ISO - Waste of time when shooting RAW?

[sk66]

I started photography with film and ASA mattered then but I am aware that what we are taught because 'it's the way we have always done it' sometimes isn't correct.

There's another thread about ISO not being part of the exposure triangle (something like that) where I explained that the exposure triangle is, and always was, SS/Ap/EV(light). The choice of using a higher ISO/film speed is a compromise to offset a lower light level (EV), and it has the same penalties... as does increasing exposure in post/pushing the film exposure.

 

[jonbeeza]

This is the biggest advantage! Set your shutter speed/aperture to a value you want, so if photographing a fast moving subject shoot at a high shutter speed, pick your aperture for DoF and leave ISO at 100 - correct the exposure in post

Years ago, I tried getting a properly exposed image, from a very under exposed file. I got the same result as I got ten minutes ago, a bad image. I took the image at base ISO, but very under exposed.

 

[Craigus]

For what it's worth I just did a test on the Fuji XT1 shooting one at 6400 and one at base, pushing the base 5 stops to match the exposure and it looked a lot noisier.

I guess it may vary from sensor to sensor but why bother? You aren't gaining anything, if anything you are hampering your ability to expose properly in the field.

 

[rookies]

For what it's worth I just did a test on the Fuji XT1 shooting one at 6400 and one at base, pushing the base 5 stops to match the exposure and it looked a lot noisier.

I guess it may vary from sensor to sensor but why bother? You aren't gaining anything, if anything you are hampering your ability to expose properly in the field.

I was thinking the same.. I would rather compose expose my images correctly less faff when processing..

 

[Fraser Euan White]

For what it's worth I just did a test on the Fuji XT1 shooting one at 6400 and one at base, pushing the base 5 stops to match the exposure and it looked a lot noisier.

I guess it may vary from sensor to sensor but why bother? You aren't gaining anything, if anything you are hampering your ability to expose properly in the field.

His argument is that you are gaining something though - better performance from the sensor! If this is the case why worry about exposure 'in the field'?

 

[Craigus]

His argument is that you are gaining something though - better performance from the sensor!

I'm yet to see any evidence to back up this claim. From the evidence I've seen so far the opposite is true.

If this is the case why worry about exposure 'in the field'?

Because it would be harder to know what your exposure (SS & F/) should be if you aren't getting any feedback from the meter. It's not impossible but unless you have very consistent shooting conditions it could be hard.

 

[Fraser Euan White]

Because it would be harder to know what your exposure (SS & F/) should be if you aren't getting any feedback from the meter. It's not impossible but unless you have very consistent shooting conditions it could be hard.

Again, the argument is it doesn't matter as long as you are shooting at the base ISO of the camera you can use whatever settings you want (within reason), therefore all we have learn't about exposure 'in camera' is 'Old Hat' in respect to RAW files. You pick your shutter speed to freeze/not freeze motion and your aperture for DoF.

The argument about extra work in Post doesn't 'wash' at all either since ALL RAW files have to be edited anyway - adjusting a slider isn't exactly hard!

 

[Gaz J]

I doubt he will be converting many photographers to his "way" of photography. Most of us want to actually see what we've taken and check focus is ok which, for me and probably many others, is far more important.

 

[Craig_85]

As Phil V says, this subject is dependent on the camera and sensor.

I won't go into detail as it is very easy to google, just type in 'ISO invariance'. Currently AFAIK only a few sensors are ISO invariant and for some of those only in a smallish range.

This.

I own a d750 which is an ISO invariant camera, it allows a different mindset altogether.

All digital cameras only really have one ISO setting in my mind. You can change the shutter speed and/or the aperture to increase or decrease the amount of light coming in. The camera has a base ISO which is the sensors native sensitivity to that light.

Now, to effectively change the exposure beyond that native ISO you can alter it yourself in post processing, or you can let the cameras processing engine do it by the use of, usually, increasing ISO to brighten the image.

