Expose to the right (ETTR) or underexpose?

[douglasf13]

Advertisement (gone after registration)

Exposing Toward the Right, without blowing highlights while using a reasonable ISO has been working just fine for me since I first picked up a D1 at the end of 1999, thanks anyway. Shooting everything at 160? no thanks. I did that with the Digilux 2.

 

You have to separate this conversation into two different categories: IQ vs. usability. It isn't really a cheat or a short cut. It's just like pushing a slow film, which may not always be ideal in use, and can actually make proper exposure more difficult, rather than less. This isn't about short cuts, just discussing the nuance to sensor design and how different types of gains react.

[Advertisement]

The Leica Q3 is available for pre-order at B&H Photo Video and Adorama

[Jaybob]

You have to separate this conversation into two different categories: IQ vs. usability. It isn't really a cheat or a short cut. It's just like pushing a slow film, which may not always be ideal in use, and can actually make proper exposure more difficult, rather than less. This isn't about short cuts, just discussing the nuance to sensor design and how different types of gains react.

 

I disagree. I'd put any D3 file, one that's been properly ETTR'ed at 640 or 800 or 1000 or 1250 up against any photon noise balanced Pentax ISO-less but shot at 160 and then post processed "exposure balanced" file you can come up with.

 

I understand WHAT you're saying, I just see no benefit to doing it that way.

 

What I'm having a hard time getting past the nuances of are statements like "It seems like you misunderstand how digital photography fundamentally works" and "You managed this far because camera makers decided to call camera gain "ISO,"" and "If the photographer doesn't understand the concept, I doubt a client would." and "I'm not going to take the time to do tests which you can do on your own..."

 

Typing makes you seem like a douchebag...is that nuanced enough for you?

Edited October 19, 2012 by Jaybob

[douglasf13]

I disagree. I'd put any D3 file, one that's been properly ETTR'ed at 640 or 800 or 1000 or 1250 up against any photon noise balanced Pentax ISO-less but shot at 160 and then post processed "exposure balanced" file you can come up with.

 

I understand WHAT you're saying, I just see no benefit to doing it that way.

 

What I'm having a hard time getting past the nuances of are statements like "It seems like you misunderstand how digital photography fundamentally works" and "You managed this far because camera makers decided to call camera gain "ISO,"" and "If the photographer doesn't understand the concept, I doubt a client would." and "I'm not going to take the time to do tests which you can do on your own..."

 

Typing makes you seem like a douchebag...is that nuanced enough for you?

 

The above clearly shows that you still don't understand the concept, yet. The point isn't about comparing the D3 to a K-5. It's about comparing a given camera's in-camera analog or digital gain to the boost from a raw converter.

 

If saying that you don't understand the concept is offensive, I apologize. I'm trying my best to explain it from every direction.

 

In regards to my initial post, you said "This statement, from beginning to end, is inaccurate, except for "don't blow highlights." All I'm doing is trying to explain why it is accurate. No need to name call.

Edited October 19, 2012 by douglasf13

[douglasf13]

Ok, I made a super quick, simple comparison with the M9. Both of these shots were captured at 1/125 and f2. One shot was ISO 1250, and the other was ISO 160 with the exposure boosted +2.9 EV in the converter (a full +3.0 EV was slightly too bright.) These shots were made under tungsten lighting and exported with the default LR4 settings to full quality jpeg.

 

Sorry for the foot shot. I wanted some skin tones in the pic.

 

Guess which is which:

 

Image 1:

 

Image 2:

Edited October 19, 2012 by douglasf13

[douglasf13]

And a couple of 100% crops:

 

 

Image 1 crop A:

 

Image 2 crop B:

 

Image 1 crop B:

 

Image 2 crop B:

 

It looks pretty close to me, and Lightroom isn't really the best raw converter for this technique. Can you tell which is which?

Edited October 19, 2012 by douglasf13

[Tonki-M]

While I understand the theory and on some of my cameras I see the difference, I don't do it on my M9. The M9's sensor acts like no other I've ever owned. The amount of detail rendered in the shadows is amazing and the highlight shoulder is very steep. I did my own personal test with some exposure bracketing using the cameras histogram as a reference. In all cases , *for me*, I preferred the colours and tonal range when I pushed the exposure in post compared to pulling. Using ETTR seemed to leave me with a harsh and nasty looking fall off in the highlights.

