Photosites have capacitors...OK

[Abitconfused]

Color me dense but I am trying to simply this for my small brain. I guess turning up the “gain” is the best analogy but I am open to a more creative comparison to everyday events should one appear. Also, turning up the gain doesn’t explain why as ISO of 400 doesn’t blow every midtone past highlights when used in a scene with adequate lighting.

[DanK]

The monitor emits light and that's analogue, just as the sun or your lamps in the kitchen. What you're referring to is the communication to the pc. That's why I wrote
When send to the monitor that signal has to be converted to an analogue signal.

Is that actually the case--that is, that there is a DAC between the computer and the monitor, or internal to the monitor? My computer uses a digital connection, so there can't be one in the computer. I know little about lcd displays, but from what I have read, their output is a step function, which is what one would expect from a digital device. Someone correct me if I am wrong.

Of course, if you add together many step functions, the result, while still a discrete scale, has so many values that it can be treated as continuous for many practical purposes.

[Manfred M]

The monitor emits light and that's analogue, just as the sun or your lamps in the kitchen. What you're referring to is the communication to the pc. That's why I wrote

You have a very unconventional view of what analogue versus digital is. Traditionally it refers to how data is handled; a series of data give the values of 0 or 1 is digital and a continuously variable signal is analogue. The output source has nothing to do with it per se.

When I look at my computer equipment, my graphics card uses DisplayPort output and nothing else. This is purely digital data.

My computer screen has two types of input; DsplayPort and HDMI. Both are purely digital data. There is no way that I can transmit analogue data to my computer screen as my graphics card cannot deliver an analogue signal and my computer screen has no way of receiving one.

I can set it to operate in three modes; B&W, 8-bit colour (sRGB - 256 colours per channel) and 10-bit colour (Adobe RGB - 1024 colours per channel). My display emits photons that have different wavelengths associated with them. I could just as well argue that the display is digital because their either is a photon emitted or no photon emitted. That being said, this has nothing to do with a device being digital or analogue.

[george013]

Is that actually the case--that is, that there is a DAC between the computer and the monitor, or internal to the monitor? My computer uses a digital connection, so there can't be one in the computer. I know little about lcd displays, but from what I have read, their output is a step function, which is what one would expect from a digital device. Someone correct me if I am wrong.

Of course, if you add together many step functions, the result, while still a discrete scale, has so many values that it can be treated as continuous for many practical purposes.

If the monitor is analogue or digital is completely irrelevant for this discussion. When a pixel has a value 255,255,255 in 8 bit, then that is supposed to be white, the maximum white of the monitor, just to stick with monitors. And if that maximum of that monitor is gray, well than it's gray.

Digital is always in steps. You should know from the discussions of editing in 8 or 16 bit, 16 bits are smaller steps.
Even an old monitor with tubes is in steps when digital managed.

George

[xpatUSA]

Also, turning up the gain doesn’t explain why an ISO of 400 doesn’t blow every mid-tone past highlights when used in a scene with adequate lighting.

Hello Ed,

Let's say that you are indeed "always in auto".

Let's keep it simple and have you shoot a gray scene such as a fully-framed Kodak R27 8x10" gray card, starting at 100 ISO. Let's say it's noon and the scene lighting is 13EV. We'll use an aperture priority of f/4 for the exercise.

At 100 ISO, your camera metering will automatically select an exposure of 13EV which at f/4 is 1/500 sec.
At 200 ISO, your camera metering will automatically select an exposure of 14EV which at f/4 is 1/1000 sec.
At 400 ISO, your camera metering will automatically select an exposure of 15EV which at f/4 is 1/2000 sec.

(go to the link, find the table and check out the column for f/4).

The bigger the camera EV, the lower the sensor exposure. The bigger the scene EV, the brighter the lighting.

Classic photography that, i.e. "EV" means two different things - it only makes sense after a while. Some authorities use "LV" for scene lighting.

At 100 ISO, let's say your sensor was perfectly exposed. At 400 ISO, your sensor will be only 1/4 exposed - and that is why mid-tones don't get "blown past highlights". To compensate for that under-exposure, the image is brightened by just four times (or two times for 200 ISO) either before the ADC or after, it makes no difference to the result. On your camera LCD, those three images will look the same (mid-gray more or less), as they will look the same also on your monitor screen.

[Abitconfused]

Right...of course. It is what it is.

[xpatUSA]

Right...of course. It is what it is.

"Right...of course. It is what it is."

Judging by the sarc, I take it that you're still a bit confused ...

[Abitconfused]

Sorry, no intent at all to be sarcastic. I simply submit to the silicone realities as they are. Yours was a great summary. But, yes, I have some doubts and I could have phrased my question better. Of course I accept that the “classic” exposure triangle applies when ISO is changed. But here is my issue. Wouldn’t it make sense for the sensor simply to stay “on” longer than turn up signal strength after the fact? And if I have enough light for an acceptable exposure but want more speed, I may increase ISO to 400 yet I see virtually no increase in grain given a D810. Granted, the sensor records 0.75 of its signal from midtones to white point but increasing ISO works so well in these tonal areas it puzzles me. I feel there must be more to it or less to it. Thanks much for your input!

