Home · Register · Join Upload & Sell

Moderated by: Fred Miranda
Username  

  New fredmiranda.com Mobile Site
  New Feature: SMS Notification alert
  New Feature: Buy & Sell Watchlist
  

FM Forums | Pro Digital Corner | Join Upload & Sell

1      
2
       end
  

Archive 2016 · Does anyone know about dynamic range and histograms?

  
 
dmcphoto
Offline
• • •
Upload & Sell: Off
p.2 #1 · p.2 #1 · Does anyone know about dynamic range and histograms?


To save some material and work you could possibly eliminate some of the middle steps. For instance, make the three brightest steps and then make four or five others that are 14, 13, 12, 11 and 10 stops darker than the brightest step. Since you're interested in total range the middle steps don't serve much purpose unless you were to test a camera with extremely limited range. You could always fill in the middle steps, or add more to the dark end if you have the need.

I'm not sure how you're going to get even illumination behind these. Each step has to be very evenly illuminated across its surface and each step must have the same illumination as all of the others. You could possibly use one light source for all of the steps or one source for each. In any case even illumination and avoidance of hot spots will be a challenge. That means the light source should be large relative to the area it illuminates. Something like a soft box could work but I'm not sure you'd get the sort of brightness you'll need from that.

Good luck!



Sep 18, 2016 at 06:27 AM
ahp-danegate
Offline

Upload & Sell: Off
p.2 #2 · p.2 #2 · Does anyone know about dynamic range and histograms?


Thanks dmcphoto, that's a useful suggestion to save some time.

To nearly eliminate hot spots I think a softbox and flash is a good idea.

I think multiple light sources would introduce further measurement uncertainty for colour.

I'll give it a go this week and see where it leads me :-)



Sep 18, 2016 at 06:36 AM
rw11
Offline
• • • •
Upload & Sell: Off
p.2 #3 · p.2 #3 · Does anyone know about dynamic range and histograms?


one trick for rough comparisons is to take photos that include an exterior window and a computer screen - as daylight fades, the DR will increase in the total scene


Sep 18, 2016 at 01:15 PM
dmacmillan
Offline
• • • • •
Upload & Sell: On
p.2 #4 · p.2 #4 · Does anyone know about dynamic range and histograms?


I'm trying to tie this to a practical professional use. It's nice to speculate, but how will this change the way you work?

BTW, in the early days of color motion pictures, they had to use a lighting ratio of about a half a stop. Going from camera negative to projection print really built up contrast.



Sep 20, 2016 at 04:57 PM
ahp-danegate
Offline

Upload & Sell: Off
p.2 #5 · p.2 #5 · Does anyone know about dynamic range and histograms?


Dmacmillan this isn’t really about professional use, rather about my original post which I can summarise as:
How come there are only 5 stops on the camera histogram? I’ve measured them with an external light meter and have proven the vertical bars represent exactly 1 stop (or 1 EV) each.

Thanks to all your interesting and useful suggestions, I’m going to knock up a simple way of seeing a little more what’s happening. Who knows where it will take me but it’s all good fun! Personally I learn best by practical methods and so I’m looking forward to getting on with this in the next couple of days.




Sep 20, 2016 at 05:25 PM
Peter Figen
Online
• • • • •
Upload & Sell: On
p.2 #6 · p.2 #6 · Does anyone know about dynamic range and histograms?


I see it's been about a week and a half since the last post here. I checked your profile and see that you're shooting Canon. Here's the deal. The Histogram is one thing and one thing only - it's a graph of the distribution of pixels from dark on the left to light on the right that that particular image contains, as defined by the Picture Style you have loaded at the time of exposure. Where the Histogram is higher, it only means there are more pixels of that value. There's no relation to f/stops. It's all just pixel value from 0 to 255. Nothing more. Your 5D3 has what, about 11-12 stops of dynamic range shooting raw when measuring from a 0,0,0 black to a 255,25,255 pure white. THAT'S what the histogram represents. So it's going to be a different range for every camera that has a different dynamic range, but still from zero to two-fifty-five.


Oct 03, 2016 at 01:15 AM
nolaguy
Offline
• • • •
Upload & Sell: Off
p.2 #7 · p.2 #7 · Does anyone know about dynamic range and histograms?


Peter Figen wrote:
There's no relation to f/stops. It's all just pixel value from 0 to 255. Nothing more. Your 5D3 has what, about 11-12 stops of dynamic range shooting raw when measuring from a 0,0,0 black to a 255,25,255 pure white. THAT'S what the histogram represents. So it's going to be a different range for every camera that has a different dynamic range, but still from zero to two-fifty-five.


That's always been my understanding, consequently it's been a little hard for me to follow this thread.

Histogram = dark/light relative to the camera's reach = baseline analysis: have I crushed the blacks or blown out the hightlights?

Dynamic range = how much breadth do I have to work with and consequently how much nuance there is across the spread of dark to light information?




