[Bf-cycles] ray depth

[David Fenner]

sorry, forgot the blend file used for testing:
http://www.pasteall.org/blend/31913


my rig:

i7- 3770
560ti
8gig ram
windows 7


2014-10-05 21:23 GMT-04:00 David Fenner <d4vidfenner at gmail.com>:

> hmm maybe I'm doing something wrong and the shader tree is still
> calculating the bounce, but for me there is almost no gain with the ray
> depth trick.
>
> 0 global bounces
> 6,77 sec
> http://www.pasteall.org/pic/show.php?id=78124
>
> 1 global bounce, light without ray depth trick so it does bounce and
> illuminates:
> 9,62 sec
> http://www.pasteall.org/pic/show.php?id=78123
>
> 1 global bounce, light with ray depth trick so it doesn't bounce:
> 9,40 sec
>  http://www.pasteall.org/pic/show.php?id=78122
>
>
> As you can see, the real gain comes with deactivating the global bounce,
> even when there is only one light and one diffuse shader for all geometry.
> Deactivating the bounce in the light should be the same as deactivating it
> globally, since there is only one light. Maybe I'm doing something wrong
> here.... pics have shader tree.
>
>
> David.
>
>
>
> 2014-10-05 21:00 GMT-04:00 Brecht Van Lommel <brechtvanlommel at pandora.be>:
>
> On Mon, Oct 6, 2014 at 1:47 AM, David Fenner <d4vidfenner at gmail.com>
>> wrote:
>> > The thing is that a scene with bounces, even if we have the ray
>> > depth trick to a light so the bounce is non existent, renders just a
>> slow as
>> > if the bounce was actually illuminating the scene.
>>
>> I don't think this is true exactly, it will skip evaluating the BSDF
>> and tracing the shadow ray, and so it will render faster on most
>> scenes. More optimizations for this specific case are possible though.
>> _______________________________________________
>> Bf-cycles mailing list
>> Bf-cycles at blender.org
>> http://lists.blender.org/mailman/listinfo/bf-cycles
>>
>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.blender.org/pipermail/bf-cycles/attachments/20141005/6d2fc43a/attachment.htm