Digital zoom and depth of field

[Bill Ferris]

Digital zoom is often regarded as effectively the same as optical zoom except that we lose a lot of pixels in the process.

It is hard to overstate just how many are lost. A 5x crop from a 30MP FF sensor leaves only 1.2MP, which doesn't even fill a FHD monitor's height at 1:1 pixel view.

What could work out nicely is something like a 400MP FF sensor and a very expensive 200mm f/1.4 prime lens heavily corrected to have low aberration wide-open (at least in the center of the image circle). Then, digital zoom can go pretty deep before you have too few pixels or not enough analog resolution. Even if there was still some aberration allowed in the corners of the full sensor (to keep the lens realistic), the aberration could still be low in the center, where you are actually doing the zooming.

However, the DoF of a digital zoom is also different from that of an optical zoom with a constant f-number. (Most zoom lenses keep the f-number constant (as far as possible) as the lens is zoomed.)

A shot taken at focal length f and f-number N and digitally zoomed to focal length nf will have the same DoF as an optical shot at focal length nf and f-number nN.

The PSF of aberration is much larger than the PSF of diffraction with many fast FF lenses (especially the lighter and more inexpensive ones) wide open, and so with such lenses, aberration kills analog resolution in digital zoom faster than diffraction does. I remember that expatusa gave a scenario about a year ago of an 18-200mm zoom, where the total blur size, including diffraction and aberration, was 2x the diameter at 200mm as it was at 18mm, making 18mm sharper, at the frame or pixel level. To some that might be seen as an indication that one should crop from the short end, but if you consider the subject-level blur, from a given distance, it will have 5.5x the diameter at 18mm than at 200mm. We could say that the angular analog resolution is about 5.5x better at 200mm, despite half the sensor-relative analog resolution.

A concrete example: suppose the lens is 25mm f/8 and you digitally zoom by 5x. The shot you get will have the same depth of field as an optical shot at 125mm f/40 (not 125mm f/8 as might be expected).

There seems to be a sizeable number of people who use smaller sensors who won't accept the f-number equivalence, but relish the focal-length equivalence. Long focal-length equivalence can be a cheap parlor trick, because to achieve it, you don't need much pixel resolution or analog resolution; you just need a limited FOV or AOV. With an EVF or LCD, this could make it easier to watch small/distant subjects and see what they are doing or gauge how well AF is sticking to the subject, with the magnified view, but nothing regarding maximum subject quality is guaranteed by long equivalent focal lengths.

Many of us have phone cameras that use digital zoom for the longer focal lengths. Phone cameras often seem to give remarkably large DoF, even at the telephoto settings. This is one of the reasons why.

Fortunately, these lenses designed for small sensors can have very low aberration. If a cell phone has a 20mm f/1.8 lens for telephoto, you can be very sure that it is much sharper lens with a small sensor than any affordable 20/1.8 designed for a FF camera adapted to a small sensor behind it, which will have much more aberration, even in the center of the image circle, despite having the same diffraction.

For more detail on working out the focal length, see this post.

I don't see pixels as being lost, the outer edges of an image are not important when you want a high degree of zoom and intelligent digital zoom has evolved to the point where the interpolation is rather good in many cases (Fuji, Olympus, Nikon) as good as a cell phone. I compared it to frontside teleconverters and in many cases intelligent digital zoom is better.

Not better; different. Interpolation at this level is more akin to graphic design than photography. The resulting image is an illustration not a photograph.

The thing about small sensors I like for birding photography (for example) is that the highly pixel packed sensors allow for higher resolution of fine detail. These cameras should be used at base ISO and in bright conditions and then they will outperform larger sensors at lower focal lengths.

The larger sensor camera is typically at a longer focal length. Cropping and digital zoom don't alter the focal length used. These practices alter angle of view. They also decrease the total light used to make a photo and, as a result make noise more prominent.

Bird photography and exposures compatible with base ISO only go together when perched birds are the subject. Photographing launch, landing, and birds in-flight is a fast shutter speed/low-light genre, even in midday light. If that activity peaks around sunrise or sunset, you're working with even less light.

