ISO RANGE

[Quarkcharmed]

Actually, the reverse is true.

Nope. Higher ISO settings in general produce less noise. That's a demonstrable fact. It's why cameras are engineered as they are.

Less noise than what?

If we study i random mages taken by random photographers at different ISOs, and without any noise reduction applied, we'll find a strong correlation "higher ISO -> higher noise". Now whether it's also a causation is a different story, but the correlation is definitely there.

Your statement is only true in a special case: a higher ISO image may produce lower visible noise compared to an image taken with the same exposure and a lower ISO setting, and then brightned up in Lightroom. So your statement implies a) same exposure b) visible noise after certain processing that equalises the brightness.

--
Is it always wrong
for one to have the hots for
Comrade Kim Yo Jong?

--
https://www.instagram.com/quarkcharmed/
https://500px.com/quarkcharmed

 

[Bill Ferris]

Neuman, we've had this conversation before. It's a reference to the output sensitivity; the lightness of the in-camera JPEG.

https://www.dpreview.com/forums/post/66229386

The ISO setting is used, as bobn2 has previously abd accurately explained, to map a relationship between scene brightness and the lightness of the output JPEG. Per ISO's own documentation, it's used to determine an output sensitivity; the lightness of an in-camera JPEG.

--
Bill Ferris Photography
Flagstaff, AZ
http://www.billferris.photoshelter.com

 

[chrisfisheye]

Neuman, we've had this conversation before. It's a reference to the output sensitivity; the lightness of the in-camera JPEG.

https://www.dpreview.com/forums/post/66229386

The ISO setting is used

This is the way this is defined, not necessarily used.

I use it as an analog gain controller as a raw shooter (because only ISO can change it) and get the ultimate SNR improvements.

The standard does not explain this, neither it can explain what means ISOless.

, as bobn2 has previously abd accurately explained, to map a relationship between scene brightness and the lightness of the output JPEG. Per ISO's own documentation, it's used to determine an output sensitivity; the lightness of an in-camera JPEG.

 

[Bill Ferris]

Neuman, we've had this conversation before. It's a reference to the output sensitivity; the lightness of the in-camera JPEG.

https://www.dpreview.com/forums/post/66229386

The ISO setting is used

This is the way this is defined, not necessarily used.

It can only be used to do what it is designed to do. How one "thinks" they're using ISO doesn't change what the setting does.

 

[Fogel70]

Actually, the reverse is true.

Nope. Higher ISO settings in general produce less noise. That's a demonstrable fact. It's why cameras are engineered as they are.

Less noise than what?

If we study i random mages taken by random photographers at different ISOs, and without any noise reduction applied, we'll find a strong correlation "higher ISO -> higher noise". Now whether it's also a causation is a different story, but the correlation is definitely there.

Your statement is only true in a special case: a higher ISO image may produce lower visible noise compared to an image taken with the same exposure and a lower ISO setting, and then brightned up in Lightroom. So your statement implies a) same exposure b) visible noise after certain processing that equalises the brightness.

It is not the higher ISO that cause of more noise, it is the lower amount of light captured that cause higher noise.

It can just as well by too fast shutter speed or stopping down the lens too much is the cause of the noise, and you have to increase ISO more than necessary.

 

[Quarkcharmed]

Actually, the reverse is true.

Nope. Higher ISO settings in general produce less noise. That's a demonstrable fact. It's why cameras are engineered as they are.

Less noise than what?

If we study i random mages taken by random photographers at different ISOs, and without any noise reduction applied, we'll find a strong correlation "higher ISO -> higher noise". Now whether it's also a causation is a different story, but the correlation is definitely there.

Your statement is only true in a special case: a higher ISO image may produce lower visible noise compared to an image taken with the same exposure and a lower ISO setting, and then brightned up in Lightroom. So your statement implies a) same exposure b) visible noise after certain processing that equalises the brightness.

It is not the higher ISO that cause of more noise, it is the lower amount of light captured that cause higher noise.

I wasn't claiming that. I was talking about correlation.

It can just as well by too fast shutter speed or stopping down the lens too much that cause the noise.

If we go pedantic, less light means less noise, more light - more noise.

What you're referring to is actually the signal-to-noise ratio, which generally proportional to the visible noise in the final rendered image.

--
https://www.instagram.com/quarkcharmed/
https://500px.com/quarkcharmed

 

[SrMi]

Actually, the reverse is true.

Nope. Higher ISO settings in general produce less noise. That's a demonstrable fact. It's why cameras are engineered as they are.

Less noise than what?

If we study i random mages taken by random photographers at different ISOs, and without any noise reduction applied, we'll find a strong correlation "higher ISO -> higher noise". Now whether it's also a causation is a different story, but the correlation is definitely there.

Your statement is only true in a special case: a higher ISO image may produce lower visible noise compared to an image taken with the same exposure and a lower ISO setting, and then brightned up in Lightroom. So your statement implies a) same exposure b) visible noise after certain processing that equalises the brightness.

Higher ISO settings produce less noise than lower ISO settings. If that were not the case, all shots would occur at base ISO, and the raw image would contain an ISO tag to brighten it in the post-processor.

If you shoot at ISO 6400 and ISO 800, same exposure and no clipping, after equalizing the image brightness, the ISO 800 image will have more noise. Therefore, lower ISO has more noise than higher ISO.

 

[fferreres]

Actually, the reverse is true.

Nope. Higher ISO settings in general produce less noise. That's a demonstrable fact. It's why cameras are engineered as they are.

Less noise than what?

If we study i random mages taken by random photographers at different ISOs, and without any noise reduction applied, we'll find a strong correlation "higher ISO -> higher noise". Now whether it's also a causation is a different story, but the correlation is definitely there.

Your statement is only true in a special case: a higher ISO image may produce lower visible noise compared to an image taken with the same exposure and a lower ISO setting, and then brightned up in Lightroom. So your statement implies a) same exposure b) visible noise after certain processing that equalises the brightness.

It is not the higher ISO that cause of more noise, it is the lower amount of light captured that cause higher noise.

It can just as well by too fast shutter speed or stopping down the lens too much is the cause of the noise, and you have to increase ISO more than necessary.

I consider the ISO more like A Licenses to Underexpose. The ISO recorded is just a tag, modified by EC. As someone said, EC can change the ISO, but I regard it as trimming this License down or up. All this mostly A. But all this is just a very complicated (in practice, it's easy) way of controlling the shutter speed for a given aperture. Below my min shutter, it switches to underexposing, as the blur (or risk of blur) would be worst than the noise according to my scene. Thus, I check the ISO, and if I feel the scene benefits, I may end up underexposing via EC, which would increase the ISO, or if ISO 100, the shutter speed.

If you follow that nonsense, it actually all means one is trying to control the blur. That's all it is. All the Auto ISO in A. The trade offs are DOF, risk of blur and Noise. Since I check the calculated ISO with half press, then it's not so much different than manual, but I like it more than manual, as the ISO reading doubling is more to my liking than the light meter.

In a few years, I will likely won't care about composition nor noise, with things like MultiNERF invading from the low end to produce 3D photos (ie. like a texture mapped 3D model but based on an actual real scene).

 

[Fogel70]

I would pay extra for a camera with no ISO control, manual or auto. :-D

:-O I bet you're the only one. ;-)

Not by a long chalk. There have been several discussions here about how a camera without ISO would work, and why it would be advantageous. The ISO paradigm leads to a method of exposure setting that is distinctly sub-optimal.

Just choose manual ISO and set it to the value you want.

The opposite. Just use auto ISO. The real problem is that the camera's UI is designed around ISO, so whilst it all works, it's inconvenient in some of the edge cases.

Yeah, I have to agree. Using auto-ISO with manual control on f-stop and SS usually get better optimized exposure than with FI aperture priority and manual ISO. At least in my experience. And having to optimize all three - f-stop, aperture and ISO, is difficult in some fast paced situations with variable light.

Most cameras are usually better optimized for controlling f-stop and SS than controlling ISO.

Using auto-ISO may not work the best if using a mode where you only control one exposure parameter,as then the camera is in control of two parameter and the user only one.

 

[Quarkcharmed]

Actually, the reverse is true.

Nope. Higher ISO settings in general produce less noise. That's a demonstrable fact. It's why cameras are engineered as they are.

Less noise than what?

If we study i random mages taken by random photographers at different ISOs, and without any noise reduction applied, we'll find a strong correlation "higher ISO -> higher noise". Now whether it's also a causation is a different story, but the correlation is definitely there.

Your statement is only true in a special case: a higher ISO image may produce lower visible noise compared to an image taken with the same exposure and a lower ISO setting, and then brightned up in Lightroom. So your statement implies a) same exposure b) visible noise after certain processing that equalises the brightness.

Higher ISO settings produce less noise than lower ISO settings.

Again if we go pedantic, higher ISO limits the exposure and therefore leads to lower noise (in raw).

If that were not the case, all shots would occur at base ISO, and the raw image would contain an ISO tag to brighten it in the post-processor.

'Equalising' the low and high ISO images is an additional implied action. It causes confusion.

If you shoot at ISO 6400 and ISO 800, same exposure and no clipping, after equalizing the image brightness, the ISO 800 image will have more noise. Therefore, lower ISO has more noise than higher ISO.

There's noise in the raw data, visible noise in rendered images and signal-to-noise ratio. Which one are you referring to? Note that equalising the brightness in postprocessing doesn't change anything about the noise in the raw data.

 

[fferreres]

Actually, the reverse is true.

Nope. Higher ISO settings in general produce less noise. That's a demonstrable fact. It's why cameras are engineered as they are.

Less noise than what?

If we study i random mages taken by random photographers at different ISOs, and without any noise reduction applied, we'll find a strong correlation "higher ISO -> higher noise". Now whether it's also a causation is a different story, but the correlation is definitely there.

Your statement is only true in a special case: a higher ISO image may produce lower visible noise compared to an image taken with the same exposure and a lower ISO setting, and then brightned up in Lightroom. So your statement implies a) same exposure b) visible noise after certain processing that equalises the brightness.

Higher ISO settings produce less noise than lower ISO settings.

Again if we go pedantic, higher ISO limits the exposure and therefore leads to lower noise (in raw).

Bewildering one-two punch to the brain. One of a few cases through history where two wrongs didn't make a right.

If that were not the case, all shots would occur at base ISO, and the raw image would contain an ISO tag to brighten it in the post-processor.

'Equalising' the low and high ISO images is an additional implied action. It causes confusion.

If you shoot at ISO 6400 and ISO 800, same exposure and no clipping, after equalizing the image brightness, the ISO 800 image will have more noise. Therefore, lower ISO has more noise than higher ISO.

There's noise in the raw data, visible noise in rendered images and signal-to-noise ratio. Which one are you referring to? Note that equalising the brightness in postprocessing doesn't change anything about the noise in the raw data.

 

[SrMi]

Actually, the reverse is true.

Nope. Higher ISO settings in general produce less noise. That's a demonstrable fact. It's why cameras are engineered as they are.

Less noise than what?

If we study i random mages taken by random photographers at different ISOs, and without any noise reduction applied, we'll find a strong correlation "higher ISO -> higher noise". Now whether it's also a causation is a different story, but the correlation is definitely there.

Your statement is only true in a special case: a higher ISO image may produce lower visible noise compared to an image taken with the same exposure and a lower ISO setting, and then brightned up in Lightroom. So your statement implies a) same exposure b) visible noise after certain processing that equalises the brightness.

Higher ISO settings produce less noise than lower ISO settings.

Again if we go pedantic, higher ISO limits the exposure and therefore leads to lower noise (in raw).

Are you agreeing that higher ISO leads to lower noise, or did I misread you?

How does higher ISO limit the exposure? Are you thinking about metering?

If that were not the case, all shots would occur at base ISO, and the raw image would contain an ISO tag to brighten it in the post-processor.

'Equalising' the low and high ISO images is an additional implied action. It causes confusion.

If you shoot at ISO 6400 and ISO 800, same exposure and no clipping, after equalizing the image brightness, the ISO 800 image will have more noise. Therefore, lower ISO has more noise than higher ISO.