 

So for me, (and you're free to disagree) my M9 is not an ETTR camera.

 

Gordon

 

nicely put, Gordon

[pgk]

Advertisement (gone after registration)

You have to separate this conversation into two different categories: IQ vs. usability. It isn't really a cheat or a short cut. It's just like pushing a slow film, which may not always be ideal in use, and can actually make proper exposure more difficult, rather than less. This isn't about short cuts, just discussing the nuance to sensor design and how different types of gains react.

To me the IQ vs. usability is the point. What surprises me is that there is so much discussion (some of which is certainly interesting) about theory when its not at all difficult to actually TRY suggested techniques. One of the maddening things about digital photography is a perception by some that understanding every last detail of theory will somehow produce better images. In my experience, getting out and taking photographs and trying things out in an old fashioned practical way is a far more productive approach. I've tried techniques which, whilst they work in practice, are too complicated and time consuming for the marginal increases in apparent (ie actually viewed) image quality, that they produce. And surely this is the point here.

 

Using ETTR is a very valid technique which I use when I consider it appropriate. But its not always appropriate and can sometimes lead to too much time dealing with an images in post processing. Fascinating as much theory is, using it isn't always as practical as might be hoped for, but fortunately we are able to think for ourselves and make decisions about how we apply technique based on our own requirements. Good thread though!

[sandymc]

The difference quoted is for 7 EV placed in the middle of a 11 EV dynamic range (which leaves the rightmost 2 EV unused) vs placed flush with the right edge. +2 EV gains you 128 additional values at the right hand side whereas you lose just 11 values on the left, resulting in a net gain of 117 values, i.e. 233 rather than just 116 values in total.

 

 

When you have just 256 values, max, you don’t want to throw away any more of those if you don’t have to.

 

 

Your original example was, and I quote, "if shifted by 1 EV to the right"

 

Sandy

[dwbell]

While I understand the theory and on some of my cameras I see the difference, I don't do it on my M9. The M9's sensor acts like no other I've ever owned. The amount of detail rendered in the shadows is amazing and the highlight shoulder is very steep. I did my own personal test with some exposure bracketing using the cameras histogram as a reference. In all cases , *for me*, I preferred the colours and tonal range when I pushed the exposure in post compared to pulling. Using ETTR seemed to leave me with a harsh and nasty looking fall off in the highlights.

 

So for me, (and you're free to disagree) my M9 is not an ETTR camera.

 

Gordon

 

Yes. I value and appreciate the ETTR discussion and examples. Did it all with my 5DII at the time as well. I understand the theory and think technical understanding is important, but does not replace "content".

 

In my practical use I expose "correctly" with the M9 using an ambient handheld meter. I don't push the histogram to the right and I'm scared of highlights. The image on the (crappy) LCD always looks too dark and underexposed. The image in Lightroom is always a very happy surprise and enjoyable revalation. I use "zeroed" as my start point which might come into play as no random contrast curve has been applied. I then know what detail I have in the shadows and increase or decrease accordingly based on the intent.

[dwbell]

Ok, I made a super quick, simple comparison with the M9. Both of these shots were captured at 1/125 and f2. One shot was ISO 1250, and the other was ISO 160 with the exposure boosted +2.9 EV in the converter (a full +3.0 EV was slightly too bright.) These shots were made under tungsten lighting and exported with the default LR4 settings to full quality jpeg.

 

That's pretty amazingly similar!

[Alberti]

That's pretty amazingly similar!

Totally agree. I like the second grain more though but that is wholly beside the point of an excellent capability of latitude balancing, if I may phrase it as such. The scene has a very flat exposure though, not any extreme highlights, what might make it easier tho pull up.

Edited October 19, 2012 by Alberti

[dwbell]

Totally agree. I like the second grain more though but that is wholly beside the point of an excellent capability of latitude balancing, if I may phrase it as such. The scene has a very flat exposure though, not any extreme highlights, what might make it easier tho pull up.

 

Good point. But still, with the 3 stops increase and the math Michael pointed out earlier you would expect to see a significant difference, especially at pixel peeping 100%! Print these two out, or reduce for web and you'll not see a jot of difference.

