rmalarz Loc: Tempe, Arizona

For those of us who actually understand the benefits of ETTR/EBTR along with utilizing all of the data captured by digital camera sensors, we'll continue to use these techniques. Additionally, it will give those whose knowledge of the techniques is somewhat lacking the opportunity to trot out numerous graphs, tables, and references in lieu of quality photographs.

As I mentioned before, and probably will again, look at the pictures.
--Bob

BebuLamar

Thanks Scotty!
As someone said I was on the high horse which I don't have but ETTR if you want to use it are.
1. Simply watch the brightest part of the image that you need to retain details (sometimes you may decide for let the brighest part go because it's not important to your image in order to get good details in the shadow) and get it as bright as you can without losing details.
2. You must use RAW because if you shoot JPEG you should expose it in a way that the image looks good straight out of the camera without any PP.
3. Use base ISO because if you don't use base ISO it's better to simply reduce the ISO.

TriX Loc: Raleigh, NC

With all due respect to both you and Scotty, I am going to have to disagree with the last statement. As many of the indoor sports examples and other low light situations show, simply lowering the ISO isn’t an option when you already have the aperture wide open and you are at the minimum shutter speed to freeze the action. In fact, in these situations where the DR of the camera may only be 5 or 6 bits (a Canon 5D3 has ~5 bits @ ISO 12,800) adding 1 stop of usable exposure (1 stop = 1 bit) is a substantial improvement in both DR and signal to noise ratio. As most Canon shooters in particular know (or should), underexposing and not using all the available DR is the kiss of death in terms of noise at high ISOs.

And don’t discount the advantage of a FF sensor in these low light/high ISO situations - the typical 1 stop advantage over a crop camera can be translated into 1/2 the ISO, twice the shutter speed or another stop of DOF - not a trivial thing when 1/2 the ISO may make the difference in visible noise or not, just like that stop that you throw away if you don’t use the full DR available by not using ETTR. I shoot HS indoor sports regularly, and it didn’t take me long to figure out that going to FF and using every bit of DR that’s available to me makes the difference between exposures that are noisey and those where the noise in not noticeable.

BebuLamar

If you can't give more exposure to get the highlight any closer to the saturation point of the sensor (that is more light to the sensor) then you simply can't use ETTR.

selmslie Loc: Fernandina Beach, FL, USA

You are implying that you understand ETTR and I don't. You should know better by now.

The reason I have posted screen prints of the RawDiger histograms is to demonstrate what the raw file actually records. Other than my 16 MP Df raw files, my 24 MP raw files are simply too big to upload to UHH. It's clear that the information presented by RawDigger cannot have been tampered with.

If you want to post credible evidence that you are not blowing your raw highlights using your approach to ETTR, I suggest that you do the same - post the RawDigger histograms as I have done. If you don't, why should we believe what you say about ETTR/EBTR?

Delderby Loc: Derby UK

Yes -it is horrible - but that is not how it looks on my viewer! I guess it must be something to do with view source to view source. Bear with me while I try again?

rmalarz Loc: Tempe, Arizona

Just simply look at the pictures.
--Bob

selmslie Loc: Fernandina Beach, FL, USA

The subject is covered very nicely in Photography Life, Exposing to the Right Explained. In that article the author states that:

"9) When to Avoid ETTR
In theory, ETTR works with every possible scene. There will always be a “best” exposure for an image — one which is as bright as possible without blowing out any of the highlights. However, in practice, this is not always true.

If you aren’t shooting at your camera’s base ISO, ETTR is all but useless."

If you read the entire article you will understand why he and many others have reached the same conclusion.

selmslie Loc: Fernandina Beach, FL, USA

Compare your results with my full sized version developed from raw and you won't see any of the problems you introduced trying to edit the JPEG.

selmslie Loc: Fernandina Beach, FL, USA

What are you afraid of? If you can't or won't show us your RawDigger histograms then you are hiding something from us.

Delderby Loc: Derby UK

But my latest certainly improves your JPG SOOC and improves over your RAW using capture 1 - especially the sky!

selmslie Loc: Fernandina Beach, FL, USA

How would anyone know that? You only posted a thumbnail.

Nevertheless, you did not recover any of the clouds that were blown in the SOOC JPEG because that is impossible.

Give it up.

selmslie Loc: Fernandina Beach, FL, USA

And here is a logical exercise I have posted before. Imagine a scene with a fairly narrow DR such as a partly overcast landscape.

1. You have your ISO set to 200 and the meter suggests 1/200 @ f/11. You take the image and find that the histogram is not on the right. So you take a second shot at 1/100 and the histogram ends up on the right.

2. You have your ISO set to 100 and the meter suggests 1/100 @ f/11. You take the image and find that the histogram is not on the right. So you take a second shot at ISO 200 and the histogram ends up on the right.

Both images ended up at the same setting - ISO 200 1/100 @ f/11. The raw histograms two raw files are identical. The exposures are identical.

Both JPEGs look like they are overexposed by 1 stop. When you start to develop the file from raw, the first thing you do is to move the Exposure slider to the left by one stop.

Since the ISO 100 version started with the same exposure (1/100 @ f/11) as the ISO 200 ETTR version, both recorded the same amount of luminance noise - with today's sensors that would be no visible noise at all. The only real difference is that the numeric values stored in the raw file are twice as large for the ETTR version.

TriX Loc: Raleigh, NC

A

I have read the entire article, but how can you deny the logic of what I just posted? Would you agree that higher DR and S/N is desirable, regardless of the ISO, and that using the max DR of the A/D satisfies that goal, whether you call it ETTR or not? And would you also agree that the JPEG derived histogram and “blinkies” may not accurately reflect just setting the MSB at the brightest zone?

There are always multiple opinions on ETTR/EBTR, ISO invariance, etc, and you can find a quote to support both sides, but I will rely on a thorough knowledge of the digitizing process, careful calibration of my equipment, and my experience in shooting in low light action photography. I get the arguement (and another poster just made the same arguement), but it ignores that amplifying the signal from the sensor prior to the A/D effectively raises the ISO and increases the usable DR of the A/D.

A photographer faces exactly the same challenge as an audio recording engineer - trying to utilize the max DR of the recording device without exceeding the MSB and clipping the peaks or highlights. You can play it “safe” and always allow a stop or a few dB of headroom, but you lose DR in the process, so accurately characterizing your recording equipment and using the max DR of the device is always desirable from a S/N perspective (which is especially important at high ISOs where noise is more visible), regardless as to whether you call it ETTR or not.

Delderby Loc: Derby UK

I see no blown clouds

(Download)