The notion that a deep crop will result in an image resolving more real detail than an uncropped photo made with a same-size or larger sensor camera is pure fantasy. The cropped photo will be made with less light. It'll be noisier, less detailed and have worse color fidelity. It will also be made with fewer pixels which will compromise image quality even more.

Is light defined at the per pixel level or over the entire sensor?

The total light energy delivered to the sensor is determined by exposure x surface area. If the image is cropped, either in-camera or in post, the surface area of the removed content is light discarded and not used to make the photo.

Doesn't intelligent digital zoom increase the brightness of the image to match the original uncropped image?

Image tonal lightness is maintained. However, cropping (aka digital zooming) reduces the light energy used to make the photo. Light energy captured by any area of the sensor where image content is removed does not contribute to the final cropped image.

Is that what causes more noise? Would downsampling down to a happy medium reduce this issue?

Shot noise in a photo is proportional to the square root of the total signal. The less total light used to make a photo the more prominent shot noise becomes. Downsampling creates a smaller image. Eventually, it becomes so small as to be useless.

Olympus cameras have a feature where they let you take both optical zoom only and +digital zoom versions in the same shot. The optical zoom only version is RAW while the +jpeg version can be +digital zoom. I find it interesting comparing the two since they are taken at the same instant of time.

Yes I fully agree with perched birds being best using this method. For example with the Nikon P900 at full tele I see feather detail I haven't seen with larger sensor cameras.

Please, share sample photos of the same subject photographed in the same light by full frame and 1/2.3-inch cameras demonstrating this. Include the circumstances under which both photos were made.

I have a little trick I use for birds landing or taking off from a feeder or perch though. Focus is set on the feeder or perch and just take pictures of them taking off and landing. The resolution isn't as great as it is with perched birds of course, but it's fun and interesting seeing the patterns on their wings when extended. I've photographed some fun Mockingbird mating dances this way, the male doesn't move closer or farther than the perch so the focal distance doesn't change and just goes up and down and does somersaults and cartwheels in the air before diving back down to the perch. I usually use my M43 gear for this in extended burst mode using the electronic shutter.

To shoot fast motion of birds in flight I sometimes use the 4K Photo mode of my Panasonic FZ300 (a nice little camera, constant f/2.8 aperture) 30 fps and continuous focusing. That might be the best bridge camera that was ever made.

Clearly, you enjoy using the gear you have. I'm not here to argue you should change systems. Keep using what you enjoy and delivers results that please you.

That said, a 1/2.3-inch camera is not resolving finer detail than a larger format system working in the same light and filling the frame with the same subject. Here's a link to the DPR Studio Scene comparing full frame, APS-C, micro four thirds, and 1/2.3-inch systems noise performance and detail when working with the same weak exposure.

https://www.dpreview.com/reviews/image-comparison/fullscreen?attr18=lowlight&attr13_0=nikon_z8&attr13_1=sony_a6700&attr13_2=omsystem_om1&attr13_3=nikon_cpp900&attr15_0=raw&attr15_1=raw&attr15_2=raw&attr15_3=jpeg&attr16_0=6400&attr16_1=6400&attr16_2=6400&attr16_3=6400&attr126_2=1&normalization=compare&widget=1&x=-0.000004640981149778198&y=-0.002585554391930641

 

[OutsideTheMatrix]

Digital zoom is often regarded as effectively the same as optical zoom except that we lose a lot of pixels in the process.

It is hard to overstate just how many are lost. A 5x crop from a 30MP FF sensor leaves only 1.2MP, which doesn't even fill a FHD monitor's height at 1:1 pixel view.

What could work out nicely is something like a 400MP FF sensor and a very expensive 200mm f/1.4 prime lens heavily corrected to have low aberration wide-open (at least in the center of the image circle). Then, digital zoom can go pretty deep before you have too few pixels or not enough analog resolution. Even if there was still some aberration allowed in the corners of the full sensor (to keep the lens realistic), the aberration could still be low in the center, where you are actually doing the zooming.

