Reading mode: Light Dark

The only thing you should be worried about ISO

Go to last post
R

Raidenorius

New member
Messages
1
Reaction score
0
Hello there.

I'm pretty new to photography, so please correct me.

One thing stills bugs me - ISO.

ISO is the only variable of the exposure that isn't analog but digital. With aperture and shutter speed you change the amount of photons you gather, but by changing ISO you just change how the sensor reacts to these photons.

ISO should be a standard (as a word "standard" is a part of "ISO" acronym"), but these days everyone knows it isn't photography standard any more.

So - why we still compare the thing we shouldn't - "ISO performance (e.g. A7RIII 12800 ISO vs Nikon Z7 12800 ISO) instead of comparing the thing that can tell us how we should handle the photos form different cameras in the post?

I mean noise of different ISO settings in one camera, which is connected to ISO Invariance and postproduction.

34a3579640ad48719e4cdc98116028b6.jpg.png

I'm X-T 20 owner, my friend uses 6d Mark II.

Look at X-T20 first. Reading the image above, I know that there is a big change how camera reacts to photons (or rather how Fuji image processor acts and change raw data) between ISO 200-640 and ISO 800-12800.

The camera is ISO Invariance between ISO 200-640 and ISO 800-12800, so when I will pull by 1EV in the post, the photo taken with ISO 200, but underexposed by 1EV, I will get prety much the same result as I would taken that photo with ISO 400 and proper exposure.

I could take photo underexposed by 4EV with ISO 800, and then pull it in post by 4EV, with the same result as photo properly exposed with ISO 12800 (the noise would be the same).

But, there would be a difference in photos taken with ISO 400 pulled by 4EV in the post, and the photo taken with ISO 6400, as the X-T20 isn't ISO Invariance between ISO 400-6400 (the noise change as the fuji change raw data at 640-800 step).

What that knowledge gives me?

Thanks to that I know that in lower light conditions I can shoot a photo with ISO 800 or higher and get lower noise but a little darker image (this is what fuji processor do to photos after iso 800) or I can take a little underexposed image with ISO 200-640 and pull it in the post, which will resulted with more noise, but won't be darken by the fuji processor.

Look now how nonlinear in ISO noise is the Canon 6d Mark II.

This gives you less flexibility in the post.

I hope that I made my point there - we should pay more attention to compare the ISO performance in the camera, not between different cameras (we can still compare the noise by different cameras).

Here is the link to the tool - http://www.photonstophotos.net/Charts/RN_ADU.htm
 
Raidenorius said:
One thing stills bugs me - ISO.

ISO is the only variable of the exposure that isn't analog but digital.
Click to expand...
Can we start with this: ISO is not a variable of exposure, the definition of photographic exposure doesn't include any ISO speed or sensitivity parameters.

ISO can be digital, but it is much more common that it is analogue, or analogue+digital.
Raidenorius said:
With aperture and shutter speed you change the amount of photons you gather, but by changing ISO you just change how the sensor reacts to these photons.
Click to expand...
No, sensor doesn't react to the change of ISO speed.

--
http://www.libraw.org/
 
Last edited:
Raidenorius said:
Hello there.

I'm pretty new to photography, so please correct me.

One thing stills bugs me - ISO.

ISO is the only variable of the exposure that isn't analog but digital. With aperture and shutter speed you change the amount of photons you gather, but by changing ISO you just change how the sensor reacts to these photons.

ISO should be a standard (as a word "standard" is a part of "ISO" acronym"), but these days everyone knows it isn't photography standard any more.

So - why we still compare the thing we shouldn't - "ISO performance (e.g. A7RIII 12800 ISO vs Nikon Z7 12800 ISO) instead of comparing the thing that can tell us how we should handle the photos form different cameras in the post?

I mean noise of different ISO settings in one camera, which is connected to ISO Invariance and postproduction.

34a3579640ad48719e4cdc98116028b6.jpg.png

I'm X-T 20 owner, my friend uses 6d Mark II.

Look at X-T20 first. Reading the image above, I know that there is a big change how camera reacts to photons (or rather how Fuji image processor acts and change raw data) between ISO 200-640 and ISO 800-12800.

The camera is ISO Invariance between ISO 200-640 and ISO 800-12800, so when I will pull by 1EV in the post, the photo taken with ISO 200, but underexposed by 1EV, I will get prety much the same result as I would taken that photo with ISO 400 and proper exposure.

I could take photo underexposed by 4EV with ISO 800, and then pull it in post by 4EV, with the same result as photo properly exposed with ISO 12800 (the noise would be the same).

But, there would be a difference in photos taken with ISO 400 pulled by 4EV in the post, and the photo taken with ISO 6400, as the X-T20 isn't ISO Invariance between ISO 400-6400 (the noise change as the fuji change raw data at 640-800 step).

What that knowledge gives me?

Thanks to that I know that in lower light conditions I can shoot a photo with ISO 800 or higher and get lower noise but a little darker image (this is what fuji processor do to photos after iso 800) or I can take a little underexposed image with ISO 200-640 and pull it in the post, which will resulted with more noise, but won't be darken by the fuji processor.

Look now how nonlinear in ISO noise is the Canon 6d Mark II.

This gives you less flexibility in the post.

I hope that I made my point there - we should pay more attention to compare the ISO performance in the camera, not between different cameras (we can still compare the noise by different cameras).

Here is the link to the tool - http://www.photonstophotos.net/Charts/RN_ADU.htm
Click to expand...
Good points, and I agree, ISO is best when used at low settings like 100 ISO. But, it all depends on the shooting conditions and what type of exposure is used. I'll shoot 3200 ISO all night when the milky way is up, and sometimes, I'll shoot 6400 ISO when it's really dark. Reason is, I need to keep shutters at their max limits without stars turning into orbs or dashes.


With Dynamic range and ISO so good on my MFT cameras, I can keep shutters at 1/1250 for wildlife, and bump ISO to 1600 and 3200 when the light gets low.


And because I shoot a lot of 4K video, the shutter speeds can only go as low as 1/40 or 1/50, at f/1.7. Thus, any exposure adjutment will be on the ISO. And when light gets low, I'll shoot at 1600 to 3200 ISO for clean 4K video at night.



Thus, it all depends on the light. If the light is bright, it's easy to keep ISO at 200 and lower shutter to as low as one second hand held, and even longer on a tripod.
 
Welcome to the forums!

There are two basic standards regarding camera exposure index: the international ISO 12232 standard and the Japanese CIPA DC-004 standard. You can find an older version of the ISO standard online, if you look hard enough, while the CIPA standard can be found on that organization’s website.

While the standards seem complex, ultimately what they define is very simple: a certain amount of light on the sensor leads to a specific midtone RGB value in a JPEG, which depends on the Exposure Index (or ISO) of the camera. Please note that this depends largely on arbitrary processing choices and may have little or nothing to do with noise, highlight headroom, or much of what we think about when considering ISO.
 
Iliah Borg said:
ISO can be digital, but it is much more common that it is analogue, or analogue+digital.
Click to expand...
The most common definition of ISO I have encountered on these fora is that it's how the voltage generated by the photons hitting the photo site is converted into a (digital) lightness value. The process is sometimes called "mapping". So would it be correct to define ISO as a variable guiding analogue to digital conversion? Not sure how that would be more analogue than digital. If we stay within the realm of digital photography.
 
Maybe the effects of climate change.
The disintegration of both institutional and interpersonal trust in society.
The possibility of being attacked by an alligator on the street.

All of the above, at least in aggregate, are slightly more worrisome than ISO.
 
Raidenorius said:
ISO should be a standard (as a word "standard" is a part of "ISO" acronym"), but these days everyone knows it isn't photography standard any more....

I hope that I made my point there - we should pay more attention to compare the ISO performance in the camera, not between different cameras (we can still compare the noise by different cameras).
Click to expand...
Well, first ISO is not an acronym, it's a word, and the S doesn't stand for "standard".

"Because 'International Organization for Standardization' would have different acronyms in different languages (IOS in English, OIN in French for Organisation internationale de normalisation), our founders decided to give it the short form ISO. ISO is derived from the Greek isos, meaning equal. Whatever the country, whatever the language, we are always ISO." - Source

I think we should compare ISO between cameras. Some really do have much cleaner images at higher ISO's than others.
 
Did you steal this from a Tony Northrup yuotube video stating the exact same thing few months back
 
It's not that ISO is digital and everything else is Analog. It's that ISO is an artificial construct, and not actually a direct factor when taking an image. The data captured by the camera is fundamentally affected by subject illumination, aperture, and shutter speed. ISO provides a context for that captured data.

Many people think of ISO as controlling noise. They think a high ISO setting causes the image to be noisier. This is not strictly the case.

The biggest factor in visible image noise is frequently the "shot noise". This is the noise inherent in the quantum nature of light. Visible shot noise is correlated to the total light captured (number of photons seen) by the camera.

"ISO" lets us talk about how the total light is distributed on the sensor. It's a holdover from the days of film, where we had a target "light per unit area" in order to hit the sweet spot of the film's S-shaped response curve.

Digital cameras don't have that sort of response curve, and there is a wide range of exposures that result in a good image.

