You get it over more pixels, too, with higher pixel density, without the small losses of contrast and light in the TC, and without making it harder for phase-detect AF to "see". That was part of my point.
Get what over more pixels too?
Define "exposure settings". Do you mean the same Av and TV values? That would not be true, and who uses the same Av value when they add a TC? Is it even possible if you were already wide open without the TC? I can see what you're saying if you mean "the same entrance pupil" (with 1.4x the f-number) and "the same shutter speed", but the entrance pupil size is not a direct part of "exposure settings".
One needs to compare apples to apples. I consider the use of a teleconverter to essentially a longer zoom- perhaps over simplified, but it helps to understand my next few statements. If I take an image of an object at something less than full zoom, the light from that object will cover a certain number of pixels. If I zoom such that the object is doubled in size on the sensor, each pixel gets only 1/4 the light. The other points about the entrance aperture are valid, but reducing the light energy over each pixel due to the increased magnification is a good starting point, and is important later in the conversation.
This link discusses how the light is spread over more pixels at higher magnifications:
http://xrm.phys.northwestern.edu …joubert_mictoday_2011.pdf
a lower magnification, an image representing a larger width
of the illuminated specimen falls on the sensor. At a higher
magnification with the same illumination, the light from a
smaller region of the illuminated specimen is spread over
the same width of sensor, so less light hits each pixel.
Did you actually mean "increase the shutter speed", or, "reduce the exposure time"?
Answered in my reply to Ray above.
In any event, the most common *wise* decision would give the same shutter speed at any same FOV (including reduced FOV if you get into cropping) since pixel count is very close with the 5D4 and 90D, and motion blur as measured in pixels is very similar. A person may choose to increase the shutter speed for a higher pixel count, but that is "meeting the pixels"; not "meeting the image". Noise vs motion blur is always a compromise. The fact that leaning towards less motion blur and more noise can be seen better with a higher pixel count does not mean that you have to do it. It is an optional compromise, but the image created with a higher pixel count and the same, slower shutter speed that might be chosen for a lower pixel count would not be any worse than the lower pixel count image, and still show more detail despite being able to isolate the motion blur better, visually.
I disagree, especially about the part about cropping. Look at my comment above about how the light energy from an object is spread out when magnification is increased. A bird exposed properly at 200 mm will be underexposed at 400 mm at the same shutter speed, ISO setting. Cropping an image is not the same as using a zoom lens. Given the differences, "wise" is not a term I would use. The rest of the statement really depends on the camera and the object being photographed.
And what kind of noise would that be? You just said in another post that you don't believe that read noise is significant.
It's not just me. Please actually read the provided links in that other post. The people who design cameras and sensors say read noise isn't significant.
Do you realize that photon noise is pretty close for all current sensors of the same size,
The size of the sensor doesn't matter with respect to photon or shot noise. It is the size of the pixel. A pixel that is double the size of another in each dimension will collect four times as many photons, reducing the shot noise. Assuming the quantum efficiency, exposure time, light energy (as in frequency), and light power, are the same, shot noise decreases as the square root of the pixel area (see equation here- sensor size isn't listed at all! https://www.pco.de …kb_noise_general_0504.pdf)
Do you have a citation for your assertion that photon noise noise is close for all current sensors of the same size?
or per square millimeter of different sensor sizes, so one camera would be just as good as the next here, based on your belief that read noise is insignificant, and you are correct, if we are talking about higher tones of lower ISO settings.
No, it does depend on the sensor as well
The cameras do vary quite a bit, however, in deep shadows and truly high ISOs, because of read noise, or "camera-added electronic noise" if you want a more semantics-proof label for the problematic noise.
It's not a matter of semantics- using incorrect terminology causes confusion and incorrect conclusions. "Camera-added electronic noise" really doesn't add any meaning, either as it isn't defined.
Analog-to-digital (ADC) noise is more significant, as shown in this link:
https://clarkvision.com …ce.summary/#dynamic_range
Anyway, I think you missed my point. My point is that zooms, whether they vary in open f-number or are consistent across their range, have smaller entrance pupils at the wide end of the zoom. This is to be avoided, if possible. Zooms at the long end can benefit from having a TC if the pixels are too large to fully tap the analog resolution of a lens, but when you zoom out with a TC still attached, to give an angle of view present in the zoom without a TC, you have a smaller entrance pupil and a higher f-number than you would with the bare lens with that same angle of view. That is why I would skip the "5D4 with 1.4x" stage in MatthewK's "reach progression".
Are you trying to describe diffraction-limited optics? What do you mean by "analog resolution of a lens"? In any case, if he is happy with his results, that is the important thing.
(
