sharkmelley

Actually my comment above about data clipping is not the whole story. If you add 100 bias exposures together (from any camera) then what do you see? The fixed pattern noise of course! Now 100 bias ...

Gamma was applied to the target being displayed on a screen to ensure that each numeral had the correct intensity.  However, the image data processed from the camera was linear - no gamma was ...

Hi Rudy, Definitely - that's the only right way to do this. Mark

The skew was to the left . Take a look at the Nikon D5300 histogram Roger posted here: https://www.dpreview.com/forums/post/62319978 Roger was plotting data with linear values - it was my data. Mark

Precisely! You have good intuition on this or you've done Prob & Stats 101. To expand on this, if it were exactly the same set of pixels that had value 588 in each and every frame then Roger would ...

You mean it makes no sense to you ! What do you know of the Central Limit Theorem, for instance? In layman's terms, the CLT says that if you add together a sufficient number of distributions of any ...

Yes I stacked the green channel.  The above RawDigger histograms are for the entire frame but an area of background wouldn't look a lot different. No it's not the case that a stack of 100 frames ...

Brilliant reply! Sorry, I hadn't realised it was the same histogram you showed earlier. Certainly the shape is very odd but I think there's a good explanation by looking at a single ISO 1600 raw ...

I think the Moire effect is probably caused by photographing the regular grid of the computer screen. The D5300 has no anti-alias filter, so perhaps I should have defocused slightly. Mark

Hi Rudy, You can be assured that ISO 400 is absolutely fine.  You won't pick up fainter dust by using ISO 800 or 1600. ISO 200 will also be fine as long as it doesn't introduce background ...

I don't understand that histogram - I've never seen anything like it. Where did you get the data and how did you produce the histogram? Remember, for instance, that the gain on each colour channel ...

My ISO 100 noise estimate (in digital units) is the measured standard deviation in the black frame.  To convert to electrons I multiplied by the gain. This noise estimate therefore includes the ...

Thanks! The same manual settings (1sec at F/11) were used in both cases.  So the exposures at ISO 100 received the same amount of light as the ISO 1600 exposures.  I scaled them differently in ...

As before here is the ISO 100 stack of 100 files: ISO 100 stack of 100 exposures With the same exposure (i.e. solar film, 1sec f/11) here is the ISO 1600 version: ISO 1600 stack of 100 exposures ...

If you want to download a single example of the 100 raw files at ISO 100, it's here: https://drive.google.com/open?id=1OkMr5oWpTkbF-jNY2eWDlR_0p9z8MJGt It might appear completely blank but it ...

Done. I've started a new thread with the results: https://www.dpreview.com/forums/post/62317339 Mark

There was discussion (actually a disagreement) on another thread about how a sensor behaves for very low photon rates. You can find the discussion starting here: https://www.dpreview.com/forums/pos ...

Wow! That is spectacular! An incredibly sharp image yet it looks so natural - none of those horrendous artefacts that we so often see from oversharpening. That Newtonian scope is certainly a great ...

I bought one - a cheap UV LED torch from Amazon. Unfortunately my wife got hold of it and started shining it in the bathrooms (I think you call them restrooms your side of the pond), the food ...

We could discuss this for hours without making much progress. But there's another big problem in your figure 10 illustration that we haven't yet discussed. Let's assume for the sake of argument ...