JPEG pictures: settings and comparisons

Following on from my last post on camera testing, I thought I'd air a few observations related to shooting JPEGs. This is particularly relevant if you have a RAW-capable camera but don't use it. And there is a final conclusion on another reason to use RAW.

The underlying premise of my testing was that RAW would be unchanged regardless of camera parameters but that these parameters, as applied to the embedded JPEGs, would affect the histogram and thus ones perception of the dynamics range and metering. And so it turned out. What has that to do with JPEG shoting? Well a lot, actually.

The thing I then came to realise is that the settings in the camera had a very strong influene of the ability of the camera to capture the dynamic range of a scene, convey accurate colours etc. There are a lot of review sites that do comparisons between even DSLRs based on their default JPEG settings. Quite frankly, these are useless test for comparison purposes, except possibly as measures of resolution erlating to pixel count. There is almost certainly more variation to be had within the parameters of any single camera thean there are differences between makes or models. If you do shoot JPEG and do not vary the parameters, you're not getting close to even the burned-in capabilities, let alone all that is lost in not using RAW.

I reckon, if I was shooting JPEG, I'd need greycard shots for all lighting situations (not just because that gives more accurate colours but also because it reduces the risk of blowing a single channel by getting it wrong), variable contrast for different lighting, likewise saturation. I'd be leaning towards the lowest contrast setting to maximise the dynamic range that can be captured. I might well need different settings for different light.

So, if you insist on sticking to JPEG, don't sweat comparisons and learn how to use the camera settings to maximise your return.

Or be a lazy shooter and go RAW. As the base data is the same for a given shot regardless of settings, you'll always get it right (providing exposure is OK). It only takes minimal RAW conversion to get from that to what you'd have got with optimal in-camera JPEG. Thus another reason for RAW - minimal fiddling around at shooting time and less chance of messing up the shot due to poor camera settings.