Ultimately, the keys to a technically good photograph is to be close and to collect a lot of light. For the documentary photographs most people want to take, t

Ultimately, the keys to a technically good photograph is to be close and to collect a lot of light. For the documentary photographs most people want to take, there is no camera that wouldn't meet their needs. They are close (literally arm's length) and there is usually plenty of light. Any camera will be good under these conditions. The challenge comes when you want to photograph something unusual. The biggest disappointment I've had in recent memory was sitting on a ferry at night, surrounded by what seemed like the blindingly bright and infinitely large city skyline. Basking in the glow, I knew I wanted to take a picture, so I busted out my iPhone. All I got was a blurry sphere or two surrounded by pitch blackness. I wasn't close and I didn't have a lot of light, so the choice in camera didn't work. (I know my a7 would have handled this admirably at ISO 1600. I kind of forgot there were stars in the sky until I started using that thing at night -- I can't see them, but the camera can.) Digital processing can help here -- AI denoising, using frames of video stacked on top of each other and carefully shifted by using accelerometer data to simulate a longer exposure, motors that move the sensor around, etc. But often you can look at one of these pictures and tell that it's synthetic -- not capturing the moment you saw as you saw it, but looking like what you saw shaped by 1000s of other similar moments composited together by an AI. I have yet to find the perfect camera. Big sensors collect a lot of light, but require big bodies and big lenses. Small sensors are always with you, but fail when you're about to take the most interesting picture. There is probably not a workaround for this, so I think you'll find a lot of people with big cameras for years to come.