Does changing shutter speed and aperture create equalized exposure?

[Ysarex]

Just a noob hear, but the ISO standard implimented in I'm thinkin in the 70's. Was to standardize film. My impression was and still is, goes back to the way I think about it. Back in the day different solutions of a silver solution on glass plates. The "faster" the solution, more of the silver grain was visible..... Low ISO/DIN fine grain ... High ISO/DIN the grain is more coarse. Then the medium changed to the "Silver" being emulsified in what we call film. Same mechanics, just simpler for the masses.
I learned the triangle, it is ingrained, I do always remember what My H.S. Teacher said of exposure though: "Compose your picture, think about what you want to capture, SEE it, create it in your mind. Then snap it. The Art you create is like a painting.. No two are ever the same.

Film and digital have separate ISO standards. They really aren't the same thing. Back in the early days of digital when so many of the new digital users were transitioning from film there was a tendency to ease that transition by presenting ISO in digital as analogous to ISO in film. That was probably unwise. Few aspects of our digital cameras are as badly misunderstood now as ISO.

 

[Ysarex]

My understanding of changing the ISO is that it does not change the exposure, as changing only the ISO does not change the amount of light on the sensor, but changing the ISO does change the camera's interpretation of that exposure. If I change aperture from f:5.6 to f: 8, a decrease in exposure of one stop, and change the ISO from 200 to 400, an increase of one stop in sensitivity, the end result will be "the same."

The two images will appear the same brightness, but the end result is not the same. Changing ISO does not change sensitivity and the one stop reduction in exposure adds noise which is not affected by the ISO change.

Is the exposure the same? No.

The misuse of "exposure" as meaning the result of the exposure, i.e. the presented image, is what's wrong with saying ISO is a leg of the exposure triangle. It's a leg of the "result" triangle.

Personally, I don't use ISO as a factor in my settings other than using appropriate shutter and aperture for the selected ISO, just like using appropriate settings for the loaded film. I use the lowest possible ISO for conditions, and I've never used auto-ISO.

 

[smoke665]

A lot of back and forth going on with bits and pieces of the same thing being said in different ways. As the OP hasn't been back lately I fear that the technical direction the thread took might have scared them off. This is after all the Beginners Forum.

 

[Rickbb]

A lot of back and forth going on with bits and pieces of the same thing being said in different ways. As the OP hasn't been back lately I fear that the technical direction the thread took might have scared them off. This is after all the Beginners Forum.

they were here just 15 mins ago, we may have just bored them. lol

 

[AlanKlein]

The end results are not the same. In the case I used with just a one stop difference there's not a lot of variance in the end result, but exposure (how much light reaches the film/sensor) matters because it's effects show in the end result. As you drop exposure you lose SNR and raising the ISO does not compensate that loss.

The point is understanding cause and effect. A shutter speed change causes a change in exposure. An aperture change causes a change in exposure. An ISO change does not cause a change in exposure and exposure matters.

You're complicating this especially for a beginner. Sure, the exposure doesn't change on the sensor when you only change the ISO setting. But from a practical standpoint, it's the end result we're concerned about. The same thing happens with film. The light falling is the same. It's just that we changed the film to a different ISO so the end result allows more or less final brightness in the picture. All three variables work together.

 

[AlanKlein]

As I noted the Exposure Triangle has a problem that involves contemporary presentation. ISO doesn't cause noise in a digital photograph, but far too many of the folks explaining the triangle on the Internet will say it does. Changing ISO has no effect on the light sensitivity of the camera's sensor, but far too many of the folks explaining the triangle on the Internet will say it does. Photographic exposure (for far longer than we've been alive) is defined as the amount of light falling on the film/sensor per unit area -- a function of scene luminance, shutter speed and aperture. Changing ISO does not change the amount of light falling on the film/sensor but far too many of the folks explaining the triangle will it say it does.

So does it apply if ISO is not a causal component of exposure? Can the triangle be presented by starting with a correct definition of exposure?

Go to youtube, do a search on exposure triangle and watch a random 20 of them. It's pretty much a guarantee that you'll be misinformed 20 times. New photographers don't realize they're being misinformed and maybe they'll be OK and little harm will be done, but why start beginners off with false info?

I've had numerous occasion to have to help someone re-learn because they were getting poorer quality photos from their cameras having learned the exposure triangle off the Internet.

