DRO (D-Range Optimizer) setting for raw shooters

[A74Me]

There is woefully little documentation about what DRO does, and I am unfamiliar with that setting. After some tests with a7cr, I believe the following describes its operation.

It seems to operate differently than other manufacturer's similar operations, as it does not affect the raw files. Raw files created with DRO on and off look identical. It does affect JPEGs as it seems to lift the shadows and mids. Most importantly, it does not affect highlight clipping. That means that zebras are equally reliable with DRO on and off.

My main issue with EVFs is that often I do not see details in deep shadows. Turning DRO on helps me with framing.

Almost correct.

Fully correct

DRO operates you have descibed. If you are shooting RAW it doesn't touch the RAW files. However, the histogram and zebras in the evf are based on the jpegs, not the RAW versions. You therefore have considerably more headroom for the highlights in RAW than is being shown by the zebras and histogram.

The op is talking about the effect of DRO. It lifts shadows and mids, this means it does not reduce headroom. Zebras are the same with DRO on or off, this is the point of the op.

No, DRO doesn't reduce headroom, yes, the zebras are the same. However (the OP heading is discussing DRO for RAW files) the point is the accuracy of the zebras and histograms for the RAW files. What I am pointing out is that there is more headroom for the highlights in the RAW files than appears to be the case when you look at the zebras and histogram/clipping on the evf histogram. ie., while it is true as the OP says that the zebras are equally reliable for highlights with DRO turned on or off, they are not reliable indicators of the highlight headroom available in the RAW files.

thats totally incorrect.

the histogram nearly mimics the raw file. its just that it doesnt have the resolution to see it with human eyes. fast raw viewer will also show you how accurate sony's zebras and histogram is shooting raw.

Here are the Fast Raw Viewer histograms for the raw and in-camera JPEG files of the same image.


raw


JPEG

The in-camera histogram will NOT mimic the raw histogram above, it will look like the green channel of the JPEG histogram above.

The author of Fast Raw Viewer explains: https://www.fastrawviewer.com/raw-histogram-for-culling, to wit, "How to Trash a Good Shot in One Step...Not sure how? It really is very simple - just rely on the JPEG histogram."

ive been through this a million times and no one can prove it doesnt . post an image of the histogram on a sony camera or better still set up your fast raw viewer without the software default margins of error. you can also set up the over exposure hot spots in fast raw viewer and they will exactly match the incamea blinkies set to 109+ .

The point is that DRO or not the situation is identical. There is no help you get if you look at highlight clipping only as the camera does not give warnings for crushing the blacks which is what DRO combats or tries to

Sony cameras do not manage that well bright scenes they seem optimised for shadows lifting so underexposing a little is not a major issue

i disagree with that and so does Dustin abbots review of the a74 compared to the latest fuji 100 ii MF as the a74 out performs it in the hightlights.

I don't need dustin abbot I have my old Panasonic GH5M2 and I put them on a tripod on the same scene and took photos with a remote shutter. The clipping point is the same the Sony manages better shadow recovery.

Only at ISO 100 you get a benefit and then at ISO 500 for my A1 the performance matches again thanks for dual gain. For the rest is very much the same soup

I have no idea what Fuji does I have zero interest in their foveon sensors they are not comparable to a bayer array

since when is the 100 meg mf sensor foveon

 

[Interceptor121]

There is woefully little documentation about what DRO does, and I am unfamiliar with that setting. After some tests with a7cr, I believe the following describes its operation.

It seems to operate differently than other manufacturer's similar operations, as it does not affect the raw files. Raw files created with DRO on and off look identical. It does affect JPEGs as it seems to lift the shadows and mids. Most importantly, it does not affect highlight clipping. That means that zebras are equally reliable with DRO on and off.

My main issue with EVFs is that often I do not see details in deep shadows. Turning DRO on helps me with framing.

Almost correct.

Fully correct

DRO operates you have descibed. If you are shooting RAW it doesn't touch the RAW files. However, the histogram and zebras in the evf are based on the jpegs, not the RAW versions. You therefore have considerably more headroom for the highlights in RAW than is being shown by the zebras and histogram.

The op is talking about the effect of DRO. It lifts shadows and mids, this means it does not reduce headroom. Zebras are the same with DRO on or off, this is the point of the op.

No, DRO doesn't reduce headroom, yes, the zebras are the same. However (the OP heading is discussing DRO for RAW files) the point is the accuracy of the zebras and histograms for the RAW files. What I am pointing out is that there is more headroom for the highlights in the RAW files than appears to be the case when you look at the zebras and histogram/clipping on the evf histogram. ie., while it is true as the OP says that the zebras are equally reliable for highlights with DRO turned on or off, they are not reliable indicators of the highlight headroom available in the RAW files.

thats totally incorrect.

the histogram nearly mimics the raw file. its just that it doesnt have the resolution to see it with human eyes. fast raw viewer will also show you how accurate sony's zebras and histogram is shooting raw.

Here are the Fast Raw Viewer histograms for the raw and in-camera JPEG files of the same image.


raw


JPEG

The in-camera histogram will NOT mimic the raw histogram above, it will look like the green channel of the JPEG histogram above.