Oct 03, 2016 at 04:25 AM
ahp-danegate
Offline

Upload & Sell: Off
p.2 #8 · p.2 #8 · Does anyone know about dynamic range and histograms?


Well interestingly when I looked at it, I found that whilst the middle vertical bar of the camera histogram represents exactly 1 stop or EV, as soon as you get to the far left bar it represents much less. Hence the darks are displayed in a very compressed way which is interesting and explains the one thing that started this all off and the reason why I posted the question in the first place.
Having tried it first on the Canon 5D3. I then tried it on the Pentax 645Z and found exactly the same thing.
Very interesting.
On further investigation is seems that it is due to the histogram being displaying in a non linear way as was pointed out earlier in the thread.
Personally I would prefer a genuine linear histogram where you see the displayed image as pixels and also the extended or potential, i.e. the lattitude.



Oct 03, 2016 at 12:38 PM
Michael White
Offline
• • • •
Upload & Sell: Off
p.2 #9 · p.2 #9 · Does anyone know about dynamic range and histograms?


I would shoot a subject at night so there's no ambient light then light the subject from the rear or opposite side of the camera the light in this test should be your pure white or brightest, then add one layer of you 1stop diffusion material take a shot this will be your 1stop photo next add a layer of the diffusion material shoot again this being the -2 stops photo at each step check the falloff with the light meter. Your actually shooting the light you just have the subject as a focal point. If each layer drops your light one stop as advertised and verified by the meter then when you bring the images into Lightroom you should see the histogram for each if desired you could cut and paste the same section from the original images into a composite and Tweek it so all fit on one slice and look at it


Oct 04, 2016 at 11:17 PM
ahp-danegate
Offline

Upload & Sell: Off
p.2 #10 · p.2 #10 · Does anyone know about dynamic range and histograms?


I like that idea Michael.
I tried a similar thing using the 1 stop diffusion material but hit a problem. When you place a second diffuser in front of one you don't get 2 stops of reduction. There must be a certain amount of air gap in between the diffusers to allow them to work properly, so things are a little more tricky.



Oct 05, 2016 at 04:17 AM
dmcphoto
Offline
• • •
Upload & Sell: Off
p.2 #11 · p.2 #11 · Does anyone know about dynamic range and histograms?


ahp-danegate wrote:
Personally I would prefer a genuine linear histogram where you see the displayed image as pixels and also the extended or potential, i.e. the lattitude.


The histogram is linear relative to the perceived brightness but not relative to f-stops.



Oct 05, 2016 at 12:18 PM
dmcphoto
Offline
• • •
Upload & Sell: Off
p.2 #12 · p.2 #12 · Does anyone know about dynamic range and histograms?


The bigger picture is:

RAW files are linear, meaning twice as much light (1 more f-stop) produces twice the pixel value. For instance, a midpoint value of 128 in an 8-bit RAW file is just one stop lower than the maximum value of 255. Note that current cameras generate 14-bit RAW files with a maximum value of more than 1E13 per color.

Everything changes when that data is converted into a particular color space like sRGB. Your camera converts RAW files into a color space to display the preview on its LCD and that preview is used to generate the histogram. Again using the 8-bit example for simplicity, the sRGB pixel value, which is what your histogram displays from the internally converted RAW file, is roughly:

sRGB value = 255^(1-1/Gamma) * RAW value^(1/Gamma)

Gamma depends on the color space and is typically 2.2 for sRGB

What I've shown here is an approximation. The full story can be seen at https://en.wikipedia.org/wiki/SRGB.



Oct 05, 2016 at 12:49 PM
henryp
Offline
• • • •
Upload & Sell: Off
p.2 #13 · p.2 #13 · Does anyone know about dynamic range and histograms?


How to Read Your Camera's Histogram

Henry Posner
[email protected]
B&H Photo-Video



Oct 10, 2016 at 02:21 PM
kdlanejr
Offline
• •
Upload & Sell: Off
p.2 #14 · p.2 #14 · Does anyone know about dynamic range and histograms?


Perhaps you would benefit from the following Sekonic videos.

http://www.sekonic.com/united-states/classroom/webinars/mastering-exposure-for-landscape-photography.aspx

and

http://www.sekonic.com/united-states/classroom/webinars/mastering-exposure-for-landscape-photography-part-ii.aspx

The first starts with calibrating your light meter to the camera. While you have not mentioned having the L-758DR, the calibration information presented and subsequent demonstration of practical use of a sekonic meter to maximize the quality of your image for the dynamic range of your cameras sensor while capturing a large dynamic range image should provide useful knowledge.

You've stated it's more about doing the experiment, however the experiment is only worth doing if you are going to benefit from the results.



Dec 11, 2016 at 11:53 AM
1      
2
       end




FM Forums | Pro Digital Corner | Join Upload & Sell

1      
2
       end
    
 

You are not logged in. Login or Register

Username       Or Reset password



This site is protected by reCAPTCHA and the Google Privacy Policy and Terms of Service apply.