"Chris R" <car### [at] comcastnet> wrote:
> "19100" <nomail@nomail> wrote:
> > "Bald Eagle" <cre### [at] netscapenet> wrote:
> > > "19100" <nomail@nomail> wrote:
> > > > This scene is my first that's (mostly) based on real world measurements, I
think
> > > > I've read enough USBC specification documents to be certified or something.
> > > >
> > > > Grab the source files here:
> > > > https://archive.org/download/19100-povray/king.zip
> > >
> > > That really came out very nice.  :)
> > >
> > > The neon lights are very cool.
> > > The alley's wood floor pattern could maybe benefit from some stretching in the
> > > z-direction, maybe with a wee bit less turbulence?
> > >
> > > Not sure about the red cinder-block wall.  Seems off somehow.  Also too
> > > CGI-perfect.
> > >
> > > The bowling ball in the foreground would benefit from a little bit of
> > > fill-lighting, and the icing on the cake would be to add just a touch of
> > > motion-blur.
> > >
> > > Maybe define a rolling function and make a few highly-transparent copies of the
> > > ball that "roll into" the 3 balls in the render.
> > >
> > > You could also possibly make the far ball actually hit and have some of the pins
> > > mid-tumble in the air.
> > >
> > > Great job - I'm sure it was a challenging exercise in measuring and modeling!
> > >
> > > - BW
> >
> > Thanks! Those are great notes to take into account for any future revisions.
> > Looking at the bricks perhaps it's the lack of weathering that makes them seem
> > off, maybe the scale too. The afterimages are a really good idea, 2.2 was before
> > scripting but the SDL is easy to generate so both that and the pin physics are
> > very doable. Getting the measurements was probably the most time consuming part
> > of it all, finding the dimensions of a specific small detail is a real pain when
> > there's so much SEO-optimized crap out there with the same handful of repeated
> > information.
>
> In my images with brick walls, I have started using iso-surfaces that use the
> brick pigment-pattern to define the shape of the bricks and the mortar as
> separate objects.  I apply turbulence to the brick pattern to get less perfect
> brick edges, and then also apply pigment patterns to the visible face of the
> brick, (for me the Z-axis), to apply a texture to them.  I then use some
> techniques I found to vary the color/texture of the bricks so they vary rather
> than having a uniform texture, but still preserving a single texture per brick.
>
> I have attached an example.

That's really interesting, thanks for explaining your technique Chris.

Post a reply to this message