Anyone here deeply familiar with the fuji's DR settings and knows how it works?

nomnomnom

Leading Member
Messages
569
Reaction score
102
Im new to fuji system (coming from Sony) and I'm trying to get into this new mindset of shooting jpeg not raws.

One of things to help get better jpegs is this DR concept in Fuji (other cameras have similar settings but I dont think they work the same way).

I've done research on this topic for a week now. Watched over 20 youtube videos and have read all the forum posts and blog posts.

Basically what I gather is that DR200/DR400 causes the camera to underexpose (via ISO) by 1 or 2 stops. And then it will selectively digitally raise the pixels exposure in post processing for shadow areas back to the original exposure.

It sounds reasonable, and that's how it is explained in most videos/blogs. But how do you actually use this thing practically?

For example, if the scene is already correctly exposed (without highlight clips), by turning on DR400, you are just introducing noise in shadow for no real gain isnt it?

Therefore, for a contract scene, it is advantageous to over expose and clip the highlights, and turn on DR400. At least in theory. I suppose one can do this with exp comp dial. But is this really the intended usage? It's seems so backwards.

---

Side topic: I also got into a lot of arguments in other forums on this topic because based on my research DR affects RAW files. But a lot of people get angry when you mention that.
 
Why not just go out and experiment to see what varying results you get. A mountain scene on a bright day with part of the mountain in shadows would work, a forest on a sunny day deep into the canopy can be challenging, even some blue hours scenarios would also work and some sunsets.
 
It underexposes, but then applies a different tone curve to the image to make it look right again.

Benefit is one or two stops of extra headroom in the highlights, at the (small) cost of some extra shadow noise. Blown highlight can look far worse than some specks in darker areas, so in high contrast scenes it is useful.

When you shoot in raw you can either use this DR function, or manually EC and then apply manual adjustments during post-processing. In jpeg mode, less so.

I find it useful when in a hurry and light changes a lot and/or high contrast scenes.
 
The Pal2Tech bloke explains it pretty thoroughly.


PS. It gets baked into the raw file, so keep that in mind if you turn it on.
 
The Pal2Tech bloke explains it pretty thoroughly.


PS. It gets baked into the raw file, so keep that in mind if you turn it on.
He explains it incorrectly. Lots of misinformation in that video.
 
Im new to fuji system (coming from Sony) and I'm trying to get into this new mindset of shooting jpeg not raws.

One of things to help get better jpegs is this DR concept in Fuji (other cameras have similar settings but I dont think they work the same way).

I've done research on this topic for a week now. Watched over 20 youtube videos and have read all the forum posts and blog posts.

Basically what I gather is that DR200/DR400 causes the camera to underexpose (via ISO) by 1 or 2 stops.
Raising ISO causes the camera to underexpose. Because the DR modes won't function at base ISO and require you to raise the ISO it's become common to claim the DR modes underexpose. Obviously we all only shoot at base ISO for everything we photograph, right? If however you're already working at say ISO 1600 then the DR modes do not underexpose; you get the exact same exposure with or without the DR modes active.

The DR modes don't require or force underexposure; they require the ISO be raised. You can set an exposure manually at base ISO, not change the exposure, but then raise ISO and activate a DR mode -- no underexposure will occur.
And then it will selectively digitally raise the pixels exposure in post processing for shadow areas back to the original exposure.
A lower contrast tone curve is applied by the camera in post processing.
It sounds reasonable, and that's how it is explained in most videos/blogs. But how do you actually use this thing practically?

Shoot camera JPEGs, it's useless for raw shooters, and in high contrast light consider activating a DR mode to get a lower contrast final image.

For example, if the scene is already correctly exposed (without highlight clips), by turning on DR400, you are just introducing noise in shadow for no real gain isnt it?
In high contrast light if the scene is already correctly exposed without highlight clipping at base ISO odds are you have very dark shadows. Keep the same exposure, raise the ISO and activate a DR mode for a better (lower contrast) JPEG.
Therefore, for a contract scene, it is advantageous to over expose and clip the highlights, and turn on DR400. At least in theory. I suppose one can do this with exp comp dial. But is this really the intended usage? It's seems so backwards.
It's a way to get a better SOOC JPEG. If you're not doing that then it's useless. Fuji engineered it to work with the camera metering system and pattern metering. If the scene contrast is really high the camera will likely set an exposure that clips highlights. The DR modes pull those back and create a lower contrast JPEG with the highlights retained.

How it works: You have have to understand how ISO works. When you raise ISO the camera meter calculates a reduced exposure. ISO then lightens that underexposed image in one of a number of ways. Most commonly the analog signal from the sensor is amplified before ADC. This signal amplification can cause highlight clipping -- we call it ISO clipping. and this is where Fuji's DR modes step in. They hold back one or two stops of that ISO amplification which then doesn't clip the highlights. The image is then lightened the required ISO amount in the image processor using the lower contrast tone curve.

If you shoot raw and can set base ISO then expose to fully utilize the sensor without highlight clipping and post process the raw file. You'll get much better results and the maximum DR your camera is capable of.
 
Last edited:
The Pal2Tech bloke explains it pretty thoroughly.


PS. It gets baked into the raw file, so keep that in mind if you turn it on.
He explains it incorrectly. Lots of misinformation in that video.
Please explain. Most people agree this is the most accurate video of out there.
He get's it completely wrong. He doesn't understand how ISO works and his nonsense that the camera is applying ISO differentially between highlights and shadows is nonsense.
 