There's noise in the raw data, visible noise in rendered images and signal-to-noise ratio. Which one are you referring to? Note that equalising the brightness in postprocessing doesn't change anything about the noise in the raw data.

If exposure is fixed, as I postulated, then SNR varies only with the noise.

Yes, equalizing the brightness in the post does not change anything about the noise in the raw data. My point is that you will see more noise in ISO 800 not because of equalizing brightness in the post but because lower ISO has more noise than higher ISO.

--
https://www.instagram.com/quarkcharmed/
https://500px.com/quarkcharmed

 

[Quarkcharmed]

Actually, the reverse is true.

Nope. Higher ISO settings in general produce less noise. That's a demonstrable fact. It's why cameras are engineered as they are.

Less noise than what?

If we study i random mages taken by random photographers at different ISOs, and without any noise reduction applied, we'll find a strong correlation "higher ISO -> higher noise". Now whether it's also a causation is a different story, but the correlation is definitely there.

Your statement is only true in a special case: a higher ISO image may produce lower visible noise compared to an image taken with the same exposure and a lower ISO setting, and then brightned up in Lightroom. So your statement implies a) same exposure b) visible noise after certain processing that equalises the brightness.

Higher ISO settings produce less noise than lower ISO settings.

Again if we go pedantic, higher ISO limits the exposure and therefore leads to lower noise (in raw).

Are you agreeing that higher ISO leads to lower noise, or did I misread you?

Not exactly agreeing. What noise are we talking about - the visible noise or noise in raw?

How does higher ISO limit the exposure? Are you thinking about metering?

In Auto ISO mode, ISO is not really a 'setting', it's an exposure index. So it doesn't limit the exposure. In manual ISO mode, the ISO setting does limit the max exposure you can use without highlight clipping.

If that were not the case, all shots would occur at base ISO, and the raw image would contain an ISO tag to brighten it in the post-processor.

'Equalising' the low and high ISO images is an additional implied action. It causes confusion.

If you shoot at ISO 6400 and ISO 800, same exposure and no clipping, after equalizing the image brightness, the ISO 800 image will have more noise. Therefore, lower ISO has more noise than higher ISO.

There's noise in the raw data, visible noise in rendered images and signal-to-noise ratio. Which one are you referring to? Note that equalising the brightness in postprocessing doesn't change anything about the noise in the raw data.

If exposure is fixed, as I postulated, then SNR varies only with the noise.

Fixed exposure wasn't explicitly postulated in the original statement: "higher ISO produces less noise". It was implied.

Yes, equalizing the brightness in the post does not change anything about the noise in the raw data. My point is that you will see more noise in ISO 800 not because of equalizing brightness in the post but because lower ISO has more noise than higher ISO.

What does "has" mean? When I go through my image library, low ISO shots have less noise than high ISO shots.

 

[bobn2]

Actually, the reverse is true.

Nope. Higher ISO settings in general produce less noise. That's a demonstrable fact. It's why cameras are engineered as they are.

Less noise than what?

If we study i random mages taken by random photographers at different ISOs, and without any noise reduction applied, we'll find a strong correlation "higher ISO -> higher noise". Now whether it's also a causation is a different story, but the correlation is definitely there.

Your statement is only true in a special case: a higher ISO image may produce lower visible noise compared to an image taken with the same exposure and a lower ISO setting, and then brightned up in Lightroom. So your statement implies a) same exposure b) visible noise after certain processing that equalises the brightness.

It's not a 'special case'. It's the default case if you want to separate out the effect of different variables. In this case we have the variable 'ISO' and the variable 'exposure' and we want to know what each variable does independently. If we change them together we don't know which is responsible for what.

DPR now provides examples of this in its 'ISO invariance' tests. So, for instance



Exposure does not change, ISO does. We can see very clearly that as the ISO increases, noise decreases.

Here are the same ISOs different exposure



(unfortunately we've added another variable here, EFC, but its effect is minimal). Now compare each patch with the same ISO patch above. It shows that the patch with more exposure is the noisier. So we have two independent variables. ISO, which has quite a small effect on noise, and its noise decreases as ISO increases; and exposure, which has a larger effect, and its noise decreases as exposure increases. So, the rule is, for lowest noise maximise exposure. Set the ISO as high as you can for that exposure.

