On dynamic range

Started 1 month ago | Discussions thread
ForumParentFirstPrevious
Flat view
bobn2
bobn2 Forum Pro • Posts: 69,811
On dynamic range
12

This is something of a continuation of a discussion on another forum. I continue it here because I know that some of those that frequent that forum are tired of these discussions.

The last post in the discussion, cut short by the thread limit was:

https://www.dpreview.com/forums/post/65503818

The topics at issue here are:

1. That DR is a good indicator of the amount of shadow noise. I suggest that it isn't, for the simple reason that DR is a fraction, and given some DR, you cannot know what the denominator is unless you know the numerator.

The counter to this is that the denominator of DR always has an SNR of 1, so the noise level is known. This is a common mistake. The low end of DR is an SNR of 0. 'Signal' is the applied 'message' to a carrier, and the noise floor is when no signal is applied, that is the SNR is 0/something, i.e., the SNR is zero. The absolute value of read noise is at issue here, since that is what is apparent in the deepest shadows when the DR becomes insufficient to cover the output tonal range.

2. That DR can be calculated simply by dividing the full scale in DN by the read noise in DN. The issue here is that it assumes linearity, and it is known that some cameras are not linear towards the upper end of the range. To properly calculate the DR both the maximum and the noise floor need to be in input referred units.

There are many other red herring raised in the linked post, but the above are the only substantive issues, as far as I can see.

-- hide signature --

Is it always wrong
for one to have the hots for
Comrade Kim Yo Jong?

ForumParentFirstPrevious
Flat view
Post (hide subjects) Posted by
K0d
K0d
ForumParentFirstPrevious
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow