Ugly Hedgehog - Photography Forum
Home Active Topics Newest Pictures Search Login Register
Main Photography Discussion
Is ETTR/EBTR a Concept Whose Time has Passed?
Page <<first <prev 4 of 4
Dec 24, 2016 18:22:36   #
imagemeister Loc: mid east Florida
 
selmslie wrote:
Yes, modern digital cameras now exceed the DR of film at low ISO and reach much higher ISO settings.

However, the best digital images are still produced at low ISO, in the 100-800 range.

Film still has the edge at extremely low ISO, 4-50. It's still more practical for medium and large format. And it's still more fun.



Reply
Dec 24, 2016 18:36:15   #
BebuLamar
 
Selmie I blame you for this. You woke it up and not it's 46 posts.

Reply
Dec 25, 2016 06:52:30   #
selmslie Loc: Fernandina Beach, FL, USA
 
Uuglypher wrote:
Scotty claims:
The difference between 1/25 sec and 1/4 sec is 2-2/3 stops, 6.25x as many photons falling on the sensor. That's the reason for the slight reduction in noise.

Again, Selmslie tries to obfuscate with bad math...no wonder he has problems with photographic calculations:
1/25sec ...1 stop > 1/12 sec...one stop > 1/6 sec ...one stop > 1/3 sec ... 1/3 stop >1/4 sec.

That is, in fact, 3 and 1/3 stops

He'll have to think on it a bit more; but, with effort, it should come to him!

I can't believe no one else has corrected him !
Scotty claims: br The difference between 1/25 sec ... (show quote)

Nobody "corrected" me because it's your math that's wrong.

1/25 +1 stop= 1/12, +1 stop=1/6, +1 stop=1/3, -1/3 stop stop=1/4. That's 1+1+1 minus 1/3 = plus 2 2/3 stops

or 1/4 -1 stop=1/8, -1 stop=1/15, -1 stop =1/30 plus 1/3 stop=1/25 That's -1-1-1+1/3= minus 2 2/3 stops

So why did nobody else correct you? I won't even hazard a guess.

Embarrassed?

Reply
 
 
Dec 25, 2016 11:04:51   #
selmslie Loc: Fernandina Beach, FL, USA
 
selmslie wrote:
Yes, modern digital cameras now exceed the DR of film at low ISO and reach much higher ISO settings.

However, the best digital images are still produced at low ISO, in the 100-800 range.

Film still has the edge at extremely low ISO, 4-50. It's still more practical for medium and large format. And it's still more fun.

Se also a fresh example at http://www.uglyhedgehog.com/t-431124-2.html#7250078 that shows a low DR scene captured at a constant exposure while varying the ISO from 100 to 6400 and adjusting in post processing.

Reply
Dec 27, 2016 23:12:07   #
rmalarz Loc: Tempe, Arizona
 
In a simple answer, No, it's time has not passed and it's a very useful practice in obtaining optimum exposure.
--Bob

selmslie wrote:
There have been numerous posts promoting the use of ETTR/EBTR (expose to the right, expose beyond the right) based on a ten year old concept developed by Michael Reichmann of the Luminous Landscape. The idea originated when Reichmann observed that his existing camera was not taking full advantage of its dynamic range (DR) thereby missing an opportunity to suppress noise and use the full DR of the sensor in capturing images.

A lot of progress made in the past ten years has widened the DR of modern sensors and found ways to suppress noise that were not part of the cameras that Reichmann originally used. By the time Reichmann passed away earlier this year he had stopped promoting the concept of ETTR/EBTR because modern sensors have pretty much eliminated the need for it.

However, there are still die-hard users of the technique who promote its use, in the absence of any remaining logical need for the convoluted process.

Today's cameras support DRs of over 14 Ev at base ISO, far exceeding the DR in the vast majority of scenes encountered by the photographer. In those rare cases where the scene's DR is too wide for the sensor, HDR can be used to combine multiple exposures.

The signal to noise ratio (S/N) is also highest at base ISO. For situations that call for the use of a higher ISO (lower S/N) there are many solutions for removing unwanted noise from the image.

ETTR/EBTR has been promoted as a way to maximize the use of the camera's DR by increasing exposure. That claim is misleading. All you need to do to accomplish that is to lower ISO and thereby increase the recommended exposure - slow down the shutter.

It has also been suggested that ETTR/EBTR can reduce S/N, noise. That is also not true. What reduces noise is the additional exposure (signal) from slowing the shutter speed. You can also get there by reducing the ISO and slowing the shutter.

Nowhere in the promotion of ETTR/EBTR has it been suggested that you can get the most out of a camera’s dynamic range by simply lowering the ISO and increasing the exposure time. Neither is it suggested that reducing the ISO and increasing the exposure is a simple way to increase S/N and reduce visible noise.

Nowhere in the promotion of ETTR/EBTR is the dynamic range of the scene compared to the dynamic range of the camera at any ISO. That would expose a critical piece of missing information – that the maximum dynamic range of the camera is only available at base ISO.

In fact, I have posted a complete list of disadvantages to the use of ETTR/EBTR. Here are just a few:

- Determining the ERADR (Extra RAW-Accessible Dynamic Range) for your camera requires careful research since it is not a constant. It is less at higher ISO settings.
- Setting up the exposure for an image is not simple since you need to look at the histogram, possibly after attempting a test image, and then calculate how much exposure you can get away with adding.
- Once the final image is captured, you often cannot review the results on your camera because the resulting JPEG representation may be washed out.
- There is always the danger that the histogram is inaccurate and that you might go too far to the right, blowing highlights.
- There may be highlight and shadow details in which you are not really interested. This complicate the interpretation of the histogram.

So not only are there significant disadvantages to the use of ETTR/EBTR, it is clear that there are no demonstrable advantages.
There have been numerous posts promoting the use o... (show quote)

Reply
Dec 27, 2016 23:58:31   #
CHG_CANON Loc: the Windy City
 
rmalarz wrote:
In a simple answer, No, it's time has not passed and it's a very useful practice in obtaining optimum exposure.
--Bob
This discussion is rather basic from DPS that I saw today in one of my feeds: http://digital-photography-school.com/exposing-to-the-right/

This UHH post is a much more advanced discussion of the same issue, where I too believe ETTR remains the best method. Note two items from the DPS discussion: 1) you collect more data in your RAW image using ETTR and 2) The 'start point' of the digital noise in your RAW image is less when ETTR-ed.

Reply
Dec 28, 2016 00:06:35   #
St3v3M Loc: 35,000 feet
 
Expose To The Right (ETTR) https://www.google.com/#q=ETTR

Reply
 
 
Dec 28, 2016 03:57:56   #
selmslie Loc: Fernandina Beach, FL, USA
 
CHG_CANON wrote:
... I too believe ETTR remains the best method. Note two items from the DPS discussion: 1) you collect more data in your RAW image using ETTR and 2) The 'start point' of the digital noise in your RAW image is less when ETTR-ed.

You don't collect more data by using ETTR/EBTR, you collect more data by using a longer exposure. See What is the Camera's Dynamic Range? and What is the Camera's Dynamic Range? - Part 2.

Here is a simple example:

1. Using ISO 400, you meter an outdoor scene in bright sunlight at f/11 and the meter suggests 1/800 second (see the Sunny 16 rule). But you notice that the histogram does not reach the right end of the display. You know from experience that you can add two stops to your exposure and get the histogram into the EBTR range without blowing your highlights so you change your exposure to 1/200 seconds. On your computer, you move the Exposure (gain) slider to the left to back out the two extra Ev of exposure and the image looks normal. Voilà! You have used EBTR and you are proud of yourself.

2. Using the same camera, I meter the same scene using ISO 100 and the camera (as well as the Sunny 16 rule and my incident meter) suggests an exposure of 1/200 seconds. On my computer, I don't need to make any adjustment. Both images end up looking the same. Both images show the same amount of visible noise - which is none if we both have decent cameras.

You went the long way around the barn and ended up in the same place I did. So what did you gain from using ETTR/EBTR other than extra work? Absolutely nothing!

What advantage did I have? Besides less work, I was able to review my JPEG on the back of the camera to check the composition. I was also able to see if the very white clouds had moved into a more attractive position so I could take another shot. You could not even see the clouds on your camera because EBTR washed them out. You could not even see the white rabbit that had hopped into the foreground.

Reply
Dec 28, 2016 04:10:07   #
selmslie Loc: Fernandina Beach, FL, USA
 
Uuglypher wrote:
Hi, John
The way I see it: ... (another convoluted explanation) ...
now...how badly have I embarassed myself? ...

Pretty badly (besides not knowing how to spell embarrassed and botching another earlier convoluted explanation claiming that you were only off by 1/3 stop) since three of us have had to explain it to you after you taunted in this post that I wasn't smart enough to figure it out.