Im new to fuji system (coming from Sony) and I'm trying to get into this new mindset of shooting jpeg not raws.

One of things to help get better jpegs is this DR concept in Fuji (other cameras have similar settings but I dont think they work the same way).

I've done research on this topic for a week now. Watched over 20 youtube videos and have read all the forum posts and blog posts.

Basically what I gather is that DR200/DR400 causes the camera to underexpose (via ISO) by 1 or 2 stops.
Not always (there is only a reduction in exposure if you choose to use a DR mode with an increased ISO setting rather than using DR100 at base ISO), and it should be clarified that ISO isn’t exposure, it can affect image lightness like exposure, but doesn’t contribute to sensor exposure in any way and must be considered independently.
And then it will selectively digitally raise the pixels exposure in post processing for shadow areas back to the original exposure.
Basically, by either a reduction in exposure (unavoidable if you choose to use a DR mode rather than base ISO), or a reduction of analog pre -D to A converter gain (which, if you’d be at a higher ISO anyway, doesn’t come with any major downsides), a one or two stop darker than normal RAW file is produced in either case to increase highlight headroom at “the right”. The resulting loss of image lightness is compensated for digitally in the jpeg processing with a highlight curve that will include detail recorded one or two stops beyond where the clipping point would have been at DR100 - flatter, but non-blown highlights.
It sounds reasonable, and that's how it is explained in most videos/blogs. But how do you actually use this thing practically?

For example, if the scene is already correctly exposed (without highlight clips), by turning on DR400, you are just introducing noise in shadow for no real gain isnt it?
Yes. In most cases, some positive exposure compensation is going to be part of the equation - at DR400 you will have an extra 2 stops of highlight headroom - up to 2 stops of positive exposure compensation to increase the shadow and mid tone lightness to where it needs to be for a good looking SOOC jpeg. IMO, DR200 (+1 stop of available headroom) will be plenty for most “normal” high DR situations. I highly recommend turning on the highlight warning blinkies which will show you when (and where) you’ve reached the limit.
Therefore, for a contract scene, it is advantageous to over expose and clip the highlights, and turn on DR400. At least in theory. I suppose one can do this with exp comp dial. But is this really the intended usage? It's seems so backwards.
You’re not overexposing, you are managing SOOC image lightness. By setting a DR mode, you are recording a darker than normal RAW, increasing exposure compensation allows you to brighten the shadows and mid tones without clipping the RAW highlights. A DR400 RAW with +2 stops of exp. comp. is functionally identical to a DR100 RAW with no exp. comp.
---

Side topic: I also got into a lot of arguments in other forums on this topic because based on my research DR affects RAW files. But a lot of people get angry when you mention that.
Using DR200/400 mode when DR100 and base ISO was feasible will indeed result in a relative 1 or 2 stop under underexposure, and a darker RAW file (with a metadata tag that usually automatically corrects the lightness in most RAW editors) but, contrary to what the Pal2Tech guy says, no manipulation of any tone curves or anything like that is going on in the RAW data. I don’t shoot jpegs at all and always use DR100, but the DR modes can help with getting better looking SOOC jpegs in high DR situations (I recommend avoiding the DR modes for jpegs too unless you actually have a high dynamic range situation that calls for them).
 
Last edited:
Im new to fuji system (coming from Sony) and I'm trying to get into this new mindset of shooting jpeg not raws.

One of things to help get better jpegs is this DR concept in Fuji (other cameras have similar settings but I dont think they work the same way).

I've done research on this topic for a week now. Watched over 20 youtube videos and have read all the forum posts and blog posts.

Basically what I gather is that DR200/DR400 causes the camera to underexpose (via ISO) by 1 or 2 stops.
Not always (there is only a reduction in exposure if you choose to use a DR mode with an increased ISO setting rather than using DR100 at base ISO), and it should be clarified that ISO isn’t exposure, it can affect image lightness like exposure, but doesn’t contribute to sensor exposure in any way and must be considered independently.
And then it will selectively digitally raise the pixels exposure in post processing for shadow areas back to the original exposure.
Basically, by either a reduction in exposure (unavoidable if you choose to use a DR mode rather than base ISO), or a reduction of analog pre -D to A converter gain (which, if you’d be at a higher ISO anyway, doesn’t come with any major downsides), a one or two stop darker than normal RAW file is produced in either case to increase highlight headroom at “the right”. The resulting loss of image lightness is compensated for digitally in the jpeg processing with a highlight curve that will include detail recorded one or two stops beyond where the clipping point would have been at DR100 - flatter, but non-blown highlights.
It sounds reasonable, and that's how it is explained in most videos/blogs. But how do you actually use this thing practically?

For example, if the scene is already correctly exposed (without highlight clips), by turning on DR400, you are just introducing noise in shadow for no real gain isnt it?
Yes. In most cases, some positive exposure compensation is going to be part of the equation - at DR400 you will have an extra 2 stops of highlight headroom - up to 2 stops of positive exposure compensation to increase the shadow and mid tone lightness to where it needs to be for a good looking SOOC jpeg. IMO, DR200 (+1 stop of available headroom) will be plenty for most “normal” high DR situations. I highly recommend turning on the highlight warning blinkies which will show you when (and where) you’ve reached the limit.
Therefore, for a contract scene, it is advantageous to over expose and clip the highlights, and turn on DR400. At least in theory. I suppose one can do this with exp comp dial. But is this really the intended usage? It's seems so backwards.
You’re not overexposing, you are managing SOOC image lightness. By setting a DR mode, you are recording a darker than normal RAW, increasing exposure compensation allows you to brighten the shadows and mid tones without clipping the RAW highlights. A DR400 RAW with +2 stops of exp. comp. is functionally identical to a DR100 RAW with no exp. comp.

Using DR200/400 mode when DR100 and base ISO was feasible will indeed result in a relative 1 or 2 stop under underexposure, and a darker RAW file (with a metadata tag that usually automatically corrects the lightness in most RAW editors) but, contrary to what the Pal2Tech guy says, no manipulation of any tone curves or anything like that is going on in the RAW data. I don’t shoot jpegs at all and always use DR100, but the DR modes can help with getting better looking SOOC jpegs in high DR situations (I recommend avoiding the DR modes for jpegs too unless you actually have a high dynamic range situation that calls for them).
Ok but let’s simplify it. If I’m trying to take a shot of a contrast scene as jpeg. How should one do it? From my understanding, it’s to overexpose by 1-2 stops (shows as wall on the far right histo) and then turn on DR400. Correct?
 
This is the right explanation there ˆ
I don’t shoot jpegs at all and always use DR100, but the DR modes can help with getting better looking SOOC jpegs in high DR situations (I recommend avoiding the DR modes for jpegs too unless you actually have a high dynamic range situation that calls for them).
I always use RAW+JPEG but often shoot DR200 to help with composition (thank you EVF), and send the occasional quick JPEG.

My belief (no proof) is that there won't be a lot of difference in noise between ISO 125 and ISO 250 for RAW processing later on.
 
Im new to fuji system (coming from Sony) and I'm trying to get into this new mindset of shooting jpeg not raws.

One of things to help get better jpegs is this DR concept in Fuji (other cameras have similar settings but I dont think they work the same way).

I've done research on this topic for a week now. Watched over 20 youtube videos and have read all the forum posts and blog posts.

Basically what I gather is that DR200/DR400 causes the camera to underexpose (via ISO) by 1 or 2 stops.
Not always (there is only a reduction in exposure if you choose to use a DR mode with an increased ISO setting rather than using DR100 at base ISO), and it should be clarified that ISO isn’t exposure, it can affect image lightness like exposure, but doesn’t contribute to sensor exposure in any way and must be considered independently.
And then it will selectively digitally raise the pixels exposure in post processing for shadow areas back to the original exposure.
Basically, by either a reduction in exposure (unavoidable if you choose to use a DR mode rather than base ISO), or a reduction of analog pre -D to A converter gain (which, if you’d be at a higher ISO anyway, doesn’t come with any major downsides), a one or two stop darker than normal RAW file is produced in either case to increase highlight headroom at “the right”. The resulting loss of image lightness is compensated for digitally in the jpeg processing with a highlight curve that will include detail recorded one or two stops beyond where the clipping point would have been at DR100 - flatter, but non-blown highlights.
It sounds reasonable, and that's how it is explained in most videos/blogs. But how do you actually use this thing practically?

For example, if the scene is already correctly exposed (without highlight clips), by turning on DR400, you are just introducing noise in shadow for no real gain isnt it?
Yes. In most cases, some positive exposure compensation is going to be part of the equation - at DR400 you will have an extra 2 stops of highlight headroom - up to 2 stops of positive exposure compensation to increase the shadow and mid tone lightness to where it needs to be for a good looking SOOC jpeg. IMO, DR200 (+1 stop of available headroom) will be plenty for most “normal” high DR situations. I highly recommend turning on the highlight warning blinkies which will show you when (and where) you’ve reached the limit.
Therefore, for a contract scene, it is advantageous to over expose and clip the highlights, and turn on DR400. At least in theory. I suppose one can do this with exp comp dial. But is this really the intended usage? It's seems so backwards.
You’re not overexposing, you are managing SOOC image lightness. By setting a DR mode, you are recording a darker than normal RAW, increasing exposure compensation allows you to brighten the shadows and mid tones without clipping the RAW highlights. A DR400 RAW with +2 stops of exp. comp. is functionally identical to a DR100 RAW with no exp. comp.

Using DR200/400 mode when DR100 and base ISO was feasible will indeed result in a relative 1 or 2 stop under underexposure, and a darker RAW file (with a metadata tag that usually automatically corrects the lightness in most RAW editors) but, contrary to what the Pal2Tech guy says, no manipulation of any tone curves or anything like that is going on in the RAW data. I don’t shoot jpegs at all and always use DR100, but the DR modes can help with getting better looking SOOC jpegs in high DR situations (I recommend avoiding the DR modes for jpegs too unless you actually have a high dynamic range situation that calls for them).
Ok but let’s simplify it. If I’m trying to take a shot of a contrast scene as jpeg. How should one do it? From my understanding, it’s to overexpose by 1-2 stops (shows as wall on the far right histo) and then turn on DR400. Correct?
Correct. But you need to back up here. What constitutes overexposure, or underexposure for that matter. How would you determine an overexposure of 1 stop? Would you adjust the camera meter till it indicated +1? Would you chimp the JPEG on the screen? Is the camera JPEG the basis you would use to determine correct/under/overexposure?

Here's what you just described. First a high contrast scene at ISO 200 DR100 with the exposure set to expose as much as possible without clipping highlights -- 1/500, f/8. The shadows are too dark, the highlights are right on the edge of clipping in the camera JPEG as well as the raw file.

