ETTR: The Ultimate Exposure Technique?

[Guest Ebrahim Saadawi]

Some old tests I dug out on Exposure. I conducted the same test on many platforms, Canon 60D/5D mk III/Canon C100 Clog/Nikon D5300/D810/Sony A7s at SLOG2, and Panasonic GH4 wat CineD & CineV and even did a quick on-set test on an Alexa, shooting 2.8K arriraw, and ALL these camera systems give the exact same results as follows:

ETTR Log ungraded


"Correct exposure or slightly under" Log ungraded


ETTR corrected to match the correctly exposed


ETTR


Correctly exposed 


ETTR to correct exposure



I mean it's such a huge difference in image quality that I never expose correctly or slightly under anymore, the higher you go in exposure and the more you feed the sensor light the better quality, thicker colour, less noise and more DR you get, to my eyes anyway.

It even works if you achieve ETTR by pumping up the ISO/gain (but it's best achieved by opening iris/lowering shutter/increasing light), a 320ISO underexposed image shows way more noise than a 3200ISO overexposed image, so it's underexposure that creates noise, not high ISOs!

Exampe for 320ISO (under) vs 3200ISO (over):

320ISO


3200ISO


Here's both when you grade them to normal point:



The first noisy one of the two graded images above is at 320 ISO, while the bottom clean one is at 3200 ISO, with the same settings/light and everything, yet the 320ISO shows significant noise because the image is underexposed, while using the ETTR technique the I can get perfectly clean images at 3200ISO using a Canon rebel (as seen on last image, at 3200ISO with a Rebel)

again to my eyes, It's underexposure that creates noise, not high ISOs. 


Plus, ETTR increases Dynamic Range (as long as you don't clip the HL) because once you start dialing down exposure, the highlights don't gain anything while the shadow noise floor is increased = lower DR, so with ETTR you're using the entire DR of your sensor, squeezing every last drop of performance)

Achieving ETTR is best done using a Waveform monitor or an RGB Histogram in-camera, you increase exposure until the image touches the upper boundary (WF) or right boundary (Histo), and Zebras if set to 100% could be used too, you increase exposure until they show up, then back down one notch. Yes the image looks ugly, flat and over exposed at first, but that's the way we shoot now anyway in a Log/LUTs dominated world. So this technique is only viable if you are doing correction to the image in post and not handing them off for direct delivery obviously, for these type of jobs get a "correct" exposure with skin at 70IRE and that stuff...

So what do you guys think? I see many professionals don't expose that way so there must be a counter argument, lets hear it. Or can I just use this simple technique for setting exposure and rest my mind in peace after all these years struggling to get "correct" exposure?

 

[neosushi]

Tried with the GH4 to get rid of the nasty noise. Turns out blown out highlights makes your video looks even more digital. Plus you don't have that great of a dynamic range either.

But thats just from my experience ! I think ETTR is a technique that depends on the combination of the camera hardware (dynamic range, sensor), its software (compression codec, picture profile - log for ex.) and your expectations in post production. Personally I wouldn't recommend a systematic use with any camera.

What is best is to try to remember that it is difficult to expose perfectly a picture. So you need to think about what your subject is. And if your subject is expose correctly, well thats more than a half of the job done. But our digital cameras don't have such great dynamic range. Its definitely good but not great. Not because of the wide range of the dynamic. But because of the way digital cameras handle noise in shadows and blown highlights. 

[Guest Ebrahim Saadawi]

Yea it increases the risk of you making a mistake and blowing the highlights (I actually did make that mistake in the tests above!), but again that's not ETTR fault, it's ours. 

ETTR means exposing as highly as possible WITHOUT clipping highlights, not even colour channels, so use an RGB histogram if available to get the most accurate ETTR. 

[Volker Schmidt]

again to my eyes, It's underexposure that creates noise, not high ISOs. 