This is where the most common methods of exposure management get it wrong. I get the impression that the most popular exposure mode is A, so lets run with that. Suppose that you're at some kind of event taking photos as you see them. I suppose that using A you'll set the aperture first, for whatever reason you choose an aperture. Then you set the ISO, either by guess or by adjusting it until you get the shutter speed you feel OK with. Whichever way you did it, by setting the ISO you have fixed the exposure that you'll be using. Thereafter the camera will adjust the shutter speed to keep the exposure constant. So now you go around snapping, and the light changes. Sometimes it will be brighter than it was when you set the ISO, and the shutter speed will increase. But, when it is brighter, you could have left the shutter speed the same, had more exposure and thus, as above, a less noisy photo. Other times the light will be darker and the shutter speed will go below what you're happy with. So this fixed ISO, automatic exposure technique is sub-optimal, because it will often lead to you having a noisier photo than you needed or a slower shutter speed than you're happy with. You could say that you can choose a higher ISO at the start to get some margin in the shutter speed, but then most of your photos will end up noisier than they need to have been.

Now contrast this to using M with auto ISO. You set the aperture you want, you set the shutter speed that' your happy with. As the light changes the ISO changes, and with it the exposure. But, that exposure is always the biggest exposure that will work with the shutter speed and aperture that you chose. Thus in every case it is the least noisy photo that those constraints allow.

Going back to the point of whether it is 'ISO' that causes noise, I think we have shown that it's not, it is exposure (or the lack of it). You see many cases here where people have tied themselves in a knot because of the misunderstanding that ISO causes noise. They set a very low ISO because they don't want noise, then they suffer camera shake or insufficient DOF, so they decide to underexpose at the low ISO (because they think that raising the ISO causes noise), and thus suffer a double whammy, the extra noise that the low ISO setting gives them plus the noise of the low exposure. Understanding the separated effects of noise and exposure is important.

As a final point, the idea that it is 'and then brightened up in Lightroom' is a misconception if you're working in raw. A raw file has no intrinsic lightness. The lightness is set when you process it. So you're not 'brightening up', you're just choosing what is the lightness that you want, and there is no additional penalty for doing so.

--

Is it always wrong
for one to have the hots for
Comrade Kim Yo Jong?

 

[bobn2]

How does higher ISO limit the exposure? Are you thinking about metering?

In Auto ISO mode, ISO is not really a 'setting', it's an exposure index. So it doesn't limit the exposure. In manual ISO mode, the ISO setting does limit the max exposure you can use without highlight clipping.

ISO is always an exposure index, by definition. The difference is that in manual ISO you set that index and it determines the exposure that you use (if you're using an auto exposure mode), in auto ISO that index informs you about the exposure that you're using. ISO is inversely proportional to exposure.

From ISO 12232:2019

exposure index

EI

numerical value that is inversely proportional to the exposure provided to an image sensor to obtain an image

And then for the two most widely used forms of 'ISO'

recommended exposure index

REI

specific exposure index value recommended by a DSC provider as a reference for adjusting photographic accessories.

standard output sensitivity

SOS

specific exposure index value for a DSC that provides a still image with a specific DSC image signal value under specified test conditions.

 

[bobn2]

Using auto-ISO may not work the best if using a mode where you only control one exposure parameter,as then the camera is in control of two parameter and the user only one.

Yes, the problem is that the design goal of auto exposure modes is wrong for digital cameras. They were designed for film, where you're trying to keep the exposure on the right part of the characteristic curve. The curve in digital is very nearly linear, so what the goal should be is maximising exposure subject to your constraints.

With modern digital cameras, there is scope for much more sensible automation. A mirrorless camera, which is reading the sensor continuously, and can without much issue estimate scene movement and distances, so exposure automation could make sensible choices about shutter speeds and apertures according to the photographer's requirements.

 

[D Cox]

Neither do I. Pictures lack consistency.

That's because you aren't using it properly. If you have a method of exposure setting with manual ISO, then there is an analogue for auto ISO that will achieve exactly the same results, but you need to think it through. In general though, it pairs most easily with a raw workflow.

