October 21, 2014, 07:14:26 AM

Author Topic: Too much dynamic range?  (Read 13611 times)

pwp

  • 1D X
  • *******
  • Posts: 1605
    • View Profile
Re: Too much dynamic range?
« Reply #15 on: November 21, 2012, 10:57:55 PM »
Great to read all the highly technical responses. You guys know your stuff! Brilliant...

At a working mans level, I'd regard having too much DR as a non issue, unless of course it goes too far and delivers ridiculously flat images. But it's always going to a simpler matter to crunch down the DR to achieve suitable output vs struggling to increase DR in post-pro.

-PW

canon rumors FORUM

Re: Too much dynamic range?
« Reply #15 on: November 21, 2012, 10:57:55 PM »

MarkII

  • PowerShot G1 X II
  • ***
  • Posts: 65
    • View Profile
Re: Too much dynamic range?
« Reply #16 on: November 22, 2012, 01:37:08 AM »
To say that dynamic range and the number of gradations in tone are equivalent, or to say that either one of them can be determined simply from the ratio between the black noise level and the white level, is as stupid as defining people's adult heights merely by their weight at birth.
Since you always have the ability to control exposure to exactly catch the highlights in an image without clipping (by aperture, shutter-speed or ND filters), the dynamic range is effectively limited by the weakest signal that you can measure relative to the 'almost-but-not-quite-clipped' highlights.

Today, that weakest signal is effectively limited by the noise floor in the A/D conversion - a combination of quantisation noise (limited precision ADC) plus all the other noise gunk added by the electronics. Hence the resolution (number of effective bits after noise) of the sensor is its dynamic range.

Eventually, you will have sufficiently good sensors that the dynamic range will be ultimately limited by the quantum nature of light itself (photon shot noise). All of this is also affected by things like the Bayer colour filter array and things like micro-lens design (how much light actually reaches the sensor). In a 5D, I think that the tonal resolution (meaning colour discrimination) is more a property of the Bayer filters than the ADC.

There is a nice article about all this here: http://theory.uchicago.edu/~ejm/pix/20d/tests/noise/

All that really matters is that you can improve image quality by exposing-to-the-right, averaging frames, HDR stacking or simply downsizing images. And you only need to do any of this seriously if you have a non-theoretical problem with an image you are trying to take.

Nathaniel Weir

  • Power Shot G7X
  • **
  • Posts: 13
    • View Profile
    • Nathaniel Weir Photography
Re: Too much dynamic range?
« Reply #17 on: November 22, 2012, 02:03:40 AM »
Don't worry about it... just start taking pictures and stop blabbing on about sensor designs, when it has little impact on your photography.  As  the great Ken Rockwell states, "You need to learn to see and compose. The more time you waste worrying about your equipment the less time you'll have to put into creating great images. Worry about your images, not your equipment."   
And...
"Your equipment DOES NOT affect the quality of your image. The less time and effort you spend worrying about your equipment the more time and effort you can spend creating great images. The right equipment just makes it easier, faster or more convenient for you to get the results you need."
Canon EOS 5D Mark III w/ grip, Canon EOS 7D with grip, Canon 17-40mm, Canon 24-105mm, Canon 70-200mm F/2.8 L IS USM II, Canon 100mm Macro L IS, Canon 40mm Pancake, Canon 85mm F/1.8, 2x 600 EX-RT Flashes and St-E3-RT. Gitzo 2452L and Manfrotto 410 head.

MarkII

  • PowerShot G1 X II
  • ***
  • Posts: 65
    • View Profile
Re: Too much dynamic range?
« Reply #18 on: November 22, 2012, 03:16:03 AM »
Quite - none of this matters unless you are doing something comparatively unusual.

Most photographs that I have seen - and taken - need better composition, lighting and subject rather than a better sensor.