However, the DoF of a digital zoom is also different from that of an optical zoom with a constant f-number. (Most zoom lenses keep the f-number constant (as far as possible) as the lens is zoomed.)

A shot taken at focal length f and f-number N and digitally zoomed to focal length nf will have the same DoF as an optical shot at focal length nf and f-number nN.

The PSF of aberration is much larger than the PSF of diffraction with many fast FF lenses (especially the lighter and more inexpensive ones) wide open, and so with such lenses, aberration kills analog resolution in digital zoom faster than diffraction does. I remember that expatusa gave a scenario about a year ago of an 18-200mm zoom, where the total blur size, including diffraction and aberration, was 2x the diameter at 200mm as it was at 18mm, making 18mm sharper, at the frame or pixel level. To some that might be seen as an indication that one should crop from the short end, but if you consider the subject-level blur, from a given distance, it will have 5.5x the diameter at 18mm than at 200mm. We could say that the angular analog resolution is about 5.5x better at 200mm, despite half the sensor-relative analog resolution.

A concrete example: suppose the lens is 25mm f/8 and you digitally zoom by 5x. The shot you get will have the same depth of field as an optical shot at 125mm f/40 (not 125mm f/8 as might be expected).

There seems to be a sizeable number of people who use smaller sensors who won't accept the f-number equivalence, but relish the focal-length equivalence. Long focal-length equivalence can be a cheap parlor trick, because to achieve it, you don't need much pixel resolution or analog resolution; you just need a limited FOV or AOV. With an EVF or LCD, this could make it easier to watch small/distant subjects and see what they are doing or gauge how well AF is sticking to the subject, with the magnified view, but nothing regarding maximum subject quality is guaranteed by long equivalent focal lengths.

Many of us have phone cameras that use digital zoom for the longer focal lengths. Phone cameras often seem to give remarkably large DoF, even at the telephoto settings. This is one of the reasons why.

Fortunately, these lenses designed for small sensors can have very low aberration. If a cell phone has a 20mm f/1.8 lens for telephoto, you can be very sure that it is much sharper lens with a small sensor than any affordable 20/1.8 designed for a FF camera adapted to a small sensor behind it, which will have much more aberration, even in the center of the image circle, despite having the same diffraction.

For more detail on working out the focal length, see this post.

I don't see pixels as being lost, the outer edges of an image are not important when you want a high degree of zoom and intelligent digital zoom has evolved to the point where the interpolation is rather good in many cases (Fuji, Olympus, Nikon) as good as a cell phone. I compared it to frontside teleconverters and in many cases intelligent digital zoom is better.

Not better; different. Interpolation at this level is more akin to graphic design than photography. The resulting image is an illustration not a photograph.

The thing about small sensors I like for birding photography (for example) is that the highly pixel packed sensors allow for higher resolution of fine detail. These cameras should be used at base ISO and in bright conditions and then they will outperform larger sensors at lower focal lengths.

The larger sensor camera is typically at a longer focal length. Cropping and digital zoom don't alter the focal length used. These practices alter angle of view. They also decrease the total light used to make a photo and, as a result make noise more prominent.

Bird photography and exposures compatible with base ISO only go together when perched birds are the subject. Photographing launch, landing, and birds in-flight is a fast shutter speed/low-light genre, even in midday light. If that activity peaks around sunrise or sunset, you're working with even less light.

The notion that a deep crop will result in an image resolving more real detail than an uncropped photo made with a same-size or larger sensor camera is pure fantasy. The cropped photo will be made with less light. It'll be noisier, less detailed and have worse color fidelity. It will also be made with fewer pixels which will compromise image quality even more.

Is light defined at the per pixel level or over the entire sensor?

The total light energy delivered to the sensor is determined by exposure x surface area. If the image is cropped, either in-camera or in post, the surface area of the removed content is light discarded and not used to make the photo.

Doesn't intelligent digital zoom increase the brightness of the image to match the original uncropped image?

Image tonal lightness is maintained. However, cropping (aka digital zooming) reduces the light energy used to make the photo. Light energy captured by any area of the sensor where image content is removed does not contribute to the final cropped image.

