Anyone ever been able to successfully use this - are there any tutorials out there?

I have successfully used this but it's a trick. You have to nuance the settings and do lots of test renders before you're happy. One thing I've also noticed is that it has this really crappy tendancy to be sensitive to the size (or aspect ratio) of the image so if you're doing tests in the modeler room, make sure you're outlining the entire image. Outlining a different shape will get you a different result. Also, you can't turn it up too much or the back blurry components will fuzz badly into the forward sharp components because the algorythm is rather stupid and simple. I'm working on writing my own plug-in for C to fix this. Odds are that I will probably release it as a stand alone little code to use as post first. The algorythm I'm planning will use a much more straight forward physical optical algorythm and will also have the same implementation that the DOF used in Raydream which ROCKED!!!!! Pray that I can make it work. -Kix

In the words of Montgomery Burns ... "Excellent" ... Thanks Kix!

The DOF controls are kinda screwey. If you're tired of messing around with them try G-buffer distance map, they can work pretty effectively in some cases. It will add a grayscale depth map to your image that will be stored in an alpha channel (white is far away from camera, black is near and gray shades are inbetween). When you open the image in Photoshop load the distance channel and add a blur or a gaussian blur. You may have to add it more than once to to get the desired amount of bluring. What's pretty cool is that you can also invert the map to reverse the effects if you want to.

So why is it DOF, and booleans, work so badly in Carrara? 90% of the manual is identical to the manual from Ray Dream 5.5 The metaball interface is practically identical. I guess code is a lot harder to cut-and-paste.... On the other hand, render engine seems much faster, and the vertex modeller has less of that annoying delay it did in RDS. Apologies for going OT here. I've been doing all my DOF with alpha renders and PhotoShop.

Several of the C components are not those used in Raydream. DOF is one of them. From the way things look, I believe soft shadows and motion blur are others. Carrara is a fusion of Inifini-D, Raydream, and some newer technology. Since this is the case, some of the models which were in Raydream didn't make it over to C for several reasons (legal issues, code integration, etc.). Metacreations did pound some effort into cranking up the rendering time in C compared to Raydream. RDS 5.5 touted a rendering speed improvement on the order of five times faster than RDS 3D. C is on the order of 11 times faster than RDS 3D. I've rendered just a couple of things on RDS 5.5 for the resource CD. While I don't really know detailed factors, RDS is WAY WAY slower! In closing, most people do their DOF in post processing. C just does not cut the mustard on this particular issue. -Kix

Uh, let me also clarify. I was a heavy duty Raydream user before I was a C user. I wanted to say that I've recently tried rendering the same file on the same machine but in different applications and the difference is VERY noticeable. -Kix

kIX - anywhere I can find a tutorial about making a post depth of field effect with Carrara and Photoshop?

Litst has posted at least one thread in this forum on what he does. You could also post the question in the photoshop forum. -Kix

Kix...get this plugin done NOW!...OR I'll hunt you down, tie you in my non-existent basement and make you program day and night till you are done!!!! Brian