Raising ISO does cause visible noise because of its amplifier in the same way musical amplifiers cause sound distortion when you raise the volume too high. Of course, depending on the camera, noise might not be visible until you reach higher ISO levels just as with better musical amps and speakers. But selecting a different ISO changes where you have to set the shutter speed and aperture.

 

[Ysarex]

Raising ISO does cause visible noise because of its amplifier in the same way musical amplifiers cause sound distortion when you raise the volume too high.

First of all, what amplifier? It's not a given that raising ISO involves an amplifier. And no raising ISO does not cause noise. ISO is either noise neutral or in fact it suppresses noise.

That's one of the cases where I've often had to help one of my students who had unfortunately already been triangulated by the Internet. I'm in class looking at my students photos and the exchange would go something like this:
ME: These photos really are too dark overall.
STUDENT: Yeah, I was trying to get photos inside without a flash and I couldn't lower my shutter speed any more -- half of them are already blurry.
ME: You raised the ISO to 1600, raising the ISO was the right thing to do you just needed to raise it higher.
STUDENT: I didn't want the photos to be any more noisy. I figured I could go with ISO 1600 noise and then try and improve the photos in Photoshop.
ME: If you had raised the ISO higher, using the same exposure you'd have less noise.
STUDENT: What? The noise get's worse as the ISO goes up right?
ME: Wrong. Where'd you get that idea?
STUDENT: Youtube tutorials.
ME: The noise get's worse as the exposure goes down -- the noise is not caused by ISO. In the case of your camera a higher ISO with the same exposure would further suppress read noise and your photos would look better.
ME: Let's do some tests with your camera and I'll show you.

Here's a photo that I took for class as part of our discussion of ISO: APS-C sensor camera (Fuji X-T2) at ISO 25,600. No special noise filtering used -- no Topaz AI or DXO deep Prime -- just standard processing through C1: ISO-25K.jpg That's the highest ISO on the camera; it's not causing noise.

Of course, depending on the camera, noise might not be visible until you reach higher ISO levels just as with better musical amps and speakers. But selecting a different ISO changes where you have to set the shutter speed and aperture.

ISO correlates with noise. Correlation is not cause. If you're shooting camera JPEGs the ISO/noise correlation is pretty strong and you can't do much to modify it -- as you note changing ISO changes the exposure you have to set, but it's the exposure that causes the noise. However if you're shooting raw the ISO/noise correlation can be pretty weak (depends on the design of the specific camera) and it may be possible to do what I did in the photo I linked -- basically break the correlation because ISO doesn't cause noise.

 

[Ysarex]

You're complicating this especially for a beginner.

I think I'm uncomplicating it. ISO is a critical function in our cameras and we need to understand what it does and use it appropriately. If we think it's doing things that it's not doing, that's a complication.

Sure, the exposure doesn't change on the sensor when you only change the ISO setting. But from a practical standpoint, it's the end result we're concerned about. The same thing happens with film. The light falling is the same. It's just that we changed the film to a different ISO so the end result allows more or less final brightness in the picture. All three variables work together.

 

[RAZKY]

Not trying to throw a monkey wrench in the works but sometimes the "best" exposure is not so simple. IE: you've metered your scene and determined the DR exceeds the camera capabilities. Do you expose for the highlights and try to recover shadows post, or the focal point knowing you'll blow the highlights? Do you adjust ISO when making your decision? Why? Are you opening up the aperture, if so will you have enough DOF? If you reduce shutter speed will you incur blur? A meter gives you parameters to work with but the photographer determines how they use those parameters.

I say it's simple because since 1954 I've looked at enough scenes to know pretty well what I can do with it before taking the camera or meter out of its case. (Note I said can do, not want to do.) I generally avoid expansive scenes - forces the eye to wander, wondering what's the purpose of the photograph. A smaller area is easier to illuminate and frame, leaving the viewer with no doubts. I've made enough internegs and masks to know that I don't like darkroom work or sitting in front of a computer, but I've kept quite a number for further processing, and they won't be easy! For me, at least.

 

[AlanKlein]

First of all, what amplifier? It's not a given that raising ISO involves an amplifier. And no raising ISO does not cause noise. ISO is either noise neutral or in fact it suppresses noise.

