Page 1 of 1

Wine bottle

Posted: Sun Apr 17, 2005 2:41 am
by Thomas An.
Well, this one is old news, but here it goes
(haven't had the chance to improve it yet)

Image

Posted: Sun Apr 17, 2005 2:45 am
by iker
:shock: The bottom of the green bottle is great!
Great render again Thomas!

Posted: Sun Apr 17, 2005 5:46 am
by tokiop
really nice frosted glass :D with sort of SSS parameter the green glass will really be optimal!

Posted: Sun Apr 17, 2005 9:56 am
by Thomas An.
Thank you iker, Thank you tokiop, I appreciate your comments :D

I am prepering a newer version, hopefully a little better :)

Posted: Sun Apr 17, 2005 11:40 am
by tom
very nice Thomas,
i prefer them without uv anyway :D

Posted: Sun Apr 17, 2005 2:01 pm
by Duncan
Good work, I like the frosted glass and the bottom of that green bottle, cool little ridges. are the modeled ?

Posted: Sun Apr 17, 2005 5:22 pm
by Thomas An.
Tom, Duncan, thank you for the comments.

Duncan, Yes everything is modelled. Rhinoll does not have texture maps or pump maps yet anyway :D

Posted: Tue Apr 19, 2005 3:21 pm
by Thomas An.
Here is a new version.

I am not too happy with the two silver coins at the bottom (they look junky / grainny). They have a grainny green hue comming from the bottle.

Image

Posted: Wed Apr 20, 2005 5:13 am
by Eric Lagman
Thomas,

I really like the frosted glass in the first image. Was this achieved with 100 percent tranparency and a high roughness value? Im trying to get a similar effect on some tests I am doing, but its taking a long time to get a good preview with dielectrics.........even on low, and small image size. Any tips on creating that frosted glass look?

Posted: Wed Apr 20, 2005 5:40 am
by Thomas An.
Eric,

This is by memory, but I think it was 0.9 absorbance and 0.7 UV rougness.
It was a high rougness value that really did the trick.

-

Posted: Tue Apr 26, 2005 3:26 am
by uniminium
Thomas An. wrote:Here is a new version.
Hey Thomas,

thats quite impressive how cool the reflections appear. A wonderful image. Please let us know your render time as well.

greetings,
clement

Posted: Tue Apr 26, 2005 3:41 am
by Thomas An.
Hi Clement,

Thank you for the nice comments !

About render time, this image took 12hours. I think the problem is that Rhinoll will not let us control the abbe value (dispersion) and Maxwell makes more calculations than needed.

-

Posted: Tue Apr 26, 2005 3:52 am
by uniminium
Thomas An. wrote:I think the problem is that Rhinoll will not let us control the abbe value (dispersion) and Maxwell makes more calculations than needed.
Hi Thomas,

well, the extra calculations seems to be worth it ! I´m quite amazed how realistic the glass color glows to the reflection of the coin beside it. And of course, the modelling detail matches the rendering quality. I found this to be one of the most important parameters when creating photorealism.

greetings and keep posting, :D
clement

Posted: Tue Apr 26, 2005 3:58 am
by Thomas An.
Yes, modeling detail is *very* important :)
You need to manually introduce imperfections (very subtle ones)

-