Deliberate overexposure article on Luminous Landscape

gwlaw99

TPF Noob!
Joined
Aug 2, 2011
Messages
4
Reaction score
0
This article suggests that you should over exposure your photographs with digital cameras (without blowing highlights) because there are more tonal values toward the right of the histogram. Then reduce exposure in a RAW editor. The result is that the darker areas of the photograph are now comprised of more tones because the original information captured in RAW has more tones. Thoughts?
Optimizing Exposure
 
It's called "exposing to the right". It's not a new concept, and some people find value in it, some people don't. Mentioning it on a forum is also usually good for several pages of argument.
 
This is traditionally known as "expose to the right", and It's usually what I do. :thumbup:
 
I didn't mean to bring up one of those horrible issues that people argue over for 20 pages. I finally learned and understood the zone system and thought I knew how to get proper exposure, but this seems to change all that. I guess I will experiment.
 
I once read an article that explained how to use the zone system with respect to ETTR and digital capture, wish I could remember where that was. In short, the zone system is totally still relevant... you've just got your headroom in a different place. If you're spanning the full dynamic range of your sensor, then you decide between blown highlights and lost shadows the same as you always would. If you've only got 5 or 6 stops to deal with though, starting at the top (by exposing just shy of blown highlights) and working your way back will ensure the most data captured.
 
I once read an article that explained how to use the zone system with respect to ETTR and digital capture, wish I could remember where that was. In short, the zone system is totally still relevant... you've just got your headroom in a different place. If you're spanning the full dynamic range of your sensor, then you decide between blown highlights and lost shadows the same as you always would. If you've only got 5 or 6 stops to deal with though, starting at the top (by exposing just shy of blown highlights) and working your way back will ensure the most data captured.

That makes a lot of sense.
 
The zone system still has value, knowing where and how to place hilights is important, less you want to trial and error exposures. Meter off zone IX and increase exposure to the limits of your cameras capabilities, usually just under 5 stops. As I had discussed in another thread, think of digital as slide film with tons of latitude and excellent color accuracy regardless of exposure and development conditions. I'm pretty sure if such a film existed, we would have all been exposing to the right long ago.
 
I don't find much value in exposing too the right these days for DSLRs and now go for exposing correct.

This is more relevant to cameras of yesteryear which had poor dynamic range, horrible noise in the shadows and a poor bitdepth in the resulting data. These days with your 14bit ADC you have tones a plenty. The amazing stuff people pull out of shadows with cameras like the D7000 and D300s combined with the low noise output renders the entire topic moot. If you have high ISO noise and you are capable of exposing to the right of where you want to be for the final image, simply lower the ISO. If you're already at the lowest ISO ignore it all, the quality will already be amazing.

The maths and the idea is very sound, but it sounds like a lot of effort for very little gain.
 
The maths and the idea is very sound, but it sounds like a lot of effort for very little gain.

Well, it depends on how much effort you put into it. I think it does have value and I often ETTR. It's almost no extra work since tweaking exposure is a normal part RAW processing.

I do think there are some people who go overboard, and that would mostly be the people who are using settings that will get them a picture that won't really resemble the scene until they process it. Namely, the guys who think you have to use a Unibal to get as accurate a histogram as possible so that you can fully maximize ETTR.

I've tried it, it works, but for me, it's just a bit of overkill.
 
^^ not if you are careful. While it is true that film falls off more gradually, everything within latitude right of Zone V is more or less equally detailed. This isn't the case with film, and data quality falls off dramatically as it reaches the outer limits. BW film handles better, of course, but even then there is a point that grain density is at maximum. It should be pretty easy to predict how the camera will clip out any given scene... How are you metering?
 
If your camera has the tone curve set to anything much over Normal, then exposing a bit to the right makes sense, since the highlights are measured off of the in-camera JPEG images, and not the raw data. "Most" cameras have pretty decent highlight recovery potential on raw files (one of the newer Pentaxes OTOH, is repeatedly dinged for its lack of highlight recovery room). If you have the tone curve set to a higher-than-normal value by NOT exposing to the right, you run the chance of under-exposing more than needed. Some of the newer sensors have AMAZING shadow recovery potential, and the overall dynamic range possible with the best of the newer sensors is remarkably better than when MR and TLL were pushing ETR so hard. On the other side of the scale, today's newer raw conversion software has highlight recovery capability that simply was NOT in existence back when those ETR articles were first written; highlight recovery is a relatively "new" software capability, and it is now better than ever before.
 
It's called "exposing to the right". It's not a new concept, and some people find value in it, some people don't. Mentioning it on a forum is also usually good for several pages of argument.

Personally I think the only reason that there is some debate about the method is because people approach their photography in different ways. Exposing to the right is a method aimed at harvesting the most potential light data from a scene without compromising the data (ie not overexposing or clipping the highlights) and then taking that data and editing it to get the end result. The method does indeed work for what it does, however it can sometimes add more editing time to the process (esp if you clipp some highlights).

For those that prefer a lesser editing phase then the expose to the right, whilst a valid method, just does not fit into their intended workflow; so its not a method they feel worth adding for its potential gains against the loss of their current working model of practice.
 

Most reactions

Back
Top