PDA

View Full Version : Relationship between Contrast Ratio and Dynamic Range

DolphinDan
31-Mar-2017, 08:18
Hi all,

I apologize if this question is blindingly obvious, but I could not find the answer just GOOGLE-ing. How does Contrast Ratio relate to Dynamic Range?

For example, according to Fuji, VELVIA 100 has the following Resolving Power:

Chart Contrast 1.6:1 = 80 lines/mm
Chart Contrast 1000:1 = 160 lines/mm

How many stops of dynamic range = a Contrast Ratio of 1.6:1? Or maybe it is 6:1?
How many stops of dynamic range = a Contrast Ratio of 1000:1?

Just curious...

Namaste
Daniel

Arne Croell
31-Mar-2017, 08:34
One stop is a ratio of 1:2 in exposure, two stops 1:4, three stops 1:8, etc (2 to the power of 1, 2, 3,...). Thus the number of stops for 1:1000 is a tad below 10 stops, since 2^10=1024. 1:1.6 is obviously less than one stop.

DolphinDan
31-Mar-2017, 11:17
Thanks Arne.

Daniel

Arne Croell
31-Mar-2017, 11:47
Thanks Arne.

Daniel

You're welcome. For exact calculations, use the logarithm base 2 (aka binary logarithm, symbols are lb or lg or log2) on the ratio number x - since most calculators don't have that, use either the base 10 logarithm (log) or the natural logarithm (ln). The conversion is:
lg(x)=log(x)/log2=ln(x)/ln2

1.6 gives you 0.678 stops, and the 1000 is exactly 9.966 stops.

Thalmees
31-Mar-2017, 13:14
Hello Daniel,
Just count down on your fingers, like me:D.
1000-500^^^^^^ 1 stop,
500 -250^^^^^^ 2 stops,
250 -125^^^^^^ 3 stops,
~.~ gap ~.~
008 -004^^^^^^ 8 stops,
004 -002^^^^^^ 9 stops,
002 -001^^^^^^10 stops.
So 1000:1, is 10 stops.
Hope this helps.

DolphinDan
2-Apr-2017, 11:59
Hello Daniel,
Just count down on your fingers, like me:D.
1000-500^^^^^^ 1 stop,
500 -250^^^^^^ 2 stops,
250 -125^^^^^^ 3 stops,
~.~ gap ~.~
008 -004^^^^^^ 8 stops,
004 -002^^^^^^ 9 stops,
002 -001^^^^^^10 stops.
So 1000:1, is 10 stops.
Hope this helps.

Thanks. Yep, that helps; nice and easy :-)

Daniel

Maris Rusis
2-Apr-2017, 15:45
I'm mystified. How can the contrast ratio of a target used to test the resolving power of a film relate to the dynamic range of a film? I always thought dynamic range was a measure of a film's ability to record bright, middle and dark areas in subject matter. Maybe I've confused terminology and arithmetic; not for the first time.

DolphinDan
12-Apr-2017, 11:42
Hi Maris,

I don't know either. Hopefully wiser souls can explain...

What puzzles me now is Fuji claims that VELVIA and PROVIA get 160 and 140 lines/mm resolution @ 1,000:1 contrast ratio, which corresponds to 10 stops of dynamic range. However, they are only capable of capturing about 5-6 stops of dynamic range, so how did Fuji get the 1,000:1 contrast ratio resolving power for these films? Extrapolating the data from 6:1 contrast ratio?

And at 10 stops of dynamic range, which B&W and color negative film can capture, it would seem that color negative film would *in practice* be capable of resolving more detail than slide film with its narrower dynamic range...

Well I am feeling pretty confused at this point. Which is why I posted these questions.

Namaste
Daniel

Nodda Duma
12-Apr-2017, 14:41
I'm mystified. How can the contrast ratio of a target used to test the resolving power of a film relate to the dynamic range of a film? I always thought dynamic range was a measure of a film's ability to record bright, middle and dark areas in subject matter. Maybe I've confused terminology and arithmetic; not for the first time.

The specification values give insight into the film's Contrast Transfer Function, which is very similar to MTF.

Imagine taking a photograph of alternating light and dark bars which can vary in contrast and size. A light and dark pair is a line pair. The size is denoted by how many line pairs are imaged onto the film (line pairs per millimeter or lp/mm).

As imaged line pairs get smaller, contrast universally decreases. So higher contrast (ratio of intensity between light and dark lines) is needed to resolve smaller line pairs. The specifications indicate this. At 160 lp/mm, a contrast ratio of 1000:1 is required to resolve the light and dark lines. At 80 lp/mm, only a contrast of 1:1.6 is necessary to resolve the difference.

Contrast is (light - dark) / (light + dark). Where light and dark are density values for film. So (1000-1)/(1000+1) and (1.6-1.0)/(1.6+1.0).

If you compare this response to the MTF of camera lenses, you'll quickly realize that this film will not normally be the limiting factor for resolution in the resulting image (assuming it's developed correctly).

DolphinDan
12-Apr-2017, 17:07
Thanks Nodda.

Daniel

EdWorkman
13-Apr-2017, 10:12
So if I want to scan, say color negs, what dynamic range is needed for a scanner?