Michael Reichman (of Luminous Landscape) is also a big advocate of ’shooting to the right’, i.e. exposing so that your histogram makes the most use of the right end of the scale.

Would be interested in anyone else’s experiences.

http://www.dxomark.com/

In addition to what Dave Wilson has already said, I’d like to add that simply increasing the bit depth of the RAW file won’t necessarily make things much better by itself, that’s a red herring. The lower bits (or higher depending whether the RAW data is big- or little-endian) of the 12/14/15 bits per channel that RAW files already contain is mostly random noise, and it’s noise that is the enemy of dynamic range (at the bottom end of the intensity spectrum anyway), as it limits the number of discernible shades of near-black. Eliminate more noise and you have more shades of black to record the image with.

One way to radically improve dynamic range is to combine high-sensitivity CCD pixels with low-sensitivity ones, as Fuji did with previous incarnations of its “SuperCCD SR” technology:
http://en.wikipedia.org/wiki/Super_CCD

Their latest version is somewhat different and even more exciting, as you’ll realise when you read about it:
http://www.dpreview.com/news/0809/08092210fujifilmEXR.asp

In the past Fuji licensed the D200 camera body from Nikon and stuck in their own SuperCCD (and software) to create the S5 Pro:
http://www.trustedreviews.com/digital-cameras/review/2007/04/30/Fujifilm-S5-Pro/p1

Hopefully they’ll do the same again with their new sensor and a new Nikon body (e.g. D3 / D700 / D90) to give us the best of both worlds… or perhaps the other way round, Nikon/Canon/someone else might licence the latest SuperCCD technology from Fuji? Fingers crossed!

If this is right, it should be possible for a manufacturer to increase the dynamic range of a sensor without changing its bit-depth. You would end up with a more quantized brightness signal, but you would still have a wider dynamic range. Hence, I wouldn’t see bit-depth increasing exponentially as you’ve suggested, because for every doubling of the dynamic range you would only need to add one bit to maintain the same quality of signal quantization.

While tone-mapped HDR imagery can produce some really awesome photos that were never possible with film, I think it’s important to remember that a wide dynamic range is only really useful when capturing high-contrast scenes. In fact, one impact of having a higher dynamic range in your camera is that scenes that don’t naturally have a wide range of light levels will appear quite flat (contrast- and colour-wise). Whenever we open up the Levels or Curves dialogs in Photoshop and pull those little arrows inwards to “give it some more punch” we’re quite often *reducing* the dynamic range of the image (although the major manipulation is over the contrast).

I love what HDR can do, but I can’t agree that it’s going to change the course of photography.

I agree 100% – the more bits of resolution I have in my source data, the more I have to work with later in Photoshop or wherever. The point of my comment, however, was that the colour resolution of RAW files already well exceeds the ability of most graphics controllers and displays to show them. To see what I mean, open a 32 bit HDR file (not the tone-mapped output, the original version with 32 bits per sample/96 bits per pixel) and see how terrible it looks on the display.

Back when I was going to school for photojournalism, two years ago, we were told not to bother with RAW. Big, slow, and with timeliness jpgs were the way to go.

So I keep my camera on Raw + JPG currently for that reason. I find Raw far better in most regards, but the times I need quick turn around (half hour at times) my computer works faster with jpgs.

Then I can turn to the raws to process these photos for the more polished versions.

Now when it comes to your last bit about film, I agree that digital will eventually take over. For the time being, the quality of a $50 film rangefinder is still astonishing, so don’t be a dick. Point and shoot digital cameras have got nothing on what was available in film.

When I upgrade my camera, I will definitely revisit RAW and the next time I am in a tricky lighting situation will flip it to RAW and see what it can do for me.

Here’s to the future!

With regard to the new frontier of high dynamic range, you should consider looking into what is really on the horizon – not just high dynamic range, but mixed or selective dynamic range.

Film has an ISO. The whole roll is stuck at that ISO (discounting pushing, pulling, etc). Digital has the advantage of changing ISO on the fly – but you are still using a single ISO for any particular image. High dynamic range just gives you a wider gamut – which is great, a step forward, but there’s something better on the horizon.

There are several patents listed and some new products pending that purport to offer mixed ISO on the same image. That is, dynamic range where the actual sensor sensitivity, and thus the low and high end of the dynamic range actually differ on the same image. This is more like the human eye sees – our eye is not stuck at a single ISO. Not only do we have ‘high dynamic range,’ but our brains also selectively enhance parts of any scene we view – and suppress others.

This will be a true step forward in dynamic range – not just ‘more’ but ‘better’ dynamic range.