NormanBates

  • Canon 7D MK II
  • *****
  • Posts: 489
  • www.similaar.com
    • View Profile
    • www.similaar.com
Re: Too much dynamic range?
« Reply #19 on: November 22, 2012, 06:14:56 AM »
True, but the geek inside me still enjoys these theoretical discussions.

From my point of view, as long as your ADC has significantly more gradations than the DR of the camera (e.g. "16 bits" for "13 stops at pixel level"), this is a non-issue: you have an ADC that has enough gradations to actually capture the read-out noise of your image, so that is your limiting factor.

Say you have a sensor with full well capacity of 20.000e-, and read-out noise of 2e-. Your DR is 20*log10(10000)=80dB, or 13.33 stops. I guess the D800 sensor is pretty similar to that.
Tie that up with 16-bit ADC, and you have absolutely no "lack of gradation" issues whatsoever: you have to count electrons, the most you'll find are 20K, and you have 65K gradations at your disposal. Even with a 14-bit ADC, you wouldn't have terrible issues: 20Ke- to count (max), 16K gradations to use; the 2e- read-out noise is still your bigger problem.
« Last Edit: November 22, 2012, 06:27:03 AM by NormanBates »

NormanBates

  • Canon 7D MK II
  • *****
  • Posts: 489
  • www.similaar.com
    • View Profile
    • www.similaar.com
Re: Too much dynamic range?
« Reply #20 on: November 22, 2012, 06:52:27 AM »
In the old days, you did have a problem.

Consider a sensor with max well capacity of 13Ke-, and rea-out noise of 13e-. DR is 60dB, or 10 stops.
Pair that with 12-bit ADC. Should be enough, right? Well, yes and no. You have 4096 gradations and you have to count up to 13000 electrons, so 892 and 895 will be the same to you. No big deal, since read-out noise means you can't really distinguish between 892 and 905, but, if you can't reduce that read-out noise, there's a small benefit if you go for a 14-bit ADC: you're getting better information about the image, and you'll be in a better position to try to average out that noise. Small, I know, but it's an improvement. if the 892 comes from a very unlucky 891 and 905 from a very unlucky 906, you're in a better shape if you can say there's a 3e- difference between them (when the real-world difference is 5e-), than if all you can say is that they're the same to you.

OTOH, if you stick to 10-bit ADC, then you clearly have a problem: your ADC-stepping will be added to your read-out noise. 892 and 904 electrons are the same to your ADC, but that 904 can come from a very unlucky 910, and that 892 can come from a very unlucky 886, and if 886 and 910 can look the same to you then you're in bad shape.
« Last Edit: November 22, 2012, 06:55:53 AM by NormanBates »

NormanBates

  • Canon 7D MK II
  • *****
  • Posts: 489
  • www.similaar.com
    • View Profile
    • www.similaar.com
Re: Too much dynamic range?
« Reply #21 on: November 22, 2012, 06:53:04 AM »
(pun: "old days", or "Canon world", however you want to put it...)

canon rumors FORUM

Re: Too much dynamic range?
« Reply #21 on: November 22, 2012, 06:53:04 AM »

jukka

  • Guest
Re: Too much dynamic range?
« Reply #22 on: November 22, 2012, 07:04:00 AM »
True, but the geek inside me still enjoys these theoretical discussions.

From my point of view, as long as your ADC has significantly more gradations than the DR of the camera (e.g. "16 bits" for "13 stops at pixel level"), this is a non-issue: you have an ADC that has enough gradations to actually capture the read-out noise of your image, so that is your limiting factor.

Say you have a sensor with full well capacity of 20.000e-, and read-out noise of 2e-. Your DR is 20*log10(10000)=80dB, or 13.33 stops. I guess the D800 sensor is pretty similar to that.
Tie that up with 16-bit ADC, and you have absolutely no "lack of gradation" issues whatsoever: you have to count electrons, the most you'll find are 20K, and you have 65K gradations at your disposal. Even with a 14-bit ADC, you wouldn't have terrible issues: 20Ke- to count (max), 16K gradations to use; the 2e- read-out noise is still your bigger problem.

