F11 lens with 2x TC = F22?

John Sheehy wrote:

zonoskar wrote:

I know that the 2x TC you lose half the light: your equivalent lens aperture becomes 2x higher, so in case of the new RF tele lenses, you get F22 light gathering. But does that still hold for images quality due to diffraction? The physical aperture is still the same size.

There's a lot of misunderstanding on this topic.

"Losing light" is a concept that was born in a world of cameras without TTL meters. If you metered externally, or judged by the weather, you would need to account for the loss of *EXPOSURE* to get the correct exposure for your film, if you added a TC, a filter, or extension tubes, as the f-number indicated on your lens would not be correct for exposure. In that sense, there was a loss that you needed to account for.

With TTL metering, you don't have to think about that (except with non-circular polarizers), and now, with the actual imaging sensor as the meter, you never have to account for anything.

In the situations in which one adds a TC to the system or uses a longer lens with the same physical aperture size as an alternative to cropping, from the same subject distance, you don't lose anything, because you would have LOST IT IN THE CROP, ANYWAY. Yes, not using the entire frame is also "losing light". People often deceive themselves about IQ based on 100% pixel views or full image frames resized to their monitor, when in fact, they can not use the entire image. When people say that they get better results cropping from a shorter focal length with a lower f-number, at a lower ISO, they are probably wrong, and are judging by full image qualities as seen on screen, or 100% pixel views. They really need to magnify based on the actual crop to be used; not the entire frame. IOW, if you stand at the same distance from the subject and shoot it at 1600/22, 800/11, and 400/5.6, for example, you should open 3 windows on your monitor, and look at the 800 at twice the magnification as the 400, and the 1600, at 4x the magnification of the 400. These "cropped" options usually fall on their faces (especially if the shutter speed is sufficient for the 1600), compared to the 1600, even if none are spectacular at higher magnifications.

When we get into focal-length-limited photography, where we are likely cropping most of the time, it makes no practical sense to worry about absolute exposure, f-numbers, and ISOs as dictating final cropped image quality.

Diffraction, noise, and background blur, all considered with a normalized subject size, from a given distance or perspective, come basically from the size of your entrance pupil, which is the same for a 200/2.8, 400/5.6, 800/11, or 1600/22. It is the 71mm that matters in this paradigm; not the f-number, or f-number-derived diffraction and DOF, and not the ISO. The focal length, combined with the pixel density, determines how well the underlying analog image (which is basically the same) is resolved.

I did a few experiments to test this out and though what you wrote here makes sense I don’t think I’m always actually seeing it in practice so maybe there are additional factors involved.

My first experiment was to make a series of photos of a heron in the distance using my EOS RP and EF 400mm f/5.6L and then another series of the same bird shot from the same location using the same setup but with the EF 2x mk3 in place. The lighting was the same and I used the same shutter speed for all of the photos. The aperture was wide open in all of the photos (f/5.6 at 400mm and f/11 at 800mm). For the 400mm shots I used ISO 800 to give me a shutter speed of 1/250s. For the 800mm shots I used ISO 3200 to get the same shutter speed. For all of the photos the rig was mounted on a solid tripod. I carefully took 10 shots with each setup trying to be as careful as possible with my technique and then picked the sharpest image from each series.

In the 800mm photos the heron filled about 80% of the frame and in the 400mm shots it filled considerably less. I cropped the 400mm shot so the framing matched that of the 800mm shot then I post processed both of the images to the best of my abilities and printed both images at 8x12”.
Both prints ended up looking pretty good but the one from the cropped 400mm shot did look a little better in terms of sharpness, detail and color. It wasn’t a huge difference but it was noticeable.

This wasn’t the result I was expecting so I decided to repeat the experiment with my EF 70-200mm f/2.8L USM at 200mm with and without the EF 2x mk3 wide open. In this case my shutter speeds were higher and ISO lower. I used ISO 400 and 1600 and a shutter speed of 1/800s.
In this case when I compared the prints the difference between the two was small but again noticeable, this time in favor of the 400mm shot.