[Jaybob]

Ok, I made a super quick, simple comparison with the M9. Both of these shots were captured at 1/125 and f2. One shot was ISO 1250, and the other was ISO 160 with the exposure boosted +2.9 EV in the converter (a full +3.0 EV was slightly too bright.) These shots were made under tungsten lighting and exported with the default LR4 settings to full quality jpeg.

 

And a couple of 100% crops:

 

It looks pretty close to me, and Lightroom isn't really the best raw converter for this technique. Can you tell which is which?

 

I clearly understand WHAT you're saying, but have no idea WHY you would want to adjust the settings later in a RAW converter, especially if there turns out to be very little difference in end quality. Just because you can do something, doesn't make it a good idea. I can get in my car and drive to New York City from my house in Maryland. The accepted way get there is straight up through New Jersey. I can be there in FOUR HOURS. I could get to NYC by driving up through Pennsylvania as well, driving back roads and avoiding tolls and listening to mix tapes and that will take me SIX hours.

 

Regardless of camera, isn't it WAY easier to set the ISO to 640, use the information your given by your light meter and histogram, and make a proper exposure, let the camera do some of the work? I spend enough time in the Develop module already.

 

Jay

[dwbell]

Jay, I understood it to be a purely technical exercise for this specific thread in this particular forum. Relevant only to this discussion. I don't think Douglas implied changing your shooting habits in any way, did he?

[Peter H]

Jay, I believe Douglas was simply trying to determine whether there were advantages in the method, not advocating it.

[Jaybob]

Jay, I understood it to be a purely technical exercise for this specific thread in this particular forum. Relevant only to this discussion. I don't think Douglas implied changing your shooting habits in any way, did he?

 

No. There were no such demands made.

I'm not sure what he wants.

[kdriceman]

 

I clearly understand WHAT you're saying' date=' but have no idea WHY you would want to adjust the settings later in a RAW converter, especially if there turns out to be very little difference in end quality. Just because you can do something, doesn't make it a good idea. I can get in my car and drive to New York City from my house in Maryland. The accepted way get there is straight up through New Jersey. I can be there in FOUR HOURS. I could get to NYC by driving up through Pennsylvania as well, driving back roads and avoiding tolls and listening to mix tapes and that will take me SIX hours.

 

Regardless of camera, isn't it WAY easier to set the ISO to 640, use the information your given by your light meter and histogram, and make a proper exposure, let the camera do some of the work? I spend enough time in the Develop module already.

 

Jay[/quote']

 

It is a bit of an academic discussion, but, for me, it is good to know that if I make an ISO setting mistake, it is likely not irreparable.

[Jaybob]

2 to 3 stops of RAW "correctablity" is hardly new news, and by no means the exclusive domain of the Leica M9. ACR is an amzing tool.

 

I have a few more questions if you'll allow me....Were the 2 foot shots "just" exposure corrected? Was the 160 ISO image boosted using just the middle or "Exposure" portion of the histogram, or were they given the full treatment, blacks, shadows, exposure, highlights, and whites until they looked "right"? Were they brightness and contrast balanced, was a curve applied, and were either images contrast and color noise reduced?

 

All the same to me, but if I was going to take a picture of my foot, the shooter in me would have split the difference and exposed it at 320 or 400 at a 60th at f2. No disrespect, but THAT's the wheelhouse of the M9 anyway. Then I'd be done with it, I would have skipped all of that post processing, AND have a perfectly exposed (and super sharp) foot photo.

Edited October 19, 2012 by Jaybob

[Advertisement]

Hallo,

Look here: Leica M

[pgk]

2 to 3 stops of RAW "correctablity", be it ISO or EV +/-, is hardly new news, and by no means the exclusive domain of the Leica M9.

Agreed BUT shadows are a problem on cameras like the 5D2 as they quickly shows lines in deep shadows when pulled too hard. In my experience, M9 files are far more tolerant in this respect.

[dwbell]

No. There were no such demands made.

I'm not sure what he wants.

 

No offence Jay, but I'm not sure what you want? The discussion was of the benefits of ETTR. He posted an interesting comparison of severe ETT*L*. What's wrong with contributing to the academic discussion as he did, in your view?