According the the ISO Digital Camera Speed standard, applying a single ISO speed to a digital camera isn't a good match for digital cameras. The reason that such a standard exists is to ease the transition from digital to film. The ISO speed setting allows the film photographer to use his film workflow in a digital camera (even though the workflow is not optimized for digital).

ISO does specify a target exposure for light per unit area on the sensor. But even this can't be directly used to estimate noise. The noise is based on total light gathered, and therefore you need to multiply the light per unit area by the sensor size to get the total light gathered. This is why at a particular exposure (light per unit area), a full frame camera produces a less noisy image than a 2X crop body. While both cameras may be set to ISO 800, the full frame has four times the sensor area, and therefore at the same light per unit area it gets four times as much total light.
 
rurikw said:
Iliah Borg said:
ISO can be digital, but it is much more common that it is analogue, or analogue+digital.
Click to expand...
The most common definition of ISO I have encountered on these fora is that it's how the voltage generated by the photons hitting the photo site
Click to expand...
It's current that is generated by photons hitting photodiodes, not voltage.

Photocurrent - Wikipedia

en.wikipedia.org en.wikipedia.org
rurikw said:
would it be correct to define ISO as a variable guiding analogue to digital conversion?
Click to expand...
No. ISO speed is what ISO standard says it is.
 
Last edited:
Michael Fryd said:
It's not that ISO is digital and everything else is Analog. It's that ISO is an artificial construct, and not actually a direct factor when taking an image. The data captured by the camera is fundamentally affected by subject illumination, aperture, and shutter speed. ISO provides a context for that captured data.

Many people think of ISO as controlling noise. They think a high ISO setting causes the image to be noisier. This is not strictly the case.

The biggest factor in visible image noise is frequently the "shot noise". This is the noise inherent in the quantum nature of light. Visible shot noise is correlated to the total light captured (number of photons seen) by the camera.

"ISO" lets us talk about how the total light is distributed on the sensor. It's a holdover from the days of film, where we had a target "light per unit area" in order to hit the sweet spot of the film's S-shaped response curve.

Digital cameras don't have that sort of response curve, and there is a wide range of exposures that result in a good image.

According the the ISO Digital Camera Speed standard, applying a single ISO speed to a digital camera isn't a good match for digital cameras. The reason that such a standard exists is to ease the transition from digital to film. The ISO speed setting allows the film photographer to use his film workflow in a digital camera (even though the workflow is not optimized for digital).

ISO does specify a target exposure for light per unit area on the sensor. But even this can't be directly used to estimate noise. The noise is based on total light gathered, and therefore you need to multiply the light per unit area by the sensor size to get the total light gathered. This is why at a particular exposure (light per unit area), a full frame camera produces a less noisy image than a 2X crop body. While both cameras may be set to ISO 800, the full frame has four times the sensor area, and therefore at the same light per unit area it gets four times as much total light.
Click to expand...
Unless you are using "quantum" to mean "particle-like" the quantum nature of light has nothing to do with shot noise - it would be there even if light was made of tiny ink-filled baloons.
 
WunWegWunDarWun said:
Unless you are using "quantum" to mean "particle-like" the quantum nature of light has nothing to do with shot noise - it would be there even if light was made of tiny ink-filled baloons.
Click to expand...
Please explain and cite your sources.
 
Mark Scott Abeln said:
WunWegWunDarWun said:
Unless you are using "quantum" to mean "particle-like" the quantum nature of light has nothing to do with shot noise - it would be there even if light was made of tiny ink-filled baloons.
Click to expand...
Please explain and cite your sources.
Click to expand...
I simulated the result of tiny ink balloons being randomly thrown on a white wall and found the result to be statistically indistinguishable from the models used to describe shot noise. Of course I might have cheated and used an underlying distribution whose analogue in nature is due to fundamental quantum physics (what isn't ?) but my virtual ink balloons interact with the virtual wall in a completely classical particle-like way.
 
Iliah Borg said:
Iliah Borg

No. ISO speed is what ISO standard says it is.
Click to expand...
I hope so. It's rather expensive at 106€ so can you recommend a good free explanation?
 
You must log in or register to reply here.

Similar threads

T
ISO test Nikon Z6
Replies
0
Views
104
Thomerik
T
N
XT-5 vs X-H2S for night photography
Replies
6
Views
35K
Northern_Amateur
N
Bill Ferris
Interesting Nikon Z9 Behavior
2 3 4
Replies
70
Views
1K
SKersting66
SKersting66
S
How reliable are “Photons to Photo” data? G9 vs Canon 6DMII
4 5 6
Replies
108
Views
3K
Jens_ac
J
BeatX
X-Trans V sensor base ISO DR test vs DR200, DR400 modes
2 3
Replies
50
Views
4K
Erik Baumgartner
Erik Baumgartner

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top