Is that what causes more noise? Would downsampling down to a happy medium reduce this issue?

Shot noise in a photo is proportional to the square root of the total signal. The less total light used to make a photo the more prominent shot noise becomes. Downsampling creates a smaller image. Eventually, it becomes so small as to be useless.

Olympus cameras have a feature where they let you take both optical zoom only and +digital zoom versions in the same shot. The optical zoom only version is RAW while the +jpeg version can be +digital zoom. I find it interesting comparing the two since they are taken at the same instant of time.

Yes I fully agree with perched birds being best using this method. For example with the Nikon P900 at full tele I see feather detail I haven't seen with larger sensor cameras.

Please, share sample photos of the same subject photographed in the same light by full frame and 1/2.3-inch cameras demonstrating this. Include the circumstances under which both photos were made.

I have a little trick I use for birds landing or taking off from a feeder or perch though. Focus is set on the feeder or perch and just take pictures of them taking off and landing. The resolution isn't as great as it is with perched birds of course, but it's fun and interesting seeing the patterns on their wings when extended. I've photographed some fun Mockingbird mating dances this way, the male doesn't move closer or farther than the perch so the focal distance doesn't change and just goes up and down and does somersaults and cartwheels in the air before diving back down to the perch. I usually use my M43 gear for this in extended burst mode using the electronic shutter.

To shoot fast motion of birds in flight I sometimes use the 4K Photo mode of my Panasonic FZ300 (a nice little camera, constant f/2.8 aperture) 30 fps and continuous focusing. That might be the best bridge camera that was ever made.

Clearly, you enjoy using the gear you have. I'm not here to argue you should change systems. Keep using what you enjoy and delivers results that please you.

That said, a 1/2.3-inch camera is not resolving finer detail than a larger format system working in the same light and filling the frame with the same subject. Here's a link to the DPR Studio Scene comparing full frame, APS-C, micro four thirds, and 1/2.3-inch systems noise performance and detail when working with the same weak exposure.

https://www.dpreview.com/reviews/im...000004640981149778198&y=-0.002585554391930641

I know full frame is better in most cases, but I'm talking about a specific use case. For example, I like astronomy so I use my larger sensor M43 gear to shoot starry landscapes. But if I want to shoot the rings of Saturn without a telescope I would use my P900. What size lens would I need to shoot Saturn's rings on full frame? I haven't seen any images of Saturn's rings (without a telescope) and only one on an APS-C camera, the Sony A6600 with the Sony 70-350mm lens.

I think the densely pixel packed sensor on the P900 helps resolve Cassini's division at 357mm (real focal length-- 2000mm EFL). I can't see it on the Sony A6600 image at 350mm.

The Nikon P1000 takes even better long zoom images especially when stacked. I have seen images of Jupiter and Saturn taken with it that I didn't think were possible without an 8 inch telescope.

 

[Bill Ferris]

Digital zoom is often regarded as effectively the same as optical zoom except that we lose a lot of pixels in the process.

It is hard to overstate just how many are lost. A 5x crop from a 30MP FF sensor leaves only 1.2MP, which doesn't even fill a FHD monitor's height at 1:1 pixel view.

What could work out nicely is something like a 400MP FF sensor and a very expensive 200mm f/1.4 prime lens heavily corrected to have low aberration wide-open (at least in the center of the image circle). Then, digital zoom can go pretty deep before you have too few pixels or not enough analog resolution. Even if there was still some aberration allowed in the corners of the full sensor (to keep the lens realistic), the aberration could still be low in the center, where you are actually doing the zooming.

However, the DoF of a digital zoom is also different from that of an optical zoom with a constant f-number. (Most zoom lenses keep the f-number constant (as far as possible) as the lens is zoomed.)

A shot taken at focal length f and f-number N and digitally zoomed to focal length nf will have the same DoF as an optical shot at focal length nf and f-number nN.

