Jump to content


Photo
- - - - -

Expose to the right?

technique exposure

  • Please log in to reply
18 replies to this topic

#1 adamhanlon

adamhanlon

    Harbor Seal

  • Admin
  • 2221 posts
  • Gender:Male
  • Location:Lancaster, UK

Posted 20 May 2015 - 12:18 PM

Hi all,

 

In the recent "Wetpixel Ask the Pros" column on wide-angle techniques, one of the questions posed was whether the pros "expose to the right"?

 

Expose to the Right (ETTR) is a technique that deliberately biases exposure towards the highlight end of histogram. The reason for doing this is that nearly 80% of the sensor information is actually recorded in one third of the tonal range-at the highlight end. By exposing the image "perfectly", distributing the sensor information uniformly throughout the histogram range, effectively we are discarding a large amount of image data.

 

For a more in-depth explanation please see Wikepedia or The Luminous Landscape.

 

The technique has generated lots of debate!

 

So, with this in mind, it would be great to know how the Wetpixel community control exposure. Do you Expose to the Right? Or do you go for a uniform distribution?

 

Personally, I try to expose to the right, particularly in scenes that are well lit. In darker scenes, I find it can be a challenge to achieve. The ability of usable high ISO with modern cameras really helps.

 

Adam


Adam Hanlon-underwater photographer and videographer
Editor-wetpixel
web | Flickr | twitter | Linkedin | Facebook


#2 LauadiBaby

LauadiBaby

    Damselfish

  • Member
  • Pip
  • 19 posts
  • Gender:Female
  • Location:Los Angeles

Posted 20 May 2015 - 05:00 PM

I always try to expose to the right, but it's not always possible to exploit the brightest stop as much as I'd like. Nonetheless, I'm always THINKING about exposing to the right, which means I'm always analyzing the histogram to see how I can maximize the potential of each image, data-wise. The histogram is a critical tool for me to get the best shot I can in camera, thus the best image to work with in post. For me the two are inextricably intertwined. My version of "nailing it in the camera" means gathering the most possible RAW data available about my shot, which results in the richest possible file and the best final image. I'm likely to spend a lot less time in post with a nice juicy ETTR file.

 

An ETTR file often looks flat and overly hot on the back of the camera, but here's the thing - the LCD preview and histogram don't accurately represent the way your RAW capture will look. They're based on an sRGB (smaller color space) preview, so the histogram may be as much as a stop off. A "properly" exposed shot, with the lion's share of the image creating a "mountain in the middle" is a shot in which you may have thrown away fully half of the available image data. I usually shoot until I see just the teensiest highlight warning blinkies and call it good.

 

Managing the dance between shutter speed, f-stop and ISO can be complex. I used to do everything possible to avoid raising the native ISO, but I've changed my ways. Though most newer cameras are built to handle high ISO magnificently, the sensor still collects data in the same linear fashion, with 2048 of a possible 4096 (12-bit RAW) tones collected in the brightest stop. That means that underexposing (no ETTR) can result in a hefty potential data loss. I'm not advocating blown highlights here. Lightroom can restore almost 2 stops of highlight data, but nothing kills a shot faster than a blasted sunball or badly clipped highlights.

 

Sensor-Data-ETTR.jpg

 

For many, "nailing it the camera" is code for not having to edit in post, so it follows that ETTR isn't at all enticing to those with an aversion to editing. I think there's a misconception held by many that all editing takes forever, and is only needed to rescue sub-par images. Not so. Finessing an ETTR exposure can take just moments in Lightroom or Adobe Camera Raw. Fluency in editing can be as much the hallmark of an accomplished shooter as their photos are. Shooting and editing inform each other, and post-production provides a set of tools like the darkroom used to, that gives the artist an extra level of control over the finished photo. 

 

 

 

 

 

 

 



#3 MikeVeitch

MikeVeitch

    1.7kbps Manta Boy

  • Senior Moderator
  • 6309 posts
  • Gender:Male
  • Location:In Bali, Indonesia but from Vancouver, BC
  • Interests:Teaching Underwater Photography

Posted 21 May 2015 - 02:09 AM

Interesting following this one along on FB and here.  Erin, could you please post a "non" edited ETTR image that shows a lot of blue water?  A typical WA scenic would be a good one to show what the blues look like when using this style before the editing process.


Join us for an Underwater Photography Workshop in the Lembeh Strait at NAD Lembeh with Doug Sloss in 2018
Blog and Photo Archive/Portfolio Site www.mikeveitchblog.com
Learn underwater photography in the ultimate classroom, Bali! or join us on a trip www.underwatertribe.com and www.baliuwphoto.com

Join us for a trip in Indonesia in Komodo or Raja Ampat


#4 adamhanlon

adamhanlon

    Harbor Seal

  • Admin
  • 2221 posts
  • Gender:Male
  • Location:Lancaster, UK

Posted 21 May 2015 - 11:51 PM

I am currently reviewing several underwater photography textbooks that have been released in the past year or so, and it is interesting to note that they advocate shooting at base ISO whenever possible. In fact one suggests stopping down is preferable to moving off 100 ISO!

