The X-Pro1: Understanding the Dynamic Range Feature and its Relationship to ISO

Scroll down to content

In the last part (click here) we looked at Jpegs and Raw, the benefits of each, and why (in my opinion) you should shoot both.

This time we’ll take a look at the Dynamic Range (DR) feature and it’s relationship to ISO.

It can be a contentious subject.. Certainly the online forums are full of questions and debate about its function and its value.

Some swear by it, others swear at it.

Am I about to end that debate? Will this article once and for all answer whether or not to use it?

No of course not!

As with everything camera related, it’s about understanding the function, then deciding if it works for you (or not)

Let’s start with a quick explanation of Dynamic Range (DR) This is the easy bit!

DR (within a photographical context) is the amount of light and dark that can be captured before detail is lost.

When the LIGHT detail exceeds the DR abilities of the camera (or film) the image will display only WHITE

When the DARK detail exceeds the DR abilities of the camera (or film) the image will display only BLACK

We’re NOT talking about, ‘oh but I can recover that detail from the RAW file’ – no, in absolutes we’re talking it’s gone, lost forever, you blew it (or blocked it!!), zilch, nada.. GONE FOR GOOD.

Of course, in none absolute terms, this might not matter… perhaps only a tiny part of the scene is lost, and that part is not important to the overall look of the photograph. Perhaps it’s only 0.1% of the shot, a tiny sunlight reflection in a pair of sunglasses, or the shadow in a wheel arch of a car.

However, I think it’s safe to say, that we should aim NOT to lose detail from our photographs, and that having the maximum amount of DR available to us is a good thing.

The more DR at our disposal, the more we can still take photographs in dynamically intense lighting conditions.

So Fujifilm (and most camera manufactures) offer us expanded DR modes.

But how can the camera exceed the PHYSICAL limitations of the light range that the camera can capture, that sounds illogical, right?

It sounds illogical because it is – the camera can’t.

Let’s take a look at some examples, and try to understand how it works, we must also look at the role of ISO in the DR expansion equation.

Fuji offer us 4 settings to chose from in regards to DR

DR100 = OFF
DR200 = 1 STOPs of DR recovery
DR400 = 2 STOPs of DR recovery
DR Auto = The camera will pick for you from options 1, 2 & 3

So let’s take a look and see what effect these settings have on a photograph.

We’ll start by looking at the DEFAULT SOOC Jpegs produced by my X-Pro1.

The standard Fujifilm Jpeg:

STD Colour profile (which is Provia) All highlight/Shadow/Colour/Sharpening/NR settings, set to ZERO (0)
Dynamic Range (DR) set to DR100 (off)

The standard DR200 Fujifilm Jpeg:

STD Colour profile (which is Provia) All highlight/Shadow/Colour/Sharpening/NR settings, set to ZERO (0)
Dynamic Range (DR) set to DR200 (1 Stop)

The standard DR400 Fujifilm Jpeg:

STD Colour profile (which is Provia) All highlight/Shadow/Colour/Sharpening/NR settings, set to ZERO (0)
Dynamic Range (DR) set to DR400 (2 Stop)

As you can see, as we increase the DR mode, we are able to retain slightly more detail in the clouds. Even with SOOC Jpegs, by the time we’re getting to DR400 (max) we’re blocking up the shadows a little, in order to darken the clouds.

But what is the camera actually doing here? As discussed in the RAW/Jpeg page, the camera ALWAYS shoots RAW and processes its own Jpeg. We just chose whether or not to make that RAW data available for our own use after the shot.

So in order to understand how the SOOC Jpeg is created using DR expansion, we must look to the RAW file to see what’s going on.

The UNPROCESSED DR200 Jpeg showing the RAW file:

Dynamic Range (DR) set to DR200 (1 Stop)

The UNPROCESSED DR400 Jpeg showing the RAW file:

Dynamic Range (DR) set to DR400 (2 Stop)

The DR modes work, by UNDEREXSPOSING the shot, then applying a brightening tone curve to the dark parts of the image.

So without the camera’s inbuilt tone brightening, we quickly see DR expansion modes for what they are, 1 or 2 stops of underexposure.

The more you increase the ISO of a digital camera, the more you UNDERexpose the image. “But Adam, don’t be daft, everybody knows when the ISO goes up, it’s because it’s dark, and we need a brighter photograph. ISO brightens a picture, it doesn’t darken it”

It’s not quite that simple! ISO gain DARKENS a shot, because it needs to REDUCE it’s sensitivity to light.

