[Bf-compositor] Clamping (Mix, Math, RGB Curves)

[Bartek Skorupa (priv)]

Hey Troy,

I hope you know I really appreciate your expertise on gazzilion matters (I've asked you several questions and appreciated your answers on matters I didn't even realized existed), but here in this thread let's try to be practical, which means (I know you hate this) let's try to "make stuff work".
Let's try to forget about all "ICCGFD, GH54321, 621111" stuff and others. Let's focus on CLAMPING.
When I say: "clamp blacks" I always mean: set every value below 0.0 to 0.0.
When I say: "clamp whites" I always mean: set every value above 1.0 to 1.0.
Every compositing artist will know what I mean.

Bartek Skorupa

www.bartekskorupa.com

On 23 maj 2014, at 20:44, Troy Sobotka <troy.sobotka at gmail.com> wrote:

> 
> On May 23, 2014 9:12 AM, "Sean Kennedy" <mack_dadd2 at hotmail.com> wrote:
> 
> >  It made sure that no matter what a compositor did, the blacks wouldn't go too dark and the brights would never get superbright.
> 
> When _working_ in the scene referred HDR domain, there is no notion of superbright. It is precisely why lift, gamma, and gain doesn't work on scene referred data.
> 
> When transferring to the display referred system at the tail end, it is necessary to assign a scene referred value to display referred RGB = 1 as diffuse white, and scale the scene referred data to it.
> 
> My only concern is assuming that 1.0 has any meaning while doing the bulk of the work in the scene referred model. It doesn't. It would _always_ limit one's working model to ~2.5 stops of latitude, which is rather silly.
> 
> A flexible clamp node however would be excellent assuming it is versatile and can be used on a full range of scene referred / HDR values. Even a clamp / scale node that allows an artist to scale values arbitrarily to slide blacks and whites to their correct printing densities.
> 
> With respect,
> TJS
> 
> _______________________________________________
> Bf-compositor mailing list
> Bf-compositor at blender.org
> http://lists.blender.org/mailman/listinfo/bf-compositor

-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.blender.org/pipermail/bf-compositor/attachments/20140523/b74fef41/attachment.htm