ETR Metering

If you would like to post, you'll need to register. Note that if you have a BCG store account, you'll need a new, separate account here (we keep the two sites separate for security purposes).

I read an article by Jared Lloyd espousing the technique of exposing to the right to eliminate noise. Steve talks about ETR in his metering book as a way to increase dynamic range but not to decrease noise. If you haven’t read it, the premise is that you can use high ISO’s, with little to no noise if you expose to the right. I’ve tried ETR to reduce noise with mixed results so I’m looking to hear some opinions on the subject, especially from Steve! 😀
 
I find ETTR doesn't really work as most hope. Once upon a time those tricks helped with older sensors, but with the newer sensors it really doesn't make a difference. ISO is a brightness control, nothing more. It doesn't add light, cranking it up doesn't make your sensor more sensitive to light (dual gain sensors can help, but there's still no extra light added). It's like if you're in Lightroom and crank up the exposure and then bring it back down. As you crank it up, the noise gets worse, as you bring it back it gets better, but the level of noise you have at say ISO 800 doesn't get any better by cranking it up and back down.

I've tested this and I think the exposure and metering book has some samples. There's no benefit to over exposing and then pulling the exposure down in post unless you are doing it at base ISO (for increased DR).
 
Last edited:
That is a really good question. My nonexpert understanding is that noise is hard to understand. There are different sources of noise and cameras handle noise differently.

For shot noise which is inherit to the way light varies randomly, the only fix is to add more exposure via opening the aperture or slowing the shutter speed, if possible or practical. So ettr will help noise if it is done first by adding actual light with exposure via f stop or shutter speed. For other kinds of noise cameras that are iso invariant past a certain iso are not going to be helped by higher iso but a camera that is not invariant will continue to improve shadows with higher iso if you can't get there by exposure alone, compared to brightening in lightroom.

So the question is whether one should use iso to push the histogram to the right beyond what is normal or only use shutter speed and f stop to move the histogram to the right? My camera is not iso invariant, so I'll use ettr via f stop and shutter speed first and primarily at base ISO if that is practical, then continue via iso if more room is needed. If I had a camera that was iso invariant at higher iso I would skip raising the iso to go higher than needed. .

The problem is that choosing a too high iso to start with can force the meter into capturing less actual light, because it makes you think a faster shutter speed or smaller aperture is required. That lower actual exposure in terms of light will lead to more noise

Just a side note, ettr is not always adding exposure, sometimes you have to reduce exposure to keep highlights from blowing out.
 
Last edited:
ISO is simply an amplifier of the data captured by the sensor. Raising ISO can reduce the bit rate of data captured even at low levels, so if anything, it hurts your image. If you plan to use ETTR, it's only with base ISO and adjustments to shutter speed and aperture. Aperture affects DOF, so it could mean you have more data but less DOF and detail into the frame, so in theory you'd need to focus stack multiple images to compensate. Shutter speeds can be longer as long as you don't run into motion issues.

ISO invariance is normally relevant when you increase ISO in post with an exposure adjustment or lifting shadows rather than in the camera. It can be useful when you need to keep your shutter speed up to prevent subject or camera motion. You are deliberately underexposing.

The catch with keeping ISO low and raising exposure in post is you are bypassing processing the camera manufacturer may incorporate in their RAW processing. For example, many cameras today use dual gain to reduce noise as ISO levels increase. If your ISO is below that level, you don't get the benefit of dual gain. You may also see some sophisticated noise reduction at the RAW level that is bypassed. Nikon does this on the Z9 according to Mark Cruz and the result is they locally apply a little NR to detail-less areas while not applying it to the subject and other detail areas. It also bypasses presets in LR that could be based on ISO.

Bottom line - as Steve said, ETTR is for situations when you can shoot at base ISO and other adjustments don't compromise the image.
 
