Poor dynamic range of A7R2 when doing long exposure (RAW files available for comparison)

[voidshatter]

It is well known that the parameters to shoot the Milky Way could be ISO 3200 and 30 seconds to get started without an equatorial mount (which is essential if you want a single exposure to include the foreground). An analysis of darkframe noise was carried out below (link to download RAW files):

 

Welcome, dear visitor! As registered member you'd see an image here…

Simply register for free here – We are always happy to welcome new members!

Analysis of darkframe noise by RawDigger

 

The standard deviation of noise σ is the key to describe the usable dynamic range. The lower the noise is, the better the dynamic range is. It can be seen that the A7R2 performs considerably worse than its predecessor A7R, not to mention that it is even worse than the Canon 5DSR.

 

Trying out different RAW decoders, it could also be seen that the A7R2 performs less than expected when compared against other cameras:

 

Capture One v8.3.2

 

RawTherapee 4.2.269

 

It is unknown whether this can be improved with a firmware update, with or without the side effect of throwing away real stars.

[Singingsnapper]

Wow I love photographs of the inside of the lens cap.  What about real world situations?

[Ben]

Isn't it also well known that bsi sensors are more prone to photo site cross talk due to the reduced boundaries between afining sites? Noise at 0 lumen is higher of course but as soon as light hits the sensors this bad behaviour does not scale linearly. Even if there is a possibility that this would lead to higher shadow noise than midtone noise, it should not concern the common astrophotographer since the low contrast nightsky is best exposed to the right.