Why are my photos noisy - discussion

"Equivalent" means "equal in value, function, meaning" and they are not, as DannoB's incorrect definition is a simplified one with a narrow application to certain types of photography. It is arguably useful to beginners like DannoB, but the same argument can be made that the exposure triangle is useful to beginners. I do not think it is as the exposure triangle incorrectly teaches that ISO is part of exposure.

I think it is better to educate beginners properly rather than improperly.

As to the concept of equivalence, assuming you mean photographic equivalence, I generally agree with it. However this thread is not about equivalence and I agree with Danno that threads should not be hijacked even by the site owner. This thread is about noisy photos which is a function of exposure but not a function of equivalence as used in this context.

Some people on DPReview said this place was going to become just like DPReview. I had hoped that was not true but it seems that's the way things are going.

Actually, it is a bit more general and covers a wider range, when the light can change during the exposure (strobes, etc.).

I'm not wrong because our definitions are the same 🙂

* exposure - amount of light per unit area that struck the sensor 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.

In a moment we will be discussing the difference between "the shutter was open" and "integration time" ;)

I have updated it to include 'per unit area'

* exposure - amount of light per unit area that struck the sensor 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.

come on Porky your letting the team down 💪💪

some cameras do not have "shutters" ( neither have their lenses )

Well, I did not mention integration on purpose but it is integration...

BTW, there should be "per unit area" there.

sometimes more exposure ( while optimal from S/N standpoint ) is not going well with a particular raw converter ... yes theoretically things on software side are supposed to be designed to allow relevant sliders to be dialed back w/o any ill effects - but it is not always the case...

Yes and yes.

Even I have my limits.

I don't find them as useful as I expected with my camera. They trigger too soon even on the highest tolerance level and they obscure the view including peaking if I try and go a little past their arrival. Works better with my Lumix G9.

Have you confirmed (with a RAW histogram) that their “arrival” is indeed pre-clipping?

So, we have hundreds of posts about exposure and ISO already across various threads. Thanks for the discussion so far (although it appears to be a subject that could be discussed forever).

I think it has improved my understanding.

But how to pull this together as an exposure strategy? I'm going to propose the following for me. Will this lead to disaster?

1. Shoot raw
2. Set camera to base ISO
3. Within constraints of available light, desired aperture and shutter speed, attempt to expose to the right while avoiding clipping of crucial highlights
4. If there is insufficient light, give priority to setting desired shutter speed and aperture even if this leads to underexposure.
5. Do not increase ISO
6. Correct tones in post and apply whatever noise reduction needed to control noise resulting from this strategy.

Justification of this approach (for me)

1. All my cameras use modern Sony sensors
2. No need to bother raising ISO or using auto-iso because boosting ISO only helps reduce read noise, not shot noise
3. Read noise is a minor component of overall noise in modern cameras, boosting ISO to reduce read noise may make a small theoretical difference but the improvement will likely be lost in the wash.

Keeps it nice and simple.

Comments (pragmatic rather than theoretical)?

Thanks

Dave

I do not believe those sensors are fully invariant, and so raising ISO speed still gives you more image quality compared to raising image brightness in post.