Snake oil in photography

[Prana Ferox]

If you think there isn't snake oil in musical instruments, start a discussion with musicians about tone woods.

Camera sensor noise for a given shot is dominated by sensor tech and pixel pitch. You can mitigate sensor noise by scaling down, which for a given sensor tech and pixel pitch is easier with a larger, higher MP sensor. If you're not printing, you're almost certainly scaling down anyway, to fit on a computer monitor or certainly a phone screen. We also have really, really good post-processing and NR these days; TBH I see people post-processing noise in all the time. Dynamic range tends to fit the same pattern.

I have an Olympus 20MP camera and a 61MP FF Sony A7R4. They're pretty similar sensor tech and pixel pitch and at a pixel level they behave pretty similarly. The Sony takes a helluva lot more pixels in a picture but it has a lot of compromises to do that. They both have their uses.

 

[Offler]

Sorry but this is not comparable. Photography is a means of creating art, so by this logic why not have snake oil in musical instruments? Both a pointless discussion in my opinion.

If you think there isn't snake oil in musical instruments, start a discussion with musicians about tone woods.

Camera sensor noise for a given shot is dominated by sensor tech and pixel pitch. You can mitigate sensor noise by scaling down, which for a given sensor tech and pixel pitch is easier with a larger, higher MP sensor. If you're not printing, you're almost certainly scaling down anyway, to fit on a computer monitor or certainly a phone screen. We also have really, really good post-processing and NR these days; TBH I see people post-processing noise in all the time. Dynamic range tends to fit the same pattern.

I have an Olympus 20MP camera and a 61MP FF Sony A7R4. They're pretty similar sensor tech and pixel pitch and at a pixel level they behave pretty similarly. The Sony takes a helluva lot more pixels in a picture but it has a lot of compromises to do that. They both have their uses.

I have two Olympuses M1x and M5MkIII. I recently started to shoot in a studio with controlled light.

Now, there is a discussion that tells people that larger sensors have larger photosensitive surface, and thus have better noise to signal ratio. SNR is something that people on this forum know a lot about. For sure, larger photosensitive surface can grant better readout, but there is plenty of other things to consider.

1. Not all of the surface of a sensor is photosensitive.
Pixel pitch is a factor, also most current CMOS sensors contain other electronics on the surface, not just photodiode. In general only 25% of currently made sensors surface is photosensitive.

2. NMOS, CMOS and CCD sensors
NMOS and CMOS actually refer to an analog amplifier type which is used for each pixel at the stage just before the signal is converted to digital. NMOS creates most noise, CCD sensors dont have per-pixel amplifiers and CMOS is middle of the road.

CMOS sensors have to sacrifice most of the surface to other electronics.

3. M43, Full-frame and Medium Format.
Take "Fullframe" as a commercial term, its just 35mm frame. People who argue about noise, or ISO performance and would never go for a small sensor (APS-C or M43) are usually not going to buy medium format system anyway, god forbid to get a CCD sensor.

4. Bit depth
This topic is usually completely omitted in the previous debate. Some people do understand that it affects dynamic range, and/or color reproduction, again lowering effects of noise.

I have not met many photographers who do understand technicalities and how sensors do work.



Now... when I get back to my first statement - studio and controlled light:

1. Lets use lowest native ISO for the sensor - its a general rule that at this setting will the sensor perform at its maximum dynamic range and lowest noise.
2. Set aperture to value where each specific lens has best MFT* readout.
3. Learn readout value of your sensor - 1/60s is commonly used.

At this value, you will get the most out of the sensor. With studio strobes, this get little bit more complicated, and even lower values (up to 1/250s without HSS) will yield excellent results, reducing noise even more.

*A value where the image is sharpest.

In general, once you are in the studio and can set the ideal light intensity, any sensor can shine.

 

[Bernard23]

If you think there isn't snake oil in musical instruments, start a discussion with musicians about tone woods.

Camera sensor noise for a given shot is dominated by sensor tech and pixel pitch. You can mitigate sensor noise by scaling down, which for a given sensor tech and pixel pitch is easier with a larger, higher MP sensor. If you're not printing, you're almost certainly scaling down anyway, to fit on a computer monitor or certainly a phone screen. We also have really, really good post-processing and NR these days; TBH I see people post-processing noise in all the time. Dynamic range tends to fit the same pattern.

I have an Olympus 20MP camera and a 61MP FF Sony A7R4. They're pretty similar sensor tech and pixel pitch and at a pixel level they behave pretty similarly. The Sony takes a helluva lot more pixels in a picture but it has a lot of compromises to do that. They both have their uses.

Except you can measure the differences in tone woods, so it's not snake oil. Snake oil is a claim without ant scientific or measurable data to support it. I still don't get the snake oil bit other than more pixels = better images but that's so obviously wrong it's yesterday's nonsense.
Just my opinion but this only matters if you expect to use a camera (as separate from photography) to recreate the truth. The majority of us don't, it's a simulation of reality that's open to interpretation. It's a very different tool to your domestic hifi system in that respect. Of course, using image capture for forensic analysis be it healthcare or manufacturing is of course a different case.