Thinking of upgrading to a full-frame camera- Z9?

[Manfred M]

Just for you:

D7200 has 3.92um pixels.

Z9 has 4.35um pixels.

Ergo D7200 has 23% more pixel density ...

Yes, but the D7200 is a 5 year old camera, so some of the technology improvements in sensor design are not captured in purely a pixel density comparison.

[Manfred M]

Srijan - the comparison of the D850 to the Z9 is not valid. The Z9 is aimed at the same market segment as the D6.

The closest current camera to the D850 is the Z7ii, which has the same sensor and same market positioning (primarily portraiture and nature / landscape).

The serious birders that I know that shoot Nikon are opting for the D6 / Z9. Bird photographers and sports photographers tend to have the most expensive gear because of high burst mode, low light and long lens requirements.

[xpatUSA]

Yes, but the D7200 is a 5 year old camera, so some of the technology improvements in sensor design are not captured in purely a pixel density comparison.

My original comment was in response to Dan's comment "IMHO, the issue with crop sensors boils down to pixel density." which was only about pixel density. Therefore, my pixel density comparison stands.

[tclune]

OK, just for you:

D7200 has 3.92um pixels.

Z9 has 4.35um pixels.

Ergo D7200 has 23% more pixel density ...

The difference in pixel density is about the same as the difference in pixel density between the D7xxx line and the D500. But the D500 is universally recognized as the superior birding camera. The reason is that difference in pixel density is not seen as significant by any serious birder I have ever encountered, while the other factors are significant. With the enthusiast-grade FF Nikons, the pixel density is approximately half that of the DX cameras. That difference is commonly recognized as worthy of consideration when choosing a camera for birding. FWIW

[Manfred M]

My original comment was in response to Dan's comment "IMHO, the issue with crop sensors boils down to pixel density." which was only about pixel density. Therefore, my pixel density comparison stands.

But is quite misleading. Most of us look at what the piece of equipment delivers in totality, rather that one specific feature.

[Manfred M]

The difference in pixel density is about the same as the difference in pixel density between the D7xxx line and the D500. But the D500 is universally recognized as the superior birding camera. The reason is that difference in pixel density is not seen as significant by any serious birder I have ever encountered, while the other factors are significant. With the enthusiast-grade FF Nikons, the pixel density is approximately half that of the DX cameras. That difference is commonly recognized as worthy of consideration when choosing a camera for birding. FWIW

The appropriate comparison is not the entire D7xxx line, but rather the D500 and its features versus the D7500 and its features as both these cameras are the same generation and in fact use the same sensor and processing engine.

[xpatUSA]

Well, there you have it, Dan. Apparently, the issue with crop sensors does not boil down to pixel density after all ...

[DanK]

Well, there you have it, Dan. Apparently, the issue with crop sensors does not boil down to pixel density after all ...

I certainly didn't suggest that one should consider only one feature. I don't know where that came from.

So let me rephrase: Ceteris paribus, the issue of sensor size for this particular point boils down to pixel density.

Or in case that still isn't narrow enough:

If you are concerned about pixels on the subject, then the issue of sensor size is really one of pixel density.

Perhaps many birders don't care about pixel density. I'm not a birder. I do, when the limitations of the equipment create a serious limitation in the size of the projected image. This is why so many macro photos don't look good: they've been cropped to the point where there isn't enough detail. I've had it happen with birds as well.

[Manfred M]

I certainly didn't suggest that one should consider only one feature. I don't know where that came from.

So let me rephrase: Ceteris paribus, the issue of sensor size for this particular point boils down to pixel density.

Or in case that still isn't narrow enough:

If you are concerned about pixels on the subject, then the issue of sensor size is really one of pixel density.

Perhaps many birders don't care about pixel density. I'm not a birder. I do, when the limitations of the equipment create a serious limitation in the size of the projected image. This is why so many macro photos don't look good: they've been cropped to the point where there isn't enough detail. I've had it happen with birds as well.

Dan - I know a number of highly regarded birders and the two most important features to them are high ISO performance and high burst speed (and of course a long, fast lens). From that respect, we can suggest that pixel density is not a critical issue as they tend to want larger, less noisy pixels. They are often shooting at maximum ISO settings.