That's one of the cases where I've often had to help one of my students who had unfortunately already been triangulated by the Internet. I'm in class looking at my students photos and the exchange would go something like this:
ME: These photos really are too dark overall.
STUDENT: Yeah, I was trying to get photos inside without a flash and I couldn't lower my shutter speed any more -- half of them are already blurry.
ME: You raised the ISO to 1600, raising the ISO was the right thing to do you just needed to raise it higher.
STUDENT: I didn't want the photos to be any more noisy. I figured I could go with ISO 1600 noise and then try and improve the photos in Photoshop.
ME: If you had raised the ISO higher, using the same exposure you'd have less noise.
STUDENT: What? The noise get's worse as the ISO goes up right?
ME: Wrong. Where'd you get that idea?
STUDENT: Youtube tutorials.
ME: The noise get's worse as the exposure goes down -- the noise is not caused by ISO. In the case of your camera a higher ISO with the same exposure would further suppress read noise and your photos would look better.
ME: Let's do some tests with your camera and I'll show you.

Here's a photo that I took for class as part of our discussion of ISO: APS-C sensor camera (Fuji X-T2) at ISO 25,600. No special noise filtering used -- no Topaz AI or DXO deep Prime -- just standard processing through C1: ISO-25K.jpg That's the highest ISO on the camera; it's not causing noise.

ISO correlates with noise. Correlation is not cause. If you're shooting camera JPEGs the ISO/noise correlation is pretty strong and you can't do much to modify it -- as you note changing ISO changes the exposure you have to set, but it's the exposure that causes the noise. However if you're shooting raw the ISO/noise correlation can be pretty weak (depends on the design of the specific camera) and it may be possible to do what I did in the photo I linked -- basically break the correlation because ISO doesn't cause noise.

Ok, let's assume it's the exposure reduction that's causing noise when you raise the ISO setting, not the ISO setting itself. What difference does that make in the final result? When you raise the ISO setting, the amount of light falling on the sensor (or film) decreases on the sensor because you're reducing the aperture size or speeding up the shutter speed to compensate for the higher ISO setting. The picture is still noisier. No?

Another question I have, if amplifiers aren't amplifying the light and changing the amplification as you change the ISO setting, what is happening in the camera?

 

[Ysarex]

Ok, let's assume it's the exposure reduction that's causing noise when you raise the ISO setting,

No need to assume -- the cause is well understood. There are two common types of noise that concern us in our photos: shot noise and read noise. Shot noise is by far and away dominant and typically what we see when we see a noisy photo. Read noise, from the camera's electronics, is less a problem and getting less all the time as the hardware improves. Shot noise is entirely a function of exposure -- SNR (signal to noise ratio) get's better with more exposure and worse with less exposure. What ISO is doing in the camera doesn't directly alter exposure and so doesn't cause shot noise. ISO implementation in the camera via amplification of the analog sensor signal suppresses read noise. What's that noise? Part one: Shedding some light on the sources of noise

not the ISO setting itself. What difference does that make in the final result? When you raise the ISO setting, the amount of light falling on the sensor (or film) decreases on the sensor because you're reducing the aperture size or speeding up the shutter speed to compensate for the higher ISO setting. The picture is still noisier. No?

Yes, ISO correlates with noise and as I noted that correlation when shooting camera JPEGs is pretty strong. But I did also note and provide an example shooting raw where that correlation is weak. The difference in the final result is in the example I gave of my student's misunderstanding -- a noisier photo because he thought ISO caused noise and so was reluctant to raise the ISO which would have in the final result made his photo less noisy. It's about understanding cause and effect. Thinking ISO causes noise is a mistake that can lead the user to getting a poorer final result. Mistaking correlation for cause and knowing the cause can matter.

Another question I have, if amplifiers aren't amplifying the light and changing the amplification as you change the ISO setting, what is happening in the camera?

Amplification is commonly used to implement ISO. The signal from the sensor is analog and it can be amplified before being digitized. Many cameras in whole or in part implement ISO using signal amplification. However there are ways to implement ISO other than signal amplification and they are also used. In some cases a camera maker may use more than one implementation at a time -- my Fuji cameras will do that.

What ISO does, as determined by the industry standard, is establish a methodology for determining the brightness in the camera output JPEG that results from a measured exposure of the camera sensor. As such ISO is a post-processing procedure in that it occurs after the exposure has been made. The industry standard does not specify implementation methods and the camera makers can do as they see fit. Amplifying the sensor signal is the most commonly used method however some cameras will use digital scaling in the ADC instead or in combination with amplification and some will simply apply the needed image brightening in the camera's image processor.

