Rendering workshop - Whisky Glass

Rich O Brien

Still struggling on this one

Not happy with the WhiskEy but i'll keep at it

Whiskey II.jpg

Al Hart

@unknownuser said:

Still struggling on this one

Not happy with the Whiskey but i'll keep at it

Much better than the first rendering. Nice ice this time.

Did you change the color of the Whiskey?

Try pouring it put and refilling the glass.

Rich O Brien

Didn't change anything?

Must be my choice of 'Glass' for the Whiskey, but i'll keep with till i get good feedback

thomthom

Another thing: how is the glass modelled?

Was doing some test over at the ASGVis forum, playing around with caustics. Learned that when rendering glass with liquids, the liquid inside the glass also needs to be a "solid", with the surfaces between the liquid and the glass slightly offset (to avoid z-fighting).

thomthom

@al hart said:

What are the refraction indices for metals used for?

I've used them in the reflection layers with fresnel maps. (...not quite sure if this is correct though..)

Gaieus

Well, it probably depends on the software. AFAIK Majid once made a nice description of this (that was Kerkythea) where he said that when there is the liquid, you use that material insteadof the glass material. Face orientation is very important of course (and even more of course, I am not even sure what I am talking about and whether I remember correctly)

Rich O Brien

@thom

I'll try that later, thanks for the tip

@gaieus

have you tried this out on Twilight? When you mention material rather liquid, which material in particular?

@the world

aaaaaaaaarrrrrrrgggggghhhhhhhhh!

Gaieus

@unknownuser said:

@gaieus
have you tried this out on Twilight?

Not yet. Will do...

@unknownuser said:

When you mention material rather liquid, which material in particular?

Well, the liquid material that is insidethe glass.

thomthom

Found my thread about glass+liquid+caustic at the ASGVis forum: http://forum.asgvis.com/index.php?topic=6135.0

Gaieus

Now this is too much light obviously... Another one (also with a bit more interesting setup, too ) is cooking ATM.

Well, the "other one" - now the cigar box could be a bit lighter but I leave it like that.

![17 passes "Easy10 Interior+" (unbiased progressive) in Twilight in a bit less than 30 minutes - original 800x600 then some noise reduction and downsample to 640x480](/uploads/imported_attachments/90u9_WhiskyGlass3.jpg "17 passes "Easy10 Interior+" (unbiased progressive) in Twilight in a bit less than 30 minutes - original 800x600 then some noise reduction and downsample to 640x480")

Al Hart

@gaieus said:

Now this is too much light obviously.

This Workshop is a good chance to study lighting and distortion.

Your ice does not seem to distort the whiskey. Did you place any refraction on the ice itself?

Gaieus

Well, to me it seems it does distort it - the "whisky-line" is different through the ice cube than on its sides.
Yes, I used a physically correctly built Kerkythea library material.

Al Hart

@gaieus said:

Well, to me it seems it does distort it - the "whisky-line" is different through the ice cube than on its sides.
Yes, I used a physically correctly built Kerkythea library material.

I guess I do see a little distortion, but not as much as in this rendering by modelhead

solo

Gaieus

Well, I have to say that I have never really seen such colourful ice cubes even in whisky.
When I made a prelim render (just to see where the lights are going), I also had something similar though.

Rich O Brien

@gaieus

care to share settings and materials used?

Great render btw

Gaieus

I'll try to find them (some of them were Kerkythea libraries so aren't just as easy as to upload the file and it brings all the settings with it).

So Ice cube > Kerkythea library > Translucent > Ice cube.
Glass > Kerkythea library > Translucent > Glass-Crown K7 (whatever it is )
For the whisky, I started from some SSS material and set the following
Alpha: 30.000
IOR 1.700
Shininess: 128.000
SSS density: 1.000 (maybe it means no SSS at all? I am totally lost when it comes to such details).

Now beside the sun (that I tuned down to 3.000), there is also a spotlight "behind" the glass and pointing through the glass. This always helps "emphasize" caustics (you need a lot of light for that) and way above, there is also a point light (set just to 0.1 strength) to give a little "global illumination" to the whole scene. Somehow like this:

Gaieus

Okay, so here is another one with a different ice material (a very slightly frosted one with fresnel procedural - which I have to confess, I don't know what will do to it).

Al Hart

@gaieus said:

Okay, so here is another one with a different ice material (a very slightly frosted one with fresnel procedural - which I have to confess, I don't know what will do to it).

Much better ice, though, don't you think.

Fresnel has to do with reflections at the very edge of the ice.

If you have time, (isn't it either very early or very late in your part of the world) - (It is 1:30 in the afternoon here in Colorado), why don't you zoom in on just the ice cube, and make two renderings - one with Fresnel on and one with it off.

(Some rendering people just know what these thing do. I find I have to try them out, compare the result, and see what I like best.)

(Also, I knew so little about Fresnel a couple of months ago when we added his reflection properties to our renderer that I had to look up on the web how to pronounce his name properly.)

@unknownuser said:

When referring to the scientist the proper pronunciation of Fresnel is " Fray-nel' "; it was his name after all. When referring to the lens the pronunciation 'frez-nel' is accepted. Like many foreign words the pronunciation has become westernized.

Al Hart

@solo said:

:shock:

I was trying to place a fly in the ice yesterday, but SketchUp crashed on me, and I didn't get back to it.

I was expecting to see some inventive changes to the model - such as turning the ice into a flaming item.