The PSF of aberration is much larger than the PSF of diffraction with many fast FF lenses (especially the lighter and more inexpensive ones) wide open, and so with such lenses, aberration kills analog resolution in digital zoom faster than diffraction does. I remember that expatusa gave a scenario about a year ago of an 18-200mm zoom, where the total blur size, including diffraction and aberration, was 2x the diameter at 200mm as it was at 18mm, making 18mm sharper, at the frame or pixel level. To some that might be seen as an indication that one should crop from the short end, but if you consider the subject-level blur, from a given distance, it will have 5.5x the diameter at 18mm than at 200mm. We could say that the angular analog resolution is about 5.5x better at 200mm, despite half the sensor-relative analog resolution.

A concrete example: suppose the lens is 25mm f/8 and you digitally zoom by 5x. The shot you get will have the same depth of field as an optical shot at 125mm f/40 (not 125mm f/8 as might be expected).

There seems to be a sizeable number of people who use smaller sensors who won't accept the f-number equivalence, but relish the focal-length equivalence. Long focal-length equivalence can be a cheap parlor trick, because to achieve it, you don't need much pixel resolution or analog resolution; you just need a limited FOV or AOV. With an EVF or LCD, this could make it easier to watch small/distant subjects and see what they are doing or gauge how well AF is sticking to the subject, with the magnified view, but nothing regarding maximum subject quality is guaranteed by long equivalent focal lengths.

Many of us have phone cameras that use digital zoom for the longer focal lengths. Phone cameras often seem to give remarkably large DoF, even at the telephoto settings. This is one of the reasons why.

Fortunately, these lenses designed for small sensors can have very low aberration. If a cell phone has a 20mm f/1.8 lens for telephoto, you can be very sure that it is much sharper lens with a small sensor than any affordable 20/1.8 designed for a FF camera adapted to a small sensor behind it, which will have much more aberration, even in the center of the image circle, despite having the same diffraction.

For more detail on working out the focal length, see this post.

I don't see pixels as being lost, the outer edges of an image are not important when you want a high degree of zoom and intelligent digital zoom has evolved to the point where the interpolation is rather good in many cases (Fuji, Olympus, Nikon) as good as a cell phone. I compared it to frontside teleconverters and in many cases intelligent digital zoom is better.

Not better; different. Interpolation at this level is more akin to graphic design than photography. The resulting image is an illustration not a photograph.

The thing about small sensors I like for birding photography (for example) is that the highly pixel packed sensors allow for higher resolution of fine detail. These cameras should be used at base ISO and in bright conditions and then they will outperform larger sensors at lower focal lengths.

The larger sensor camera is typically at a longer focal length. Cropping and digital zoom don't alter the focal length used. These practices alter angle of view. They also decrease the total light used to make a photo and, as a result make noise more prominent.

Bird photography and exposures compatible with base ISO only go together when perched birds are the subject. Photographing launch, landing, and birds in-flight is a fast shutter speed/low-light genre, even in midday light. If that activity peaks around sunrise or sunset, you're working with even less light.

The notion that a deep crop will result in an image resolving more real detail than an uncropped photo made with a same-size or larger sensor camera is pure fantasy. The cropped photo will be made with less light. It'll be noisier, less detailed and have worse color fidelity. It will also be made with fewer pixels which will compromise image quality even more.

Is light defined at the per pixel level or over the entire sensor?

The total light energy delivered to the sensor is determined by exposure x surface area. If the image is cropped, either in-camera or in post, the surface area of the removed content is light discarded and not used to make the photo.

Doesn't intelligent digital zoom increase the brightness of the image to match the original uncropped image?

Image tonal lightness is maintained. However, cropping (aka digital zooming) reduces the light energy used to make the photo. Light energy captured by any area of the sensor where image content is removed does not contribute to the final cropped image.

Is that what causes more noise? Would downsampling down to a happy medium reduce this issue?

Shot noise in a photo is proportional to the square root of the total signal. The less total light used to make a photo the more prominent shot noise becomes. Downsampling creates a smaller image. Eventually, it becomes so small as to be useless.

Olympus cameras have a feature where they let you take both optical zoom only and +digital zoom versions in the same shot. The optical zoom only version is RAW while the +jpeg version can be +digital zoom. I find it interesting comparing the two since they are taken at the same instant of time.