 

They are written by folks that did a lot of film shooting. 

 

I think that with newer cameras, ISO is another exposure control, and this really opens up the possibilities with ETTR. Reichman et al were using this technique for landscape photography with cameraS on tripods, which are not really a viable option for underwater shooting. The fact that we can get clean images at ISO 1600 and beyond now means that this is now viable.

 

I confess that I still find it difficult to deliberately expose "wrongly". :)

 

Adam


Adam Hanlon-underwater photographer and videographer
Editor-wetpixel
web | Flickr | twitter | Linkedin | Facebook


#5 SwiftFF5

SwiftFF5

    Manta Ray

  • Member
  • PipPipPip
  • 494 posts
  • Gender:Male
  • Location:Virginia, USA
  • Interests:Water Rescue (Public Safety Diving) - yeah, I'm a mud crawler.

Posted 22 May 2015 - 04:02 AM

I am currently reviewing several underwater photography textbooks that have been released in the past year or so, and it is interesting to note that they advocate shooting at base ISO whenever possible. In fact one suggests stopping down is preferable to moving off 100 ISO!

 

They are written by folks that did a lot of film shooting. 

 

<snip>

 

 

That's an interesting observation - and probably a very important one.  My main background is in film photography, and I am still learning how different the digital world is.  A lot of the skills that I had learned just aren't needed or are wrong, so digital can be a challenge still.  But, that's also what makes it interesting, and why I appreciate all the great advice here on Wetpixel. 


Canon Vixia HF-S30, L&M Bluefin Pro
Public Safety Diver

#6 buddy

buddy

    Sting Ray

  • Member
  • PipPipPip
  • 292 posts
  • Gender:Male
  • Location:Switzerland
  • Interests:diving, photography, golf

Posted 22 May 2015 - 02:37 PM

...

I think that with newer cameras, ISO is another exposure control, and this really opens up the possibilities with ETTR. Reichman et al were using this technique for landscape photography with cameraS on tripods, which are not really a viable option for underwater shooting. The fact that we can get clean images at ISO 1600 and beyond now means that this is now viable.

...

 

Adam

 

Exactly my opinion!

 

I used to shoot underwater at base ISO, let's say ISO 100. Today, I usually shoot with ISO 400 or higher. That let me get a smaller f-stop (f11 or smaller), a faster shutter speed (1/125 sec or faster), lower flash output (1/4 power or less, which means less backscatter) and faster flash recycle.

 

When I was recently in Cat Island for the close-up action of the oceanic white tip sharks I thus was able to shoot 5 images per second (to select the best image and throw away the others in the same series...)

 

And funny was  -after 1 day of shooting-  the other photographers did the same from the second day on....

 

What a wonderful world!

Attached Images

  • 1505-Bahamas-0121-1.jpg
  • 1505-Bahamas-1682.jpg

Juerg
www.jvpictures.com
Nikon D4s, Nikon 16-35 /f4 VR II, 70-180,105mm f/2.8, Subal ND4 housing and ports, 2 Subtronic Fusion (flash and video lights), Sigma 15mm /f2.8, Subal 4" Minidome


#7 LauadiBaby

LauadiBaby

    Damselfish

  • Member
  • Pip
  • 19 posts
  • Gender:Female
  • Location:Los Angeles

Posted 23 May 2015 - 06:44 PM

Because I edit and print, I don't mind a washed out image straight from the camera. Better that than a dark one, by far.

Attached Images

  • EQ_ETTA-1.jpg
  • EQ_ETTA-2.jpg
  • EQ_ETTA-3.jpg

Edited by LauadiBaby, 23 May 2015 - 06:47 PM.


#8 LauadiBaby

LauadiBaby

    Damselfish

  • Member
  • Pip
  • 19 posts
  • Gender:Female
  • Location:Los Angeles

Posted 23 May 2015 - 06:48 PM

Here's one that wouldn't upload in the last post...I'll post the edited versions when I get a minute :-)

Attached Images

  • EQ_ETTA-7.jpg

Edited by LauadiBaby, 23 May 2015 - 06:50 PM.


#9 frogfone

frogfone

    Triggerfish

  • Member
  • Pip
  • 48 posts
  • Gender:Male
  • Location:Livingston, UK

Posted 24 May 2015 - 02:04 PM

Is anyone using this with a canon digic 4 processor? these are 14 bit so even the darkest tone will have 256 levels available. i will still give it a go with some landscape/cloud shots but any feedback from actual underwater would be useful.

