Sensor size and MPixel

A forum to ask questions, post setups, and generally discuss anything having to do with photomacrography and photomicroscopy.

Moderators: rjlittlefield, ChrisR, Chris S., Pau

UncleChip
Posts: 59
Joined: Wed Jan 11, 2023 1:56 pm
Location: Uk

Sensor size and MPixel

Post by UncleChip »

Can someone either explain or point me in a direction where I can understand the importance of resolution in relation to sensor size?
For example a 24mp full frame and 24mp crop camera will have a different pixel pitch, this I understand but is there a maximum and minimum pixel pitch to capture fine detail? I am presuming that at some point a pixel becomes too big to resolve fine detail, is there any effect the other way, are you just waisting file size below a certain pitch?

JKT
Posts: 420
Joined: Fri Oct 28, 2011 9:29 am
Location: Finland
Contact:

Re: Sensor size and MPixel

Post by JKT »

The pixel pitch defines how small details you can capture. The optical resulution defines how small details there can be for the sensor to capture. The limiting factor can be either one. If the minimums are relatively close to each other you are using both optimally. If not, you are wasting either file size or optical capability, BUT the result can be sufficient to your purpose either way.

blekenbleu
Posts: 146
Joined: Sat May 10, 2008 5:37 pm
Location: U.S.
Contact:

Re: Sensor size and MPixel

Post by blekenbleu »

Turning the question around, suppose a traditional microscope with a selection of relay/projection oculars for the trinocular port,
such that one ideally projects to APS-C while the other projects to 35mm.
Then, given equal quality optics and sensors of the same megapixel count, each should be capable of the same image quality,
since the same photon rate will be available to either.

Many years ago, larger sensors had advantages because of e.g.
  • so-called fill factor, where there were appreciable insensitive regions between sensors' photosensitive regions
  • electronics (e.g. thermal) noise dominated photon statistics
Those issues are insignificant for modern digital camera sensors.
Practically, for the same pixel count, smaller modern sensors have an advantage:
  • faster read-out time
This allows for shorter exposure with electronic shutter,
which generates negligible mechanical vibration and consequent motion blur.
Of course, motion blur can be worked around using high speed flash illumination....
Metaphot, Optiphot 1, 66; AO 10, 120, and EPIStar 2571
https://blekenbleu.github.io/microscope

UncleChip
Posts: 59
Joined: Wed Jan 11, 2023 1:56 pm
Location: Uk

Re: Sensor size and MPixel

Post by UncleChip »

Thx

For a practical example let’s say you are using the Mitutoyo 5x Mplan, and a FF camera, you can get a 12meg up to a 65meg modern mirrorless camera, the same sensor size but a huge difference in resolution, 12 meg must only resolve a certain size, and do you hit a point where their is no gain in detail, so for example anything over 30 meg will show no additional detail?

Beatsy
Posts: 2105
Joined: Fri Jul 05, 2013 3:10 am
Location: Malvern, UK

Re: Sensor size and MPixel

Post by Beatsy »

UncleChip wrote:
Tue Mar 28, 2023 11:05 am
...
A practical example let’s say you are using the Mitutoyo 5x Mplan, and a FF camera, you can get a 12meg up to a 65meg modern mirrorless camera, the same sensor size but a huge difference in resolution, 12 meg must only resolve a certain size, and do you hit a point where their is no gain in detail, so for example anything over 30 meg will show no additional detail?
The best way to look at this is to consider the smallest distance a lens can resolve multiplied by its magnification factor - then compare that to the size of the pixels on the camera sensor. That will give you the image scale.

A simple formula for calculating the resolution of an objective is 0.61*wavelength/NA (NA is numerical aperture). To keep things simple, we'll choose a middling wavelength of 550 nm (green light). So the formula becomes 335/NA.

Using the 5x Mitty as an example. It has an NA of 0.14, so the smallest resolvable distance on the subject is 335/0.14 = 2393nm roughly 2.4 microns.

Now multiply this by the magnification (5x) and that becomes the size of the smallest resolvable spot on the sensor namely 12 microns (assuming you're using a 200mm tube lens behind the Mitty)

You will want this 12 micron spot to cover something like a 2-3-pixel diameter on the sensor (some may recommend differently), so the pixels need to be 4 to 6 microns pitch - in this example.

A Sony A7rii is 42 megapixels full frame and individual pixels on the sensor are 4.5 microns. Just about ideal for this application (with the 5x mitty).

Hope that helps

UncleChip
Posts: 59
Joined: Wed Jan 11, 2023 1:56 pm
Location: Uk

Re: Sensor size and MPixel

Post by UncleChip »

Thx for that Beatsy

So the Apsc 24meg is about the same pitch,

At the moment I have the A9 and A7iii both 24meg, so these are a bit low in resolution for an ideal macro camera, it is my intention to swap the A9 for the A7r5, but I have also been contemplating the A7rii as a macro only camera, due to the high shutter count of macro work,

Beatsy
Posts: 2105
Joined: Fri Jul 05, 2013 3:10 am
Location: Malvern, UK

Re: Sensor size and MPixel

Post by Beatsy »

You can shoot with full electronic shutter on the Sony cameras (silent shooting) and not use the shutter at all. Trouble with the A7rii is it drops to 12-bit RAW output instead of 14-bit in silent shooting mode. Don't know why. If you need to do heavy contrast stretching, it's quite noticeable. At least it is when compared to the same image done with 14-bit output. I still use my A7rii for extreme macro though. As long as you control the lighting to reduce the need for really heavy contrast stretching in post, the image are fine. I'm just being picky really. However...

I think your A7iii will do the job just fine! With 5.9 micron pixels, you're getting just over 2 pixels diameter under each resolvable spot - that will work. Unless you're printing big, you'll never notice the difference in sampling - certainly not at typical screen resolutions. And happily, the A7iii retains 14-bit RAW output in silent shooting mode. I'd go with that. Your A9 is suitable too - but possibly not *quite* as good IQ as that one's built for speed.

Cheers
Beats

UncleChip
Posts: 59
Joined: Wed Jan 11, 2023 1:56 pm
Location: Uk

Re: Sensor size and MPixel

Post by UncleChip »

Thx again
I use flash as i already have a large strobe setup so electronic shutter isn’t an option, I also forgot about the small batteries and poor viewfinder on the rii, will use the a7iii for now and see what difference the R5 makes when I get it,
Anyhow I have a better understanding now so thx,


Marcepstein
Posts: 293
Joined: Thu Oct 13, 2022 8:39 pm
Location: USA
Contact:

Re: Sensor size and MPixel

Post by Marcepstein »

You can shoot with full electronic shutter on the Sony cameras (silent shooting) and not use the shutter at all. Trouble with the A7rii is it drops to 12-bit RAW output instead of 14-bit in silent shooting mode.
I believe this is also the case for Canon R series. On the R7, there are 2 bugs I found. 1) flash can be used with efcs or mechanical shutter but cannot be coupled with the integrated timer for stacking, and 2). The autofocus stays engaged when focus bracketing and doesn’t always pick the right starting point for stacking. There are 2 work arounds for the latter. CPS is aware of the bugs so maybe they will get addressed in a firmware update.

Marc

Post Reply Previous topicNext topic