It needs a high quality to
> animate well, so I would use it subtle. Besides it probably won't benefit
> this particular scene, because you are using highly reflective materials.
Yeah, I think you're right. I'm trying a radiosity version at the moment, and it
really doesn't gain much for a simple scene like this (well, apart from a huge
increase in the render time!). If I want to use the paint in more complex
environments I might use radiosity.
> But it looks impressive.
> Good luck in your further work!
Thank you :)
--
Tek
http://www.evilsuperbrain.com
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> Just like real cars the reflection would be perfectly sharp.
Huh, it's sharp? :o) Well.. perhaps.. I haven't thought about it, except
that sharp reflections in computer renderings tend to look unnatural.. But
thinking about it, this may also be due to a compressed range of contrast.
That's why we have the reflection_exponent keyword in newer versions of
POV-Ray (although it's still not the real solution).
Regards,
Hugo
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Well you're right that perfectly smooth surfaces look more artificial, but it's
also a desirable quality in automobiles :)
The compressed contrast range should be handled pretty realistically by the HDR,
though at some point I'll try megapov's film response curve simulation which
should make it possible to simulate accurately.
--
Tek
http://www.evilsuperbrain.com
"Hugo Asm" <hua### [at] post3teledk> wrote in message
news:3eceb771$1@news.povray.org...
> > Just like real cars the reflection would be perfectly sharp.
>
> Huh, it's sharp? :o) Well.. perhaps.. I haven't thought about it, except
> that sharp reflections in computer renderings tend to look unnatural.. But
> thinking about it, this may also be due to a compressed range of contrast.
> That's why we have the reflection_exponent keyword in newer versions of
> POV-Ray (although it's still not the real solution).
>
> Regards,
> Hugo
>
>
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> contrast range should be handled pretty
> realistically by the HDR
It's still a question of mine. According to the docs, POV uses a limited
reflection formula that clips values above 1.. I don't know if this was
changed in MLPOV but I suppose not. If the problem was trivial, it would
probably already have been solved in the offical POV.
Radiosity supports the HDR range, yes, but reflection?
Maybe you can test it...
Regards,
Hugo
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Hmm, I've not heard of that restriction, but it's pretty easy to test...
...I did a test with 3 spheres with different reflection values, and the camera
inside a black sphere with transmit 0.1. The reflections showed all of the
details on the HDR image, with no flattening of colours until they hit full
white. So I can't find any limitation on the brightness of reflections:
That was using ML pov, but the same test scene replaced with a very brightly
coloured sky (rgb 20) in standard POV 3.5 also showed no capping of reflections.
--
Tek
http://www.evilsuperbrain.com
"Hugo Asm" <hua### [at] post3teledk> wrote in message
news:3ed0ff89@news.povray.org...
> > contrast range should be handled pretty
> > realistically by the HDR
>
> It's still a question of mine. According to the docs, POV uses a limited
> reflection formula that clips values above 1.. I don't know if this was
> changed in MLPOV but I suppose not. If the problem was trivial, it would
> probably already have been solved in the offical POV.
>
> Radiosity supports the HDR range, yes, but reflection?
> Maybe you can test it...
>
> Regards,
> Hugo
>
>
Post a reply to this message
Attachments:
Download 'hdr.jpg' (11 KB)
Preview of image 'hdr.jpg'
|
|
| |
| |
|
|
|
|
| |
| |
|
|
> That was using ML pov, but the same test scene replaced with
> a very brightly coloured sky (rgb 20) in standard POV 3.5 also
> showed no capping of reflections.
Thank you. I did a quick experiment on my own, and you're right. The docs
have confused me. It says the reflection_exponent keyword compensates for "a
limited light model that cannot distinguish between objects which are simply
brightly colored and objects which are extremely bright."
Reading the following sentence closely, it says "A white piece of paper, a
light bulb, the sun, and a supernova, all would be modeled as rgb<1,1,1> and
slightly off-white objects would be only slightly darker." But who on earth
would model a supernova as <1,1,1> ?? So in fact, it's not POV-Ray that
uses a limited light model, it's a USER who doesn't distinguish between
objects which are simply brightly colored, and objects which are extremely
bright.
Regards,
Hugo
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
|
|
