Jaime Vives Piqueres wrote:
> Almost perfect! ...and I say "almost" only because the upper border of 
> the coup makes it look like it is too flat. In the reference photo I can 
> see a little the edge highlight, suggesting some thickness. Anyhow, that 
> was a great improvement on the metal texture: just don't touch it! Well, 
> you can tweak the color pigment, but the finish and normals seem perfect 
> like they are now.
> 
> -- 
> Jaime
Thanks Jaime. Yes, that top edge highlight is on the to do list.  As I 
am sure to realize, the present look is because the upper edge is really 
created by the isosurface's bounding box, so I'll need to bend the SOR 
in, then clip it just above with the bounding box.

I also want to get more out of the embossed lettering on the cup if 
possible.

Post a reply to this message

Bob H wrote:

> Something I couldn't stop thinking about is reflection 'exponent' and 
> whether or not that might be plausible to use for the metal if you hadn't 
> already done so.

Wow, that looks like just the thing I was looking for!  Must try it! 
I'd completely forgot about it though I now remember reading about it in 
the manual, in the past.  Thing is, with this model it is too much work 
to edit the mesh so a material with ior can be used.  At least too much 
at the moment.  I *had* been trying to remember if "fresnel" took an 
exponent value, (while sitting in a cafe,... not at my computer :P ). 
Now I look at the docs I see that I might try the conserve_energy 
keyword too.

Post a reply to this message

Bob H wrote:

> that texture, though, since it looks pretty amazing already. Which makes me 
> ask, are your render times for this reasonable (hours not days)?

Usually just under 2 hrs, for 600x600 and most of that is on the iso 
surface cup especially where the lettering is displaced.  At higher 
resolutions though, I expect more time will be needed around the 
filligree at the base.

Post a reply to this message

Jim Charter <jrc### [at] msncom> wrote:
> Stephen wrote:
> > Jim Charter <jrc### [at] msncom> wrote:
>

>
> It is a pragmatic ideal because "light" meshes can be rendered faster in
> games and in full movie animations also consume less resources.  Also
> less complexity allows for easier morphing at joints and in the faces.

Oh! Games, I forgot about those. There is no time for frivolity in my brand


only need a couple of hundred triangles to enter the gates of realism. Then
who am I to argue.

output to look like.


>
> I LOVED those brushes.  As a general rule I love humble things.  But not
> when it's used as a moral stricture.

Oh! Nice brushes BTW


humble way:-)

Stephen

Post a reply to this message

Stephen wrote:
> Jim Charter <jrc### [at] msncom> wrote:
> 
>>Stephen wrote:
>>


My father's mother was born in Scotland and he and his two brothers and 
six sisters made a big deal of it.  And of course they kept their 
hardwood polished. Polished.


> Oh! Nice brushes BTW

I can't tell you enough.  They were thin you see, so they didn't suck up 
a lot of expensive paint.  And those long, irregular, limp bristles. 
Perfect.  Especially for oil paint of the certain viscosity that I 
liked.  I used a lot of varnish and stand oil in my medium.  Funny they 
are sold as gesso brushes.  The hog's hair swells with water which makes 
them useless.


> 

> humble way:-)
> 

No opinion due to my ignorance of that tale.  Maybe I can get back to 
you on it tho'.

Post a reply to this message

Jim Charter <jrc### [at] msncom> wrote:

> Thing is, with this model it is too much work
> to edit the mesh so a material with ior can be used.  At least too much
> at the moment.

Probably I am making wrong assumptions, but if you are keeping this mesh in
a separate file and #including it in your main scene file, and it is too
big to comfortably edit, can't that be as simple as using something like
this in the main scene?

object {
  #include "Chalice.mesh" // or whatever;

  interior {
    ior 10
  }
}

No editing of the mesh file required, just an appropriate fresnel reflection
statement for the mesh. Possibly the exponent keyword will be more useful to
you, though.