I think I'll stay where I am. And I do RAW only.

Well, of course it's your affair how you work, but it would be interesting to discuss the reasoning behind it.

I think you overestimate me. I use simple processes, some of them are simply force of habit. For example, I always shoot RAW, I never use Auto ISO, I never use Spot Metering, 99% of time it is Aperture priority, simple things like that. I would not even attempt discussing them, that's the way I like them. You may call me limited, I am perfectly fine with it.

Pictures taken at different ISOs are not the same, if there were, there would be no upper limit to ISO.

Maybe we could discuss the ways that you think pictures taken at different ISOs are not the same.

Not sure what there is to discuss. That they are not different? Or, why are they different? I think you may need a better partner for your discussion. I am satisfied with simple: They are different.

It's true that higher ISO settings tend to produce less noise,

Actually, the reverse is true.

He means higher ISO settings with the same shutter speed and aperture. These might have less read noise. I doubt if the difference would be apparent, and this technique requires you to set speed and aperture manually -- by which time the bird has flown.

I think your method is the best.

Don

 

[D Cox]

Neither do I. Pictures lack consistency.

That's because you aren't using it properly. If you have a method of exposure setting with manual ISO, then there is an analogue for auto ISO that will achieve exactly the same results, but you need to think it through. In general though, it pairs most easily with a raw workflow.

I think I'll stay where I am. And I do RAW only.

Well, of course it's your affair how you work, but it would be interesting to discuss the reasoning behind it.

I think you overestimate me. I use simple processes, some of them are simply force of habit. For example, I always shoot RAW, I never use Auto ISO, I never use Spot Metering, 99% of time it is Aperture priority, simple things like that. I would not even attempt discussing them, that's the way I like them. You may call me limited, I am perfectly fine with it.

I haven't called you limited, but if it's just 'that's my preference because it suits me' there's not much anyone else can take from that. Their preferences could be different and their's no reason to think that they might be better or worse. So coming back to the thing being discussed here, whether or not auto ISO is a good idea, there's no reason to think that what you choose illuminates the discussion.

Pictures taken at different ISOs are not the same, if there were, there would be no upper limit to ISO.

Maybe we could discuss the ways that you think pictures taken at different ISOs are not the same.

Not sure what there is to discuss. That they are not different? Or, why are they different?

I was thinking about how they are different, which might lead to understanding whether they are significantly different.

I think you may need a better partner for your discussion. I am satisfied with simple: They are different.

The real point I think is that for you 'it's my personal preference' is a satisfactory reason for doing things, but as above it's not something anyone else would be interested in.

It's true that higher ISO settings tend to produce less noise,

Actually, the reverse is true.

Nope. Higher ISO settings in general produce less noise. That's a demonstrable fact. It's why cameras are engineered as they are.

It's demonstrable with manual settings and JPG output. It's not demonstrable with A, S or P modes and raw files. When higher ISO settings lead to less exposure (because you want a faster shutter speed), then shot noise will increase. The only thing a camera can do to reduce shot noise is to apply (crude) noise reduction to JPG files.

(I hear rumours that one company applies noise reduction to "raw" files. If so, they are not raw files.)

Don

 

[D Cox]

If there had been a technical reason for calling it 'ISO sensitivity' they would have given it, in fine detail and with formulae that explained it. The very fact that they say that it is for historical reasons means that there is no technical reason, but a historical one.

So we are all agreed: the so-called ISO control is the sensitivity setting.

It's not a 'so-called ISO control'. It is an ISO control, which also has side-effects not mandated in the standard. But all of the ISO controls that I know of do actually control the ISO.

My issue with calling it a 'sensitivity setting' is that it's a very misleading term, with respect to what the control actually does. Most people, when they encounter a technical term don't go and read the documents defining that term, they take a shot at what is meant by from a more generalised view of the meaning of the word. If you take that approach to the use of the word 'sensitivity' with respect to ISO you end up completely misunderstanding what ISO is and how it operates. This has led to a whole load of web 'experts' taking this approach and producing material that is just plain wrong.

For instance, this is quite typical:

What is ISO in photography?