Canons, and the majority of cameras, give you a cleaner Raw file with the most information when they are exposed to the right. This extends to increasing the ISO setting in camera to get the brightest possible image in camera to maximise signal to noise ratio. (Although this hits a ceiling where they become ISO invariant too)

Conversely Nikon d750 exposed 'correctly' in camera with a given shutter speed and aperture at say iso800 is identical in image quality to the shot exposed 3 stops darker by keeping the shutter speed/aperture the same but shooting at iso100 and pushing 3 stops in your raw converter.

But what's the advantage you may ask?

Wide dynamic range scenes for a start. At ISO100 the camera is achieving its greatest dynamic range coverage. Added to that when the cameras processing engine increases exposure through you upping the ISO, it universally doubles the brightness for each stop of ISO. So your ISO800 shot of a wide dynamic range scene starts to suffer compared to your ISO100 3 stop 'under exposed' shot. Because not only does base ISO have greater dynamic range but also when you increase the brightness by 3 stops in raw conversion you can also pull back the white point and highlights selectively, giving you more usable dynamic range. The cameras processing engine does not have this finesse when upping the ISO.

Another reason and real world example is more to do with moving subjects and or handholding situations. Imagine your lens is at maximum aperture, wide open. You are shooting a moving object that will blur completely beneath a certain shutter speed. With a normal sensor you up your ISO to keep the shutter speed up.

With an ISO invariant sensor you leave it at ISO100 and record a darker image which you can push in post. Other than the dynamic range issues mentioned already you also have way more highlight headroom doing this. Upping the ISO, either in manual or aperture priority to keep the image as recorded 'bright' does in some cases risk overexposure if the light changes drastically and tricks the meter as the subject moves along. You have less of a safety net with the highlights with the higher ISO, but at ISO100 there's loads of headroom.

Downsides of shooting at base ISO with an ISO invariant sensor is dark images on the playback review.

 

[Fraser Euan White]

This.

I own a d750 which is an ISO invariant camera, it allows a different mindset altogether.

All digital cameras only really have one ISO setting in my mind. You can change the shutter speed and/or the aperture to increase or decrease the amount of light coming in. The camera has a base ISO which is the sensors native sensitivity to that light.

Now, to effectively change the exposure beyond that native ISO you can alter it yourself in post processing, or you can let the cameras processing engine do it by the use of, usually, increasing ISO to brighten the image.

Canons, and the majority of cameras, give you a cleaner Raw file with the most information when they are exposed to the right. This extends to increasing the ISO setting in camera to get the brightest possible image in camera to maximise signal to noise ratio. (Although this hits a ceiling where they become ISO invariant too)

Conversely Nikon d750 exposed 'correctly' in camera with a given shutter speed and aperture at say iso800 is identical in image quality to the shot exposed 3 stops darker by keeping the shutter speed/aperture the same but shooting at iso100 and pushing 3 stops in your raw converter.

But what's the advantage you may ask?

Wide dynamic range scenes for a start. At ISO100 the camera is achieving its greatest dynamic range coverage. Added to that when the cameras processing engine increases exposure through you upping the ISO, it universally doubles the brightness for each stop of ISO. So your ISO800 shot of a wide dynamic range scene starts to suffer compared to your ISO100 3 stop 'under exposed' shot. Because not only does base ISO have greater dynamic range but also when you increase the brightness by 3 stops in raw conversion you can also pull back the white point and highlights selectively, giving you more usable dynamic range. The cameras processing engine does not have this finesse when upping the ISO.

Another reason and real world example is more to do with moving subjects and or handholding situations. Imagine your lens is at maximum aperture, wide open. You are shooting a moving object that will blur completely beneath a certain shutter speed. With a normal sensor you up your ISO to keep the shutter speed up.