The author of Fast Raw Viewer explains: https://www.fastrawviewer.com/raw-histogram-for-culling, to wit, "How to Trash a Good Shot in One Step...Not sure how? It really is very simple - just rely on the JPEG histogram."

ive been through this a million times and no one can prove it doesnt . post an image of the histogram on a sony camera or better still set up your fast raw viewer without the software default margins of error. you can also set up the over exposure hot spots in fast raw viewer and they will exactly match the incamea blinkies set to 109+ .

The point is that DRO or not the situation is identical. There is no help you get if you look at highlight clipping only as the camera does not give warnings for crushing the blacks which is what DRO combats or tries to

Sony cameras do not manage that well bright scenes they seem optimised for shadows lifting so underexposing a little is not a major issue

i disagree with that and so does Dustin abbots review of the a74 compared to the latest fuji 100 ii MF as the a74 out performs it in the hightlights.

I don't need dustin abbot I have my old Panasonic GH5M2 and I put them on a tripod on the same scene and took photos with a remote shutter. The clipping point is the same the Sony manages better shadow recovery.

Only at ISO 100 you get a benefit and then at ISO 500 for my A1 the performance matches again thanks for dual gain. For the rest is very much the same soup

I have no idea what Fuji does I have zero interest in their foveon sensors they are not comparable to a bayer array

since when is the 100 meg mf sensor foveon

I don't know and I don't care. As I said I shot the cameras I have side by side I don't need Dustin Abbot nor your to point me at something you don't have or you have not tried

When you have a few cameras give it a go it is instructive

Edit just checked the data of the fuji 100 and in fact the well capacity seems low too lower than a Sony A7 IV

Seems like lots of manufacturers are concerned about scene DR not absolute DR

--
If you like my image I would appreciate if you follow me on social media
instagram http://instagram.com/interceptor121
My flickr sets http://www.flickr.com/photos/interceptor121/
Youtube channel http://www.youtube.com/interceptor121
Underwater Photo and Video Blog http://interceptor121.com
Deer Photography workshops https://interceptor121.com/2021/09/26/2021-22-deer-photography-workshops-in-woburn/
If you want to get in touch don't send me a PM rather contact me directly at my website/social media

 

[Interceptor121]

Just tested on my A1 the DRO bracket functionality set to high

This is a hint on what the functionality does

The camera in fact takes one shot and then produces 3 HEIF files out of it with DRO 1,3,5 so in the case of the A1 the DRO now does not seem to change exposure but only affect the settings

The change of DRO also affects the EVF real time which is I guess what the op is hinting so you do see brighter shadows if you set it to 5

The situation has changed over the years since the original implementation and I would not exclude that the tuning has been updated at set intervals. In particular the last generation of sensor may not have a decrease of exposure compared to older ones that were not back illuminated

That is reflects or not for RAW I think is irrelevant as the cameras do not have a shadow warning on the live view so you don't know if you crushed the shadows or not

From my perspective this functionality is somewhat irrelevant but I can see if you like to shoot JPEG/HEIF this could be helpful

With regards to the EVF if you set live view to off you can see also in the dark but of course you don't have any functioning zebra warnings

 

[SrMi]

Just tested on my A1 the DRO bracket functionality set to high

This is a hint on what the functionality does

The camera in fact takes one shot and then produces 3 HEIF files out of it with DRO 1,3,5 so in the case of the A1 the DRO now does not seem to change exposure but only affect the settings

The change of DRO also affects the EVF real time which is I guess what the op is hinting so you do see brighter shadows if you set it to 5

The situation has changed over the years since the original implementation and I would not exclude that the tuning has been updated at set intervals. In particular the last generation of sensor may not have a decrease of exposure compared to older ones that were not back illuminated

That is reflects or not for RAW I think is irrelevant as the cameras do not have a shadow warning on the live view so you don't know if you crushed the shadows or not

From my perspective this functionality is somewhat irrelevant but I can see if you like to shoot JPEG/HEIF this could be helpful

With regards to the EVF if you set live view to off you can see also in the dark but of course you don't have any functioning zebra warnings

I have the Live View Disp. Select assigned to a button. It helps in high-contrast situations when highlights must be preserved, and the mids and shadows become very dark. Unfortunately, it is easy to forget to turn it back on, and there is no feedback in EVF and the "For Viewfinder" display.

 

[Neil12345]

Hello

I have an A7C and use DRO all the time, for jpgs and raw photos. If DRO is used in camera, the jpg preview shows the resulting exposure, I use DRO Auto as default, the camera judges a level to use based on the scene.

The DRO level data is stored in the Raw file, but is only interpreted by the Sony Imaging Edge Desktop software, which shows a raw image with the DRO setting as made in camera. This means that any other editing software is simply not showing the benefit of the DRO setting. In IED, the DRO level can be adjusted to off, auto or a level up to DRO 10. The max level in camera is DRO 5. Note that DRO can be applied to a Raw photo in IED, even if the photo was taken with DRO off.

DRO is not just changing shadows / highlights / contrast across the photo, it does a multiple section analysis of the photo to adjust brightness and contrast, on a pixel by pixel basis.

For high backlight scenes, I often expand to DRO 10 in post. This is equivalent to adding approx 6EV of dynamic range.

On the more recent Sony sensors, the previous issue of slight colour loss at high DRO has been largely corrected.