ISO refers to your camera’s sensitivity to light. The higher the ISO, the more sensitive your camera sensor becomes, and the brighter your photos appear.

...

By increasing the ISO, you make your photos brighter.

That’s why ISO is important.

In other words, ISO works alongside the other two exposure variables – aperture and shutter speed – to determine the overall brightness level of an image.

and so on. That's one of very many.

I agree. The ISO setting is not a "sensitivity" (or QE) setting. It's a setting to reduce exposure and attempt to ameliorate the resulting increase in shot noise.

Almost always, you raise the ISO number because you want a faster shutter speed for hand-held shooting. For still life work in a studio with the camera on a tripod or copy stand, there's no reason to ever use any ISO setting other than base.

Don

 

[bobn2]

Neither do I. Pictures lack consistency.

That's because you aren't using it properly. If you have a method of exposure setting with manual ISO, then there is an analogue for auto ISO that will achieve exactly the same results, but you need to think it through. In general though, it pairs most easily with a raw workflow.

I think I'll stay where I am. And I do RAW only.

Well, of course it's your affair how you work, but it would be interesting to discuss the reasoning behind it.

I think you overestimate me. I use simple processes, some of them are simply force of habit. For example, I always shoot RAW, I never use Auto ISO, I never use Spot Metering, 99% of time it is Aperture priority, simple things like that. I would not even attempt discussing them, that's the way I like them. You may call me limited, I am perfectly fine with it.

Pictures taken at different ISOs are not the same, if there were, there would be no upper limit to ISO.

Maybe we could discuss the ways that you think pictures taken at different ISOs are not the same.

Not sure what there is to discuss. That they are not different? Or, why are they different? I think you may need a better partner for your discussion. I am satisfied with simple: They are different.

It's true that higher ISO settings tend to produce less noise,

Actually, the reverse is true.

He means higher ISO settings with the same shutter speed and aperture. These might have less read noise. I doubt if the difference would be apparent, and this technique requires you to set speed and aperture manually -- by which time the bird has flown.

I don't think that's the case. By and large, if you're using auto ISO you preset the shutter speed and aperture according to your limits for the situation in which you're shooting. What's happening is that rather than fixing exposure with the ISO control, then trying to juggle the shutter and aperture, you're letting the exposure change and freeing up control of shutter and aperture.

So, let's think about what you'd do, photographing the bird in the different scenarios.

First, let's look at the ISO-first methods.

Presumably for the bird, you're using a long lens and maybe want a bit of wing blur, so you choose a shutter speed accordingly, and use S (Tv) mode. Then you have to choose an ISO. You change the ISO until you have an aperture that will give you sufficient DOF to capture the bird. No you go around photographing. It's OK until you get a bird in darker light, the aperture increases and you no longer have enough DOF, so you have to decrease the shutter speed to open the aperture. Now is it too slow to prevent shake? Alternatively you can change the ISO to reduce exposure and so keep the shutter and aperture within limits. Changing the ISO is generally slower (unless you can dedicate a control ring on the lens, which would be sensible here). In any case, you've suddenly been plunged into making a quick decision and adjusting at least one control, in which case the bird might have flown.

The same is true in A (Av), except now it's the shutter that you might find has gone outside limits, with the same results as above.

Now lets look at an auto ISO mode. You start off deciding the widest aperture that will get you the DOF you need (either enough to keep all of the bird in focus or give adequate background blur, as you wish). Now you decide on the slowest shutter speed that you need that will prevent shake, or give adequate wing motion blur, depending on what you're trying to do. Now you set auto ISO. You can go and photograph the birds without touching the aperture and shutter, and you'll never find yourself having to make the on-the fly decisions above. The one time that you might is when the light is bright enough that the ISO has gone as low as it can and you need to reduce the exposure, so one wheel touch will do that. Alternatively you could have set up the camera to do that automatically. Put it in A (Av) mode, select the aperture as above. Set the minimum shutter speed in the auto ISO menu with the shutter speed as above. Now the A mode takes priority over auto ISO, so if the light is bright enough for base ISO the camera adjusts the shutter speed to stop overexposure. If it's not bright enough the shutter speed hits the limit and the auto ISO raises the ISO.