With an ISO invariant sensor you leave it at ISO100 and record a darker image which you can push in post. Other than the dynamic range issues mentioned already you also have way more highlight headroom doing this. Upping the ISO, either in manual or aperture priority to keep the image as recorded 'bright' does in some cases risk overexposure if the light changes drastically and tricks the meter as the subject moves along. You have less of a safety net with the highlights with the higher ISO, but at ISO100 there's loads of headroom.

Downsides of shooting at base ISO with an ISO invariant sensor is dark images on the playback review.

Great explanation Craig - thanks for taking the time to write this.

 

[Fraser Euan White]

I doubt he will be converting many photographers to his "way" of photography. Most of us want to actually see what we've taken and check focus is ok which, for me and probably many others, is far more important.

Why bother checking focus - you've taken the image so you can't adjust it; the 'moment' has gone; it's either tack sharp or in the bin. You can't preview film images. (Isn't 'chimp-ing' a bad habit where you miss so many images whilst looking at the back of your camera?)

 

[gramps]

Why bother checking focus - you've taken the image so you can't adjust it; the 'moment' has gone; it's either tack sharp or in the bin. You can't preview film images. (Isn't 'chimp-ing' a bad habit where you miss so many images whilst looking at the back of your camera?)

I don't think it's quite so simplistic, the 'moment' may not have gone but the opportunity might have if you can't see if you have 'got the shot' ... it's sometimes very difficult or perhaps impossible to repeat.

 

[Fraser Euan White]

I don't think it's quite so simplistic, the 'moment' may not have gone but the opportunity might have if you can't see if you have 'got the shot' ... it's sometimes very difficult or perhaps impossible to repeat.

Can you give an example? What did film photographers do when their was no LCD on the back of their cameras and no autofocus either?

 

[gramps]

Can you give an example? What did film photographers do when their was no LCD on the back of their cameras and no autofocus either?

I'm not talking about film, I'm talking about digital, which is what is being discussed.
There are many instances of photo opportunities which present themselves for a short period of time, allowing for a re-take if necessary ... the fact you couldn't tell with film is irrelevant now

 

[Fraser Euan White]

I'm not talking about film, I'm talking about digital, which is what is being discussed.
There are many instances of photo opportunities which present themselves for a short period of time, allowing for a re-take if necessary ... the fact you couldn't tell with film is irrelevant now

...........and probably even more examples of people missing even better shots because they were looking at the back of their camera! If the first shot is out of focus then it is down to poor technique; how do you make an auto focus lens focus more accurately in a split second? (quite often it is down to image blur rather than focus) This technique allows you to select the correct shutter speed & aperture to prevent camera shake or improve DoF for infocus images; "f8 and be there!"

 

[maarten.dhaese]

According to this article by Michael Reichmann which I read a long time ago:

https://luminous-landscape.com/expose-right/

You need to expose to the right to capture the maximum amount of tonal values in your RAW file.

Quote from the article:
"A 12 bit image is capable of recording 4,096 (2^12) discrete tonal values. One would think that therefore each F/Stop of the 5 stop range would be able to record some 850 (4096 / 5) of these steps. But, alas, this is not the case. The way that it really works is that the first (brightest) stop’s worth of data contains 2048 of these steps — fully half of those available. (...) This realization carries with it a number of important lessons, the most important of them being that if you do not use the right-hand fifth of the histogram for recording some of your image you are in fact wasting fully half of the available encoding levels of your camera."

So my understanding is that yes you can underexpose and fix in post, but it dramatically reduces the amount of information you could have captured had you exposed to the right.

 

[Craig_85]

According to this article by Michael Reichmann which I read a long time ago:

https://luminous-landscape.com/expose-right/

You need to expose to the right to capture the maximum amount of tonal values in your RAW file.

Quote from the article:
"A 12 bit image is capable of recording 4,096 (2^12) discrete tonal values. One would think that therefore each F/Stop of the 5 stop range would be able to record some 850 (4096 / 5) of these steps. But, alas, this is not the case. The way that it really works is that the first (brightest) stop’s worth of data contains 2048 of these steps — fully half of those available. (...) This realization carries with it a number of important lessons, the most important of them being that if you do not use the right-hand fifth of the histogram for recording some of your image you are in fact wasting fully half of the available encoding levels of your camera."

