ISO discussion, digital SLR compared to film SLR

[Designer]

I think this video is correct. Tony does get off into the weeds toward the second half of the video, but as I understand it, he is correct in his concepts and terminology.

 

[Ysarex]

if i am understanding ysarex right, he is asserting that increasing exposure (eg extending shutter speed) will get more shadow detail whereas increasing the ISO doesnt. this is probably true for newer cameras with ISO invariance.

It is. It is also true for older digital cameras that are not ISO invariant. Exposure determines the data you will record. ISO brightening is a post exposure process that can't create more data out of nothing. In a non ISO invariant camera ISO brightening of the sensor signal will advantageously suppress noise.

Joe

testing my camera which is about 7 years old ISO works pretty much like film. taking identical images, one with a longer shutter speed at ISO 100 and the other with a shorter shutter speed at ISO 1600 reveals zero difference in shadow details. diffing the images shows differences on the edges and in the midtones. likely due to increased noise in the ISO 1600 shot and some noise reduction code.

in other words my camera was engineered so that the ISO setting worked much like film ISO. the sensor is by modern standards pretty noisy, so the readout at ISO 100 leaves a lot low level signal behind (below the LSB) in favor of getting a clean signal. upping the ISO does indeed recover more shadow detail at the expense of reading out additional noise. then the software does god knows what, and in the end, it's a wash and ISO works a lot like film's ISO.

point is you can engineer these things to work any way you like. modern systems favor ISO invariance which, as a consequence, means that upping the ISO will treat shadow detail differently from opening the aperture or increasing the shutter speed. which, if you want, I suppose you can define "sensitivity" to mean "not the thing ISO does" and then you get to grump at videos.

if you care about shadow detail, or color fidelity, or highlight detail, or haptics, or how easy it is to clean the sensor, or any of those things it will behoove you to learn about your camera a bit. if you don't care about one thing or another, just let it slide.

ISO make pichers brighter.

if that's not enough detail for you do figure out what ISO does on your camera until you know enough to satisfy you. every camera is gonna be a lil differet.

 

[Bill The Lurker]

It is. It is also true for older digital cameras that are not ISO invariant. Exposure determines the data you will record. ISO brightening is a post exposure process that can't create more data out of nothing.

it may not "create more data" but it's certainly finding more data in the well and putting it into my file. it's just nicely blended with some noise. you can assert all you like that it's not true and yet here i am looking at these image files.

do you really think those blokes put the analog amplifiers in there because they're dumb? or because they just love amps?

what, exactly, is a non-ISO-invariant sensor, if the analog ISO circuitry "can't create more data" in some user-visible way?

 

[Bill The Lurker]

Just remember that noise is "additive" in that it always results in a photosite reporting a HIGHER numeric value than it should have reported. There is no "anti-noise" where the photosite reported a LOWER numeric value (at least if there is, I have yet to encounter it.)

huh? no

take a picture of a white piece of paper at high ISO, zoom in on it, and observe in astonishment.

discussing noise is extremely difficult and not one i would dare approach in a page or two. there are many sources of noise each with their own properties and each with their own appropriate models for analysis and so on. sadly it's also critical to understanding in any real
technical detail any of the issues here, which is why all the efforts to nupack what ISO settigs actually do are so hilarious and yet dull

 

[Ysarex]

It is. It is also true for older digital cameras that are not ISO invariant. Exposure determines the data you will record. ISO brightening is a post exposure process that can't create more data out of nothing.

it may not "create more data" but it's certainly finding more data in the well and putting it into my file. it's just nicely blended with some noise. you can assert all you like that it's not true and yet here i am looking at these image files.

I have not and would not assert that is not true. You're getting extremely pedantic here but, yes, the gain to the sensor signal improves the system efficiency and saves recorded data that would otherwise be lost. This is happening post exposure however and does not constitute a sensitivity increase. Rather it's an efficiency improvement that prevents data loss in the system. You're just doing a better job extracting what's already there.

Joe

do you really think those blokes put the analog amplifiers in there because they're dumb? or because they just love amps?

what, exactly, is a non-ISO-invariant sensor, if the analog ISO circuitry "can't create more data" in some user-visible way?

 

[Ysarex]

far be it from me to judge how a persons thought process leads them to the asnwer and if yours takes a trip through details of sensor electrocics then so be it, more power to ya

I'm a photographer. My thought process in this regard has always been anchored in understanding my hardware so that I can take better photographs.

Joe

me, i just expose for whatever highlights i want to keep, s ame as with slide film, same as any digital camera

 

[Bill The Lurker]

so let's see if i have this right.

my camera, a typical mainstream non-ISO-invarient camera of a few years ago. changing the ISO doesn't change the recorded data,
that is, to be precise, the data recorded in the wells of the sensor where it will reside for a few milliseconds before evaporating, the data which I do not care about in the slightest. that data. the data i can't even access.

changing the ISO does change the data that is recorded in the file, though. the data that lasts for sometimes thousands or even millions of milliseconds. the data i can actually access. the data i care about. the picture i just took. that data, changing the ISO does give me more data there.

so you might almost say that the camera becomes more sensitive as a recording instrument, in terms of the bits it actually puts where i can access them, when i turn up the ISO on it.

now, you're certainly right, if you draw a red line on the chip that includes these transistors but not those transistors, and declare that set of analog circuitry to be "the sensor", that system doesn't record any more data. in fact, setting the iso doesn't even effect that circuitry at all. so, yeah, the iso doesn't do a single darn thing to the parts of the circuit its not connected to. you got me there.

but the camera considered as a whole system, w ell, the iso setting makes it, you imght almost say, "more sensitive"

 

[Ysarex]

But back to that "upstream" amplification... this results in a gain being applied BEFORE digital conversion (before ADC) and as a result it doesn't lose much in the way of dynamic range.