Yes I fully agree with perched birds being best using this method. For example with the Nikon P900 at full tele I see feather detail I haven't seen with larger sensor cameras.

Please, share sample photos of the same subject photographed in the same light by full frame and 1/2.3-inch cameras demonstrating this. Include the circumstances under which both photos were made.

I have a little trick I use for birds landing or taking off from a feeder or perch though. Focus is set on the feeder or perch and just take pictures of them taking off and landing. The resolution isn't as great as it is with perched birds of course, but it's fun and interesting seeing the patterns on their wings when extended. I've photographed some fun Mockingbird mating dances this way, the male doesn't move closer or farther than the perch so the focal distance doesn't change and just goes up and down and does somersaults and cartwheels in the air before diving back down to the perch. I usually use my M43 gear for this in extended burst mode using the electronic shutter.

To shoot fast motion of birds in flight I sometimes use the 4K Photo mode of my Panasonic FZ300 (a nice little camera, constant f/2.8 aperture) 30 fps and continuous focusing. That might be the best bridge camera that was ever made.

Clearly, you enjoy using the gear you have. I'm not here to argue you should change systems. Keep using what you enjoy and delivers results that please you.

That said, a 1/2.3-inch camera is not resolving finer detail than a larger format system working in the same light and filling the frame with the same subject. Here's a link to the DPR Studio Scene comparing full frame, APS-C, micro four thirds, and 1/2.3-inch systems noise performance and detail when working with the same weak exposure.

https://www.dpreview.com/reviews/im...000004640981149778198&y=-0.002585554391930641

I know full frame is better in most cases, but I'm talking about a specific use case. For example, I like astronomy so I use my larger sensor M43 gear to shoot starry landscapes. But if I want to shoot the rings of Saturn without a telescope I would use my P900. What size lens would I need to shoot Saturn's rings on full frame?

I'd reach out to my friends at Lowell Observatory and ask about getting some time on the 25-inch Clark

 

[OutsideTheMatrix]

Digital zoom is often regarded as effectively the same as optical zoom except that we lose a lot of pixels in the process.

It is hard to overstate just how many are lost. A 5x crop from a 30MP FF sensor leaves only 1.2MP, which doesn't even fill a FHD monitor's height at 1:1 pixel view.

What could work out nicely is something like a 400MP FF sensor and a very expensive 200mm f/1.4 prime lens heavily corrected to have low aberration wide-open (at least in the center of the image circle). Then, digital zoom can go pretty deep before you have too few pixels or not enough analog resolution. Even if there was still some aberration allowed in the corners of the full sensor (to keep the lens realistic), the aberration could still be low in the center, where you are actually doing the zooming.

However, the DoF of a digital zoom is also different from that of an optical zoom with a constant f-number. (Most zoom lenses keep the f-number constant (as far as possible) as the lens is zoomed.)

A shot taken at focal length f and f-number N and digitally zoomed to focal length nf will have the same DoF as an optical shot at focal length nf and f-number nN.

The PSF of aberration is much larger than the PSF of diffraction with many fast FF lenses (especially the lighter and more inexpensive ones) wide open, and so with such lenses, aberration kills analog resolution in digital zoom faster than diffraction does. I remember that expatusa gave a scenario about a year ago of an 18-200mm zoom, where the total blur size, including diffraction and aberration, was 2x the diameter at 200mm as it was at 18mm, making 18mm sharper, at the frame or pixel level. To some that might be seen as an indication that one should crop from the short end, but if you consider the subject-level blur, from a given distance, it will have 5.5x the diameter at 18mm than at 200mm. We could say that the angular analog resolution is about 5.5x better at 200mm, despite half the sensor-relative analog resolution.

A concrete example: suppose the lens is 25mm f/8 and you digitally zoom by 5x. The shot you get will have the same depth of field as an optical shot at 125mm f/40 (not 125mm f/8 as might be expected).