So my understanding is that yes you can underexpose and fix in post, but it dramatically reduces the amount of information you could have captured had you exposed to the right.

Very true. But where it gets another level of complicated is the difference between exposing to the right, including ISOing to the right with an ISO variant sensor, or the actual greater iq achievable at base ISO with an ISO less sensor and pushing it in a controlled manner in raw conversion...

 

[gramps]

...........and probably even more examples of people missing even better shots because they were looking at the back of their camera!

I really doubt that.

If the first shot is out of focus then it is down to poor technique; how do you make an auto focus lens focus more accurately in a split second? (quite often it is down to image blur rather than focus) This technique allows you to select the correct shutter speed & aperture to prevent camera shake or improve DoF for infocus images; f8 and be there!

You mean it's never possible to get it wrong even with the correct shutter speed and aperture?

 

[sk66]

At ISO100 the camera is achieving its greatest dynamic range coverage.

This is wrong if the image is underexposed. As I explained earlier, it is *only* true for a correct/comparable exposure achieved at base ISO. And it is because the sensor is more evenly/fully saturated with light than a comparable image/exposure taken with a higher ISO.

If the image is underexposed 3 stops and recovered in post the DR will be the same as if a 3 stop higher ISO had been used in camera... that is, assuming the highlights are not clipped. The only benefit is the ability to underexpose in order to save highlights and then selectively bring up the shadows, as opposed to universally increasing exposure for the shadows.
This only applies to ISO invariant cameras and only in situations where it is necessary to underexpose for the highlights. In any other case, underexposing the image and recovering in post *reduces* the DR capability/recorded or has no effect on DR total (only where it is placed).

A digital image is still a recording of light. And every single sensor characteristic (DR/color depth/noise/etc) is dependent on receiving light. If you want the maximum image quality it is necessary to saturate the sensor as much as possible (i.e. collect as much light as possible w/o clipping).

 

[Bollygum]

It strikes me that while the signal to noise ratio may stay close to constant, the signal strength drops considerably and thus the dynamic range. Signal to noise is not the only thing to worry about.
Imagine that a bright, green pixel registers 200 at ISO 3200, it will register about 6 at iso 100. This will give 6 possible shades of pure green as against 200. It's likely that even my tired eyes could see the difference there.

 

[Craigus]

But what if you underexposed even for the base ISO? No feedback on the screen and this is entirely possible given the scenarios about setting shutter speed and aperture for the subject.

I'm at a wedding, it's the first dance, I'd like to freeze the movement so set the shutter speed to 1/500 and I'd like everything in focus so I set the aperture to f/8, after all I can recover it in post.

Oh wait, I can't, it's far too underexposed and now after pushing 11 stops it looks like a shot taken on my phone from 2006.

Extreme example yes, but proves the point.

 

[Fraser Euan White]

But what if you underexposed even for the base ISO? No feedback on the screen and this is entirely possible given the scenarios about setting shutter speed and aperture for the subject.

I'm at a wedding, it's the first dance, I'd like to freeze the movement so set the shutter speed to 1/500 and I'd like everything in focus so I set the aperture to f/8, after all I can recover it in post.

Oh wait, I can't, it's far too underexposed and now after pushing 11 stops it looks like a shot taken on my phone from 2006.

Extreme example yes, but proves the point.

It would only prove you are incredibly stupid to try that! (I'd also like to see your image at 11 stops higher ISO) Everything in engineering has tolerances.

 

[Fraser Euan White]

You mean it's never possible to get it wrong even with the correct shutter speed and aperture?

If the shutter speed & aperture are correct for the first shot what else may be wrong that you can adjust?

 

[Craigus]

It would only prove you are incredibly stupid to try that! (I'd also like to see your image at 11 stops higher ISO) Everything in engineering has tolerances.