 

 

Roddy



#10 LauadiBaby

LauadiBaby

    Damselfish

  • Member
  • Pip
  • 19 posts
  • Gender:Female
  • Location:Los Angeles

Posted 24 May 2015 - 03:16 PM

I'm shooting a Canon 1DX - I'll ask my pal at Canon what he thinks. Even though the Digic 4 is 14 bit, I think the sensor still collects data in the same fashion, so although the numbers of available tones would change, the percentage collected for each stop would remain the same. 

In the meanwhile here are some before and afters, histograms included.

 

20090612_Bahamas_UW_1343_ETTRc.jpg

 

20131208_Komodo_00640_ETTRc.jpg

 

20150428_Bonaire_0656_ETTRc.jpg

 

20150430_Bonaire_0872_ETTRc.jpg


Edited by LauadiBaby, 24 May 2015 - 03:19 PM.


#11 adamhanlon

adamhanlon

    Harbor Seal

  • Admin
  • 2221 posts
  • Gender:Male
  • Location:Lancaster, UK

Posted 25 May 2015 - 03:51 AM

Thanks Erin,

 

How are you "correcting" the exposure in post?

 

Adam


Adam Hanlon-underwater photographer and videographer
Editor-wetpixel
web | Flickr | twitter | Linkedin | Facebook


#12 errbrr

errbrr

    Eagle Ray

  • Member
  • PipPipPip
  • 388 posts
  • Gender:Female
  • Location:Melbourne, Australia

Posted 25 May 2015 - 07:00 PM

Interesting topic. I find I'm much more tolerant of too dark spots in my final images than I am of blown out highlights. As such I tend to take the highlights as far up as I'm comfortable with and use that as my measure of the "correct" exposure in camera. Then I routinely pull the highlights back just a little bit using Lightroom. I also think that overly dark images in camera can often reveal way too much backscatter when lightened up afterwards. 

 

And the fact that I'm normally shooting in the dark means the glowing camera screen can make the image reviews look brighter than they are. After tripping over that fact a few times I now try to increase the exposure in camera to compensate, especially when I've just taken a shot I really like. 

 

So overall yes, I do aim for exposure a little bit right of the middle, but not particularly far. I like your post-processing Erin - a good view of leaving the subject bright and clear and pulling the background down a bit.



#13 frogfone

frogfone

    Triggerfish

  • Member
  • Pip
  • 48 posts
  • Gender:Male
  • Location:Livingston, UK

Posted 28 May 2015 - 05:42 AM

Laudibaby i think your before and after images speak for themselves are they all just exposure corrected as the shark shot looks sharper and the Nudi seems more vibrant. would be wonderful if that was all from just a simple adjustment to how we shoot. i will give it a try this weekend.

 

thank you so much.  :notworthy:  :notworthy:  :notworthy:



#14 adamhanlon

adamhanlon

    Harbor Seal

  • Admin
  • 2221 posts
  • Gender:Male
  • Location:Lancaster, UK

Posted 29 May 2015 - 12:48 PM

I think to is important to note that in Erin's images above, she has not blown the highlights. All the image info is contained within the histogram, there is no clipping. The principle of ETTR is that exposure is shifted towards the right, but not clipped. 

 

I don't think that an LCD is suitable for judging exposure. In order to get accurate exposure, it is crucial to consult the histogram.


Adam Hanlon-underwater photographer and videographer
Editor-wetpixel
web | Flickr | twitter | Linkedin | Facebook


#15 Paul Kay

Paul Kay

    Giant Squid

  • Industry
  • PipPipPipPipPip
  • 1793 posts
  • Gender:Male
  • Location:North Wales, UK

Posted 30 May 2015 - 12:02 PM

I have tried experimenting with ETTR and 'normal' exposure. My personal conclusion is that ETTR offers little if any advantages in practice. My opinion FWIW, is that any increased information supplied by the biased exposure is IMO often undermined by subtle but often irritating shifts in tonality produced when readjusting exposure in post, which then require additional adjustment and on occasion I have found that they cannot be thoroughly compensated for. Blown highlights on the other hand are blown. In situations where the contrast is far too high for the sensor to handle I tend to try to decide what can be blown, if anything, and expose accordingly.

 

On the subject of ISO, I tend, on my land cameras (digital Leicas), never to adjust from base ISO (160) and adjust/compensate underexposure in post. If you think about it, increasing ISO is pre-setting another adjustment in-camera - in this case 'gain'. Software has become good enough to allow 'gain' to be applied afterwards, to an underexposed shot. In many circumstances anyway. Problems do occur if too much ''gain is applied to the deepest shadows where banding can occur. With the Leicas this is almost negligible, on my Canons it can be a problem - clearly they are slightly different somehow (CCD vs. CMOS?).