There seems to be a sizeable number of people who use smaller sensors who won't accept the f-number equivalence, but relish the focal-length equivalence. Long focal-length equivalence can be a cheap parlor trick, because to achieve it, you don't need much pixel resolution or analog resolution; you just need a limited FOV or AOV. With an EVF or LCD, this could make it easier to watch small/distant subjects and see what they are doing or gauge how well AF is sticking to the subject, with the magnified view, but nothing regarding maximum subject quality is guaranteed by long equivalent focal lengths.

Many of us have phone cameras that use digital zoom for the longer focal lengths. Phone cameras often seem to give remarkably large DoF, even at the telephoto settings. This is one of the reasons why.

Fortunately, these lenses designed for small sensors can have very low aberration. If a cell phone has a 20mm f/1.8 lens for telephoto, you can be very sure that it is much sharper lens with a small sensor than any affordable 20/1.8 designed for a FF camera adapted to a small sensor behind it, which will have much more aberration, even in the center of the image circle, despite having the same diffraction.

For more detail on working out the focal length, see this post.

I don't see pixels as being lost, the outer edges of an image are not important when you want a high degree of zoom and intelligent digital zoom has evolved to the point where the interpolation is rather good in many cases (Fuji, Olympus, Nikon) as good as a cell phone. I compared it to frontside teleconverters and in many cases intelligent digital zoom is better.

Not better; different. Interpolation at this level is more akin to graphic design than photography. The resulting image is an illustration not a photograph.

The thing about small sensors I like for birding photography (for example) is that the highly pixel packed sensors allow for higher resolution of fine detail. These cameras should be used at base ISO and in bright conditions and then they will outperform larger sensors at lower focal lengths.

The larger sensor camera is typically at a longer focal length. Cropping and digital zoom don't alter the focal length used. These practices alter angle of view. They also decrease the total light used to make a photo and, as a result make noise more prominent.

Bird photography and exposures compatible with base ISO only go together when perched birds are the subject. Photographing launch, landing, and birds in-flight is a fast shutter speed/low-light genre, even in midday light. If that activity peaks around sunrise or sunset, you're working with even less light.

The notion that a deep crop will result in an image resolving more real detail than an uncropped photo made with a same-size or larger sensor camera is pure fantasy. The cropped photo will be made with less light. It'll be noisier, less detailed and have worse color fidelity. It will also be made with fewer pixels which will compromise image quality even more.

Is light defined at the per pixel level or over the entire sensor?

The total light energy delivered to the sensor is determined by exposure x surface area. If the image is cropped, either in-camera or in post, the surface area of the removed content is light discarded and not used to make the photo.

Doesn't intelligent digital zoom increase the brightness of the image to match the original uncropped image?

Image tonal lightness is maintained. However, cropping (aka digital zooming) reduces the light energy used to make the photo. Light energy captured by any area of the sensor where image content is removed does not contribute to the final cropped image.

Is that what causes more noise? Would downsampling down to a happy medium reduce this issue?

Shot noise in a photo is proportional to the square root of the total signal. The less total light used to make a photo the more prominent shot noise becomes. Downsampling creates a smaller image. Eventually, it becomes so small as to be useless.

Olympus cameras have a feature where they let you take both optical zoom only and +digital zoom versions in the same shot. The optical zoom only version is RAW while the +jpeg version can be +digital zoom. I find it interesting comparing the two since they are taken at the same instant of time.

Yes I fully agree with perched birds being best using this method. For example with the Nikon P900 at full tele I see feather detail I haven't seen with larger sensor cameras.

Please, share sample photos of the same subject photographed in the same light by full frame and 1/2.3-inch cameras demonstrating this. Include the circumstances under which both photos were made.

I have a little trick I use for birds landing or taking off from a feeder or perch though. Focus is set on the feeder or perch and just take pictures of them taking off and landing. The resolution isn't as great as it is with perched birds of course, but it's fun and interesting seeing the patterns on their wings when extended. I've photographed some fun Mockingbird mating dances this way, the male doesn't move closer or farther than the perch so the focal distance doesn't change and just goes up and down and does somersaults and cartwheels in the air before diving back down to the perch. I usually use my M43 gear for this in extended burst mode using the electronic shutter.

