Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

bright colors rendered with ACES OCIO-Config v1.0.3 don't match ACES CTLRENDER #14

Open
michaeldsmith opened this issue Mar 29, 2018 · 1 comment

Comments

@michaeldsmith
Copy link

Using ACES OCIO-Config v.1.0.3 to render ACES ST2065-1 AP0 EXR files to Output-REC709 does not seem to match the result of CTLRENDER. The bright ACES colors stay saturated and colorful in the OCIO-Config render, while in the CTLRENDER the bright ACES colors desaturate to white.

The differences are visible when processing the ACES-dev test image syntheticChart.01.exr which contains bright colors. See the snip below that shows the rendered output differences:

image

I used the following OCIO command-line to make the above example image using the OCIO-Config:

export OCIO=./ocio/aces_1.0.3/config.ocio

ocioconvert  ./input/syntheticChart.01.exr "ACES - ACES2065-1" ./output/syntheticChart.01.exr.OutputRE709.tif "Output - Rec.709"

I used the following CTLRENDER command-line to make the above example image using CTLRENDER:

ctlrender -force -ctl RRT.ctl -ctl ODT.Academy.Rec709_100nits_dim.ctl syntheticChart.01.exr -format tiff16 syntheticChart.01.ctl_render_to_rec709.tif`

I've also replicated the same OCIO result using the OCIO-Config ACES 1.0.3 built into Nuke 11.1, and also using the "baked" LUT Rec.709 for ACES2065-1 Maya.csp that is included in the folder /aces_1.0.3/baked/maya

I've also replicated the same CTLRENDER result using the built-in ACES workflow in Resolve 14.

To further confirm and research this issue, I made some test pattern images with RGB color-cube sweeps from -12 to +12 stops centered on mid-gray 0.18, and used OCIO and CTLRENDER to render Output-Rec709 using command-lines similar to what is described above. Rendering of these test patterns also shows the differences in rendering output of bright colors. Some examples are shown below with OCIO render on the left-side and CTLRENDER on the right-side.

-12 to +12 stops sweep in Red/Green with Blue=0.18 (OCIO on left CTLRENDER on right)
tiled_ocio_v_ctlrender_rg 13

-12 to +12 stops sweep in Red/Blue with Green=0.18 (OCIO on left CTLRENDER on right)
tiled_ocio_v_ctlrender_rb 13

-12 to +12 stops sweep in Green/Blue with Red=0.18 (OCIO on left CTLRENDER on right)
tiled_ocio_v_ctlrender_gb 13

Below are DropBox links to zips containing the input EXR test patterns I created to generate the above images, and the resulting TIF renders from OCIO-Config ACES 1.0.3 using Nuke 11.1 and CTLRENDER represented by the above images:

EXR -12 to +12 RGB cube sweeps

OCIO Output-Rec709 render of EXR -12 to +12 RGB cube sweeps

CTLRENDER Rec709 render of EXR -12 to +12 RGB cube sweeps

@KelSolaar
Copy link
Contributor

Hi @michaeldsmith,

This is a known issue: http://acescentral.com/t/colour-artefacts-or-breakup-using-aces/520/60

Cheers,

Thomas

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants