Welcome to our community

Be a part of something great, join today!

Log in
  • Hey all, just changed over the backend after 15 years I figured time to give it a bit of an update, its probably gonna be a bit weird for most of you and i am sure there is a few bugs to work out but it should kinda work the same as before... hopefully :)

ISO100 Color Science

Got it.

And so am I correct in my understanding that the mid gray for the Dragon is "tuned" for 800 ISO?
 
THere's a valid range of mid-grey code values that produce low noise and great highlight protection. ISO800 is in that range, but so is ISO250.

Graeme
 
Aaaah. Interesting.

Test, we must.
 
Mark Toia said:
I shoot dragon at 250 iso all day everyday... highlights only blow out when you see the traffic lights come on. Sometimes I shoot over exposed because Im more interested in the subject in the foreground more than a hot back light.
What I love is, that I have noise less blacks when i start to push...

DRAGON @ 250iso
https://vimeo.com/94276293

EPIC MX@ 320iso
https://vimeo.com/86507926
Click to expand...

Great shots Mark as always. What lenses did you use with the DRAGON?
 
Last edited:
Think of it like the levels tool or curves tool in Photoshop. Your black and your white remain the same. All ISO does is redistribute the mid point of the gray between those two points.*

*(When working in REDGamma space. RLF and Linear are different and actually shift the entire data "left" and "right" to maintain linearity. Even in REDGamma it's a bit more complicated than that but it's essentially what's happening.)
 
Gavin Greenwalt said:
Think of it like the levels tool or curves tool in Photoshop. Your black and your white remain the same. All ISO does is redistribute the mid point of the gray between those two points (When working in REDGamma space, RLF and Linear actually shift the entire data "left" and "right").
Click to expand...

That is a great way of explaining it, thank you!
 
Extended Edition Pro-Level Explanation (Graeme you can correct me if I get something wrong.):

ISO simply multiplies the values in linear space. So 0.0 - > 1.0 * 2.0 (increase ISO by 2.0) = 0.0 -> 2.0. This 2.0 is then converted into Cineon (RLF/log) space. This compresses the values to 0.09 -> .74. This 0.09 -> 0.74 range now has an S-Curve (REDGamma) applied to it which gives it a nice black to white appearance on sRGB and rec709 monitors.

So the important step on there is the Linear conversion to Cineon/REDLogFilm/Log space.

In Log Space white is at 0.66, super white (1 stop over white) is 0.75, Superduperwhite (4 stops over) is still only 1.0. So if you go from 250 ISO * 2^4 stops (16) = 4,000 ISO in linear space (0 -> 1.0 to 0 ->16.0 you've only shifted your white point from 0.66 on the S-Curve to 1.0.

cineon-curve-650.jpg


So looking at an S-Curve like this:

SCurve.JPG


Which is similar to REDGamma you can see as increase the ISO (multiply the values), more and more of the the data falls into the top of that highlight protection as opposed to being on the "flats" in between the top and bottom. So technically if you increase your ISO to 320 ISO * ~14.0 (clipping point in Cineon) = ~4,300 ISO a cineon/RLF image will start to clip with a standard gamma curve, however RED seems to deviating from Cineon above 4300 ISO and applying a bit of a compression to the highlights in order to protect the highlights even at very high ISOs. So even at very high ISOs your white point will continue to approach white (in log space) without actually reaching it.

So the white and black points are moving, but they're moving very slowly since they're in LOG space before being converted. The end result is "Pretend it's like the midpoint slider in Photoshop". It's technically wrong, but it's an easy way to think of what's happening without introducing 4 full gammaspace transformations haha.
 
That's pretty good Gavin.

Graeme
 
You must log in or register to reply here.
Back
Top