Using teleconverters

[Ray]

Only if you force the camera to use the same shutter speed. If you are using auto-metering the shutter speed lengthens to accommodate the reduced aperture

Whilst there's some confusion about the necessity to increase shutter speed when using a teleconverter, over and above what one would consider the ideal shutter speed without converter, if one intended to crop the image; one would definitely not recommend reducing the shutter speed when using a converter, unless the scene were static and one was using a tripod, or unless the light were so bright that the shutter speed without converter would be unnecessarily fast at base ISO.

My example related to the usual situation, when using a telephoto lens, of having to increase ISO to achieve the optimal shutter speed, especially at 400mm. Increasing ISO does not of course increase the number of photons the sensor receives.

This is where it could get interesting and is something I had not considered before: if you crop an image and blow it up, you are taking a fixed amount of light information and spreqding it over a larger area. Is this analogous to the teleconverter taking a smaller amount of light from the 'lens output' and spreading it over a larger area? It makes sense but I hardly see my images fading on cropping but maybe the computer program accounts for it? I guess one way to look at it is how did it used to work in the days of film processing - I never processed my own stuff so don't know.
vs teleconverter then when you crop and blow that crop to the same size as the teleconverter image you are also desaturing the amount of light available?

As I understand, the converter grabs the same amount of light as would exist in the crop of the image without converter, assuming the exposure is the same, and in the process of enlarging the image to cover the whole sensor, reduces the light intensity to a half for each pixel, using a 1.4x converter, and to a quarter for each pixel when using a 2x converter.

However, if you take that same crop without converter and enlarge the image or file size in photoshop, the process of interpolation essentially duplicates each correctly exposed pixel with its full DR and noise characteristics. There is no desaturation nor any fading.

[spidermike]

the process of interpolation essentially duplicates each correctly exposed pixel with its full DR and noise characteristics. There is no desaturation nor any fading.

Does it interpolate when on the screen? It will certainly downscale if I look at an image 'full screen' on a screen less than the pixel limits of the camera (for example a 2000x1500 screen showing a 4000x3000 image). But as soon as I look at it at 100% where one screen pixel=one image pixel and then increase it again there is no interpolation if I look at it 200%....is there? Or at least no interpolation in the sense of creating an image for printing super large: it would stay as 4000x3000 unless I specifically ran an interpolation program.

Whilst there's some confusion about the necessity to increase shutter speed

There's no confusion - if you are using a tc you need to shorten shutter speed to reduce the effect of camera shake. Just like an image cna look wonderfully sharp on the camera LCD only to find it is blurred to heck when viewed on the computer monitor. As ever, the combination of image viewing size and viewing distance dictate what it looks like  - the tc gives you a 'crop' and magnifies it like looking at a billboard from 20 feet and 100 feet.

[Petrus]

There's no confusion - if you are using a tc you need to shorten shutter speed to reduce the effect of camera shake. Just like an image cna look wonderfully sharp on the camera LCD only to find it is blurred to heck when viewed on the computer monitor. As ever, the combination of image viewing size and viewing distance dictate what it looks like  - the tc gives you a 'crop' and magnifies it like looking at a billboard from 20 feet and 100 feet.

Using a teleconverter means you need to raise the shutter speed to minimize camera shake induced blur. If you make the same crop in post, you also need to raise the shutter speed just as much, as you are using the same picture angle, "equivalent" focal length, so to speak. Simple.

It is the image angle which determines the slowest "safe" shutter speed, not focal length. It is just that with 135 systems the rule of thumb used to be 1/focal length. If we use crop factor 1.4x (compared to 135 size) there same rule is 1/(1.4xfocal length). It does not matter if we use teleconverter or cropping or smaller sensor. With a TC it becomes 1/(1.4xoriginal focal length).

[BJL]

So far, according to the points you have raised or agreed with, BJL. it's looking very favourable for the converter.
(1) No loss or change in DoF despite the drop in f/stop number.
(2) No requirement for a faster shutter speed than what you would use without a converter, for a given size image/print of same FoV.
(3) No increase in noise, or reduction in DR, as a result of increasing ISO to accommodate the drop in f/stop, because the larger area of the recorded image offsets such increase in noise.

Wow! Everyone should get a teleconverter. These are amazing devices.

