 |
|
|
|
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Hi,
Thought I was getting used to using SSLT but I think I have something wrong
somewhere as my poor ant looks like he is about to spontaneously combust..
I am probably doing something stupid so feel free to point it out, I have had a
few attempts at adjusting the SSLT settings but I seem to go from not much
effect to burning him, any ideas/tips would be really appreciated (the most
appreciated thing would be faster SSLT though ;-)
I am trying to make something similar to
http://jonlieffmd.com/wp-content/uploads/2012/02/Ant-individual-.jpg where the
light can be seen to pass through parts of the ants body.
The ant is made of a few blobs (don't think they are the issue though).
My fear is that it is just not enough samples..
Subsurface settings in global_settings:
global_settings {
#if(useSubSurface)
mm_per_unit 10
subsurface { samples 50, 10 radiosity on}// do another render with this on..
#end
}
Texture for the part that seems to glow:
#declare antMat = material {
texture
{
gradient y
scale 2
turbulence 0.025
translate <0, 0.075, 0>
omega 0.7
lambda 2.8
triangle_wave
texture_map
{
[ 0.45 pigment{ granite triangle_wave colour_map { [ 0.0 color rgb
antColourLight*0.8 ] [ 0.0 color rgb antColourLight*0.2 ] [ 1.0 rgb
antColourLight*0.7 ] } turbulence 0.2 omega 0.65 lambda 3.2 scale 0.1 }
finish { specular albedo 0.1 roughness 0.0015 diffuse albedo 0.6,
0.2 brilliance 0.9 reflection { 0.005, 0.025 fresnel on } conserve_energy
#if(useSubSurface)
subsurface { translucency <.4, .6, 0.1>*9 }
#end
}
normal { granite 0.2 scallop_wave scale 0.007 turbulence 0.3 omega
0.7 }
]
[ 0.6 pigment{ granite triangle_wave colour_map { [ 0.0 rgb antColour*0.2
] [ 1.0 rgb antColour*0.1 ] } turbulence 0.5 omega 0.8 lambda 2.5 scale 0.1 }
finish { specular albedo 0.1 roughness 0.0015 diffuse albedo 0.5,
0.1 brilliance 1.2 reflection { 0.005, 0.035 fresnel on } conserve_energy
#if(useSubSurface)
subsurface { translucency <.4, .6, 0.1>*2 }
#end
}
normal { granite 0.2 scallop_wave scale 0.007 turbulence 0.3 omega
0.7 }
]
}
scale <0.35, 1, 2>
}
interior { ior 1.6 }
}
Lighting is an HDR probe plus 1 area light.
Any help/tips greatly appreciated.
Thanks
Sean
Post a reply to this message
Attachments:
Download 'ant.png' (447 KB)
Preview of image 'ant.png'
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 02.06.2014 19:07, schrieb s.day:
> The ant is made of a few blobs (don't think they are the issue though).
Blobs? Are you serious? That's one hot hell of an ant you have there!
(No surprise it's about to combust ;-)) I'd have expected it to be some
commercial mesh model.
> My fear is that it is just not enough samples..
No, the samples are probably ok, they just affect how noisy the result is.
Are those perhaps transitions between multiple separate blob objects?
Probably a major obstacle you're facing is that ants /are/ pretty
translucent inside, it's mostly the exoskeleton that gives them their
tint. Maybe layering a semi-transparent non-SSLT texture (or a
semi-transparent SSLT texture with low translucency) over a
high-translucency texture might do the trick.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 2014-06-02 12:07, s.day wrote:
> Thought I was getting used to using SSLT but I think I have something wrong
> somewhere as my poor ant looks like he is about to spontaneously combust..
Is a...fire ant, ja?
*puts on shades*
...that's made of 'a few blobs'? Whoah!
--
Tim Cook
http://empyrean.sjcook.com
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
clipka <ano### [at] anonymous org> wrote:
> Am 02.06.2014 19:07, schrieb s.day:
>
> > The ant is made of a few blobs (don't think they are the issue though).
>
> Blobs? Are you serious? That's one hot hell of an ant you have there!