Just a follow up question. This quote is from the photography life article linked below, which I generally trust to be a source of accurate info. It seems to indicate there is a benefit in terms of noise in some cases by using higher iso. Your thoughts?

  1. If a particular ISO is not simulated, don’t be afraid to use it, especially if your camera isn’t close to being ISO invariant! You’ll get better noise performance from using a “real” ISO 1600 photo rather than brightening an ISO 100 photo in post-processing, especially if your camera has a lot of back-end read noise at ISO 100.
 
Just a follow up question. This quote is from the photography life article linked below, which I generally trust to be a source of accurate info. It seems to indicate there is a benefit in terms of noise in some cases by using higher iso. Your thoughts?

  1. If a particular ISO is not simulated, don’t be afraid to use it, especially if your camera isn’t close to being ISO invariant! You’ll get better noise performance from using a “real” ISO 1600 photo rather than brightening an ISO 100 photo in post-processing, especially if your camera has a lot of back-end read noise at ISO 100.
What he's saying is that if the camera is applying analog and digital amplification at ISO 1600, then shooting ISO 1600 makes more sense than shooting a 4 stop underexposure at ISO 100 and bringing it up in post with digital amplification. He's not really advocating higher ISOs instead of lower ISOs, he's just giving an example of why you may want to use in-camera amplification instead of digital-only after the fact.
 
Last edited:
In a world comprised entirely of gray cards, ETTR would make sense. For a gray card, make use of the stop or two headroom to overwhelm electronic noise and increase photon counting statistics. But counting statistics scale logarithmically (slowly) so there is never much to be gained, and "overwhelming" is not what's in the cards.

The world is not a gray card. Blowing highlights is the greater problem. If it's a fleeting capture and I have doubts, i will reduce exposure to be safe. If there is time I will bracket and chimp.

If there is plenty of time, then fill the entire dynamic range. But for modern cameras, ETTR is usually not a useful concept for everyday shooting. In the time spent putzing around with it, the moment will be gone.
 
  • Like
Reactions: Hut
FYI, I did find one of the examples from my book where I did some ETTR tests. In this case, the proper ISO was 800 and the ETTR ISO was 2500 and then brought back down in post.

These are 100% crops from my test chart, I can't see any difference.

800-iso-ettr-patch.jpg
You can only see EXIF info for this image if you are logged in.


This was in the shadows from that same shot of a b&w box in the background (shallow DoF used so you can see the noise pattern). Even in the shadow areas, ETTR just ins't helping as far as I can tell.

800-iso-ettr-shadows.jpg
You can only see EXIF info for this image if you are logged in.


In my opinion, you're not seeing any gain use ISO for ETTR but you are taking a far greater risk of clipped highlights and poorer color fidelity.
 
FYI, I did find one of the examples from my book where I did some ETTR tests. In this case, the proper ISO was 800 and the ETTR ISO was 2500 and then brought back down in post.

These are 100% crops from my test chart, I can't see any difference.

View attachment 30597

This was in the shadows from that same shot of a b&w box in the background (shallow DoF used so you can see the noise pattern). Even in the shadow areas, ETTR just ins't helping as far as I can tell.

View attachment 30596

In my opinion, you're not seeing any gain use ISO for ETTR but you are taking a far greater risk of clipped highlights and poorer color fidelity.

Looks the same on my monitor. You didn't say which camera. If it a camera that is basically invariant at 1600 then it would have to be the same. For example the R5 is flat past 400 in the chart below, but the RP for example keeps improving shadows (relative to Lightroom) throughout the native range. Do you think it matters if the camera is not invariant?

 
Last edited:
Looks the same on my monitor. You didn't say which camera. If it a camera that is basically invariant at 1600 then it would have to be the same. For example the R5 is flat past 400 in the chart below, but the RP for example keeps improving shadows (relative to Lightroom) throughout the native range. Do you think it matters if the camera is not invariant?


Sorry, D850.