If you were to test the dynamic range of a camera (using a test target) what you'd probably notice is that as you boost ISO, you don't seem to be losing much dynamic range... but there's a limit to this... and then suddenly you hit an ISO where you get a linear drop off in dynamic range for each boost in ISO beyond that point.

What you're describing here is peculiar to Canon engineered cameras. This is a Canon only behavior that does not apply to the rest of the photo world. You need to preface this with that disclaimer. The basic rule is DR drops linearly with increasing ISO. Here's a visual on that looking at the Canon 5D III and Nikon D810: Photographic Dynamic Range versus ISO Setting

Joe

 

[Bill The Lurker]

lest I lose the thread

every bit of technical twaddle I have trotted out here was merely to demonstrate that the technical fiddly bits go on forever

nobody in this thread actually understands how these things work every one of us is saying stuff that would make the sensor team at sony take to drink, or at least sigh loudly. i understand these issues better than the other people because I have a moderately relevant background, and am therefore in a position to illustrate how to nitpick endlessly

what you need to actually take pictuers is a mental model that is good enough for your work. it's going to be simplified and wrong, unless you were actually on the team that designed the sensor system you're using. but if you're lucky it's going to be good enough.

fussing about whether 'more sensitive but noisier' is strictly speaking true is silly. you can "correct" that but you'll still be wrong. ysarex has, in my opinion, failed to illustrate how that particular wrong mental model is going to produce worse pictures than his mental model, w hich is also wrong in some details here and there.

 

[Ysarex]

so let's see if i have this right.

my camera, a typical mainstream non-ISO-invarient camera of a few years ago. changing the ISO doesn't change the recorded data,
that is, to be precise, the data recorded in the wells of the sensor where it will reside for a few milliseconds before evaporating, the data which I do not care about in the slightest. that data. the data i can't even access.

changing the ISO does change the data that is recorded in the file, though. the data that lasts for sometimes thousands or even millions of milliseconds. the data i can actually access. the data i care about. the picture i just took. that data, changing the ISO does give me more data there.

so you might almost say that the camera becomes more sensitive as a recording instrument, in terms of the bits it actually puts where i can access them, when i turn up the ISO on it.

You might say that but you'd be wrong. What you're describing instead is an improvement in system efficiency that's preventing slight data loss. Raising ISO on a digital camera does not record more data from the original scene -- sensor light sensitivity does not increase.

Joe

now, you're certainly right, if you draw a red line on the chip that includes these transistors but not those transistors, and declare that set of analog circuitry to be "the sensor", that system doesn't record any more data. in fact, setting the iso doesn't even effect that circuitry at all. so, yeah, the iso doesn't do a single darn thing to the parts of the circuit its not connected to. you got me there.

but the camera considered as a whole system, w ell, the iso setting makes it, you imght almost say, "more sensitive"

 

[Ysarex]

lest I lose the thread

every bit of technical twaddle I have trotted out here was merely to demonstrate that the technical fiddly bits go on forever

nobody in this thread actually understands how these things work every one of us is saying stuff that would make the sensor team at sony take to drink, or at least sigh loudly. i understand these issues better than the other people because I have a moderately relevant background, and am therefore in a position to illustrate how to nitpick endlessly

what you need to actually take pictuers is a mental model that is good enough for your work. it's going to be simplified and wrong, unless you were actually on the team that designed the sensor system you're using. but if you're lucky it's going to be good enough.

fussing about whether 'more sensitive but noisier' is strictly speaking true is silly. you can "correct" that but you'll still be wrong. ysarex has, in my opinion, failed to illustrate how that particular wrong mental model is going to produce worse pictures than his mental model, w hich is also wrong in some details here and there.

Well, you know I answered what and when I did because the OP asked:

How does a "fixed" sensor in a digital camera become more or less sensitive to light by just selecting a different ISO? I'd like to know what it's doing so I know how to use it (and use it to my advantage when I can).

Good question! How does it do that? And then he asked specifically about ISO invariant cameras.

I gave him accurate information pertinent to his questions. You decided that "pedantic fiddly bits" weren't appropriate for him or helpful and he should just stick to taking pichers using a false mental model that works for millions. I'll trust he can sort those options out.

Joe

 

[Designer]

I think the OP gave up. He probably went to some other photo discussion site to ask his question.

 

[waday]

 

[Bill The Lurker]

You decided that "pedantic fiddly bits" weren't appropriate for him or helpful

no, i actually decided that they're wrong and not helpful for anyone. saying that changing the iso doesn't alter the sensor's sensitivity has got to be one of the single most confusing bits of twaddle trotted out by "photography experts" today. it's only true at all if you squint and qualify the words very very carefully, and it presents absolutely forward path for useful action.

in fact, it strongly suggests to the unannointed that adjusting the iso serves no purpose, and yet, even the greenest tyro can see that changing the iso most certainly does do something. how can they not be confused at this pont? but the tyro certainly knows who the cleverest fellow in the rom is, a nd taht's the importantpoint, isn't it?

 

[Braineack]

EDIT: Scott cleared it up for me- I was confused how colors were detected and identified if not by the sensor and I'm now told that they bayer filter does this. Awesome!

yeah, bayer array. the video gives you a good general idea better than most try to explain. the specifics, as shown by this thread, are boring.