- this is my experience also! 
I have just tested two Minolta-lenses (on my 5dmk2-MLraw) with a external sekonic exposure-meter and I was very surprised by the quality of the pro res 422 Export files!
No noise and a film-like grain - with 800/1600 ISO!
Precise exposure is definitely a guarantee to avoid Noise - more than ISO Settings!
Here´s the test, but the H264 - and Vimeocompression, is a bit disappointing:).

 

- 

 

 

 

[Nick Hughes]

A strong counter argument to ETTR is that it can yield wildly different looking clips, which becomes a problem when you have a complex shoot with a lot of footage. You have to then go back and wrangle all the clips into a cohesive look when grading, often starting from scratch with each new clip. Inexperienced colorist may have trouble matching everything up. It can be difficult for the colorist to determine how the footage 'should' look when everything looks different. 

If you shoot your footage the way you want it to look, you can avoid these issues. You can also shoot with a combination of these philosophies. Overexposure by a consistent amount will yield more continuity among all of your footage, though you may not be exposing completely to the right in each clip.

[Don Kotlos]

ETTR with 8 bit is asking for trouble. 

[Axel]

All compressed video has some curve baked it. With ETTR, you 'fill the well' of your sensor and 'avoid the noise floor', but depending on the said curve (picture style), you may also have thinned out your values. The common S-curve will also try to avoid the noisy depths and favor skin tones, midtones. By not exposing to the center, you risk getting very doughy skintones when changing that curve in post.

A strong counter argument to ETTR is that it can yield wildly different looking clips, which becomes a problem when you have a complex shoot with a lot of footage.

 At least with 8-bit, but also with 10-bit, it's possible that the one shot you need has too little definition in the areas that count (usually the skin), and you are forced to de-grade the good shots to match the worst. Especially when you shot in very bright places. For 8-bit, it's a good idea to light flat.

ETTR is a no-brainer with raw. Then even zebra suffices. 

​

[Guest Ebrahim Saadawi]

Yes it's the only counter argument I'm seeing on the forums, that you somehow lose skin tone detail and highlight 'tonality' when exposing them up the curve, which is the same argument that highlights are more compressed parts of the codec, 

Fortunately I have some free time so I am testing for skin tones and highlight tonality today, to see if I lose any when ETTR and adjusting in post using an 8bit codec, will post back

Other tests are HUGELY welcome. There isn't much information about this on the web other than theory 

We've already established that ETTR gives much cleaner images on terms of noise, now lets look at ''highlight/skin issues''. You can simply shoot someone with a high expsoure (putting his skin into the supossedly most compressed part of the codec) then shoot the same with a darker normal exposure and compare if the adjusted ETTR skin shows less information or more break up. Easy test really. 

 

 

[neosushi]

Its personal but between noise and skin I choose 10 times skin. 

But it also depends how clean your camera is. Had tons of issues on the GH4, haven't asked myself this question once with the NX1.

[Nikkor]

again to my eyes, It's underexposure that creates noise, not high ISOs. 

- this is my experience also! 
I have just tested two Minolta-lenses (on my 5dmk2-MLraw) with a external sekonic exposure-meter and I was very surprised by the quality of the pro res 422 Export files!
No noise and a film-like grain - with 800/1600 ISO!
Precise exposure is definitely a guarantee to avoid Noise - more than ISO Settings!
Here´s the test, but the H264 - and Vimeocompression, is a bit disappointing:).

 

- 

 

 

 

​Your really mastered that camera, those shots reminded me of Shame

[JazzBox]

Dear Ebrahim, 

after some tests with different cameras (Canon 600D, 60D, 70D, 5D MkIII, Panasonic G6 and GH4 and my father's Olympus E-M10) I came to the same conclusions!

Some years ago I started shooting searching to use the lowest ISO possible, now I try to ETTR whenever is possible. 
Also Robin Wong on his blog talks about clean high ISO: http://robinwong.blogspot.it/2014/11/about-high-iso-shooting-with-olympus.html

Just a question: how do you correct the exposure in post? I ask you because on that argument there are some different schools of thought and I find your images very convincing. 