Post a reply to this message

Tom York wrote:
> Jim Charter <jrc### [at] msncom> wrote:
> 
> 
>>Thing is, with this model it is too much work
>>to edit the mesh so a material with ior can be used.  At least too much
>>at the moment.
> 
> 
> Probably I am making wrong assumptions, but if you are keeping this mesh in
> a separate file and #including it in your main scene file, and it is too
> big to comfortably edit, can't that be as simple as using something like
> this in the main scene?
> 
> object {
>   #include "Chalice.mesh" // or whatever;
> 
>   interior {
>     ior 10
>   }
> }
> 
> No editing of the mesh file required, just an appropriate fresnel reflection
> statement for the mesh. Possibly the exponent keyword will be more useful to
> you, though.
> 
> 
Yes I believe you are right.  Anyway to be sure I thought it time for a 
quick test.

Eggs from left to right
1. Mesh2 with original texture applied in the texture list
2. Mesh2 with exponent .125 added to original texture and applied in 
texture list
3. Mesh2 with interior { ior 1.18 } applied in object wrapper as you suggest
4. Mesh2 with texture list removed and material{texture{...} interior 
{ior 1.18}} applied in object wrapper
5. Scaled sphere primitive with original texture
6. Scaled sphere with exponent added to texture
7. Scaled sphere with material applied

Post a reply to this message

Attachments:
Download 'img.0134.jpg' (101 KB)

Preview of image 'img.0134.jpg'

Jim Charter nous apporta ses lumieres en ce 02/04/2006 14:17:
> Tweeks:
>     
> on the metal
>     reflection raised significantly esp the lower or coincident
>     value ( .1 .6  ->  .6 .9 )
>     specular roughness tweeked up by ~ 50x ( now .12 )
>     color changed to try and match the brass in the reference
>     (rather than gold) ( rgb CHSL2RGB( <70,.12,.29> ) )
> 
> on the enamel
>     desaturated the colors by 50% and tinted the white marks
>     lowered specular 3x ( -> .2 ) but increased specular roughness      
> 400x     ( -> .4)
>     increased reflection slightly ( .1 .3 -> .2 .4 )
> 
> changed the photo behind the camera and increased its ambient
> 
> desaturated the lights so they still vary, but across a smaller color 
> range and are all much closer to white
> 
> included a shot of the reference
> 
> 
> ------------------------------------------------------------------------
> 
> 
> ------------------------------------------------------------------------
> 
Very impressive! Looks like the original may have looked like when almost new. Save
your file and 
try aging it some. Darkening the enamels and adding some grime over the metal parts.

-- 
Alain
-------------------------------------------------
You can go anywhere you want if you look serious and carry a clipboard.

Post a reply to this message

I hadn't thought a texture on a mesh would look different from same used 
sphere but your test seems to show the 2 and 6 are maybe a little different. 
Interesting how those two have such enhanced highlights from neighboring 
eggs and a bit of a brightnened area on the checkers below the light(?).

You said fresnel before so I wanted to clarify I wasn't suggesting it; only 
that the old keyword reflection_exponent was exchanged with the shortened 
keyword exponent and isn't related to fresnel, directly anyhow AFAIK. At 
least, it (the old keyword) existed before fresnel did, IIRC. Keyword 
falloff in the reflection block might be something similar yet still not the 
same thing.

Didn't want people reading here to get confused about this, hopefully not 
anyhow.

Post a reply to this message

Jim Charter <jrc### [at] msncom> wrote:
> Stephen wrote:
> > Jim Charter <jrc### [at] msncom> wrote:
> >
> >>Stephen wrote:
> >>


> My father's mother was born in Scotland and he and his two brothers and
> six sisters made a big deal of it.  And of course they kept their
> hardwood polished. Polished.
>



> > Oh! Nice brushes BTW
>
> I can't tell you enough.  They were thin you see, so they didn't suck up
> a lot of expensive paint.  And those long, irregular, limp bristles.
> Perfect.  Especially for oil paint of the certain viscosity that I
> liked.  I used a lot of varnish and stand oil in my medium.  Funny they
> are sold as gesso brushes.  The hog's hair swells with water which makes
> them useless.
>

When you are working with your hands. The importance of good tools cannot be
over emphasised.

> >

> > humble way:-)
> >
>
> No opinion due to my ignorance of that tale.  Maybe I can get back to
> you on it tho'.



thought. JC the New York cabbie who plays classics to his clients.




Not so fanciful as my wife has said it to me after a 400 mile journey.

Stephen

Post a reply to this message