To shoot fast motion of birds in flight I sometimes use the 4K Photo mode of my Panasonic FZ300 (a nice little camera, constant f/2.8 aperture) 30 fps and continuous focusing. That might be the best bridge camera that was ever made.

Clearly, you enjoy using the gear you have. I'm not here to argue you should change systems. Keep using what you enjoy and delivers results that please you.

That said, a 1/2.3-inch camera is not resolving finer detail than a larger format system working in the same light and filling the frame with the same subject. Here's a link to the DPR Studio Scene comparing full frame, APS-C, micro four thirds, and 1/2.3-inch systems noise performance and detail when working with the same weak exposure.

https://www.dpreview.com/reviews/im...000004640981149778198&y=-0.002585554391930641

I know full frame is better in most cases, but I'm talking about a specific use case. For example, I like astronomy so I use my larger sensor M43 gear to shoot starry landscapes. But if I want to shoot the rings of Saturn without a telescope I would use my P900. What size lens would I need to shoot Saturn's rings on full frame?

I'd reach out to my friends at Lowell Observatory and ask about getting some time on the 25-inch Clark

I love telescopes I have an 8 inch one myself but use it strictly for viewing. I like downsizing, so large equipment usually just sits there unused. I also don't like changing lenses at all and need dust resistant equipment (hence the reason why I have the FZ300 and S1, two of the only dust and weather resistant bridge cameras.) When I had more ILCs I found I would just buy duplicate bodies so I would have a body for every lens I owned so I'd never have to change lenses.

 

[Jon_T]

... (hence the reason why I have the FZ300 and S1, two of the only dust and weather resistant bridge cameras.) ...

As usual Incorrect: all the Sony RX10 I, II, III, and IV; the Canon PowerShot G3 X just to note the few I'm aware of off the top of my head. ;-)

The Panasonic FZ30/ FZ50 both had internal zoom lens (manual lens zoom ring) which would prevent dust being drawn into the camera from lens zooming action, the main source of dust intrusion with multiple sections zoom telescoping lenses.

 

[OutsideTheMatrix]

... (hence the reason why I have the FZ300 and S1, two of the only dust and weather resistant bridge cameras.) ...

As usual Incorrect: all the Sony RX10 I, II, III, and IV; the Canon PowerShot G3 X just to note the few I'm aware of off the top of my head. ;-)

The Panasonic FZ30/ FZ50 both had internal zoom lens (manual lens zoom ring) which would prevent dust being drawn into the camera from lens zooming action, the main source of dust intrusion with multiple sections zoom telescoping lenses.

The Sonys and Canon are larger sensor cameras and thus in a different category my friend.

Thanks, for the info on the internal zoom cameras, I've been looking for cameras like that to buy. Are those Panasonic cameras the longest zoom cameras with an internally zooming lens? Thanks.

 

[Jon_T]

... (hence the reason why I have the FZ300 and S1, two of the only dust and weather resistant bridge cameras.) ...

As usual Incorrect: all the Sony RX10 I, II, III, and IV; the Canon PowerShot G3 X just to note the few I'm aware of off the top of my head. ;-)

The Panasonic FZ30/ FZ50 both had internal zoom lens (manual lens zoom ring) which would prevent dust being drawn into the camera from lens zooming action, the main source of dust intrusion with multiple sections zoom telescoping lenses.

The Sonys and Canon are larger sensor cameras and thus in a different category my friend. ...

Note in your post you (or others) were discussion limited to 1/2.3" sensor cameras

Also unless a camera lens and/ or TC is bad, and the user knows what he/ she is doing optical TC will "always" provide better IQ.

As we discussed in your post of the Cooper's Hawk in the Panasonic Talk forum. Don't agree with your statement that the OOC 2x intelligent digital zoom are providing greater detail.

As i posted (with image) and explained "I" can get far better IQ upsizing a image with Post Processing than any camera's Digital Zoom. :-|

Or your post HERE :

"... I find it weird they separate mirrorless cameras from compact cameras, compact cameras ARE mirrorless cameras. ..."