[george013]

Sorry, no intent at all to be sarcastic. I simply submit to the silicone realities as they are. Yours was a great summary. But, yes, I have some doubts and I could have phrased my question better. Of course I accept that the “classic” exposure triangle applies when ISO is changed. But here is my issue. Wouldn’t it make sense for the sensor simply to stay “on” longer than turn up signal strength after the fact? And if I have enough light for an acceptable exposure but want more speed, I may increase ISO to 400 yet I see virtually no increase in grain given a D810. Granted, the sensor records 0.75 of its signal from midtones to white point but increasing ISO works so well in these tonal areas it puzzles me. I feel there must be more to it or less to it. Thanks much for your input!

You want to know why the quality of the D810 is so good?

George

[Abitconfused]

Why?

[george013]

Why?

I may increase ISO to 400 yet I see virtually no increase in grain given a D810.

Try the D80. You won't be happy at 400 iso.

George

[Manfred M]

The D80 came out in 2006 and the D810 in 2014. By my count there are at least 3 generations of technology between them as well as the D80 being a crop factor consumer camera and the D810 being a full-frame pro camera. Each generation resulted in improved technology.

No wonder that one sees a difference. Now the D850 has come out and it definitely out performs the D810.

[xpatUSA]

Sorry, no intent at all to be sarcastic. I simply submit to the silicone realities as they are. Yours was a great summary. But, yes, I have some doubts and I could have phrased my question better. Of course I accept that the “classic” exposure triangle applies when ISO is changed. But here is my issue. Wouldn’t it make sense for the sensor simply to stay “on” longer than turn up signal strength after the fact? And if I have enough light for an acceptable exposure but want more speed, I may increase ISO to 400 yet I see virtually no increase in grain given a D810. Granted, the sensor records 0.75 of its signal from midtones to white point but increasing ISO works so well in these tonal areas it puzzles me. I feel there must be more to it or less to it. Thanks much for your input!

Thank you, Ed, and no offense taken.

I reject the Exposure Triangle and live in less dumbed-down world, where I think only of sensor exposure as if it were "the film". My ISO is always set to 100 and the sensor exposure is set by the combination of aperture and shutter speed. I set the aperture for the desired DOF and if the resulting shutter "speed" is not high enough, I don't take the shot.

So, my answer to your question "Wouldn’t it make [more] sense for the sensor simply to stay “on” longer than turn up signal strength after the fact?" is YES, it would make more sense to anybody like ourselves. But "most people" here don't think that way, seeking ever high ISO settings so that speeding animals or objects can be frozen in motion with mega-fast shutter speeds.

As to what your D810 does when you deliberately under-expose the sensor by increasing the ISO, I have no comment.

[george013]

The D80 came out in 2006 and the D810 in 2014. By my count there are at least 3 generations of technology between them as well as the D80 being a crop factor consumer camera and the D810 being a full-frame pro camera. Each generation resulted in improved technology.

No wonder that one sees a difference. Now the D850 has come out and it definitely out performs the D810.

I don't know anymore what the question is.
Is the "why" of Ed asking why I formulated his question or is it a why to explain the differences?

George

[xpatUSA]

I don't know anymore what the question is.
Is the "why" of Ed asking why I formulated his question or is it a why to explain the differences?

George

You asked Ed:

[do] You want to know why the quality of the D810 is so good?

By responding with "why?", he is asking you to tell him why the quality of the D810 is so good.

[george013]

The answer should be then "Yes. Why".

George

[xpatUSA]

The answer should be then "Yes. Why".

Thank you for the English grammar lesson.

Will you actually be telling Ed why the quality of the D810 is so good? . . . . . . . . . .

[george013]

Thank you for the English grammar lesson.

Will you actually be telling Ed why the quality of the D810 is so good? . . . . . . . . . .

I doubt if this is a matter of grammar.

And I don't have the answer.

George

[Abitconfused]

My thinking is that with a low light photograph we may set our shutter speed to 1/15 second but we have no idea how long the sensor actually stays active. It could be any time up to 1/15s. So an engineer may have greater latitude than we might imagine with regard to sensors. Even at 1/8000s we may believe the sensor is active throughout the entire time span but it may be considerably less. Look at movies taken at 60 FPS using digital cameras. That sensor must be far more capable than we realize. So an ISO of any amounts is a more imaginary concept than we think. As to the D810, I am more interested in the D850 not because I need it bet because I love technology. Or as CPO said in the original Star Wars, “Curse my metal body. I wasn’t fast enough.”

[Manfred M]

My thinking is that with a low light photograph we may set our shutter speed to 1/15 second but we have no idea how long the sensor actually stays active. It could be any time up to 1/15s

Actually we do know; the sensor will be active for the length of time that it is exposed to light and that will always be at least the sync speed of the camera, as that is roughly the minimum time that the shutter of the camera is actuated. Once one gets up to anything slower than the sync speed, the exposure time will define how long the sensor is actuated. In your 1/15s example, it will be just over 1/15th of a second.

There is one tiny exception to this, sort of, and that depends on if you have the long exposure noise reduction functionality activated. Here the camera takes the picture and once that is done, the sensor is activated without the shutter opening for the same same length of time as for the original shot. This is usually referred to as a "dark frame exposure" and any uniform digital noise can be recorded and subtracted from the image, so effectively the sensor will be on roughly twice as long as the exposure setting.