   D800   2.7e read noise      FWC   44972   = 14.0 stop



NormanBates

  • Canon 7D MK II
  • *****
  • Posts: 489
  • www.similaar.com
    • View Profile
    • www.similaar.com
Re: Too much dynamic range?
« Reply #23 on: November 22, 2012, 09:22:35 AM »
True, but the geek inside me still enjoys these theoretical discussions.

From my point of view, as long as your ADC has significantly more gradations than the DR of the camera (e.g. "16 bits" for "13 stops at pixel level"), this is a non-issue: you have an ADC that has enough gradations to actually capture the read-out noise of your image, so that is your limiting factor.

Say you have a sensor with full well capacity of 20.000e-, and read-out noise of 2e-. Your DR is 20*log10(10000)=80dB, or 13.33 stops. I guess the D800 sensor is pretty similar to that.
Tie that up with 16-bit ADC, and you have absolutely no "lack of gradation" issues whatsoever: you have to count electrons, the most you'll find are 20K, and you have 65K gradations at your disposal. Even with a 14-bit ADC, you wouldn't have terrible issues: 20Ke- to count (max), 16K gradations to use; the 2e- read-out noise is still your bigger problem.

   D800   2.7e read noise      FWC   44972   = 14.0 stop

Nice. And the ADC is 14-bit, right?

So it has 16K values to count up to 45K electrons, and read noise is close to 3 electrons. Not ideal (16-bit ADC would be better), but not bad at all.
What's sure is that you can't say "I wish it had higher read noise or lower FWC, so I could get a better use of my 14-bit DAC"

Radiating

  • Canon 6D
  • *****
  • Posts: 336
    • View Profile
Re: Too much dynamic range?
« Reply #24 on: November 22, 2012, 11:52:19 AM »
An interesting thought came to me before I went to bed. Below you'll find an assumption that came suddenly to my head, so please don't take it too seriously.

So... Let's assume there are two cameras with similar color tones reproduction abilities, but with different possible lightness level capturing ability. For example:
- sensor of camera A has 12 stops of DR, 16 billion tones it can distinguish
- sensor of camera B has 10 stops of DR, 16 billion tones it can distinguish

Having a flat scene (i.e. low DR scene) on a shot we'll push an image with, say, 8 DR to be captured with both sensors. And then both images will be edited in post to retrieve lacking contrast. So we need to add:
- 4 stops for 12-stop camera
- 2 stops for 10-stop camera

So my point is: with lower DR camera we'll have lower tone delta (difference of the initial color tone in the scene with reproduced tone by the sensor) when processing the low DR shot made using lower DR sensor. That happens because of decreased amount of modifications made to the file to achieve required result.

What do you guys think about that?

Yeah it doesn't work like that at all, whatsoever. The range of dynamic range is not determined by the camera, but by the data format.

both Canon CR2 and Nikon NEF files have 14 bit depth, or 14 stops.

When you measure a CAMERA'S dynamic range that has nothing to do with how much data it can record from maximum through minimum, that is going to be 14 stops either way. It has to do with taking those 14 stops you start with and subtracting the NOISE floor. So you take your original 14 stops and subtract how many stops are going to be noise, such as say 4.5 and you get a 9.5 stop camera.

Having more dynamic range is never bad because it means there is less noise from the get go. The tone delta is always identical.

NormanBates

  • Canon 7D MK II
  • *****
  • Posts: 489
  • www.similaar.com
    • View Profile
    • www.similaar.com
Re: Too much dynamic range?
« Reply #25 on: November 22, 2012, 04:21:46 PM »
An interesting thought came to me before I went to bed. Below you'll find an assumption that came suddenly to my head, so please don't take it too seriously.

