[Tig] camera dynamic range quotes are getting silly!
richard at filmlight.ltd.uk
Thu Feb 17 21:38:41 GMT 2011
> Bob Friesenhahn <bfriesen at simple.dallas.tx.us> said....
> I don't think that the human eye can view an HDR image so I guess it
> would not be so pretty, or at least it could not be appreciated. The
> human eye is rather poor with simultaneous contrast. Contrast ratios
> of 350:1 are still keeping the eye rather satisified.
The inside of the eye is not black. We do not have an anti-reflective
coating like a cat does. So a fraction of the light entering the eye
will bounce around and give us flare. Even if your photoreceptors were
able to measure to enough bits, this would limit our appreciation of
simultaneous contrast to 100:1.
However, that isn't the whole story. Our eyes scan images, so we may
be looking in a dark corner of an image, and seeing a local 100:1
contrast where the average lightness was only 10% of the white point.
In general, I found you needed about 1400:1 contrast to give a good
'film-like' image in a grading studio. Of course, it depends on the
material, but less than 1000:1 was not usually good enough. However,
it is possible on some images to trade long-range variation for local
contrast - so you subtly raise the broad shadows and drop the broad
highlights, and that can give you a 'super-contrast' effect. However,
this doesn't work for all subjects: try doing that to text and you
On top of that, it would be nice to have brilliant highlights. I don't
think you want to have the sun at full brightness in your scene, but
it is good to have specular highlights. These are usually tiny, and we
are not good at judging the absolute brightness of tiny highlights as
they can be only a retinal cell or two across. Film rolls off these
highlights, and it does make the image look 'flatter' but we are used
to looking at images like that, and we don't complain. But it is one
of the things that still separates images from real life. I think HDR
images look good.
Old CRT video used to permit small excursions all the way to peak
white provided they were small and they moved. If they were large then
you drew too much beam current for the CRT HT supply, and if they
stayed in one place, they could overheat the phosphor. But they could
be used to add 'sparkle'. No change of getting that though QA these
More information about the Tig