• Welcome to the Lightroom Queen Forums! We're a friendly bunch, so please feel free to register and join in the conversation. If you're not familiar with forums, you'll find step by step instructions on how to post your first thread under Help at the bottom of the page. You're also welcome to download our free Lightroom Quick Start eBooks and explore our other FAQ resources.
  • Stop struggling with Lightroom! There's no need to spend hours hunting for the answers to your Lightroom Classic questions. All the information you need is in Adobe Lightroom Classic - The Missing FAQ!

    To help you get started, there's a series of easy tutorials to guide you through a simple workflow. As you grow in confidence, the book switches to a conversational FAQ format, so you can quickly find answers to advanced questions. And better still, the eBooks are updated for every release, so it's always up to date.
  • Dark mode now has a single preference for the whole site! It's a simple toggle switch in the bottom right-hand corner of any page. As it uses a cookie to store your preference, you may need to dismiss the cookie banner before you can see it. Any problems, please let us know!

Super Resolution

Status
Not open for further replies.

BobT

Active Member
Joined
Apr 7, 2009
Messages
632
Location
Australia
Lightroom Experience
Advanced
Lightroom Version
Classic
Lightroom Version Number
10.4
Operating System
  1. Windows 10
In DPR someone stated:
"Adobe is very explicit in their release notes that Super Resolution works best on camera-original raw files, since those provide more data for the AI to work on."
Is this correct? Does it mean that converting from the native camera raw format to DNG will not give optimum Super Resolution results?
 
I would think that RAW is RAW. It would not matter if the RAW data come from a proprietary RAW file like NEF or CR2 or a converted RAW DNG or a RAW DNG native to say a Pentax or other camera that outputs RAW DNG files.
 
I would think that RAW is RAW. It would not matter if the RAW data come from a proprietary RAW file like NEF or CR2 or a converted RAW DNG or a RAW DNG native to say a Pentax or other camera that outputs RAW DNG files.
I agree. That's why I was surprised at the comment I quoted above. I wondered if there was more to it.
 
I have never seen an Adobe statement that proprietary raw would be better than raw converted to dng. What I have seen and understand is that super resolution works better on raw data than on RGB data.
 
S
What I have seen and understand is that super resolution works better on raw data than on RGB data
So, since RAW is not an image, but a collection of sensor readings, doesn’t Super Resolution have to convert to some RGB format?
 
S

So, since RAW is not an image, but a collection of sensor readings, doesn’t Super Resolution have to convert to some RGB format?
In the end, yes. But if you look at the dialog, you will see that super resolution on a raw file is actually two things. It is the somewhat older ‘Enhance Details’, followed by the newer Super Resolution upscaling. That is what makes it different from super resolution on an RGB file: ’Enhance Details’ is a raw demosaicing technique.
 
S

So, since RAW is not an image, but a collection of sensor readings, doesn’t Super Resolution have to convert to some RGB format?
Yes, the end result is a RGB DNG file. Whether Super Resolution upscales the photo sites before Converting to RGB or Converts to RGB and then upscales is anyones guess. The Key point here is that the develop setting applied are applied after the upscaling. If you work from a Lossy compressed JPEG, the RGB file will be expanded, lossy data lost and develop data will have been lost in the compression process. All before the u[scaling can occur.
 
I understand that it is possible to super-enhance a raw image once only, in that it will then become a dng and cannot be super-enhanced again as a dng. However, if you convert to another file type then there is nothing to stop that being done a second or more times. This is what I picked up today on the Adobe livestream from their guru Terry White.
 
I understand that it is possible to super-enhance a raw image once only, in that it will then become a dng and cannot be super-enhanced again as a dng. However, if you convert to another file type then there is nothing to stop that being done a second or more times. This is what I picked up today on the Adobe livestream from their guru Terry White.

There is nothing sacrosanct about the DNG file type. I don’t see why you would be limited to one super enhance simply because it is a DNG file type.

I can see why you might not want to super enhance a file that has been already super enhanced because eventually you will get image degradation and lose any benefits of the Super Resolution.

If your RAW file is a DNG file you can do super resolution on it just as you would any proprietary RAW file. The output file from any Super Resolution process is an RGB DNG file type. I suppose it is possible for the Super Resolution process to add some indicator in the DNG header so that any future attempt at Super Resolution would detect the indicator bit and know that additional Super Resolution would be of limited value.


Sent from my iPad using Tapatalk
 
I understand that it is possible to super-enhance a raw image once only, in that it will then become a dng and cannot be super-enhanced again as a dng. However, if you convert to another file type then there is nothing to stop that being done a second or more times. This is what I picked up today on the Adobe livestream from their guru Terry White.
Do not confuse two things. There is ‘Enhance Details’ which is a special demosaicing and so it is only available for raw data. I specificly say ‘raw data’ and not ‘a raw file’, because a proprietary raw file that is converted to DNG still contains raw data, so you can use ‘Enhance Details’ on it too. Enhance Details creates a linear RGB DNG, which means this DNG does not contain raw data and so you cannot use Enhance Details twice.

And then there is Super Resolution. Super Resolution is an AI based upsizing algorithm. It works on RGB data. If you use Super Resolution on raw data, then it is a two-step process. First Enhance Details will be used, then the result will be upsized with Super Resolution. Because you can use Super Resolution on any type of RGB file, you can use it again (and again and again) on the resulting file.
 
Status
Not open for further replies.
Back
Top