So... Let's assume there are two cameras with similar color tones reproduction abilities, but with different possible lightness level capturing ability. For example:
- sensor of camera A has 12 stops of DR, 16 billion tones it can distinguish
- sensor of camera B has 10 stops of DR, 16 billion tones it can distinguish

Having a flat scene (i.e. low DR scene) on a shot we'll push an image with, say, 8 DR to be captured with both sensors. And then both images will be edited in post to retrieve lacking contrast. So we need to add:
- 4 stops for 12-stop camera
- 2 stops for 10-stop camera

So my point is: with lower DR camera we'll have lower tone delta (difference of the initial color tone in the scene with reproduced tone by the sensor) when processing the low DR shot made using lower DR sensor. That happens because of decreased amount of modifications made to the file to achieve required result.

What do you guys think about that?

Yeah it doesn't work like that at all, whatsoever. The range of dynamic range is not determined by the camera, but by the data format.

both Canon CR2 and Nikon NEF files have 14 bit depth, or 14 stops.

When you measure a CAMERA'S dynamic range that has nothing to do with how much data it can record from maximum through minimum, that is going to be 14 stops either way. It has to do with taking those 14 stops you start with and subtracting the NOISE floor. So you take your original 14 stops and subtract how many stops are going to be noise, such as say 4.5 and you get a 9.5 stop camera.

Having more dynamic range is never bad because it means there is less noise from the get go. The tone delta is always identical.

dynamic range is not how many shades you have on your color space
it is related to real-world things: how much brighter can one object be than another, while the camera still captures them both correctly at the same time

nightbreath

  • Canon 7D MK II
  • *****
  • Posts: 456
    • View Profile
    • Свадебный фотограф в Днепропетровске
Re: Too much dynamic range?
« Reply #26 on: November 22, 2012, 04:46:22 PM »
dynamic range is not how many shades you have on your color space
it is related to real-world things: how much brighter can one object be than another, while the camera still captures them both correctly at the same time
And that's why I ask  :). Why would I need DR for portraits? It is what I mainly do with my cameras as wedding photographer, so number of shades sensor produces is more important than DR  ;)
Wedding photography. My personal website: http://luxuryphoto.com.ua

bdunbar79

  • Canon EF 300mm f/2.8L IS II
  • *******
  • Posts: 2600
    • View Profile
Re: Too much dynamic range?
« Reply #27 on: November 22, 2012, 05:26:57 PM »
Don't worry about it... just start taking pictures and stop blabbing on about sensor designs, when it has little impact on your photography.  As  the great Ken Rockwell states, "You need to learn to see and compose. The more time you waste worrying about your equipment the less time you'll have to put into creating great images. Worry about your images, not your equipment."   
And...
"Your equipment DOES NOT affect the quality of your image. The less time and effort you spend worrying about your equipment the more time and effort you can spend creating great images. The right equipment just makes it easier, faster or more convenient for you to get the results you need."

You have GOT to be kidding me!
2 x 1DX
Big Ten, GLIAC, NCAC

canon rumors FORUM

Re: Too much dynamic range?
« Reply #27 on: November 22, 2012, 05:26:57 PM »

NormanBates

  • Canon 7D MK II
  • *****
  • Posts: 489
  • www.similaar.com
    • View Profile
    • www.similaar.com
Re: Too much dynamic range?
« Reply #28 on: November 23, 2012, 03:35:41 AM »
dynamic range is not how many shades you have on your color space
it is related to real-world things: how much brighter can one object be than another, while the camera still captures them both correctly at the same time
And that's why I ask  :). Why would I need DR for portraits? It is what I mainly do with my cameras as wedding photographer, so number of shades sensor produces is more important than DR  ;)

There are some usage models for which DR is important, there are many usage models for which it doesn't matter at all. If the scene you have in front of you doesn't require more than 8 stops of DR, you're fine with a camera that can capture that, no point in going for one that is the same in every respect but will record 14 stops of DR.

