Is the E-M5 sensor response nonlinear?

Started Jul 10, 2012 | Discussions thread
crames
Regular MemberPosts: 188
Like?
Re: A word of advice.
In reply to Iliah Borg, Jul 16, 2012

Iliah Borg wrote:

The last column is most interesting. Photon noise dominates in midtones and highlights. This type of noise follows Poisson distribution, with dispersion equal to signal. So we can expect that the ratio of the signal to dispersion in midtones and highlights is pretty constant.

It looks like you're measuring a lot more than photon noise with that kind of target. Paper texture, dust, glue, paint chip flatness, etc. are also contributing to the variances. For example, the right side of the white chip has a sigma about 10 higher than the right side, so your ratios are going to vary quite a bit depending on which area of the chip you sample from.

Hard to draw any conclusions from it.

Reply   Reply with quote   Complain
Post (hide subjects)Posted by
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark post MMy threads
Color scheme? Blue / Yellow