Author Topic: ACES OCIO Config  (Read 5951 times)

2022-06-06, 18:38:09

Steven_IAS

  • Active Users
  • **
  • Posts: 5
    • View Profile
Hello,

I am trying to get a proper workflow set up with the ACES OT transform. We render out or passes and recomp the beauty in Nuke. The one thing I am having an issue finding information on is the config that this new ACES OT uses. Is it ACEScg 1.0.3, spi-anim, spi-vfx? We need this information to be able to set up our colour workflow properly. I have not been able to get a 1:1 match out of the VFB to Nuke.

Has anyone figured this out, or can offer some advice?

Thank you,
Steven

2022-06-06, 19:08:04
Reply #1

TomG

  • Administrator
  • Active Users
  • *****
  • Posts: 5434
    • View Profile
While I leave it to the devs to answer the question specifically, in case it is useful anyway, I did just want to add in the reminder that applying the ACES OT does not make the output compatible with an ACES color managed workflow. It simply adds a tone mapping curve that is used in ACES as a "sweetener". This may or may not affect your question, but I did want to make sure that was clarified.
Tom Grimes | chaos-corona.com
Product Manager | contact us

2022-06-06, 20:00:13
Reply #2

davemahi

  • Active Users
  • **
  • Posts: 192
    • View Profile
    • iamstatic
"While I leave it to the devs to answer the question specifically, in case it is useful anyway, I did just want to add in the reminder that applying the ACES OT does not make the output compatible with an ACES color managed workflow. It simply adds a tone mapping curve that is used in ACES as a "sweetener". This may or may not affect your question, but I did want to make sure that was clarified."

Hmmm that is a bit disappointing.

2022-06-06, 20:49:48
Reply #3

TomG

  • Administrator
  • Active Users
  • *****
  • Posts: 5434
    • View Profile
Sorry it is disappointing, but it was carefully described in the release info, "For the more technical readers, it applies the curve that “sweetens” the end result in ACES workflows when taking an image from a wide to a narrow color space." so it was never described as being ACES workflow compatible (OCIO) output.
Tom Grimes | chaos-corona.com
Product Manager | contact us

2022-06-06, 22:01:22
Reply #4

Steven_IAS

  • Active Users
  • **
  • Posts: 5
    • View Profile
Correct me if I am wrong. This means what the ACES OT is not an actually ACES workflow, but a LUT that is applied to the image?

If this is the case, is there anyway to set up Corona to work properly with the ACES industry standard?

2022-06-06, 22:28:12
Reply #5

burnin

  • Active Users
  • **
  • Posts: 1532
    • View Profile

2022-06-06, 22:34:30
Reply #6

davemahi

  • Active Users
  • **
  • Posts: 192
    • View Profile
    • iamstatic
Yes we can render LWF......

But when you work with people/studios doing comp and VFX and they are asking to render in ACES and you say sorry I can't, that is not the best.

2022-06-06, 22:36:46
Reply #7

Steven_IAS

  • Active Users
  • **
  • Posts: 5
    • View Profile
Linear is ancient at this point. Every render engine has a solution for ACES. Even if we could just get an OCIO loader of some kind? Vray has it figured out.

2022-06-07, 10:23:47
Reply #8

burnin

  • Active Users
  • **
  • Posts: 1532
    • View Profile
No, you provide unclamped raw HDRI, and any pro should easily take that into their ACES pipeline. Asking you to conform to their OT, is you  doing their job for them and they basically have no clue what ACES is about.

2022-06-07, 11:02:19
Reply #9

piotrus3333

  • Active Users
  • **
  • Posts: 230
    • View Profile
ACES OT matches closely ACES OCIO 1.2 lin srgb to srgb display transform. the difference is around gamma 2.2 vs srgb I guess. not measured but looks to me like that is the case.
Marcin Piotrowski
youtube

2022-06-07, 11:40:55
Reply #10

piotrus3333

  • Active Users
  • **
  • Posts: 230
    • View Profile
Yes we can render LWF......

But when you work with people/studios doing comp and VFX and they are asking to render in ACES and you say sorry I can't, that is not the best.

do you have access to advanced/debug settings in Corona for C4D?

edit: ok, I can see now that settings are the same in C4D. anyway - no need for changing the space here, output is lin srgb one way or the other and default wide srgb is wide enough.
just listen to @burnin and convert lin srgb to acescg if you really need to.


« Last Edit: 2022-06-07, 11:57:40 by piotrus3333 »
Marcin Piotrowski
youtube

2022-06-07, 15:46:32
Reply #11

Steven_IAS

  • Active Users
  • **
  • Posts: 5
    • View Profile
piotrus3333 Thank you for that I had no idea that this was hidden in the UI. This is a big help and I am going to dig into this. Most helpful. I will start to work on this. Thanks again!

2022-06-07, 15:58:15
Reply #12

TomG

  • Administrator
  • Active Users
  • *****
  • Posts: 5434
    • View Profile
Also just a reminder, using the ACEScg internal color space does not change the saved output - so again, saved output is not part of an ACES pipeline even if using this as the internal color space where rendering calculations are done. It also doesn't make a big difference, and the only visible differences will be in niche cases such as strong colored volumetrics, or highly saturated lights shining onto highly saturated surfaces.
Tom Grimes | chaos-corona.com
Product Manager | contact us

2022-06-07, 16:01:30
Reply #13

TomG

  • Administrator
  • Active Users
  • *****
  • Posts: 5434
    • View Profile
(the difference is not noticeable in most cases because Corona always used a wide RGB internal color space while rendering - so while for some engines that were using RGB color space for calculations, having ACEScg as the internal space made a huge difference, it really does not make a big difference in Corona's case :) )
Tom Grimes | chaos-corona.com
Product Manager | contact us

2022-08-16, 23:24:27
Reply #14

katzenwaffe

  • Active Users
  • **
  • Posts: 42
    • View Profile
    • Katzenwaffe.se
Did you or anyone figure this out Steven?

Hello,

I am trying to get a proper workflow set up with the ACES OT transform. We render out or passes and recomp the beauty in Nuke. The one thing I am having an issue finding information on is the config that this new ACES OT uses. Is it ACEScg 1.0.3, spi-anim, spi-vfx? We need this information to be able to set up our colour workflow properly. I have not been able to get a 1:1 match out of the VFB to Nuke.

Has anyone figured this out, or can offer some advice?

Thank you,
Steven