So if your camera is set to base ISO (which is 200 on the X-Pro1) and the automated shutter speed selects 1/30, you may think to yourself ‘Hmmm I don’t fancy handholding this at 1/30’ so you increase the ISO to 400, and you’re rewarded with a automatically selected shutter speed of 1/60

This is because ISO darkens, it tells the camera that we don’t mind losing (in this example) a stop of light during image capture, so needing less light, the camera can select a higher shutter speed.

On the X-Pro1:
ISO / Stops
200 / 0
400 / -1
800 / -2
1600 / -3
3200 / -4
6400 / -5

This underexposure is then dealt with in two ways (depending on the SOOC or the RAW)

The SOOC Jpeg is then ‘artificially’ brightened in the camera, so that it’s not too dark

The fact that the image has been underexposed (in this case a stop) is also hardcoded into the RAW file, so that your PP software of choice knows to brighten it.

Underexposure is GENERALLY a bad idea. When we underexpose we lose dynamic range and details.

But let’s have a little bit of fun!!

It’s a very bright scene, and I’ve deliberated blown it out (by using F1.4) so that we can see how the camera and 3 RAW convertors handle it.

So here’s the Jpeg, that’s SOOC and DR has been set to 400 (2 stops)


EPIC FAIL! Even with DR400 there was far too much light and the shot is ruined

OK, lets try the RAW file from the same shot (NB: NO PP WAS USED, SIMPLY OPENED AND EXPORTED TO JPEG)

First up it’s Light Room CC.


Again, EPIC FAIL! LR has read the imbedded instruction in the RAW file to brighten the shot and the net result is about as lousy as the SOOC Jpeg.

Next up it’s Capture One.


Yup, again, EPIC FAIL! C1 has read the imbedded instruction in the RAW file to brighten the shot and the net result is about as lousy as the SOOC Jpeg.

Lastly we have Photo Ninja.


Now PN can be set to ignore all the instructions embedded in the RAW file. As a general rule of thumb this isn’t a great idea, the camera doesn’t embed info in the RAW file for the fun of it, it’s important stuff! But in this rather forced, contrite example, those 2 stops of underexposure have (give or take) just about saved the image.

So if you’re a PN user, habitually like to over expose images and don’t own an ND filter, this might be of interest to you!!!

Hopefully this article has given a little insight into how digital ISO works, and most importantly how the DR expansion modes work.

Now I realise that I’ve (believe it or not) made this as simplistic as possible. There are many RAW convertors and PP software, and of course each has it’s own abilities to recover highlights and boost shadows.

None of my example shots are designed to do anything other than demonstrate what’s going on, and none of them have had any recovery PP work done on them.

But for me?

I leave DR OFF (which on Fujifilm X cameras means setting DR100)

-The maximum dynamic range is base ISO (remember raising ISO darkens the image and forces us to artificially brighten)
-The maximum resolution is base ISO (when we darken an image, we generally lose detail)
-More often or not I go with the RAW file

If I was 100% a Jpeg shooter, I’d use the DR expansion modes, but I probably wouldn’t automate them (ie I would pick when I wanted to use them, and NOT select DR-Auto) but I’m not (a 100% Jpeg shooter) so I don’t use DR.

I go about highlight and shadow recovery in PP, and do my best to expose the shot as I want it within the camera

Of course, as ever, what works for me, needn’t work for you, so go out and find your own settings. But hopefully this piece helps you with doing that.

PREVIOUS PAGE
NEXT PAGE

======================================
A lot of time and effort goes into this site.. Hopefully it’s helped you? Perhaps you’d consider helping me?

One way you could help me is if you want to buy from Amazon, if you do so using the links below, then I will receive a small percentage of your expenditure, and you will pay NO MORE than you would have paid anyway.

Shop at Amazon USA
Fujifilm X-Pro2 ¦ Fujifilm X-Pro2 Handgrip ¦ Shop for Fujifilm X-Pro ¦ Shop for Fujifilm X-T ¦ Fujifilm XF Lenses ¦ Fujifilm XF Acessories

Shop at Amazon UK
Fujifilm X-Pro2 ¦ Fujifilm X-Pro2 Handgrip ¦ Shop for Fujifilm X-Pro ¦ Shop for Fujifilm X-T ¦ Fujifilm XF Lenses ¦ Fujifilm XF Acessories

If there’s a different product you’re considering, then perhaps you’d drop me a line and I can send you an associate link for it?