Have a look at Daniel Parent's Flickr, page to see which body and lenses he uses. https://www.flickr.com/photos/11107124@N02/

He gets a lot of kudos and awards for his wildlife work, especially his birds.

[DanK]

I think I should drop out of this conversation. I was not attempting to suggest what high end birders should buy. I don't have the expertise to say. I was simply stating what I believe is an emprical fact, which is that if you are far enough away for a given focal length that you can only fill a small part of the frame, sufficiently lower pixel density can create a visibly inferior image. I have seen this concretely when I photograph birds because I am usually too far away for my focal length. It's a serious issue in macro photography because minimum working distance is fixed, and if the creature is small at MWD, you're stuck with few pixels.

I doubt it was an accident that in the era of the 7DII, many wildlife photographers chose it over the 5D III, and that people discussing the rumors of an APS-C R7 often write that they expect it to be aimed at wildlife photographers. But that's a path I won't go down.

[xpatUSA]

I think I should drop out of this conversation.

I think I'll join you ...

[xpatUSA]

... However, I'll almost certainly wait until I am satisfied with a mirrorless APS-C camera, or perhaps even go with a micro four thirds like the new Olympus OM-1. That would avoid the bulk of the FF camera, give me the pixel density I want, and give me the specific features of current mirrorless cameras that I want for my purposes.

I have two cameras with greatly different pixel densities - 20MP Lumix DC-G9 (µ4/3) and 3.4MP Sigma SD9 (below APS-C). The Lumix has an image height of 3888px and a sensor height of 13mm; The Sigma has an image height of 1512px and a sensor height of 13.8mm. Therefore, densities:

Lumix G9: 90,000px/sq.mm
Sigma SD9: 12,023px/sq.mm

However, when cropping but not re-sampling an object from the Lumix to the same pixel size as that from the Sigma, it can be found that the Lumix crop has less IQ e.g. contrast than the Sigma. That is because the Lumix has a significantly lower MTF taken on a per-pixel basis.

[DanK]

I have two cameras with greatly different pixel densities - 20MP Lumix DC-G9 (µ4/3) and 3.4MP Sigma SD9 (below APS-C). The Lumix has an image height of 3888px and sensor height of 13mm; The Sigma has an image height of 1512px and sensor height of 13.8mm. Therefore, densities:

Lumix G9: 90,000px/sq.mm
Sigma SD9: 12,023px/sq.mm

However, when cropping but not re-sampling an object from the Lumix to the same pixel size as that from the Sigma, it can be found that the Lumix crop has less IQ e.g. contrast than the Sigma. That is because the Lumix has a significantly lower MTF taken on a per-pixel basis.

Good point. What I have been interpreting as a problem of pixel density in highly cropped images may be at least in part an issue of MTF/px. The issue is the same for purposes of choosing a camera, but my explanation may have been off.

The effect, whatever it is, can be quite striking. I've ended up discarding the majority of my bird shots for this reason, and I no longer try extreme crops with macro shots, as they never look good.

[pschlute]

Changing from aps-c to FF may create a better image....depending on pixel pitch and tech advances, and of course the photographers ability.

I use FF now, but my previous camera was aps-c with a lower pixel pitch than the FF. Anyone who claims that aps-c (without clarification) has "longer reach" is talking nonsense.

Have a look at the pictures that get posted on your favourite hosting sites. Look at the EXIF of the ones that really grab you. Do you find yourself saying "that is a lovely image but I wish it had more pixels"?

A good photo is made by the cameraman, not the camera.

[xpatUSA]

Changing from APS-C to FF may create a better image....depending on pixel-pitch and technical advances, and of course the photographer's ability.

Equally, it may not with those very same dependencies.

I use FF now, but my previous camera was APS-C with a lower pixel pitch than the FF. Anyone who claims that APS-C (without clarification) has "longer reach" [than FF] is talking nonsense.

I've never liked the term "longer reach" but I do think that it is based on geometry, not spatial resolution.