That does not sound so amazing; it is just a list of some respects in which there is a _rough_ equivalence between what you get with the TC vs what you get by cropping instead.  If that were the whole story, one would be better off cropping: one less piece of gear to buy and carry, and no time spent swapping the TC on and off.  (which is why I have yet to find a reason to use a TC.) The main potential advantage of a TC lies elsewhere, in increased resolution by using more pixels: twice as many with a 1.4x TC.  (And my preferred alternative to that would be having a sensor that out-resolves the lens!)

But I do agree that the equivalence is not exact (and no one said it was exact, to please avoid straw man exaggerations of the positions that you are disagreeing with. In particular, I am happy to agree with your point about the possibility that sensor noise is worse with a TC than with a crop, due to spreading the same light over more photo-sites at lower intensity of illumination: same signal (photons gathered from the subject), so same photon shot noise, but likely more dark current and read noise, due to having a greater area of silicon involved and more photo-sites to read out.

I am happy to agree partly because this is a version of something I said many years ago in discussions with you: that the low light noise level advantage of a larger format (here, TC vs crop) depends _entirely_ on also using using a lens of larger aperture diameter, which then gathers light faster from the subject and so delivers "more photons per bird" in the same exposure time.  If instead the larger format is used with a lens of roughly the same size and weight by increasing the minimum f-stop in proportion to focal length and format size (like going from 200mm f/2.8 in "APS-C" to about 300mm f/4.2 in 36x24mm) one gains nothing as far as low light handling, and indeed one likely loses a little due to increased electronic noise produced within the camera.

(3) No increase in noise, or reduction in DR, as a result of increasing ISO to accommodate the drop in f/stop, because the larger area of the recorded image offsets such increase in noise.
...
However, (sorry to introduce a bit of negativity), I'm still not sure about point #3. ...

For example, let's consider the effects of a 1.4x converter. The converter doesn't add any light. It enlarges the image with a corresponding reduction in light per pixel. Each pixel receives just half of the number of photons it would have received without converter, comparing images of the same FoV. ...

[BJL]

Thank you; this should be bookmarked, especially the part that I underlined!  So simple, yet so often misunderstood.

Using a teleconverter means you need to raise the shutter speed to minimize camera shake induced blur. If you make the same crop in post, you also need to raise the shutter speed just as much, as you are using the same picture angle, "equivalent" focal length, so to speak. Simple.

It is the image angle which determines the slowest "safe" shutter speed, not focal length.

[EricV]

... possibility that sensor noise is worse with a TC than with a crop, due to spreading the same light over more photo-sites at lower intensity of illumination: same signal (photons gathered from the subject), so same photon shot noise, but likely more dark current and read noise, due to having a greater area of silicon involved and more photo-sites to read out.

Spreading the light across more pixels is a *good* thing, not a bad thing.  In fact, that is a nice summary of the entire benefit of a TC.  Assuming the sensor does not out-resolve the lens (which is the situation where a TC is better than cropping), spreading the light across more pixels clearly increases overall resolution.  As you point out, if exposure is kept the same, there is no overall light loss, only a light loss per pixel, and there will be some increase in readout noise from the extra pixels.  But in this situation you can improve the result by increasing exposure, providing more total photons and hence less overall noise.  So spreading the image across more pixels is a win-win situation, unless you are constrained to under-expose the larger image.

[BJL]

Spreading the light across more pixels is a *good* thing, not a bad thing.

As I tried to explain, using more pixels can be a good thing is some ways (resolution) but possibly a bad thing in others (more sensor noise in situations that call for an elevated exposure index, so giving the sensor far less than full exposure). As you say:

... there will be some increase in readout noise from the extra pixels. ... spreading the image across more pixels is a win-win situation, unless you are constrained to under-expose the larger image.

Agreed, but such a constraint to underexposing is rather common when using a TC, due to the combination of the higher minimum f-stop it gives with the high shutter speeds often needed with the narrow image angles that TCs are typically used to achieve.  

But let's keep this in proportion: I expect that in many typical cases, the gain in resolution wins over any loss in SNR.  More so because (in my narrow angle photography anyway) there is usually not a huge subject brightness range or deep shadows that I wish to lift, so that the relevant noise is mainly photon shot noise rather than sensor noise, and for that, it is a dead-heat between TC and crop.
(Perhaps I have erred in letting this discussion wander into the often irrelevant domain of total dynamic range or "engineering SNR", which can be quite a different consideration than the local SNR in the relevant parts of the image.)