Another way you could help, is by making a donation. The donate button can be found on the link below

Donate

Thank You Very Much!
======================================

The X-Pro Series Content: Referenced and All In One Place

The X-Pro Series Content: A Contents Page with Page Numbers and a Brief Description

My Blog Click Blog Table of Contents

Other Stories Click Pages and Articles

Advertisements

12 Replies to “The X-Pro1: Understanding the Dynamic Range Feature and its Relationship to ISO”

  1. Adam, love your commentary on this topic, well explained. Gotten better results by using DR100 and PP editing. Im still clinging to my xpro1, although the new model is calling me!

    Like

    1. Hi Joe,

      Thanks very much! max DR is always at base ISO, but DR can be good for SOOC jpegs.

      The longer you can hold out for the newer X-Pro, the cheaper it’ll get!

      Thanks again
      Adam

      Like

  2. Excuse me, I haven’t studied what your saying here yet (I intend to) but you wrote “The maximum dynamic range is base ISO (remember raising ISO darkens the image and orces us to artificially brighten)” and i’m not following. Did you mean to type forces instead of “orces”? and the phrasing is incomplete I think, brighten what?Not nit-picking, trying to understand.
    TIA
    cvt, Florida

    Like

    1. Hi Chunk,

      Thanks – yup that was a typo, you’re the first person to notice, and I wrote that months ago!

      You’re correct, it should of read ‘forces’ and I’ve now changed it.

      The answer too “brightens what?” is in the paragraphs above the part you’ve quoted.

      Basically digital ISO works by underexposing the shot, then making it brighter afterwards (the same principle as turning the volume up on an amplifier for a bit of music that you want to play louder)

      It can sound counter intuitive, because when we see a dark scene, we tend too think ‘Hmmm there’s not much light, I need to raise ISO sensitivity’

      But in reality, what we do when we raise ISO is that we make the camera LESS sensitive to light (because there’s less light for it to be sensitive too), the shot is then artificially brightened by the camera during the RAW/Jpeg creation (or after in the case of ISOless sensors)

      Like

  3. While I appreciate your effort to share your knowledge and help others in their learning journey, I can’t help but think some of your explanation is inaccurate (and so may mislead people who come here to learn).

    The idea I have issue with is: “ISO gain DARKENS a shot, because it needs to REDUCE it’s sensitivity to light.”

    Where are you getting this information?

    Put your camera into Manual mode, Raw file format. Set you ISO at base (200 in the case of FUJI). Take a shot. Now increase the ISO a stop or two (or more, to make the effect more obvious). Take another shot. What happens?

    It gets brighter. The same amount of light came in past the aperture and through the shutter, and onto the sensor. But then ISO came into play – ISO boosts the live sensor data before it gets to a processing engine and is written to the card as a file. ISO doesn’t actually CHANGE the sensor’s sensitivity to light. It amplifies the signal coming from the sensor.

    This makes weak signals stronger. This can be good or bad, depending on the situation. If the scene contains little light (night time), and the aperture cannot open anymore, and slowing the shutter would introduce motion blur, then the only way to make the final image bright enouph, is to amplify what little light was hitting the sensor.

    This is why increasing ISO tends to introduce noise. The noise is always there, but when the signal (light data) is strong, it overpowers the noise. Kind of like the static that is always coming out of your speakers from the stereo system, but you only hear it when you crank the volume and have no actual music playing.

    Increasing ISO is necessary, and worth it, when you have very little light to play with. But if you already have enouph light, it will amplify what should not be amplified, and you get “blown out” highlights . The light is amplified to the point that it can no longer be represented by distinct digital values – they’re all at full blast. In audio, you’d call this clipping.

    Anyway, it may seem like a side-point, or a non-point. But a subject like DR is founded on fundamentals like ISO, and the fundamental must be understood correctly before any related subject can be understood.

    Sorry for the diatribe. Otherwise, it was a very well-done article. Keep on sharing!

