RavineII - the Return

And now what can I add here. An excellent image! I'm not going to repeat anyone's comment, because Horo and Rashad seem to have revealed all the "mystery" and I fully agree with them. About the plop-rendering, yes, I'm using it sometimes, but you'll have to be really careful not to leave any shadows or other tiny elements unrendered while doing the trick. Well, at the end the postwork isn't forbidden after all.

plop render One of the features I like most about 5.5 and I'm sure 6 is that the plop render is fixed. It no longer leaves a line of unrendered pixels on the plop render edge. It seemslessly fits in now. Since the changes you are making are in the bottom half of the image only, a plop render of just the bottom half should do the trick and cut days out of your render time.

Thank you, yes... I missed the hovering root. The issue, as you can probably guess, is that with such a slow render it is more or less impossible to preview the scene. And I was too content with tweeking the volume grass... Maybe a more realistic shadow within the vol mat might answer some concerns. I'll set one going on that, but given the ramping in render complexity, it could be a week before it completes... we shall see. Thank you for your constructive comments - with a scene that is getting this complex (at least from a technical perspective) it is sometimes difficult to decide what direction to take when it comes to planning improvements. Aye... grass and fur... and fluff (if you recally my fluff rendering project?) these are all things to challenge the render engine and our own imagination.

It's a fiver. The volumetric grass at right is fantastic. I've compared the pictures - swapping several times.

Yup - much better. Funny tree at left midground with one of its roots out over the water like a fishing rod. I prefered the tree setup at right in the previous version more, because it showed the hill better. The grass is very nice, I have less quibbles about the volumetrics, except in front of the second second tree at right. The first tree at right is even more fantastic, now that it is nearer. A great picture.

improvements All around improvement. Much better water. Transparency is key. Grass models are exactly right. Of course it only takes a few grass models before you realize that you cannot possibly populate your entire scene with real 3d grass. To solve this issue you chose to go the volumetric route, a very clever approach. The idea is right but in the end this particular implentation fails to convince me, as it is just not consistent with the real grass in the foreground. A tweak of this idea just may work. The only problem with volumetrics in this case is that you will have an obvious flat top to your grass as the volumetrics will trace out the shape of the cube and flatten on top. The next big mystery to solve in bryce is with issues like grass. There needs to be some reliable way to handle these types of issues without memory overloads. Great work.