[Thomas An.]

i never want to see that scene again..it looked fake in EVERY renderer including waxmell

There is only one geometric optics solution to this scene; given the provided area lights.

In other words, all engines should produce the exact same caustic pattern.

About "fakeness". This maybe a materials issue. It was a requirement of this experiment to have no textures/bumbs at all. The resulting clean objects might contribute to what perceive as fake (but it was on perpuse).

[giacob]

i have always been wondering.. why nobody has ever started a thread entitled... "how to make a mental ray looking render with maxwell or vray?"

[DELETED]

DELETED

[Thomas An.]

sorry i was so harsh..but yes, it's the perfectly clean surface on objects (coins, rings) that even if they come freshly out the stamp are never this perfect. so even if the image is 100%correct it'll always look fake..

Right, but if we can't (isolate and) get past the geometrics optics aspect, then the rest (texturing effort) would be futile.

[jeffpatton]

Ahh, thanks for pointing that out with the examples Thomas. You're right about the bounced effect. I'll have to research that to see if it was my light settings/placement or if indeed mental ray has a problem with bounced caustics.

I'm fairly certain the other issues you mention are releated to the scene setup and not mental ray issues.

Thanks for taking the time to clarify the issues.

8etty - The abberation effect is from a photoshop filter. I added that, the grain, and some color correction with Photoshop. The filter name is "PTlens" from here: http://epaperpress.com/ptlens/

Wiz - I think the render time was around 20 minutes, but I'd have to re-render the scene to say for 100% sure since it's an older test.

Aitraaz - thank you. I'll have to figure out what's going on with the bounced caustics as Thomas pointed out.

[ludi]

..... I added that, the grain, and some color correction with Photoshop....

hehe => "he add grain...."
Man, that sounds almost like a sarcastic comment in a maxwellforum *lol*

[Thomas An.]

Jeff,

This is a very difficult scene... it is almost like a "rite of passage" type of thing for many engines.

I have honestly been looking for engines that can do reflective caustics faster than Maxwell, but experience pulls me back here.
The only other image that impressed me was the Vray image by Thev. However, keep in mind that image was done personally by the developer (Thev = Vlado)

Also, IMHO, V1.1 seems superior to beta (or Alpha for that matter).
This is a preview from Version 02 of this scene ... containing more detailed caustics elements.

[glypticmax]

Version 02??
Not only did your image stop me in my socks, but that hint made me stop breathing for a second.
Are you talking about Maxwell V1.2 or Maxwell V2? Or some other secret?
I thought the first Maxwell images of your test were incredible, but this one is totally off the charts.
I understand about NDA's and all that stuff, but a mere clarification of your number reference would be nice.
Thanks for posting the image. Its an inspiration.

[DELETED]

DELETED

[glypticmax]

Well, thats cool too, if Maxwell V1.1 can get all that stuff out of the scene.
*I* just can't yet all that out of Maxwell.
But that has more to do with personal problems, lack of active neural synapses and stuff like work getting in the way of life.
Back to the salt mine.

[Thomas An.]

Version 02??

Well, this scene has an original version (ver01) as seen here:
http://mywebpages.comcast.net/doehtal-r ... eIntro.htm

I am preparing a version 2 of the scene which has more details.

but, the render itself was done in Maxwell 1.1 (yes, this is Maxwell 1.1)

(P.S. Glad you like it ! ... thank you for the comments ... keep in mind that Maxwell can do even better if subtle textures are applied to the materials ... and I haven't applied any glare yet. All cameras have some associated glare however small)