 

And when all is said and done I've also tried experimenting with 'inadequate exposure' and find that post processing can actually produce very acceptable results from mis-judged exposures. My own personal rules on photographic technique are that there are no rules. Photography is practical, if in doubt about what something will produce, try it and be objective about the results. It can be surprising and disconcerting.  


Paul Kay,Canon EOS5DII SEACAM c/w S45, 8-15, 24L,35L, 60/2.8 (+Ext12II) & 100/2.8 Macros - Sony A7II SEACAM 28/2 & 50/2.8 Macro - UK/Ireland Seacam Sales -see  marinewildlife


#16 Hal

Hal

    Starfish

  • Member
  • Pip
  • 9 posts
  • Gender:Male
  • Location:New York

Posted 03 June 2015 - 02:50 AM

The information stated here so far is correct, but what I feel is the most important reason for "ETTR" was not mentioned, and that is digital noise. We all know the limited light conditions underwater, and the inherent problem of underexposure. When digital information is underexposed, chips will strain to record information in the deeper values, and in doing so will produce digital noise. This noise is intensified as you move darker on the value scale. The theory of "exposing the right" gives more light to the shadow areas, thus reducing digital noise in the darker values. In post production, one has the ability to darken the shadow areas or the entire image. Reducing luminosity will not increase digital noise. Conversely, brightening an underexposed file will intensify the digital noise in an image. There is a great book out there for those interested in reading. The Digital Negative by Jeff Schewe. In my opinion, Jeff Schewe is a true digital guru, so much that Adobe hires him as a consultant. Best, Hal www.silvermanstudio.com



#17 adamhanlon

adamhanlon

    Harbor Seal

  • Admin
  • 2221 posts
  • Gender:Male
  • Location:Lancaster, UK

Posted 03 June 2015 - 10:46 AM

This increase in noise is particularly noticeable when the the "shadows" slider is used in Lightroom/Camera Raw to brighten dark/underexposed areas of the image. 

 

So, is the increase in sensor information available by exposing to the right solely a side effect? 

 

Adam


Adam Hanlon-underwater photographer and videographer
Editor-wetpixel
web | Flickr | twitter | Linkedin | Facebook


#18 MikeVeitch

MikeVeitch

    1.7kbps Manta Boy

  • Senior Moderator
  • 6309 posts
  • Gender:Male
  • Location:In Bali, Indonesia but from Vancouver, BC
  • Interests:Teaching Underwater Photography

Posted 05 June 2015 - 05:52 AM

Erin, nice set of before and afters for sure.  You have really done a good job to bring back the colours. I am also interested to know your methodology.  I would simply just pull the Highlights and exposure sliders down but I have a feeling there is a lot more to it than that! 


Join us for an Underwater Photography Workshop in the Lembeh Strait at NAD Lembeh with Doug Sloss in 2018
Blog and Photo Archive/Portfolio Site www.mikeveitchblog.com
Learn underwater photography in the ultimate classroom, Bali! or join us on a trip www.underwatertribe.com and www.baliuwphoto.com

Join us for a trip in Indonesia in Komodo or Raja Ampat


#19 Paul Kay

Paul Kay

    Giant Squid

  • Industry
  • PipPipPipPipPip
  • 1793 posts
  • Gender:Male
  • Location:North Wales, UK

Posted 06 June 2015 - 01:20 AM

When digital information is underexposed, chips will strain to record information in the deeper values, and in doing so will produce digital noise.

Ummmm. The darkest areas in the image will be just that - dark. Noise is only produced when you increase the gain (i.e. amplify the signal), so brightening/lightening shadow areas will increase their noise level - very simple. If they are important (more so than highlights) then underexposing them will lead to increased noise when adjusting later. Its got nothing to do with ETTR and lots to do with exposing for your preconceived requirement of output. ETTR in itself is in essence overexposing and as I have said before, can lead to tonal anomalies. So correct exposure must be made on the basis of the subject. Until we get sensors able to record much wider contrast ratios we are always going to be compromising somewhere. I don't see any real advantage to ETTR even in the low contrast conditions (which we compensate for by using flash) underwater. Gurus are great but actual practice is even better IMO.


Paul Kay,Canon EOS5DII SEACAM c/w S45, 8-15, 24L,35L, 60/2.8 (+Ext12II) & 100/2.8 Macros - Sony A7II SEACAM 28/2 & 50/2.8 Macro - UK/Ireland Seacam Sales -see  marinewildlife






Also tagged with one or more of these keywords: technique, exposure