It's not just advanced amateurs who read this forum.

 

[Fraser Euan White]

It's not just advanced amateurs who read this forum.

??? -sorry I don't follow the meaning of that?

 

[Box Brownie]

I feeling a tad lost in the depth of the technology discussion but to me it seems to summarise as follows.

A) if a camera has an ISO invariant sensor then the photographer has additional tools in his/her armoury at the point of taking the picture to vastly underexpose yet in post production yield acceptable output.

B) a non ISO invariant sensor camera requires the ETTR approach to producing the best final image.

So which make & models fall into which category, because it seems to me that the informed photographer needs to know what and how to get the best for them from a camera........makes wonder why is this approach & pp methodology not discussed in camera reviews e.g. imaging resource and dp review???

Edit ~ most rear LCD have a setting to increase brightness and also show the histogram......so checking and chimp confirmation should be possible whichever type of camera one has, doesn't it???

 

[flavio]

BS. Like the shutter speed doesn't matter, like focal length of the lens makes no difference, like subject distance and sensor size make no difference. These people need to stop trying to mess with what I know to be correct, I will remain sceptical, understanding of the fundamentals of photography whether film or digital remain the same and will ensure that I'm able to capture the images I want and how I want to capture and express them.
thanks.

 

[Fraser Euan White]

BS.

Nice helpful contribution - well done


......................



It is also more helpful that you edited your previous post I quoted in full to make a contribution to the discussion. Nobody has said shutter speed/aperture/focal length/sensor size doesn't matter.

Technologies advance and ***maybe** what we know and understand might not be applicable any more to get the best results?

I don't know the answer I just found the debate on the other forum interesting and will make my own conclusions from my testing - it doesn't cost anything to try unlike the days when I wasted rolls of film trying out things.

 

[woof woof]

BS. Like the shutter speed doesn't matter, like focal length of the lens makes no difference, like subject distance and sensor size make no difference. These people need to stop trying to mess with what I know to be correct, I will remain sceptical, understanding of the fundamentals of photography whether film or digital remain the same and will ensure that I'm able to capture the images I want and how I want to capture and express them.
thanks.

What's BS? If you think ISO invariance is BS it isn't. Google and you shall find.

 

[KIPAX]

At the end of the day its .......... Take a bad picture and fix later .........Somehting I totally disagree with....

 

[KIPAX]

(Isn't 'chimp-ing' a bad habit where you miss so many images whilst looking at the back of your camera?)

not if you have a good reason no..

I don't chimp to see if soemhting is sharp.. I chimp to make sure my exposure is correct when the light keeps changing.. or i chimp to see if I want to lock the file for quick download.... but mostly exposure...

 

[flavio]

What's BS? If you think ISO invariance is BS it isn't. Google and you shall find.

Lol, I can only but laugh out loud. Stop trying to shortcut the fundamentals of photography which to all and intense purposes are all to do with light, rather the quantity and quality of light.
The originator of this thread thinks he's hit on something new, well sorry but I don't think so.
With regard to digital cameras, use them in the normal manner and expose to the right and you'll not go far wrong, think you've hit on a new phenomenon and you'll screw up.

 

[flavio]

What's BS? If you think ISO invariance is BS it isn't. Google and you shall find.

You think you've discovered something new, grow up, try and grasp the fundamentals, then come back to me and admit you were sidetracked, then I'll have a discussion with you!

 

[Fraser Euan White]

I really can't believe so many people are missing the point of this thread!

I am the OP here but I clearly state in the first post that it is a thread running on another forum - so I don't think I have found something new but I find the concept worth investigating.

technology changes over time so why are we so damn stubborn we think the old way is always right?

The debate is that the sensor produces it's best image at it's base ISO (This may or may not be true, I don't know), therefore to get the best IQ expose at base ISO; it is not fixing in post it is a way of getting a better IQ.

