This blurb is prompted by the continued emphasis in this forum on basing camera comparison on ISO numbers. I call this the “ISO-centric view”. It may be driven by the notion that the first question that comes to mind for a lot of photographers is what ISO to use for a given scene, before deciding on any of the other aspects that affect the final image. That notion may be useful for film cameras where changing film is cumbersome, but I contend that it is misdirected effort when it comes to digital photography.
Instead, I would suggest to carry out camera comparisons based on fundamental photographic aspects that have more to do with the artistic vision than with the technicalities of how a sensor reacts to incoming light.
Here is what I propose, and I’ll use an example to illustrate the process:
Say, you’re shooting wedding portraits at a windy location against a nice backdrop. Before you even touch a camera, you decide on the framing, the DoF, the amount of motion blur and even on a potential size that the resulting image is going to be printed.
Then you set up the gear to achieve that vision: focal length, subject distance, aperture, shutter speed. You have different cameras, MF, FF, APS-C, M43, etc. You set them all up in the same manner. In the end you snap basically identical images. You are interested in the IQ from the different cameras. You adjust the brightness to match, print or view the images at the same size and then assess SNR, absolute noise, DR, color, etc., whatever you’re looking for.
Why keeping these aspects constant for camera comparisons?
Shutter speed: maintains the same amount of motion blur
DoF: avoids confusion between image sharpness/detail and out-of-focus effects
Framing: avoids having to crop, which would introduce another variable, but could be acceptable
Print size: because that’s how the picture is going to hang on a wall eventually
Where does ISO come in here? In the best case, it doesn’t. Just use Auto-ISO, and let the camera get you into the proper image brightness range. Or you set it manually to make sure important highlights aren’t clipped. Or you digitally adjust image brightness in post. Or you do a combination of that. The point is that, with this approach, ISO (or in the more general sense, image brightness) becomes a secondary aspect, and it plays a minor role compared to the other aspects mentioned above. Not the other way around.
What then if the IQ is not acceptable for a given artistic vision? Then one either has to switch equipment or sacrifice one or more of the artistic aspects in order to collect more light. Or increase scene luminance.
I realize that there are valid comparisons where the photographic parameters are not kept constant. For example, looking at images at 100% will allow pixel-level assessments, or to see just how large an image can be printed before IQ suffers for a given viewing distance.
I’d like to hear if anything speaks against the mentioned approach and/or in favor of an approach based on ISO numbers. Which aspects matter most to you when comparing cameras and why?
Instead, I would suggest to carry out camera comparisons based on fundamental photographic aspects that have more to do with the artistic vision than with the technicalities of how a sensor reacts to incoming light.
Here is what I propose, and I’ll use an example to illustrate the process:
Say, you’re shooting wedding portraits at a windy location against a nice backdrop. Before you even touch a camera, you decide on the framing, the DoF, the amount of motion blur and even on a potential size that the resulting image is going to be printed.
Then you set up the gear to achieve that vision: focal length, subject distance, aperture, shutter speed. You have different cameras, MF, FF, APS-C, M43, etc. You set them all up in the same manner. In the end you snap basically identical images. You are interested in the IQ from the different cameras. You adjust the brightness to match, print or view the images at the same size and then assess SNR, absolute noise, DR, color, etc., whatever you’re looking for.
Why keeping these aspects constant for camera comparisons?
Shutter speed: maintains the same amount of motion blur
DoF: avoids confusion between image sharpness/detail and out-of-focus effects
Framing: avoids having to crop, which would introduce another variable, but could be acceptable
Print size: because that’s how the picture is going to hang on a wall eventually
Where does ISO come in here? In the best case, it doesn’t. Just use Auto-ISO, and let the camera get you into the proper image brightness range. Or you set it manually to make sure important highlights aren’t clipped. Or you digitally adjust image brightness in post. Or you do a combination of that. The point is that, with this approach, ISO (or in the more general sense, image brightness) becomes a secondary aspect, and it plays a minor role compared to the other aspects mentioned above. Not the other way around.
What then if the IQ is not acceptable for a given artistic vision? Then one either has to switch equipment or sacrifice one or more of the artistic aspects in order to collect more light. Or increase scene luminance.
I realize that there are valid comparisons where the photographic parameters are not kept constant. For example, looking at images at 100% will allow pixel-level assessments, or to see just how large an image can be printed before IQ suffers for a given viewing distance.
I’d like to hear if anything speaks against the mentioned approach and/or in favor of an approach based on ISO numbers. Which aspects matter most to you when comparing cameras and why?