If your portraits are in a studio, with a standard backdrop, your DR needs will probably be pretty modest. If your portraits happen in other less-controlled locations, you may have very high DR needs (e.g. if you want to take a portrait of someone in their bedroom, and there's a window in an interesting area). Wedding photographers take lots of portraits, and, not having a lot of control over their shooting scenarios, they usually need a lot of DR (for this reason, a friend of mine was still using his Fuji S3 pro as his backup body up until the D800 came out: a 12 mpix camera from 2005... with 13.5 stops of DR as measured by dxomark).

In any case, ADC precision will only be a problem if the manufacturer screws up the sensor-ADC matching. No current camera has that issue AFAIK.

Radiating

  • Canon 6D
  • *****
  • Posts: 336
    • View Profile
Re: Too much dynamic range?
« Reply #29 on: November 23, 2012, 04:13:41 AM »
An interesting thought came to me before I went to bed. Below you'll find an assumption that came suddenly to my head, so please don't take it too seriously.

So... Let's assume there are two cameras with similar color tones reproduction abilities, but with different possible lightness level capturing ability. For example:
- sensor of camera A has 12 stops of DR, 16 billion tones it can distinguish
- sensor of camera B has 10 stops of DR, 16 billion tones it can distinguish

Having a flat scene (i.e. low DR scene) on a shot we'll push an image with, say, 8 DR to be captured with both sensors. And then both images will be edited in post to retrieve lacking contrast. So we need to add:
- 4 stops for 12-stop camera
- 2 stops for 10-stop camera

So my point is: with lower DR camera we'll have lower tone delta (difference of the initial color tone in the scene with reproduced tone by the sensor) when processing the low DR shot made using lower DR sensor. That happens because of decreased amount of modifications made to the file to achieve required result.

What do you guys think about that?

Yeah it doesn't work like that at all, whatsoever. The range of dynamic range is not determined by the camera, but by the data format.

both Canon CR2 and Nikon NEF files have 14 bit depth, or 14 stops.

When you measure a CAMERA'S dynamic range that has nothing to do with how much data it can record from maximum through minimum, that is going to be 14 stops either way. It has to do with taking those 14 stops you start with and subtracting the NOISE floor. So you take your original 14 stops and subtract how many stops are going to be noise, such as say 4.5 and you get a 9.5 stop camera.

Having more dynamic range is never bad because it means there is less noise from the get go. The tone delta is always identical.

dynamic range is not how many shades you have on your color space
it is related to real-world things: how much brighter can one object be than another, while the camera still captures them both correctly at the same time

Face palm. No. No. No.

Raw images are captured in bits by intensity at the photo site. The simplest version would be a 1 bit photo site that either registers full of photons or empty.

So with a simple 2 bit system we can have:

00 = 0-100 photons in a pixel
01 = 100-200 photons in a pixel
10 = 200-400 photons in a pixel
11 = 200-infinity photons in a pixel

Then for different ISO settungs we multiply or divide the photons to produce different exposures.

This gives us 2 stops of dynamic range from 100 photons to 400 (or multiples of that). A stop is a doubling of light so 2x2=4.

This is how cameras work. A cameras dynamic range rating is essentially the theoretical dynamic range minus how ever many stops in the shadows are unreadable information. So in our 2 stop example if photons from 0-200 ISO were too noisy to determine what is supposed to be there then our theoretical camera has 1 stop of DR. You can think of noise as a random number generator that's added to the photon count. So out count of 0-400+ would have a number from 0-100 randomly added or subtracted from it. This is the noise you see when you put fill light to max. Anyways if a ranom number from 0-100 is added or subtracted it is mathematically impossible to determine how many photons were in our pixel in the 1st stop. Literally all you'd see is something resembling TV static if you tried to make a picture from it.

So cameras with more dynamic range have the static come in at a lower stop.

<---- is an engineer.

« Last Edit: November 23, 2012, 04:16:49 AM by Radiating »

canon rumors FORUM

Re: Too much dynamic range?
« Reply #29 on: November 23, 2012, 04:13:41 AM »