[Ray]

Using a teleconverter means you need to raise the shutter speed to minimize camera shake induced blur. If you make the same crop in post, you also need to raise the shutter speed just as much, as you are using the same picture angle, "equivalent" focal length, so to speak. Simple.

It is the image angle which determines the slowest "safe" shutter speed, not focal length. It is just that with 135 systems the rule of thumb used to be 1/focal length. If we use crop factor 1.4x (compared to 135 size) there same rule is 1/(1.4xfocal length). It does not matter if we use teleconverter or cropping or smaller sensor. With a TC it becomes 1/(1.4xoriginal focal length).

Thank you; this should be bookmarked, especially the part that I underlined!  So simple, yet so often misunderstood.

BJL, you should know that this is true only with regard to the concept of a "safe' shutter speed, ie. one that gets you a traditional size print of, say, A4 or A3 or A3+ size that is acceptably sharp.

If you want a shutter speed which maximizes the full resolution potential of one's sensor, then pixel count needs to be taken into consideration as well as FoV. The 1/FL rule for shutter speed applied specifically to 35mm prints at the traditional 8"x10" size, which was considered by many before the digital era, to be the maximum size (approximately) that 35mm film could produce whilst retaining  an acceptably sharp and acceptably grain-free image.

Things have changed. I recall shortly after Michael reviewed the Nikon D800, a number of authoritative photographers stressed the fact that in order to benefit from the significantly increased 'potential' resolution of the 36mp sensor,  it would be necessary to use a significantly faster shutter speed than what one was used to using with previous cameras of significantly lower pixel count, even when the lens had VR.

If we take the example of a 2x converter attached to a lens used with the Nikon D800, the crop of the image shot without converter will be a mere 9mp. If one were using the same quality of lens of double the focal length instead of the shorter lens plus converter, that is, a lens of double the focal length which also had the same MTF response as the shorter lens at all frequencies, then in order to maximize image sharpness from the longer lens it would definitely be necessary to increase shutter speed, simply because 36mp is more demanding than 9mp.

The reason for my comment in post #5 expressing doubt about the need to use a faster shutter speed, is due to the unavoidable fact that the insertion of a number of additional and separate glass elements between the lens and the camera will degrade the image to some extent. There will be a drop in MTF response across all frequencies. One is effectively using a lower quality lens with a corresponding lower need for a faster shutter speed to maximise image quality.

However, such loss of quality, hopefully will be more than compensated by the additional resolving power of the 36mp sensor. I would speculate that the degree to which the higher megapixel count more than compensates for the initial degradation, is the degree to which the shutter speed needs to be faster, when using a converter.

This situation seems very analogous to the general advantages of the sensor with the higher pixel count, in the sense that any mediocre lens used with a high megapixel camera becomes effectively a 'good' lens, and any 'good' lens becomes effectively a 'very good' lens, compared with results from a previous model of camera with significantly lower megapixel count, assuming individual pixel quality is similar on both sensors.

This is why I'm rather excited about the announcement of the new Canon 50mp camera. I don't bother selling my old equipment when I buy new equipment. I've still got my first Canon full-frame, the 12.7mp 5D. It'll be interesting to see how much sharper a 12.7mp crop from the 5DS is, when compared with the full 5D image using 2x converter with the same lens. 

[BJL]

BJL, you should know that this is true only with regard to the concept of a "safe' shutter speed, ie. one that gets you a traditional size print of, say, A4 or A3 or A3+ size that is acceptably sharp.

Agreed; and that is the most relevant criterion here: we are comparing the two alternatives of cropping vs using a TC, so IQ comparisons should indeed be done on the basis of equal [apparent] image size, and for that, the same shutter speed gives the same degree of visibility of motion blurring.

On the other hand, to fulfill the TC's potential for a sharper, more detailed image could require increasing the shutter speed over what is needed for the lower resolution image possible without it.  This is the same as the fact that getting the benefit from an increase in sensor or lens resolution often needs an increase in shutter speed, and I have never seen that used as an argument against using a camera or lens that offers higher resolution.