2

I am a 3D artist and I use a lot of textures for bump mapping etc...

If I take a regular 8bit black and white jpg image and add a blur to it with PS set to 16bit. Am I Gaining information? (I realize my image will be blurry, and I will lose sharpness detail) but will I GAIN pixel information with the blur gradients that are created?

So if I saved it out as Tiff, EXR, etc.. in 16bit, would it be a true 16bit image?

EDIT:

Added some test (have no idea if this was a fair comparision). Clearly converting an 8 bit to 16bit is better. So something is being added. enter image description here

1
  • 1
    You have the same amount of information, but it's just blurred more. Sort of the difference between "Yes" and "yeeee-aaa---hhhh". Photoshop cannot create new information. As to your "so...": yeaaa--aa-ah ... if you save it as such, then why shouldn't it?
    – Jongware
    Nov 28 '18 at 19:11
4

Take a look at this: What's the point of capturing 14 bit images and editing on 8 bit monitors

enter image description here

In this case, yes and no... it depends.

No, because your initial 8-bit image will have the same gaps as image B, regardless that you transform it into a 16-bit image. The information will not change.

enter image description here

The difference will be noticeable if you "invent" new in between information, for example, smudging the image. In this scenario, you will have new information filling the space you prepared by transforming the image into 16 bits.

enter image description here

But this new information will not be yet noticeable until you apply a similar transformation as image C and D on my original image.


But if your render can be saved as 16 bits, sure, save it as such. Some render engines generate an image of more than a 24 bits (RGB) file.

2
  • Thanks for the answer, I would be exporting OUT of PS as a 16bit TIFF to import INTO a 3D program to use as height information. So to be clear I AM making NEW information by using the blur which is essentially "filling in the gaps". So that leads me to wonder if I can use the frequency separation technique to have a mostly 16bit TIFF converted from 8bit image, to fill in the gaps? sorry if that is confusing.
    – icYou520
    Nov 28 '18 at 20:24
  • 1
    Not confusing at all... I would say... try it. I am not sure if your 3D program will detect it. It is probably better that you refine your own workflow.
    – Rafael
    Nov 28 '18 at 20:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.