ok, ok ๐ Now you are really over complicating things. I'm sticking with just Good data to Bad data ratio for the sake of maintaining what sanity I still have at my 60+ years of age ๐
You don't need to stick with wrong. It is very important to realize that signal is everything, including noise.
As to over-complication, I'm not suggesting that you dive into why, under certain circumstances and for certain purposes, noise isn't bad data.
Let's see if at my 68 years I understand this. Signal is overall data. Noise is bad data. SNR is an indicator of how much bad data in overall data or how much good data in overall data.
SNR - Signal to Noise Ratio - is a ratio or a fraction if you like - S/N. If the numerator (top number, S) is larger than the denominator (bottom number, N) then there is more Signal than Noise in the data.
ok, I think I'll just go back to Plan A and just remember that to minimise visible noise I need to maximise the SNR in my raw data by maximising the exposure* within my artistic constraints and without clipping highlights without having to understand the nuts and bolts of what SNR actually is in detail.
* exposure - amount of light that struck the sensor per unit area while the shutter was open
** optimal exposure - the maximum exposure* within dof and motion blur requirements without clipping important highlights.
*** under exposed - more exposure* could have been added with the DOF and blur constraints still being met without clipping important highlights.
Yep ๐ ๐ especially to software manufacturers. Some do it better than others but that is not what this thread is about so let's not go there please.
While they have the same units and can be divided, they are still different animals. First, there are some assumptions: a signal of constant amplitude, and additive zero mean noise. The "additive" is not true even for shot noise but there is a way around it. Then S is the mean, and N is the standard deviation of the noise typically. Not the "level"/mean of the noise, which is zero by assumption.
I agree that the discussions are more advanced than necessary for beginners. However, it is often hard to put yourself in other people's skin (knowledge-wise). Therefore beginners need to ask questions about unclear explanations so that knowledgeable people understand what needs to be adequately explained.
The summary for "how to minimize noise" is all in the first post
Maximize exposure:
Aperture as wide as possible unless you need it set narrower to expand depth of field
Shutter speed set to the lowest value consistent with acceptable subject movement
Add light if appropriate and practical
Adjust ISO to achieve acceptable image lightness, but not so high as to blow out the bright areas of the images. That's much more important for JPGs than for raw.
(or just use Auto ISO, perhaps with a bit of negative EC for safety.
A successful answerer does need to be able to adopt the mindset of a beginner. As far as noise is concerned, beginners may not even notice it. But when they do, the question is likely to be: "What can I do to avoid the speckly effects on the black parts of this photo?" The appropriate answer might something like be: "Those speckles are noise and are due to not enough light entering the lens. The way to avoid, or reduce the noise is to do this." (Better illumination is not a good answer, as it may not be possible!) Of course, an acceptable answer may not be possible without asking for more details of the settings that were used.
The signal to noise ratio of this thread is far too low!
Today's beginners are not using Kodak Brownie cameras, they are using highly sophisticated devices, like the canon R series. They are faced with more buttons, knobs and menu choices than they know what to do with. Thus the best advice is for them to start by setting the camera to automatic. On my R6, choosing A will set exposure and a simple focus mode. The first question that the beginner is likely to pose, is "how do I make the camera focus on Aunt Sally, rather than choosing to focus on the bookcase?"
In fact, one of the best night photos I have taken was done with the automatic night scene mode on my RP, which I believe combines multiple shots, and thus reduces noise.
Proceeding from there it would be good to deliberately take some lousy photos, with the controls set inappropriately. Then take the same scene with controls optimised. In this way, one can show, not only noisy photos, but blurred photos, too light and too dark photos, etc. The comparisons need to be very obvious.
Remember that a picture is worth 1000 words, particularly to a beginner, and an area might be set up here that demonstrates how to avoid these mistakes, and explains the controls. For the Canon EOS R, I always recommend the series of cheap Ebooks written by Nina Bailey. I dont always agree with all she write, but these books can take a beginner a long way in the right direction. She does not talk down to her audience, but explains things simply without any theory.
Are you a beginner, or are you explaining what a beginner would ask?
IMO, to get back on track (OP), we need beginners asking questions, not advanced users figuring out what a beginner would ask.