Have a look at the pictures that get posted on your favourite hosting sites. Look at the EXIF of the ones that really grab you. Do you find yourself saying "that is a lovely image but I wish it had more pixels"?

Not I.

A good photo is made by the cameraman, not the camera.

How true.

[DanK]

Have a look at the pictures that get posted on your favourite hosting sites. Look at the EXIF of the ones that really grab you. Do you find yourself saying "that is a lovely image but I wish it had more pixels"?

Not usually, but (1) many people don't post the images that don't look good, and (2) for most of us, that's a very low resolution.

If you want to see the effects of extreme cropping, look for macro and wildlife photos where were constrained by their equipment to get a small projected image. I'd show you some of mine, but I've thrown them all away.

"Reach" is a bad term because it conflates two things, pixels and angle of view. And as Ted pointed out, the effects of severe cropping on detail may be optics, not pixel count. But whatever the cause, the effect of extreme cropping is real enough. I go fairly often to a nearby refuge and almost always take a camera and long lens, but usually not long enough given the distances, and because of cropping, I've ended up throwing almost all of them out.

[xpatUSA]

I've never liked the term "longer reach" but I do think that it is based on geometry, not spatial resolution.

I guess one could think of "reach" as the distance required for a given focal length to fill the frame, esp. with a particular object.

One could stand at a distance S1 with an "FF" camera that fills the frame with summat. Then pick up one's APS-C camera with the same focal length lens on it and walk backward until that summat again fills the frame at a distance S2.

Walking backward increases the distance thereby increasing the "reach" to S2 compared with S1.

It is no coincidence that S2 over S1 equals the so-called "crop factor" (another horrible phrase) of the APS-C camera.

The above also applies to a µ4/3 camera or a cellphone provided that the focal lengths are equal (not equivalent). Also applies to a medium frame or a view camera except that one would have to walk forward, i.e., end up with "less reach".

P.S. Reach is not a property of a lens.

P.P.S. Reach is not a property of a sensor.

[DanK]

I guess one could think of "reach" as the distance required for a given focal length to fill the frame, esp. with a particular object.

That's precisely how the term is most often used.

P.P.S. Reach is not a property of a sensor.

Not sure what you mean by this, since what you wrote above seems to contradict it. Isn't the smaller image circle on a smaller sensor precisly the reason why the angle of view is smaller and the reach is longer? Seems like basic trig.

[pschlute]

I guess one could think of "reach" as the distance required for a given focal length to fill the frame, esp. with a particular object.

I actually think that most who think that an aps-c camera gives longer "reach" (with the same FL lens) are thinking more along the lines of : "the aps-c camera gives a larger image of the subject than a FF camera, when taking a picture from the same distance, with the same FL lens."

This assumption is correct. However, it leads to a few misconceptions. The major one being that the FL has somehow increased. This has not been helped by the use of the term equivalence by the camera industry. We would all agree I think that if I take an image with any camera and say a 100mm lens, then crop the resultant image so the subject is 1.5 times the size, that it is nonsense to suggest I have used a 150mm lens? Yet this assumption comes up regularly when comparing aps-c to FF.

My first aps-c digital camera was 6MP. My current FF camera is 36MP. The pixel pitch of the FF sensor means that an aps-c sized portion of the image it captures is approx 16MP. So if i took two images of the same subject and cropped the FF one to the same FOV as the aps-c one, my cropped FF image will have 2.6 times the number of pixels. It will be a higher resolution image. If I did the same with a more modern 24MP aps-c camera, it would be a lower resolution image. But in both examples this is down solely to the pixel pitch of the sensor, and little to do with the format.

Equally, I have a "aps-c" mode on my FF camera which produces a cropped aps-c sized image. Whether I use that mode or crop the image myself, the result is absolutely identical.

I do not like the term "reach" as it implies magnification which does not occur at all.

[pschlute]

Isn't the smaller image circle on a smaller sensor precisly the reason why the angle of view is smaller and the reach is longer? Seems like basic trig.

The image circle produced by a lens is identical across all formats. It is the rectangular size of the smaller aps-c sensor that takes a smaller bite out of the image circle that creates the narrower FOV and gives a larger image, and the impression of "longer reach"