I'm currently trying to bring in some 3D renders from Blender into Affinity Photo/Designer (have tried Krita and Photoshop as well), but I'm finding that the colors are quite dull due to color profile issues (using sRGB/Filmic view transform), which seem to be an issue for many. I've followed quite a few suggestions, and also followed this video (multiple times) https://www.youtube.com/watch?v=K-Z-ha9D9Ro ... I do exactly as stated, but when the OCIO adjustment layers are put on and the image is exported as a png, the colors look exactly the same when opening up the exported PNG in any editing application. I can use things like 32-bit previews, and see the files display correctly in some apps, but the problem is it's totally non-destructive so as soon as I open the image in another application or copy/paste it somewhere, the same issues persist. My export settings are as follows, but I've tried PNG, TIFF etc. as well:

File Format: OpenEXR, Color: RGBA, Color Depth: Float(full) 32-bit

Is there something I can do to get these renders to convert/save with the same colors as shown in the render view inside blender? Ideally, what I'm trying to do in the end is to bring these images over to a design tool to work with for some web mockups and be able to use them in some other apps as well, having them look identical to when I've rendered them in Blender. For the life of me, I just can't seem to get the colors to match in a permanent way.

  • 1
    $\begingroup$ You are thinking about this wrong, I believe. The problem is that applications are extremely poor at handling what scene referred EXRs represent. There is a portion of “image” in them, and a portion is derived by rolling the values through a [Display | Camera] Rendering Transform. The only way to use them is with appropriate applications such as Fusion or Nuke. Affinity can be coaxed to work, but inherits the baggage from other legacy woeful approaches such as PS. This is why the Filmic Log Encoding base is important to use, along with a simple 1D LUT for contrast. $\endgroup$ – troy_s May 8 '20 at 14:50

Your Answer

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

Browse other questions tagged or ask your own question.