Maxwell Render | Official public forum

I need some advice guys. I was recently hired at a firm that does high end restaurants and bars. Needless to say wine bottles and liquor bottles are everywhere. So here is my question. I need advice for texturing bottles in a way that looks convincing but renders fast. (I never render with caustics due to the high SL needed to get decent results.) I have started to texture the bottles with the AGS glass texture. It looks ok, but it can't hold a candle to the high grade glass material. If I’m not rendering with caustics does it matter which glass I use as far as render time/noise? Also these wine bottles will be back lit as well as bottom lit so emitters will be very close to these bottles. And I was also wondering about liquids inside of the bottles. For bottles far away I don’t really think I would need it but my hero bottles will defiantly need it, any advice in regards to that? So any advice you guys have would be very much appreciated. Here are some examples of what I’m up against. As well as a test rendering of the bottles with the AGS material.

AGS Bottles that i have started to texture
http://www.flickr.com/photos/90744975@N ... hotostream

Examples of what I'm talking about
http://www.flickr.com/photos/90744975@N ... hotostream
http://www.flickr.com/photos/90744975@N ... hotostream
http://www.flickr.com/photos/90744975@N ... hotostream
http://www.flickr.com/photos/90744975@N ... hotostream
http://www.flickr.com/photos/90744975@N ... hotostream

interesting... i clicked on this topic to read your tutorial...

I think you really ought to stick with AGS...or you could use something like a shiny opaque surface, if the view is wide enough.

Why don't you create two separate materials, one (real glass) for any potential close-ups and one (AGS or opaque shiny) for sufficiently zoomed out views?

Just use glass with ROUGHNESS 0, will render quickly.

Your AGS bottles look good. They should work well for background bottles.

Any close-up bottles, use Maxwell's full power. Use Thomas An.'s modeling method for liquid in glass.

get away with whatever you can....... best way to get away with that is to test, test, and then test again until you have figured out the best solution.

i had similar dilema for client in New York that wanted a rendering of upscale bar in East End (which is about as superficial as it gets). Here's a few of the tests i did prior to start:



hope it helps...

Hello gadzooks,

Definitely I would go with a 3-qualities approach:

- Real dielectric bottles should be used (in your case) only in the front row of bottles on each shelf. Use Roughness 0 as Eric in mentioning to get a cleaner and faster result.

- AGS bottles only for bottles in the second row and beyond, and for far objects. They can look a bit weird, to just use them for bulk purposes.

- Solid material for red wine bottles and dark full bottles (like Ballantine's whiskey). When those kind of bottles are full of liquid, the look like opaque, so you can use a plastic-like structure instead (color BSDF + specular reflection BSDF), like the ones in the Rusteberg's example.

And pay special attention to avoid "dielectric behind a dielectric", like a bottle behind a glass shelf, because the caustics in that situation may last forever.

In addition to that, disable the reflection and refraction caustics in the Render Settings panel. This won't skip Maxwell from calculating the caustics, but will prevent it from painting it in the final image, so you will get rid of a lot of caustic noise.

I guess that with these indications and with an smart disposition of the glass qualities, you will get the fastest render possible.

Cheers!

Dario Lanza

Thanks everyone for taking the time to reply to my question. I have a few more question:

Tom
that is exactly what I'm after.
1. The glass material you used in option 1 (HDRI+Backlit) was it the high grade glass? with roughness at 0?
2. To me option one looks great, was that the one u used for the project?
3. also check your inbox I sent you a PM.

dariolanza

Solid material for red wine bottles and dark full bottles (like Ballantine's whiskey). When those kind of bottles are full of liquid, the look like opaque, so you can use a plastic-like structure instead (color BSDF + specular reflection BSDF), like the ones in the Rusteberg's example.

So your saying don't even use glass for the dark bottles but use a plastic solid color.... correct?
What about liquid in the bottles any considerations I should be aware of? And by adding a liquid in the bottles will that dramatically increase render times? The liquid will only be on the first row of bottles and not all of them.

Once again many thanks to you guys for helping me out.

i think it was low grade glass with roughness zero and a low (maybe around 20-30mm) attenuation value.

Hello gadzooks,

Yes, I meant that for that kind of bottles (red wine, or dark glass like Ballentine's bottle) you can skip the transparency of the material and use an opaque plastic-like structure (specially for those in not prominent positions).
the real thing should be using a dielectric material with a dark Transmittance color and a low Attenuation value, that would make it looking opaque) but using directly a pure opaque solid material would provide a similar look with less render calculations.

And regarding liquid in the bottles: for those bottles you want to show half full, keep in mind Thomas Ann's method for liquid in a glass (I'm pasting here).

Greetings

Dario Lanza

Thanks very much Dario! Will start testing and post some results.

Should this test use instances? I mean, you may have what 5 to 7 different types of bottles? Use instances, this should render fast in Maxwell and then you can use real glass materials for the bottles. AGS was not intended for the purpose of what you have here for a scene.