> (No surprise it's about to combust ;-)) I'd have expected it to be some
> commercial mesh model.
>
Thanks, I guess it is actually 13 blobs, one for each leg, antennae and mandible
then the head and each section of the body. The eyes are just spheres and the
hairs are cones so not 100% a blob ant but close.
> > My fear is that it is just not enough samples..
>
> No, the samples are probably ok, they just affect how noisy the result is.
Ahh, that is good to know as now I have started using SSLT I seem to find images
without it a bit less interesting but did not want to have to increase the
samples too much due to lack of patience ;-)
>
> Are those perhaps transitions between multiple separate blob objects?
>
No, that section is made from a single blob, I did have some artifacts at one
point where the mandibles met the head with an earlier version but that has been
corrected with the latest mandibles (still need to do a bit of work on the
texture of these I think they are too shiny).
> Probably a major obstacle you're facing is that ants /are/ pretty
> translucent inside, it's mostly the exoskeleton that gives them their
> tint. Maybe layering a semi-transparent non-SSLT texture (or a
> semi-transparent SSLT texture with low translucency) over a
> high-translucency texture might do the trick.
Now I know it is not samples maybe I can lower them a bit and try some different
options/variations as it will be a bit faster, it seems increasing the
mm_per_unit helps to reduce the glow but getting it just right is the problem as
if I reduce it too much the effect is no longer as good.
Thanks again for responding your advice is always very helpful.
Sean
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Tim Cook <z99### [at] gmailcom> wrote:
> On 2014-06-02 12:07, s.day wrote:
> > Thought I was getting used to using SSLT but I think I have something wrong
> > somewhere as my poor ant looks like he is about to spontaneously combust..
>
> Is a...fire ant, ja?
>
> *puts on shades*
>
> ...that's made of 'a few blobs'? Whoah!
>
> --
> Tim Cook
> http://empyrean.sjcook.com
Thanks, I am now thinking I need to model a child with a magnifying glass as I
already have the effect so I just need the cause.
People are way beyond my blob modelling skills though ;-)
Sean
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Am 03.06.2014 00:49, schrieb s.day:
> Now I know it is not samples maybe I can lower them a bit and try some different
> options/variations as it will be a bit faster, it seems increasing the
> mm_per_unit helps to reduce the glow but getting it just right is the problem as
> if I reduce it too much the effect is no longer as good.
Leave the mm_per_unit setting be; it should always match the actual
scale of the object.
Any effect that can be achieved by tweaking mm_per_unit can also be
achieved by tweaking the materials' translucency.
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
>s.day on date 02/06/2014 19.07 wrote:
> Hi,
>
> Thought I was getting used to using SSLT but I think I have something wrong
> somewhere as my poor ant looks like he is about to spontaneously combust..
A great ant!
;-)
Paolo
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
Paolo Gibellini <p.g### [at] gmailcom> wrote:
> >s.day on date 02/06/2014 19.07 wrote:
> > Hi,
> >
> > Thought I was getting used to using SSLT but I think I have something wrong
> > somewhere as my poor ant looks like he is about to spontaneously combust..
>
> A great ant!
> ;-)
> Paolo
Thanks Paolo, still may add a few improvements to it yet and hoping to stop it
going into meltdown but experimenting with SSLT is time consuming ;-)
Sean
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
On 02/06/2014 18:07, s.day wrote:
> Thought I was getting used to using SSLT but I think I have something wrong
> somewhere as my poor ant looks like he is about to spontaneously combust..
I think it is a great ant, for a fire ant. ;-)
Seriously great modelling.
--
Regards
Stephen
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
| |
|
|
"s.day" <s.d### [at] uelacuk> wrote:
>
> Any help/tips greatly appreciated.
>
> Thanks
>
> Sean
Why you don`t adapt your gradient texture?
From my perspective, it's a povray prob (of sslt), not your adaption.
Personally speaking I always use gray colors for sslt, but I'm far of being
experienced like you... (I admire your starfish...)
Norbert Kern
Post a reply to this message
|
|
| |
| |
|
|
|
|
| |