I reacted when you first mentioned amplification because that's another common misunderstanding about ISO. Because the sensor signal amplification method is the most used there's been a tendency to ignore the other options and then make the extra leap to defining ISO as amplification which can lead to trouble -- shows up in a lot of those youtube videos that misinform when they try and explain ISO.

 

[Don Fischer]

So I'm shooting pretty much all digiytal and my light meter measure's ISO. How do you compare the digital equlavent to ISO?

 

[greybeard]

The 2nd image would be 2 stops over exposed. 1/500 and f/16 - 1/250 and f/22 would be the same exposure.

 

[Ysarex]

So I'm shooting pretty much all digiytal and my light meter measure's ISO. How do you compare the digital equlavent to ISO?

Just as with film previously you were advised to test your meter and film to determine accuracy, the same applies with digital. In digital ISO sets the brightness of the camera processed image (JPEG) relative to a measured exposure. We have a twist in operation with digital cameras as unfortunately our camera makers were unable to agree on a single standardized methodology with which to determine the output image brightness and so they agreed to disagree and wrote a requirement into the industry standard that they inform users as to which method they are using. In the EXIF data for your photos your camera will self identify whether ISO is REI (relative exposure index) or SOS (standard output sensitivity). The two tend to be close (about 1/2 stop) but they're not identical. SOS camera brands tend as a group to agree with each other more so than REI camera brands.

I have four different brand cameras and two are ISO/REI and two are ISO/SOS. If I were to use an external light meter I'd have to run tests to calibrate each camera with the meter.

 

[AlanKlein]

No need to assume -- the cause is well understood. There are two common types of noise that concern us in our photos: shot noise and read noise. Shot noise is by far and away dominant and typically what we see when we see a noisy photo. Read noise, from the camera's electronics, is less a problem and getting less all the time as the hardware improves. Shot noise is entirely a function of exposure -- SNR (signal to noise ratio) get's better with more exposure and worse with less exposure. What ISO is doing in the camera doesn't directly alter exposure and so doesn't cause shot noise. ISO implementation in the camera via amplification of the analog sensor signal suppresses read noise. What's that noise? Part one: Shedding some light on the sources of noise

Yes, ISO correlates with noise and as I noted that correlation when shooting camera JPEGs is pretty strong. But I did also note and provide an example shooting raw where that correlation is weak. The difference in the final result is in the example I gave of my student's misunderstanding -- a noisier photo because he thought ISO caused noise and so was reluctant to raise the ISO which would have in the final result made his photo less noisy. It's about understanding cause and effect. Thinking ISO causes noise is a mistake that can lead the user to getting a poorer final result. Mistaking correlation for cause and knowing the cause can matter.

Amplification is commonly used to implement ISO. The signal from the sensor is analog and it can be amplified before being digitized. Many cameras in whole or in part implement ISO using signal amplification. However there are ways to implement ISO other than signal amplification and they are also used. In some cases a camera maker may use more than one implementation at a time -- my Fuji cameras will do that.

What ISO does, as determined by the industry standard, is establish a methodology for determining the brightness in the camera output JPEG that results from a measured exposure of the camera sensor. As such ISO is a post-processing procedure in that it occurs after the exposure has been made. The industry standard does not specify implementation methods and the camera makers can do as they see fit. Amplifying the sensor signal is the most commonly used method however some cameras will use digital scaling in the ADC instead or in combination with amplification and some will simply apply the needed image brightening in the camera's image processor.

I reacted when you first mentioned amplification because that's another common misunderstanding about ISO. Because the sensor signal amplification method is the most used there's been a tendency to ignore the other options and then make the extra leap to defining ISO as amplification which can lead to trouble -- shows up in a lot of those youtube videos that misinform when they try and explain ISO.

You seem to be contradicting yourself and confusing me and your student. On the one hand you say that raising ISO doesn't have a direct effect on noise that it's because you reduce light to the sensor. Well, if you raise the ISO, you have to change the aperture and shutter settings to reduce light to the sensor. So, ipso facto, raising ISO causes noise. So your student is right.

You also now agree that changing ISO itself causes noise due to the amplification of the signal. Since this is the most used method, noise occurs most often in most cameras when you raise the ISO. It seems that the problem is compounded because there is a concurrent reduction in the amount of light on the sensor as well.