I guess it would also be too embarrassing to simply apologize an say, "Scotty was right, I was wrong and I was rude to make such an insulting remark."

Reply
Dec 28, 2016 04:35:43   #
selmslie Loc: Fernandina Beach, FL, USA
 
rmalarz wrote:
In a simple answer, No, it's time has not passed and it's a very useful practice in obtaining optimum exposure.
--Bob

There is no simple answer.

Cameras 13 years ago may have needed it. Today's cameras don't. I am in the process of demonstrating this in a couple of other threads.

Reply
Dec 28, 2016 04:42:25   #
selmslie Loc: Fernandina Beach, FL, USA
 
St3v3M wrote:
Expose To The Right (ETTR) https://www.google.com/#q=ETTR

Many of us know all about what ETTR is and how to go about it. The question remains, why bother?

It's not to get less noise at high ISO. You get that by slowing the shutter down. Lowering the ISO does the same thing.

Eventually you reach base ISO. If you still have noise by then, there is something seriously wrong or you are using a 13 year old camera.

Reply
 
 
Dec 28, 2016 08:43:50   #
rmalarz Loc: Tempe, Arizona
 
"...The question remains, why bother?" The answer still remains, to obtain optimum exposure.
--Bob

selmslie wrote:
Many of us know all about what ETTR is and how to go about it. The question remains, why bother?

It's not to get less noise at high ISO. You get that by slowing the shutter down. Lowering the ISO does the same thing.

Eventually you reach base ISO. If you still have noise by then, there is something seriously wrong or you are using a 13 year old camera.

Reply
Dec 28, 2016 09:08:37   #
St3v3M Loc: 35,000 feet
 
selmslie wrote:
Many of us know all about what ETTR is and how to go about it. The question remains, why bother?

It's not to get less noise at high ISO. You get that by slowing the shutter down. Lowering the ISO does the same thing.

Eventually you reach base ISO. If you still have noise by then, there is something seriously wrong or you are using a 13 year old camera.

And many of us do not, but the links give us the opportunity to learn, and even if you don't believe in it I imagine there must be value there otherwise why are there so many articles from well-respected photography sites extolling the value of the process? One man's belief is another man's quest to learn. S-

Reply
Dec 28, 2016 09:20:06   #
selmslie Loc: Fernandina Beach, FL, USA
 
rmalarz wrote:
"...The question remains, why bother?" The answer still remains, to obtain optimum exposure.
--Bob

How do you define "optimum exposure?"

Is is something you can actually see in the final image? If you can get the same result with and without ETTR/EBTR, the question remains ...

If it is only something you can speculate about by doing the math but you can't demonstrate it in an actual image then we are talking about data processing, not photography.

Reply
Dec 28, 2016 10:25:05   #
selmslie Loc: Fernandina Beach, FL, USA
 
rmalarz wrote:
"...The question remains, why bother?" The answer still remains, to obtain optimum exposure.
--Bob

If it's data processing that you want to see, take a look at the attached image.

One of the often repeated mantras of ETTR/EBTR is that you should strive to use the cameras's maximum dynamic range. But what that usually means is to utilize the highest possible numeric values in the raw file.

Let's assume that we have a camera with 1 stop (for the sake of simplicity) of ERADR. If you only use ETTR you will be missing half of the raw file's numeric values, the top 8192 values for a 14-bit raw file. If you use EBTR you will use that extra stop and include the missing 8192 values. That's going to happen regardless of the camera's ISO setting, assuming that the 1 stop of ERADR stays with you as you raise the ISO.

But how much of the camera's sensor capacity is actually used with EBTR? At base ISO, almost all of it. But as you increase the ISO you will be using less and less.

The attached table shows that, for a scene containing a DR of 7 stops, nearly all of the sensor's capacity can be used at base ISO. Each time you double the ISO that capacity gets cut in half. Pretty soon the noise becomes visible in the darker zones.

So if you really want to maximize the use of the camera's DR, you must use base ISO. Under any other condition you are simply attempting to "optimize" the little bit of sensor capacity you have left for yourself when you increased the ISO.


(Download)

Reply
Page <<first <prev 4 of 4
If you want to reply, then register here. Registration is free and your account is created instantly, so you can post right away.
Main Photography Discussion
UglyHedgehog.com - Forum
Copyright 2011-2024 Ugly Hedgehog, Inc.