Another question: how do you ETTR when you shoot in night or in a very dark environment?

Thank you very much, I really appreciate your post!

 

[Guest Ebrahim Saadawi]

After initial testing on how ETTR affects skin tones, I am seeing what I saw earlier, it gives better everything, less noise, more colour thickness. 

I am really not seeing any skin issues others claim that supposedly come from placing the skin in the "most compressed area of the curve", in theory yes but to my eyes the ETTR skin looks better.... I've seen a lot of examples when people try to ETTR, they clip the Red channel, and when they try to bring back the exposure in post it turns all funky with loss of tonality. So perhaps that's the reason. ETTR means not to clip anything, not Luma, not colour channels, nothing. Use an RGB histogram if available while shooting.

Here's skin with normal exposure:

Here's with ETTR:


same skin colour, just a better less noisy image and you can push the image much farther due to the lack of noise. 

But again, before I declare this my ultimate exposure technique I will run more in depth testing on the Skin issue. 

[Cinegain]

Well, the samples speak for themselves. Seems ETTR gives you great results indeed. Interesting. Definitly need to test this for myself.

[Guest Ebrahim Saadawi]

Please do and share what you find. It'd be great if someone else confirmed what I am seeing with ETTR. 

[Bioskop.Inc]

Did this about a year ago with ML RAW on a 60D - been doing some more filming with ETTR, so will post once I've got something to show:

 

 

Did have some really good articles bookmarked somewhere, but can't seem to find them...

[Axel]

 

Did this about a year ago with ML RAW on a 60D - been doing some more filming with ETTR, so will post once I've got something to show:

​Of course ETTR is right for raw, nobody disagreed. This is about whether it's also right for compressed codecs, which contradicts common beliefs.

[Nikkor]

ETTR works if you are working with a linear file, before any tone curve is applied.

Raw is linear, 8bit dslr files are not (8bit linear would be horrible,btw)

[Inazuma]

So Ebrahim, you're saying that pushing the exposure up (to just before clipping) even if it means using much higher ISO's will yield better results?

[Axel]

After initial testing on how ETTR affects skin tones, I am seeing what I saw earlier, it gives better everything, less noise, more colour thickness. 

I am really not seeing any skin issues others claim that supposedly come from placing the skin in the "most compressed area of the curve", in theory yes but to my eyes the ETTR skin looks better.... I've seen a lot of examples when people try to ETTR, they clip the Red channel, and when they try to bring back the exposure in post it turns all funky with loss of tonality. So perhaps that's the reason. ETTR means not to clip anything, not Luma, not colour channels, nothing. Use an RGB histogram if available while shooting.

Here's skin with normal exposure:

Here's with ETTR:


same skin colour, just a better less noisy image and you can push the image much farther due to the lack of noise. 

But again, before I declare this my ultimate exposure technique I will run more in depth testing on the Skin issue. 

​Yes, it's the same skin color at first glance. Yet it looks as if the skin was softened in the ETTR image. You really can't judge that if you take just one shot. Shoot an entire sequence with two or more people hit by the light sources in different angles. Then try to match those in post. One looks perfect, the other 'posterized' (old Photoshop effect), and you can't do anything about it. 

Carpets and vases have no 'memory colors'. They can be reduced to, say, five nuances, and nobody will notice. A face will need - I don't know - fifteen? to look alive.

[Guest Ebrahim Saadawi]

So Ebrahim, you're saying that pushing the exposure up (to just before clipping) even if it means using much higher ISO's will yield better results?

​Exactly. The above examples are all ETTR by raising the ISO. In all my testing, a 320 ISO underexposed image is waaay more noisy than the same image overexposed image (just before clipping) at 3200 ISO. 

The results I am getting with ETTR in lowlight are simply astonishing, I can shoot clean 6400 ISO images with a Canon rebel, whild before I figured out the technique 1600 was as high as I can get. Just push that exposure up and youll get clean images with thick colours, even if it means clipping a few unimportant elements in the frame.