[jotero]

WOW very nice thomas

[Jozvex]

Amazing! Hopefully under 100 hours hehe.

[tom]

cool.

[sandykoufax]

Yes great.

I want to know SL and RT.

Hey Jozvex, You must not be so grasping.

He cooked it more than 1000hours with beta.

[Jozvex]

Ohhh, I missed that bit. Well, just call me super optimistic.

[jomaga]

Simply great, Thomas!

[Thomas An.]

Amazing! Hopefully under 100 hours hehe.

Nothing to be exited about. Sorry.

This is meant to be an equivalent test using the same old slave PC that was used for the beta image (its a four year old single core Athlon).
Both images are about the same time. The beta was 1024hr, the V1.1 is about 810 or so at SL 26.27. Still no caustics in the reflections.

Here is a dim light version.

[Rickyx]

1000 hour???

More than a month... you have to forget your computer in a cave...

(I should try...)

[Thomas An.]

1000 hour???

More than a month... you have to forget your computer in a cave...

(I should try...)

Yes, it is a decommissioned computer (no one uses it here), so I get it to render something long like this in the background and forget about it.

[numerobis]

...810 hours...Still no caustics in the reflections...

[Thomas An.]

...810 hours...Still no caustics in the reflections...

Well we should clarify.

Direct caustics in reflections are not a problem. Maxwell can do those very well. For example a mirror will reflect a direct caustic in a simple scenario.

However, the caustics in the above image are much more complex.
Firstly, the reflected image you see in the left prism is actually due to total internal reflection.
Secondly, the caustic you see under the colimator has jumped many hoops ... it is actually a ray that has been refracted and then internaly relfected inside the prism and then refracted again as it exited the prism and then it hits the metalic support and it is being reflected on the horizontal part of the metalic leg (so you have a caustic reflection there). Now this battered ray needs to pass through the left prism, be refracted again and then be internally reflected and then refracted again as it exits and comes towards the camera ....

... this is a long darn process for a ray to go through. I am willing to say that no commercial spectral engine currently known to man is capable of showing that reflection.

[Tim Ellis]

Hi all,

Since some people from time to time send mail asking about the setup of this scene... Here is a sample mxs file for Maxwell 1.0 to get you started.

http://rapidshare.de/files/20124664/Sam ... 0.mxs.html

Thanks for the file Thomas.

I was playing with it over the weekend and found that activating Auto exposure causes MXCL to crash, either if selected in Studio or during a render.

Can you confirm this?

Cheers,

Tim.

[DELETED]

DELETED

[b-kandor]

and why is it again those beams are curved? (on the wall)

W not 8

Answered in first post.

@thomas> v1.1 versions look awesome!!!!!!!

[Tim Ellis]

and why is it again those beams are curved? (on the wall)

W not 8

Why is the first grouping arched instead of vertical, while the second and third grouping appear more straight ?
EDIT: This has been mostly resolved. The behavior is correct as it apears in nature as well. It relates to the height of the colimator slit producing rays that are slightly not parallel as they reach the prism surfaces, thus bending at minutely different angles.
Ref Photo: http://www.rit.edu/~andpph/photofile-c/ ... m_8661.jpg

Can anyone else confirm the Auto exposure crash please?

Cheers,

Tim.