00d6ebe8bc0b403d910f501c2c52b1f9.jpg

I raised the ISO to 800 and set DR400 then exposed at 1/500, f/11 and here's that camera JPEG. It's a much better camera JPEG. Did I overexpose or underexpose, based on what? Based on my previous ISO 200 exposure an increase of 2 stops ISO would be compensated by a two stop exposure change. Assuming the camera meter was zeroed for the ISO 200 shot then for the ISO 800 shot the meter would have indicated f/16. At f/11 am I overexposing? The JPEG doesn't look it. Compared with the ISO 200 exposure the sensor is being exposed 1 stop less. And is in fact underexposed given the sensor's actual recording capacity.

3b6f55234f1d4776864f52594f849eae.jpg

The first photo, the ISO DR100 shot. Is that over or underexposed? It's a bad camera JPEG because of the lighting contrast. Do all the too dark shadows make it an underexposure? The sensor exposure is in fact perfect in that it makes maximum use of the sensor's recording capacity and as such records the maximum possible DR. Here's the processed raw file. Is it over or underexposed?



de9b918b1d914fdbac6c3adda7d6ccb6.jpg
 

Attachments

  • d032b1d9c1614c33a62e10e32406a59e.jpg
    d032b1d9c1614c33a62e10e32406a59e.jpg
    537 KB · Views: 0
Last edited:
Im new to fuji system (coming from Sony) and I'm trying to get into this new mindset of shooting jpeg not raws.

One of things to help get better jpegs is this DR concept in Fuji (other cameras have similar settings but I dont think they work the same way).

I've done research on this topic for a week now. Watched over 20 youtube videos and have read all the forum posts and blog posts.

Basically what I gather is that DR200/DR400 causes the camera to underexpose (via ISO) by 1 or 2 stops.
Not always (there is only a reduction in exposure if you choose to use a DR mode with an increased ISO setting rather than using DR100 at base ISO), and it should be clarified that ISO isn’t exposure, it can affect image lightness like exposure, but doesn’t contribute to sensor exposure in any way and must be considered independently.
And then it will selectively digitally raise the pixels exposure in post processing for shadow areas back to the original exposure.
Basically, by either a reduction in exposure (unavoidable if you choose to use a DR mode rather than base ISO), or a reduction of analog pre -D to A converter gain (which, if you’d be at a higher ISO anyway, doesn’t come with any major downsides), a one or two stop darker than normal RAW file is produced in either case to increase highlight headroom at “the right”. The resulting loss of image lightness is compensated for digitally in the jpeg processing with a highlight curve that will include detail recorded one or two stops beyond where the clipping point would have been at DR100 - flatter, but non-blown highlights.
It sounds reasonable, and that's how it is explained in most videos/blogs. But how do you actually use this thing practically?

For example, if the scene is already correctly exposed (without highlight clips), by turning on DR400, you are just introducing noise in shadow for no real gain isnt it?
Yes. In most cases, some positive exposure compensation is going to be part of the equation - at DR400 you will have an extra 2 stops of highlight headroom - up to 2 stops of positive exposure compensation to increase the shadow and mid tone lightness to where it needs to be for a good looking SOOC jpeg. IMO, DR200 (+1 stop of available headroom) will be plenty for most “normal” high DR situations. I highly recommend turning on the highlight warning blinkies which will show you when (and where) you’ve reached the limit.
Therefore, for a contract scene, it is advantageous to over expose and clip the highlights, and turn on DR400. At least in theory. I suppose one can do this with exp comp dial. But is this really the intended usage? It's seems so backwards.
You’re not overexposing, you are managing SOOC image lightness. By setting a DR mode, you are recording a darker than normal RAW, increasing exposure compensation allows you to brighten the shadows and mid tones without clipping the RAW highlights. A DR400 RAW with +2 stops of exp. comp. is functionally identical to a DR100 RAW with no exp. comp.

Using DR200/400 mode when DR100 and base ISO was feasible will indeed result in a relative 1 or 2 stop under underexposure, and a darker RAW file (with a metadata tag that usually automatically corrects the lightness in most RAW editors) but, contrary to what the Pal2Tech guy says, no manipulation of any tone curves or anything like that is going on in the RAW data. I don’t shoot jpegs at all and always use DR100, but the DR modes can help with getting better looking SOOC jpegs in high DR situations (I recommend avoiding the DR modes for jpegs too unless you actually have a high dynamic range situation that calls for them).
Ok but let’s simplify it. If I’m trying to take a shot of a contrast scene as jpeg. How should one do it? From my understanding, it’s to overexpose by 1-2 stops (shows as wall on the far right histo) and then turn on DR400. Correct?
I would set the correct exposure (exposing to the right) at DR100 manually, or with exposure compensation using the highlight warning blinkies to gauge highlight exposure (very accurate in Fuji cameras - with Natural Live view off). The metering got his one right with zero exposure compensation. If the shadows/midtones look too dark in EVF with the highlights exposed correctly, I'd switch to DR200 and bump up the exposure compensation (up to +1 additional stop) until the shadows and midtones are bright enough and shoot. If +1 stop isn't enough to get the shadows and mid tones right, I'd switch to DR400 and bump up the EC up to one more stop.

Here's a DR100 vs. a DR400 shot with the highlights correctly exposed for DR100, as you can see, they look about the same except for the flatter highlights at DR800 (and noisier shadows).