The thing is, most (all?) modern cameras are ISO Invariant to some extent, which is why ETTR isn't all that helpful. As I mentioned in my first post, I think older cameras may have seen a benefit beyond base ISO with this technique, but I have't come across any modern bodies where I thought it made any real difference. Most ETTR shooters using ISO to boost the exposure are needlessly risking highlights.
 
Sorry, D850.

The thing is, most (all?) modern cameras are ISO Invariant to some extent, which is why ETTR isn't all that helpful. As I mentioned in my first post, I think older cameras may have seen a benefit beyond base ISO with this technique, but I have't come across any modern bodies where I thought it made any real difference. Most ETTR shooters using ISO to boost the exposure are needlessly risking highlights.

Good to know. I guess the rule of thumb is don't bother with shooting to the right unless you are at base iso or at least below wherever your camera flattens out, or maybe just don't bother with ettr at all and worry about composition instead!
 
Good to know. I guess the rule of thumb is don't bother with shooting to the right unless you are at base iso or at least below wherever your camera flattens out, or maybe just don't bother with ettr at all and worry about composition instead!

I think so too - at least, that's how I play it. :)

There are people who swear by ETTR and others that shoot at base ISO and raise the exposure in post. I've tired all the tricks and the truth is, in my experience, just shooting a proper exposure is all you really need 99% of the time :)
 
So, what is the meaning of "Expose to the Right" (ETTR)?

I thought it was to exploit the full dynamic range of the sensor by placing the brightest (colored) pixel (or the highlight to be preserved) at magnitude 255 (8-bit color). There are various ways to do this, including changing ISO.

ISO invariance is a related, but separate topic, is it not?

Hope no one minds this naive question...I'm just trying to clarify in my own mind what is ETTR.
 
The world is not a gray card. Blowing highlights is the greater problem. If it's a fleeting capture and I have doubts, i will reduce exposure to be safe. If there is time I will bracket and chimp.

If there is plenty of time, then fill the entire dynamic range. But for modern cameras, ETTR is usually not a useful concept for everyday shooting. In the time spent putzing around with it, the moment will be gone.

depends on the subject. given sometimes i have black dogs against a bright sky, i'm totally willing to blow out highlights in order to get a good exposure on my subject.

just know your priorities and prioritize that in the exposure

and yah, i'd rather get the shot with a less than perfect exposure than lose the moment
 
So, what is the meaning of "Expose to the Right" (ETTR)?

i think the meaning is different to different people.

i think a lot of ettr detractors, assume it means you expose bright at the expense of highlights.

as far as i can tell, most of us modern ettr advocates suggest you expose as bright as you can _without loosing any highlights YOU CARE ABOUT_. that is to say, you only sacrifice highlights if that's a sacrifice you want or need to make.

i think you could maybe say ETTR means exposing as bright as you need in order to capture the shadow detail you care about.

ISO invariance is a related, but separate topic, is it not?

agree
 
So, what is the meaning of "Expose to the Right" (ETTR)?

I thought it was to exploit the full dynamic range of the sensor by placing the brightest (colored) pixel (or the highlight to be preserved) at magnitude 255 (8-bit color). There are various ways to do this, including changing ISO.

ISO invariance is a related, but separate topic, is it not?

Hope no one minds this naive question...I'm just trying to clarify in my own mind what is ETTR.

That's the right track. The idea is to maximize dynamic range (DR) by decreasing noise in the shadows. The reason we lose DR as ISO increases isn't because of the highlights, it's because we lose shadow detail. The idea with ETTR is to get more exposure in the shadow areas, thus improving the single-to-noise ratio in those areas and increasing the overall DR of the image.

However, you can only increase single to noise ratio by adding light and the only way to do that is with shutter speed and aperture. ISO increases brightness by increasing gain, but it doesn't add any light. It can only work with what's there. It's like if you have a radio station with static - in order to increase the quality of the sound, you don't turn up the volume, instead you need a better signal-to-noise ratio, so a more powerful antenna or getting closer to the tower is the only way to really fix it.

