• Global community
    • Language:
      • Deutsch
      • English
      • Español
      • Français
      • Português
  • 日本語コミュニティ
    Dedicated community for Japanese speakers
  • 한국 커뮤니티
    Dedicated community for Korean speakers
Exit
0

why are dng files produced bt denoise ai 6 times larger than the orignal

Explorer ,
Apr 24, 2023 Apr 24, 2023

Copy link to clipboard

Copied

After using lightroom denoise AI the size of the Denoise generated dng file is 6 times larger than my original 20 mg OM-Systems ORF file. Is that standard workflow? Topaz photo AI produces dng that is 5 times larger than my original ORF file. Can anyone tell me why these files are so much larger than the original ORF raw files?

TOPICS
macOS

Views

1.5K

Translate

Translate

Report

Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines
Community Expert ,
Apr 24, 2023 Apr 24, 2023

Copy link to clipboard

Copied

They are demosaiced. One grayscale channel is encoded into three RGB channels.

 

It's also possible that the bit depth is higher (haven't checked that).

Votes

Translate

Translate

Report

Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines
Community Expert ,
Apr 24, 2023 Apr 24, 2023

Copy link to clipboard

Copied

The Enhanced DNG file contains a demosaiced file, which is typically 3 times the original file size as it now contains RGB channels. Additionally, the Enhanced DNG file also contains a copy of the original mosaic data (i.e. a copy of the original raw file).

 

 

Votes

Translate

Translate

Report

Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines
Contributor ,
Jun 13, 2023 Jun 13, 2023

Copy link to clipboard

Copied

i have just started to look at the AI denoiser.

i have some old DNG files that have lots of noise (small point and shoot camera).

 

how can i access the original raw data in the DNG if i need to?

if we have acess to that original data in the denoised DNG i could maybe delete the original DNG.

 

i mean i see no way to get back to the original data once i have denoised the file.

but that sure would be handy.

 

 

 

Votes

Translate

Translate

Report

Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines
Community Expert ,
Jun 13, 2023 Jun 13, 2023

Copy link to clipboard

Copied

No, there's no way to get to the original mosaic raw data. So you can't delete the original file.

 

There is in fact a copy of the mosaic data embedded in the Denoise file - but it's only there for future improvements in the algorithm. You can't extract it. I don't know why they made this decision, but that's how it is.

Votes

Translate

Translate

Report

Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines
Contributor ,
Jun 13, 2023 Jun 13, 2023

Copy link to clipboard

Copied

LATEST

ok.. thanks for the answer.

Votes

Translate

Translate

Report

Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines
LEGEND ,
Apr 24, 2023 Apr 24, 2023

Copy link to clipboard

Copied

If you want to dive deeper into the two correct answers about the whys:

http://www.barrypearson.co.uk/articles/dng/linear.htm

The bit depth is still high bit. 

Author “Color Management for Photographers" & "Photoshop CC Color Management/pluralsight"

Votes

Translate

Translate

Report

Report
Community guidelines
Be kind and respectful, give credit to the original source of content, and search for duplicates before posting. Learn more
community guidelines