With no exp. comp. the DR100 image (L) definitely looks better, but both are darker than I'd like (this was very bright midday sun). No point in using a DR mode in this case without any positive EC.
With no exp. comp. the DR100 image (L) definitely looks better, but both are darker than I'd like (this was very bright midday sun). No point in using a DR mode in this case without any positive EC.

So what do you do? I'd just process the optimally exposed DR100 RAW file...

Lightroom processed RAW, DR100 base ISO.
Lightroom processed RAW, DR100 base ISO.

But, for a SOOC jpeg, I'd switch to DR200 and bump up the exposure compensation...

This is DR400 with +1 stop of exposure compensation (+2/3 would have been better). Note: at +1 stop of exposure compensation at DR400, the exposure for this shot (and the resulting RAW file) is functionally identical to the DR100 base ISO shot.
This is DR400 with +1 stop of exposure compensation (+2/3 would have been better). Note: at +1 stop of exposure compensation at DR400, the exposure for this shot (and the resulting RAW file) is functionally identical to the DR100 base ISO shot.
 
Last edited:
Ok but let’s simplify it. If I’m trying to take a shot of a contrast scene as jpeg. How should one do it? From my understanding, it’s to overexpose by 1-2 stops (shows as wall on the far right histo) and then turn on DR400. Correct?
Simple answer:

No. The whole point of the DR modes is that you don't need to do anything different in terms of exposure.

If you're in a higher contrast scene, then DR200 will give you an extra stop of highlights to cope with it. DR400 will give you two, if it's really high contrast.

Explanation:

The DR modes change the relationship between amplification and exposure. (To avoid confusion ISO does not equate to amplification).

- In a bright light situation, where you would usually choose base ISO (let's assume that's ISO 100: I'm not sure which model you're using), you notice the highlights are clipping.

You turn on DR200. It can't reduce amplification (you're already at base) so it has to reduce exposure. So it chooses a reduced exposure, consistent with ISO 200, to capture an extra stop of highlights. But it doesn't increase the amplification: it remains at its base state (otherwise the amplification would push the extra highlights back to clipping). The extra highlights are then incorporated into the JPEG.

You haven't had to apply any exposure comp: moving from DR100 to DR200 changed the amplication/exposure relationship to capture and accommodate another stop of highlights.

- In low light settings, say you want to make sure the bright areas in the scene don't clip.

Let's say you're at ISO 3200. You move from DR100 to DR200. Unlike the base ISO situation, the camera is already applying a lot of amplification, so it's now able to reduce the amplification by a stop and maintain the same exposure. The effect is still the same, though: the capture/retention of an additional stop of highlight data and the use of the DR200 JPEG tone curve to incorporate them into the JPEG.

In both instances you don't need to change exposure: the camera will change exposure or amplification to give a combination of the two that captures and retains an extra stop of highlight information.

In bright conditions, where the camera has no choice but to reduce exposure (none of these examples involve under- or over-exposure), then yes, you'll pay a 1 stop noise cost in the shadows. Noise reduction is usually increased in the JPEG to hide this.

This is unavoidable. There is no way, on any camera, to increase highlight capture at base ISO without reducing exposure and hence increasing noise. You have to make that choice: are the less clipped hightlights worth the extra shadow noise.

Richard - DPReview.com
 
Last edited:
Im new to fuji system (coming from Sony) and I'm trying to get into this new mindset of shooting jpeg not raws.

One of things to help get better jpegs is this DR concept in Fuji (other cameras have similar settings but I dont think they work the same way).

I've done research on this topic for a week now. Watched over 20 youtube videos and have read all the forum posts and blog posts.

Basically what I gather is that DR200/DR400 causes the camera to underexpose (via ISO) by 1 or 2 stops.
Not always (there is only a reduction in exposure if you choose to use a DR mode with an increased ISO setting rather than using DR100 at base ISO), and it should be clarified that ISO isn’t exposure, it can affect image lightness like exposure, but doesn’t contribute to sensor exposure in any way and must be considered independently.
And then it will selectively digitally raise the pixels exposure in post processing for shadow areas back to the original exposure.
Basically, by either a reduction in exposure (unavoidable if you choose to use a DR mode rather than base ISO), or a reduction of analog pre -D to A converter gain (which, if you’d be at a higher ISO anyway, doesn’t come with any major downsides), a one or two stop darker than normal RAW file is produced in either case to increase highlight headroom at “the right”. The resulting loss of image lightness is compensated for digitally in the jpeg processing with a highlight curve that will include detail recorded one or two stops beyond where the clipping point would have been at DR100 - flatter, but non-blown highlights.
It sounds reasonable, and that's how it is explained in most videos/blogs. But how do you actually use this thing practically?

For example, if the scene is already correctly exposed (without highlight clips), by turning on DR400, you are just introducing noise in shadow for no real gain isnt it?
Yes. In most cases, some positive exposure compensation is going to be part of the equation - at DR400 you will have an extra 2 stops of highlight headroom - up to 2 stops of positive exposure compensation to increase the shadow and mid tone lightness to where it needs to be for a good looking SOOC jpeg. IMO, DR200 (+1 stop of available headroom) will be plenty for most “normal” high DR situations. I highly recommend turning on the highlight warning blinkies which will show you when (and where) you’ve reached the limit.
Therefore, for a contract scene, it is advantageous to over expose and clip the highlights, and turn on DR400. At least in theory. I suppose one can do this with exp comp dial. But is this really the intended usage? It's seems so backwards.
You’re not overexposing, you are managing SOOC image lightness. By setting a DR mode, you are recording a darker than normal RAW, increasing exposure compensation allows you to brighten the shadows and mid tones without clipping the RAW highlights. A DR400 RAW with +2 stops of exp. comp. is functionally identical to a DR100 RAW with no exp. comp.

Using DR200/400 mode when DR100 and base ISO was feasible will indeed result in a relative 1 or 2 stop under underexposure, and a darker RAW file (with a metadata tag that usually automatically corrects the lightness in most RAW editors) but, contrary to what the Pal2Tech guy says, no manipulation of any tone curves or anything like that is going on in the RAW data. I don’t shoot jpegs at all and always use DR100, but the DR modes can help with getting better looking SOOC jpegs in high DR situations (I recommend avoiding the DR modes for jpegs too unless you actually have a high dynamic range situation that calls for them).
Ok but let’s simplify it. If I’m trying to take a shot of a contrast scene as jpeg. How should one do it? From my understanding, it’s to overexpose by 1-2 stops (shows as wall on the far right histo) and then turn on DR400. Correct?
Correct. But you need to back up here. What constitutes overexposure, or underexposure for that matter. How would you determine an overexposure of 1 stop? Would you adjust the camera meter till it indicated +1? Would you chimp the JPEG on the screen? Is the camera JPEG the basis you would use to determine correct/under/overexposure?

Here's what you just described. First a high contrast scene at ISO 200 DR100 with the exposure set to expose as much as possible without clipping highlights -- 1/500, f/8. The shadows are too dark, the highlights are right on the edge of clipping in the camera JPEG as well as the raw file.

00d6ebe8bc0b403d910f501c2c52b1f9.jpg

I raised the ISO to 800 and set DR400 then exposed at 1/500, f/11 and here's that camera JPEG. It's a much better camera JPEG. Did I overexpose or underexpose, based on what? Based on my previous ISO 200 exposure an increase of 2 stops ISO would be compensated by a two stop exposure change. Assuming the camera meter was zeroed for the ISO 200 shot then for the ISO 800 shot the meter would have indicated f/16. At f/11 am I overexposing? The JPEG doesn't look it. Compared with the ISO 200 exposure the sensor is being exposed 1 stop less. And is in fact underexposed given the sensor's actual recording capacity.

3b6f55234f1d4776864f52594f849eae.jpg

The first photo, the ISO DR100 shot. Is that over or underexposed? It's a bad camera JPEG because of the lighting contrast. Do all the too dark shadows make it an underexposure? The sensor exposure is in fact perfect in that it makes maximum use of the sensor's recording capacity and as such records the maximum possible DR. Here's the processed raw file. Is it over or underexposed?

de9b918b1d914fdbac6c3adda7d6ccb6.jpg
well no, I wouldnt use the EV value or anything. I would just look at the histogram. I cant see the histogram of your sample images so I cant speak to that. But a high contrast scene like this one, I am basically choosing between clipped shadows or clipped highlights (evident in histogram - wall on left, or wall on right). In this case with the DR function, Im asking, is it correct to do wall on right, and then set to DR400 for jpeg shooting.
 
Im new to fuji system (coming from Sony) and I'm trying to get into this new mindset of shooting jpeg not raws.

One of things to help get better jpegs is this DR concept in Fuji (other cameras have similar settings but I dont think they work the same way).

I've done research on this topic for a week now. Watched over 20 youtube videos and have read all the forum posts and blog posts.

Basically what I gather is that DR200/DR400 causes the camera to underexpose (via ISO) by 1 or 2 stops.
Not always (there is only a reduction in exposure if you choose to use a DR mode with an increased ISO setting rather than using DR100 at base ISO), and it should be clarified that ISO isn’t exposure, it can affect image lightness like exposure, but doesn’t contribute to sensor exposure in any way and must be considered independently.
And then it will selectively digitally raise the pixels exposure in post processing for shadow areas back to the original exposure.
Basically, by either a reduction in exposure (unavoidable if you choose to use a DR mode rather than base ISO), or a reduction of analog pre -D to A converter gain (which, if you’d be at a higher ISO anyway, doesn’t come with any major downsides), a one or two stop darker than normal RAW file is produced in either case to increase highlight headroom at “the right”. The resulting loss of image lightness is compensated for digitally in the jpeg processing with a highlight curve that will include detail recorded one or two stops beyond where the clipping point would have been at DR100 - flatter, but non-blown highlights.
It sounds reasonable, and that's how it is explained in most videos/blogs. But how do you actually use this thing practically?

For example, if the scene is already correctly exposed (without highlight clips), by turning on DR400, you are just introducing noise in shadow for no real gain isnt it?
Yes. In most cases, some positive exposure compensation is going to be part of the equation - at DR400 you will have an extra 2 stops of highlight headroom - up to 2 stops of positive exposure compensation to increase the shadow and mid tone lightness to where it needs to be for a good looking SOOC jpeg. IMO, DR200 (+1 stop of available headroom) will be plenty for most “normal” high DR situations. I highly recommend turning on the highlight warning blinkies which will show you when (and where) you’ve reached the limit.
Therefore, for a contract scene, it is advantageous to over expose and clip the highlights, and turn on DR400. At least in theory. I suppose one can do this with exp comp dial. But is this really the intended usage? It's seems so backwards.
You’re not overexposing, you are managing SOOC image lightness. By setting a DR mode, you are recording a darker than normal RAW, increasing exposure compensation allows you to brighten the shadows and mid tones without clipping the RAW highlights. A DR400 RAW with +2 stops of exp. comp. is functionally identical to a DR100 RAW with no exp. comp.

Using DR200/400 mode when DR100 and base ISO was feasible will indeed result in a relative 1 or 2 stop under underexposure, and a darker RAW file (with a metadata tag that usually automatically corrects the lightness in most RAW editors) but, contrary to what the Pal2Tech guy says, no manipulation of any tone curves or anything like that is going on in the RAW data. I don’t shoot jpegs at all and always use DR100, but the DR modes can help with getting better looking SOOC jpegs in high DR situations (I recommend avoiding the DR modes for jpegs too unless you actually have a high dynamic range situation that calls for them).
Ok but let’s simplify it. If I’m trying to take a shot of a contrast scene as jpeg. How should one do it? From my understanding, it’s to overexpose by 1-2 stops (shows as wall on the far right histo) and then turn on DR400. Correct?
Correct. But you need to back up here. What constitutes overexposure, or underexposure for that matter. How would you determine an overexposure of 1 stop? Would you adjust the camera meter till it indicated +1? Would you chimp the JPEG on the screen? Is the camera JPEG the basis you would use to determine correct/under/overexposure?

Here's what you just described. First a high contrast scene at ISO 200 DR100 with the exposure set to expose as much as possible without clipping highlights -- 1/500, f/8. The shadows are too dark, the highlights are right on the edge of clipping in the camera JPEG as well as the raw file.

00d6ebe8bc0b403d910f501c2c52b1f9.jpg

I raised the ISO to 800 and set DR400 then exposed at 1/500, f/11 and here's that camera JPEG. It's a much better camera JPEG. Did I overexpose or underexpose, based on what? Based on my previous ISO 200 exposure an increase of 2 stops ISO would be compensated by a two stop exposure change. Assuming the camera meter was zeroed for the ISO 200 shot then for the ISO 800 shot the meter would have indicated f/16. At f/11 am I overexposing? The JPEG doesn't look it. Compared with the ISO 200 exposure the sensor is being exposed 1 stop less. And is in fact underexposed given the sensor's actual recording capacity.

3b6f55234f1d4776864f52594f849eae.jpg

The first photo, the ISO DR100 shot. Is that over or underexposed? It's a bad camera JPEG because of the lighting contrast. Do all the too dark shadows make it an underexposure? The sensor exposure is in fact perfect in that it makes maximum use of the sensor's recording capacity and as such records the maximum possible DR. Here's the processed raw file. Is it over or underexposed?

de9b918b1d914fdbac6c3adda7d6ccb6.jpg
well no, I wouldnt use the EV value or anything. I would just look at the histogram. I cant see the histogram of your sample images so I cant speak to that. But a high contrast scene like this one, I am basically choosing between clipped shadows or clipped highlights (evident in histogram - wall on left, or wall on right).
So you're shooting camera JPEGs? Shooting raw you don't have to make that choice, you can have both.
In this case with the DR function, Im asking, is it correct to do wall on right, and then set to DR400 for jpeg shooting.
Yes.

But, what histogram are you using to determine exposure? The single live-view histogram in the Fuji display is a luminosity histogram. Not a great choice as an exposure aid as you'll often clip color channels. Fuji also provides a larger optional RGB histogram that you can assign to a function button. Are you looking at those RGB histograms?

In either case the camera histograms apply to the JPEG the camera will create and not to the raw file. If you want camera JPEGs without post processing the histogram should be a supplemental check -- determine exposure by assessing the viewfinder image.
 
Im new to fuji system (coming from Sony) and I'm trying to get into this new mindset of shooting jpeg not raws.

One of things to help get better jpegs is this DR concept in Fuji (other cameras have similar settings but I dont think they work the same way).

I've done research on this topic for a week now. Watched over 20 youtube videos and have read all the forum posts and blog posts.

Basically what I gather is that DR200/DR400 causes the camera to underexpose (via ISO) by 1 or 2 stops.
Not always (there is only a reduction in exposure if you choose to use a DR mode with an increased ISO setting rather than using DR100 at base ISO), and it should be clarified that ISO isn’t exposure, it can affect image lightness like exposure, but doesn’t contribute to sensor exposure in any way and must be considered independently.
And then it will selectively digitally raise the pixels exposure in post processing for shadow areas back to the original exposure.
Basically, by either a reduction in exposure (unavoidable if you choose to use a DR mode rather than base ISO), or a reduction of analog pre -D to A converter gain (which, if you’d be at a higher ISO anyway, doesn’t come with any major downsides), a one or two stop darker than normal RAW file is produced in either case to increase highlight headroom at “the right”. The resulting loss of image lightness is compensated for digitally in the jpeg processing with a highlight curve that will include detail recorded one or two stops beyond where the clipping point would have been at DR100 - flatter, but non-blown highlights.
It sounds reasonable, and that's how it is explained in most videos/blogs. But how do you actually use this thing practically?

For example, if the scene is already correctly exposed (without highlight clips), by turning on DR400, you are just introducing noise in shadow for no real gain isnt it?
Yes. In most cases, some positive exposure compensation is going to be part of the equation - at DR400 you will have an extra 2 stops of highlight headroom - up to 2 stops of positive exposure compensation to increase the shadow and mid tone lightness to where it needs to be for a good looking SOOC jpeg. IMO, DR200 (+1 stop of available headroom) will be plenty for most “normal” high DR situations. I highly recommend turning on the highlight warning blinkies which will show you when (and where) you’ve reached the limit.
Therefore, for a contract scene, it is advantageous to over expose and clip the highlights, and turn on DR400. At least in theory. I suppose one can do this with exp comp dial. But is this really the intended usage? It's seems so backwards.
You’re not overexposing, you are managing SOOC image lightness. By setting a DR mode, you are recording a darker than normal RAW, increasing exposure compensation allows you to brighten the shadows and mid tones without clipping the RAW highlights. A DR400 RAW with +2 stops of exp. comp. is functionally identical to a DR100 RAW with no exp. comp.

Using DR200/400 mode when DR100 and base ISO was feasible will indeed result in a relative 1 or 2 stop under underexposure, and a darker RAW file (with a metadata tag that usually automatically corrects the lightness in most RAW editors) but, contrary to what the Pal2Tech guy says, no manipulation of any tone curves or anything like that is going on in the RAW data. I don’t shoot jpegs at all and always use DR100, but the DR modes can help with getting better looking SOOC jpegs in high DR situations (I recommend avoiding the DR modes for jpegs too unless you actually have a high dynamic range situation that calls for them).
Ok but let’s simplify it. If I’m trying to take a shot of a contrast scene as jpeg. How should one do it? From my understanding, it’s to overexpose by 1-2 stops (shows as wall on the far right histo) and then turn on DR400. Correct?
Correct. But you need to back up here. What constitutes overexposure, or underexposure for that matter. How would you determine an overexposure of 1 stop? Would you adjust the camera meter till it indicated +1? Would you chimp the JPEG on the screen? Is the camera JPEG the basis you would use to determine correct/under/overexposure?

Here's what you just described. First a high contrast scene at ISO 200 DR100 with the exposure set to expose as much as possible without clipping highlights -- 1/500, f/8. The shadows are too dark, the highlights are right on the edge of clipping in the camera JPEG as well as the raw file.

00d6ebe8bc0b403d910f501c2c52b1f9.jpg

I raised the ISO to 800 and set DR400 then exposed at 1/500, f/11 and here's that camera JPEG. It's a much better camera JPEG. Did I overexpose or underexpose, based on what? Based on my previous ISO 200 exposure an increase of 2 stops ISO would be compensated by a two stop exposure change. Assuming the camera meter was zeroed for the ISO 200 shot then for the ISO 800 shot the meter would have indicated f/16. At f/11 am I overexposing? The JPEG doesn't look it. Compared with the ISO 200 exposure the sensor is being exposed 1 stop less. And is in fact underexposed given the sensor's actual recording capacity.

3b6f55234f1d4776864f52594f849eae.jpg

The first photo, the ISO DR100 shot. Is that over or underexposed? It's a bad camera JPEG because of the lighting contrast. Do all the too dark shadows make it an underexposure? The sensor exposure is in fact perfect in that it makes maximum use of the sensor's recording capacity and as such records the maximum possible DR. Here's the processed raw file. Is it over or underexposed?

de9b918b1d914fdbac6c3adda7d6ccb6.jpg
well no, I wouldnt use the EV value or anything. I would just look at the histogram. I cant see the histogram of your sample images so I cant speak to that. But a high contrast scene like this one, I am basically choosing between clipped shadows or clipped highlights (evident in histogram - wall on left, or wall on right).
So you're shooting camera JPEGs? Shooting raw you don't have to make that choice, you can have both.
In this case with the DR function, Im asking, is it correct to do wall on right, and then set to DR400 for jpeg shooting.
Yes.

But, what histogram are you using to determine exposure? The single live-view histogram in the Fuji display is a luminosity histogram. Not a great choice as an exposure aid as you'll often clip color channels. Fuji also provides a larger optional RGB histogram that you can assign to a function button. Are you looking at those RGB histograms?

In either case the camera histograms apply to the JPEG the camera will create and not to the raw file. If you want camera JPEGs without post processing the histogram should be a supplemental check -- determine exposure by assessing the viewfinder image.
I dont care about raw files, I dont know why you keep coming back to raw files. The thread was started with a discussion about jpeg files. With a digression about raw that was clearly labelled as "side discussion".

But yes you correct, I am only looking at the luminosity histo. I will switch to the color ones for this strategy. But mostly I'm just looking for confirmation on the strategy itself, as it seems a bit awkward to have to overexpose.
 
Ok but let’s simplify it. If I’m trying to take a shot of a contrast scene as jpeg. How should one do it? From my understanding, it’s to overexpose by 1-2 stops (shows as wall on the far right histo) and then turn on DR400. Correct?
Simple answer:

No. The whole point of the DR modes is that you don't need to do anything different in terms of exposure.

If you're in a higher contrast scene, then DR200 will give you an extra stop of highlights to cope with it. DR400 will give you two, if it's really high contrast.
Without trying to analyze your detailed explanation (which I admit I dont fully follow), I dont get why this is a no?

Say you are in a high contrast scene. Typically (without any DR magic), you have to choose between clipping the shadows or clipping the highlights. You can't win both for a sooc jpeg.

Now, with DR magic, you can set it to DR400. Great, but the question still remains, do you expose it such that you are clipping the shadow or highlights? It has to be the highlights.

In fact, this is easily testable. I went to the window of my house where the outdoor is blasted with the sun and the inside wall is dark. If I set it to DR400 and expose it such that the highlights are all retained while clipping the shadows. DR400 did not have an effect. If I overexpose and clip the highlights instead with DR400, the highlights would get some details back while the shadows are lifted.
 

Keyboard shortcuts

Back
Top