Have you ever considered how much of the image your lens projects gets thrown away because there is no sensor to capture it? I started playing around with an FX lens on a DX body and somehow my mind meandered to this topic. It would naturally follow from the opposite arrangement, a DX lens on an FX body, but hey, my noodle is a bit askew.

Lenses are round. Photographic images, at least as they're usually presented, are rectangles. We also know from grade school math that squares are rectangles too. The great grandfathers of today's camera designers knew one of the most mechanically efficient ways to capture the image coming through a round lens was to use a square negative.

Cameras that emulate the 35mm style have sensors with a 3 to 2 dimensional ratio. Nikon's FX sensor is 36 x 24 mm and the DX is 24 x 16 mm. To cast an image on the corner of the rectangle of an FX image, the lens projects a circle that is slightly larger than 43mm. (see photo drawing) Now whether all that image makes it through the camera body, and where it gets clipped is a design issue. The point is the potential image size for those lenses is there.

Now let's think about sensor production. I think it would be safe to say that after re-tooling costs were factored out, a square sensor and a rectangular sensor of the same area, pixel concentration, and technology would cost roughly the same to produce. A square sensor that had about the same area as an Nikon FX sensor would be 29mm across.

My question to you camera guys out there is: Would you consider a camera that had a square sensor of about the same area as your current body if the cost was the same? It would probably be mirrorless; since that would be easier to develop - no new shutters to design, tool, etc. If properly designed, it might be able to use an existing family of lenses.

In my mind, since I flip my body from portrait to landscape constantly, taking that decision away would be a plus. As I stated above, a 29mm square would have equal area to the FX. Make the sensor a bit smaller, say 27mm and issues like vignetting start to diminish if we keep those FX size lenses.

Now let's really go outside the box. What if your sensor array was round? Well not exactly, it would actually be rows and columns of sensors that approximated a round shape. The output could be buffered to make the raw file square (black in the corners), but the body would be designed to capture everything projected by the lens, warts and all. You wouldn't display these images in their native form, but the real estate would be there and you could decide where to crop it.

What do you think?

I love my new/used Pentax K3 and have spent quite a bit of time going through the manual and taking test photos trying to learn all the new features and menu options. Many of my test photos have been taken in "P" mode strictly for convenience, but somehow I frequently end up with "AV-hyper" as the mode description, rather than the "P" mode I selected, at the top left corner of the screen. I have no idea how I got there, why it changed, or even what this mode really is and I don't seem to find any mention of this mode in the manual. Yet, I find myself with this unknown mode selected frequently and have no idea why. Could one of you more experienced Pentax users shed any light of this situation for me? Thanks and good shooting to all!