    Like

    1. ISO (in RAW mode) = file is underexposed x number of stops, and the instruction that the file needs a global signal amplification is written into the RAW file for the external RAW software to act upon

      ISO (in camera JPEG mode) = file is underexposed x number of stops, and the camera then applies a global signal amplification to create the resultant jpeg

      Fuji DR Expansion modes = RAW is underexposed 1 or 2 stops. The resultant in camera JPEG then has a selective dark and mid tone amplification applied to it. However, the RAW file remains 1 or 2 stops underexposed (as it lacks the metadata to provide a global signal amplification.

      Your statement here

      “But then ISO came into play – ISO boosts the live sensor data before it gets to a processing engine and is written to the card as a file. ISO doesn’t actually CHANGE the sensor’s sensitivity to light. It amplifies the signal coming from the sensor.

      It’s not as binary as this. Modern ISO invariant cameras work like this, more or less (sometimes base ISO is a mixture of amplification) using POST ADC amplification – but older cameras (such as the X-Pro1 this article is about) use PRE ADC amplification. Boosting pre ADC amplification after ADC leads to more noise than if you’d raised ISO in the camera

      That’s how ISO works and how Fuji’s DR modes work.

      Your question/comment

      The idea I have issue with is: “ISO gain DARKENS a shot, because it needs to REDUCE it’s sensitivity to light.”

      For example, you set ISO 200. The camera tells you the SS will be (say) 1/30th. This is bad, you need 1/60th. So you raise ISO to 400, and shoot at 1/60th. This is underexposure and the file needs amplification as per the conditions I set out above.

      Thanks for reading my site, no apology necessary 🙂

      Like

  4. “It’s not quite that simple! ISO gain DARKENS a shot, because it needs to REDUCE it’s sensitivity to light.”

    This is wrong. Perhaps you meant: “Iso darkens the raw mechanical capture of light, since the shutter speed reduces to compensate for the increased electronic sensitivity to light (ISO) “. This is the only interpretation of your statement that is correct.

    Increasing ISO gain has the net effect of brightening a shot because ISO increases the electronic sensitivity (amplification) of light; it records all the light hitting the sensor then artificially scales the light electronically onto the final image (amplification). Therefore even the darker regions of where there is more noise gets amplified.

    The reason shutter speed goes up when ISO goes up (assuming shutter speed is in ‘auto”) is because the camera knows that “ok we’re gaining light electronically now, so if we keep the same shutter speed we’re going to white wash the image. Better compensate by decreasing the shutter speed).
    The purpose of ISO is when you don’t want to/can’t decrease shutter speed anymore and want to get that extra sensitivity to light.

    Like

    1. No.

      ISO is under exposure combined with digital or analogue amplification.

      You can see this clearly with the ISOless sensor cameras. If the camera wants (say) ISO 3200 and 1/60th, and instead you shoot at ISO 200 and 1/60th, you’ll need to add 4 stops of amplification in your RAW editor.

      What do you get if you don’t add this amplification?

      A very dark image. An underexposed image in fact.

      The sensor can ONLY EVER RECORD LIGHT AT BASE ISO, and if you stray from the optimum SS for the base ISO, then the shot is underexposed and brightened artificially.

      When there’s not enough light for your desired SS and base ISO, you’re not making the sensor more light sensitive, that’s not physically possible.

      Instead you’re “telling” the sensor that it doesn’t need as much light as it thinks it does, which allows it to select a higher shutter speed. (That’s a higher shutter speed, not a “reduced SS” as wrote in your comment – reducing the SS makes the image brighter, not darker)

      This is REDUCING the sensitive to light, and the reduction comes about because you’re substituting captured light for amplification that will be applied post capture.

      Like

      1. Really, isn’t it time to replace “ISO” as an measure of sensitivity with something else?
        Sensors have (as far as I am aware?) one basic sensitivity. Everything else is an amplification of the light energy captured. And some amplification circuits do better than others.
        Unless somebody’s changed the laws of physics since my education took place?
        ‘Course I’ve always supected that ISO, ASA, BS,and DIN were polite fiction at best! (Hey, I’m very old school, too much CD3 fogging at the neurons!)

        Like

      2. Yes, the digital world is a little antiquated in it’s terminology and a more current nomenclature would help people like the previous two commenters understand what was happening.

        A sensor can only record at full light sensitivity at base (as you say), to raise “ISO” is to throw away light gathering ability

        That’s why available dynamic range decreases with a raised ISO.

        Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

w

Connecting to %s

%d bloggers like this: