[Bf-cycles] Normalized Emission shader

Thu Nov 13 15:24:41 CET 2014

The thing here is, both of new node type and option wouldn't make much
sense if the shader tree is used by lamp. What is more tricky, the same
shader tree could be hared between lamps and meshes.

That makes making an UI which wouldn't be confusing a bit hard.

Brecht, don't really think question which node to use is so huge an issue?
Just give a clean name and artists will figure out which node they need i
guess.

On Thu, Nov 13, 2014 at 3:02 PM, David Black <db4tech at yahoo.co.uk> wrote:

>  Hi Brecht,
>
> Great to have your input.
>
> In relation to lamp and emission plane calculation differences, can a
> light type be identified internally from its lamp ID / emission plane? If
> yes, would a more artist friendly solution be, instead of extra nodes, add
> a Normalize check box to the Emission node with a helpful tool-tip,
> something like...
> "Light output remains constant while light source size changes"
>
> Hope things are going well for you.
>
> David
>
> 3d-designs-davidblack.blogspot.com
> <http://www.3d-designs-davidblack.blogspot.com>
>  On 12/11/2014 18:52, Brecht Van Lommel wrote:
>
> There's no significant performance hit, the surface area of the mesh
> is already computed and stored for all objects so it's just a matter
> of dividing by that.
>
> The main reason I didn't expose this option initially is because it
> raises a bunch of UI questions as you guys found. There's a single
> emission node for both meshes and lamps, so if they have different
> defaults depending if they're added to material or lamp nodes, that's
> a bit of an odd case not supported by the node system.
>
> Also related to this is how to show strength units for the emission
> nodes, it's Watts for lamps normalized by surface area and Watts/m^2
> for meshes without normalization. And there's the confusion around
> "shadeless materials", which really are just emission nodes with
> strength 1 but people keep asking how to do them, so perhaps there's a
> way to improve the UI here.
>
> Perhaps separate nodes for normalized and non-normalized emission
> could simplify things, and make it possible to show the right units,
> but then people might get confused about which node to use, so it's
> not ideal either.
>
> Brecht.
>
> On Wed, Nov 12, 2014 at 4:05 PM, Greg Zaal <gregzzmail at gmail.com> <gregzzmail at gmail.com> wrote:
>
>  Hey,
>
> I agree this could be useful. I'm guessing it'll work by checking the total
> surface area which emits light and calculate the intensity based on that
> maybe? Would there be any significant performance hit when (and when not)
> using this feature?
>
> Personally I would prefer if this were not the default behavior, but we can
> have that argument when/if it's implemented.
>
> Similarly, it might be useful to have the opposite behavior for regular
> lamps - increasing the intensity as the size increases.
>
> Cheers,
> Greg
>
> On 12 November 2014 16:27, Marco G <marco.gzt at gmail.com> <marco.gzt at gmail.com> wrote:
>
>  Hi, i'm happy other people thinks it's useful. Thomas said it could
> implement it *eventually*, IF there is enough interest.
> @Marc Dion, actually it's geared more towards another use case, mostly
> flat mesh lights out of camera frustum to light subjects, withtout the need
> to touch anymore its strength and instead focus on size/position etc to get
> desired result...
>
> Hope to hear from other Cycles module member Greg and Matthew too. It's
> one of those optional little features which improve daily work a lot.
>
> MG
>
> 2014-11-09 21:03 GMT+01:00 David Black <db4tech at yahoo.co.uk> <db4tech at yahoo.co.uk>:
>
>  Hi Marco,
>
> This is something I feel would be very helpful as the default behaviour.
>
> While improving Sweet Home 3D's render engine results (user db4tech),
> lights is one of the areas I looked at, allowing size (for soft shadows) and
> intensity changes, previously lights had a small fixed size and intensity.
> One of my goals, to make it easier for artists, was to try and make sure
> lights intensity remained constant while size was adjusted. Certainly made
> light manipulation a lot easier and received very positive feedback.
>
> Link: (render results image links are no longer working)http://www.sweethome3d.com/support/forum/viewthread_thread,1688_offset,0
>
> David
> 3d-designs-davidblack.blogspot.com
>
> On 09/11/2014 17:41, Marco G wrote:
>
> Hi members,
>
> months ago a feature regarding the Emission node has been removed, the
> problem is that it was never exposed to the UI but actually pretty useful,
> so i'm writing to ask to bring it back if possible and if other members
> agree.
>
> With this option enabled the total amount of emitted light is the same
> regardless of the mesh size. (it would still be optional of course).
> Biggest advantage it would be that once you're satisfied with how much
> light you have in your scene you can resize the light without affecting its
> power, for example to make it bigger to soften shadows or get bigger
> reflections, without the need to adjust the strenght since it would stay
> equal
>
> If reimplemented, maybe a flag "Normalize" exposed in the emission node
> would do the job?
> Tooltip: "If checked, the total amount of emitted light is the same
> regardless of the mesh size"
>
> Opinions?
>
> Thanks,
> MG
>
>
> _______________________________________________
> Bf-cycles mailing listBf-cycles at blender.orghttp://lists.blender.org/mailman/listinfo/bf-cycles
>
>
>
> _______________________________________________
> Bf-cycles mailing listBf-cycles at blender.orghttp://lists.blender.org/mailman/listinfo/bf-cycles
>
>
> _______________________________________________
> Bf-cycles mailing listBf-cycles at blender.orghttp://lists.blender.org/mailman/listinfo/bf-cycles
>
>
> _______________________________________________
> Bf-cycles mailing listBf-cycles at blender.orghttp://lists.blender.org/mailman/listinfo/bf-cycles
>
>  _______________________________________________
> Bf-cycles mailing listBf-cycles at blender.orghttp://lists.blender.org/mailman/listinfo/bf-cycles
>
>
>
> _______________________________________________
> Bf-cycles mailing list
> Bf-cycles at blender.org
> http://lists.blender.org/mailman/listinfo/bf-cycles
>
>

-- 
With best regards, Sergey Sharybin
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.blender.org/pipermail/bf-cycles/attachments/20141113/9f8f878d/attachment.htm 