That's why ETTR is really only effective at base ISO. If I make a scene brighter with a slower shutter speed or faster F/stop, I'm improving the signal-to-noise ratio for the entire image - and especially the shadows. If I am above base ISO, I'm just turning up the volume knob for brightness, I'm hot improving the signal-to-noise ratio.

In essence, your sensor only has one ISO - base ISO (newer sensors kind of have two base ISOs with dual gain technology, but let's keep this simple). This means that everything is captured at, say ISO 100. What you get at ISO 100 is then brightened using ISO. Increasing ISO is only taking what you already captured and increasing those values to make them brighter, but it's not improving the signal-to-noise ratio.
 
So, what is the meaning of "Expose to the Right" (ETTR)?

I thought it was to exploit the full dynamic range of the sensor by placing the brightest (colored) pixel (or the highlight to be preserved) at magnitude 255 (8-bit color). There are various ways to do this, including changing ISO.

ISO invariance is a related, but separate topic, is it not?

Hope no one minds this naive question...I'm just trying to clarify in my own mind what is ETTR.

All good questions. At the core I think is the idea that exposure has only to do with photons hitting the sensor as in f number and shutter speed. ISO is brightening after the exposure is made. The sensor is only one sensitivity, it doesn't get more sensitive with iso. So if a camera was absolutely iso invariant it doesn't matter if you add iso in camera or raise the brightness with the exposure slider in Lightroom. No extra noise is added either way and the dynamic range doesn't get better by raising the iso in camera, ettr would only happen via f number and shutter speed. If the camera is not invariant then raising iso in camera can reduce noise compared to sliding the exposure slider in Lightroom, so in that case iso in camera is helping get better dynamic range. I think what is being said is there is little point in raising iso higher than a normal exposure if you are just going to reduce the exposure slider in lightroom.

However I'm still unclear about a camera like mine, which is neither invariant nor dual gain, whether it helps a little since you can then pull down highlights but keep shadows high. Steve seems to be saying there is no value to bothering with it.
 
Last edited:
SO increases brightness by increasing gain, but it doesn't add any light. It can only work with what's there.

sure, but there's another variable. what's recorded to the file. if the data isn't recorded to the file then it isn't recorded, period. depending on the situation, adding iso can push stuff that would have fallen off the left side of the histogram, onto the histogram, hence you have _something_.
 
sure, but there's another variable. what's recorded to the file. if the data isn't recorded to the file then it isn't recorded, period. depending on the situation, adding iso can push stuff that would have fallen off the left side of the histogram, onto the histogram, hence you have _something_.
ISO can't add anything that wasn't there. So, if it was there, it's recorded regardless. ISO wouldn't allow it to record something that normally wouldn't record. It'll just make it brighter. (I'm assuming RAW files, JPEG will toss out info)
 
Steve said "That's why ETTR is really only effective at base ISO. If I make a scene brighter with a slower shutter speed or faster F/stop, I'm improving the signal-to-noise ratio for the entire image - and especially the shadows. If I am above base ISO, I'm just turning up the volume knob for brightness, I'm hot improving the signal-to-noise ratio."

This doesn't make sense to me. From practical shooting experience, i can often get away with significant underexposure (i.e. correct in postprocessing) at base ISO. However if I underexpose a low light scene when shooting at a high ISO, then I'm stuck with a lot of noise. So, for night photography, for example, i am especially careful to try to expose to the right to capture as much signal as I can. Said another way, if I increase the shutter speed or open the aperture to bring the exposure to the right side of the histogram., then I am increasing the signal no matter what ISO I happen to be using at the time..
 
Steve said "That's why ETTR is really only effective at base ISO. If I make a scene brighter with a slower shutter speed or faster F/stop, I'm improving the signal-to-noise ratio for the entire image - and especially the shadows. If I am above base ISO, I'm just turning up the volume knob for brightness, I'm hot improving the signal-to-noise ratio."