Kipax - if you are chimping to see if your exposure is correct, if (and I appreciate it is a BIG if) the theory is correct that will make that unnecessary

Flavio: The way you expose film and digital are completely the opposite so that has changed in the last 20yrs; why shouldn't something else also change? (your responses on this thread are, so far, quite rude!)

 

[flavio]

Nice helpful contribution - well done


......................



It is also more helpful that you edited your previous post I quoted in full to make a contribution to the discussion. Nobody has said shutter speed/aperture/focal length/sensor size doesn't matter.

Technologies advance and ***maybe** what we know and understand might not be applicable any more to get the best results?

I don't know the answer I just found the debate on the other forum interesting and will make my own conclusions from my testing - it doesn't cost anything to try unlike the days when I wasted rolls of film trying out things.

I'm sorry to dissapoint you old chap but don't come here with your news of a major revelation and feel let down when others who've been around a while and who also understand these things aren't as impressed as you are. Understand the fundamentals of photography, apply them to modern technology
and use them to your advantage. You are not fully understanding what you are reporting, there is no quick fix and there are many reasons why, I doubt though you have sufficient grasp of the concepts to comprehend therefore you are not able to effectively convince others of your major revelation. Sorry but don't be so naive!

 

[Fraser Euan White]

I'm sorry to dissapoint you old chap but don't come here with your news of a major revelation and feel let down when others who've been around a while and who also understand these things aren't as impressed as you are. Understand the fundamentals of photography, apply them to modern technology
and use them to your advantage. You are not fully understanding what you are reporting, there is no quick fix and there are many reasons why, I doubt though you have sufficient grasp of the concepts to comprehend therefore you are not able to effectively convince others of your major revelation. Sorry but don't be so naive!

I won't lower myself to your level of insults as you don't know me nor I you - that way the discussion can stay useful Again - it is not my revelation!

 

[flavio]

I won't lower myself to your level of insults as you don't know me nor I you - that way the discussion can stay useful Again - it is not my revelation!

And that is why you don't sufficiently understand it so why try and promote it!

 

[Fraser Euan White]

............but Tony, in your replies you only choose to insult other posters and myself so why should I assume you know more than me/us. I work on evidence and you show very little of it, just a propensity to be insulting. At least others that have 'rebuked' the claim have explained theories which is how a discussion forum 'should' work.

Likewise there are posts on this thread that state the claim may also be true.



Anyway, here is a good article (evidence Tony) about the subject which clearly explains that shooting at a lower ISO and altering exposure in post will give better IQ due to an increase in the dynamic range (assuming your camera is ISO-invariant which he states numerous modern cameras are):

https://photographylife.com/iso-invariance-explained

 

[HoppyUK]

I'm sorry to dissapoint you old chap but don't come here with your news of a major revelation and feel let down when others who've been around a while and who also understand these things aren't as impressed as you are. Understand the fundamentals of photography, apply them to modern technology
and use them to your advantage. You are not fully understanding what you are reporting, there is no quick fix and there are many reasons why, I doubt though you have sufficient grasp of the concepts to comprehend therefore you are not able to effectively convince others of your major revelation. Sorry but don't be so naive!

Having been 'around a while' puts you at a disadvantage on the subject of ISO-invariance. It's a very recent development and pretty much confined to some Sony sensors that are also used by Nikon. It has not made ISO irrelevant (and the very real need for some 'signal gain' to be applied at some point in the chain) but it has certainly re-written the way ISO has traditionally been used at the moment of capture.

 

[ecoleman]

Having been 'around a while' puts you at a disadvantage on the subject of ISO-invariance. It's a very recent development and pretty much confined to some Sony sensors that are also used by Nikon. It has not made ISO irrelevant (and the very real need for some 'signal gain' to be applied at some point in the chain) but it has certainly re-written the way ISO has traditionally been used at the moment of capture.

To Canon shooters, it’s all snake oil. We’re still using sensors from the 50’s and probably will be for a while yet