This doesn't make sense to me. From practical shooting experience, i can often get away with significant underexposure (i.e. correct in postprocessing) at base ISO. However if I underexpose a low light scene when shooting at a high ISO, then I'm stuck with a lot of noise. So, for night photography, for example, i am especially careful to try to expose to the right to capture as much signal as I can. Said another way, if I increase the shutter speed or open the aperture to bring the exposure to the right side of the histogram., then I am increasing the signal no matter what ISO I happen to be using at the time..

It's all about signal-to-noise ratio. If you underexpose by three stops at base ISO with a relatively ISO Invariant sensor, you can add three stops and the shot will look like it was shot at ISO 800. The reason it works is that you're starting with a decent signal-to-noise ratio. However, in dim light the signal-to-noise ratio is poor and if you'r already at a higher ISO, the camera is already adding a lot of gain to brighten it up. If you push an underexposed ISO 1600 image by three stops, it looks like ISO 12,800.

However, if, as you say, you add more light using shutter speed or F-stop, then you are improving the signal-to-noise ratio and getting more light to the sensor. If that leads to an overexposure that you have to bring down a stop in post, then you could have shot at the equivalent ISO in the field (i.e. one stop less) and had the same results. The reason you're getting better results is because of the extra light from the shutter speed or F/stop - not the ISO.

It may help to think of your sensor as what it is - a photon counter. Shutter speed and aperture work together to send photons to the sensor and these photons produce a change at each photosite. The difference in these charges and the way the photons hit the sensor (the poisson effect) is the primary cause of noise. The greater the signal - the more the pixel wells fill - the lower the noise. Increasing these charges unilaterally with ISO will make everything brighter, but it doesn't improve the SNR.
 
It's all about signal-to-noise ratio. If you underexpose by three stops at base ISO with a relatively ISO Invariant sensor, you can add three stops and the shot will look like it was shot at ISO 800. The reason it works is that you're starting with a decent signal-to-noise ratio. However, in dim light the signal-to-noise ratio is poor and if you'r already at a higher ISO, the camera is already adding a lot of gain to brighten it up. If you push an underexposed ISO 1600 image by three stops, it looks like ISO 12,800.

That is all nice and correct in theory, but that is not how most of us shoot in the field. I use manual exposure in auto ISO mode a lot to shoot wildlife. I base my exposure by the in-camera histogram in the viewfinder and the setting level for zebra- stripes in my Sony A1 viewfinder. I adjust shutter speed and aperture to " expose to the right" to maximize the exposure. So ETTR is important no matter what ISO you use. I think all the discussion that restricts ETTR to only base ISO just confuses people. It is valid at all ISOs as long as one remembers that the"signal" in signal/noise ratio (or the "Exposure" in ETTR) is dependent on shutter speed and aperture.
I know I can get away with more underexposure at base ISO, because I am looking at data that is not being artificially amplified by a higher ISO. However, significant underexposure when one happens to be necessarily shooting at a high ISO can be devastating. So to me that means ETTR is even more critical at high ISOs.
 
Last edited:
ISO can't add anything that wasn't there. So, if it was there, it's recorded regardless. ISO wouldn't allow it to record something that normally wouldn't record. It'll just make it brighter. (I'm assuming RAW files, JPEG will toss out info)

while i agree it has to be there for iso to amplify, the file size is finite and so basically the file only captures a “window” of the data. if the data falls outside of that window it will not be recorded, thus lost forever.
 
sure, but there's another variable. what's recorded to the file. if the data isn't recorded to the file then it isn't recorded, period. depending on the situation, adding iso can push stuff that would have fallen off the left side of the histogram, onto the histogram, hence you have _something_.
Perhaps the raised ISO is not adding information, but just showing you a bit better what data is deep in the shadows?

--Ken
 
Back
Top