Notice of Meeting - Output Transforms Architecture VWG - October 27, 2021

Output Transforms Architecture VWG Meeting

Wednesday, October 27, 2021
1pm Pacific time Add to calendar (.ics)

Dropbox Paper homepage for this group:

We use the same GoToMeeting URL and phone numbers for all VWG meetings.

You may join via computer/smartphone (preferred) which will allow you to see any presentations or documents that are shared, or you can join using a telephone which will be an audio only experience.

Please note that meetings are recorded and transcribed and open to the public. By participating you are agreeing to the ACESCentral Virtual Working Group Participation Guidelines.

Audio + Video

Please join my meeting from your computer, tablet or smartphone.

Audio Only

You can also dial in using your phone.

Dial the closest number to your location and then follow the prompts to enter the access code.

** United States: +1 (669) 224-3319 **
** Access Code: 241-798-885 **

More phone numbers:

Australia: +61 2 8355 1038
Austria: +43 7 2081 5337
Belgium: +32 28 93 7002
Canada: +1 (647) 497-9379
Denmark: +45 32 72 03 69
Finland: +358 923 17 0556
France: +33 170 950 590
Germany: +49 692 5736 7300
Ireland: +353 15 360 756
Italy: +39 0 230 57 81 80
Netherlands: +31 207 941 375
New Zealand: +64 9 913 2226
Norway: +47 21 93 37 37
Spain: +34 932 75 1230
Sweden: +46 853 527 818
Switzerland: +41 225 4599 60
United Kingdom: +44 330 221 0097

Meeting #29 - 27th October


Scott to discuss modified SSTS

Params modified to match Kevin’s production averaged data (And any other changes).

Jed to discuss Open DRT vs JzDRT and other developments

Differences and thought processes.

Alex Forsythe

Possibility of use of Image Color Appearance Models as the new core output transform.

Alright. I promised written comments from my team about OpenDRT 0.90b2 vs JzDT and I was waiting from my techart lead to write the longer version of his comment before posting but I’ll have to make do with copy/pasting what I currently have in a Slack discussion:

Tech art lead:

ACES clearly shows its issues in these shots, clipping, yellow shift, and even light bounce loss.
It’s 090b2 vs JzDT for me. While in the lego image JzDT looks desaturated, the highlights on the face example hold up much better.
I like the colors on 090b2 more and highlights on JzDT more

Env-art lead:>

  1. Lego:

-sRGB (ACES): Too much contrast, blue lego brick is pushed towards purple.

-sRGB (OpenDRT v0.0.83b2): Blue and red brick become too flat, blue brick is too saturated

-sRGB (OpenDRT v0.0.90b2): My favorite overall image, decent amount of saturation

-sRGB (JzDT): blue brick too desaturated. same with red figure in background


-sRGB (ACES): blue is pushed too much towards purple, overall image a lot of contrast, I like the warm feeling of the red light that shifts towards orange

-sRGB (OpenDRT v0.0.83b2): overall my favorite, the gradients of the lights are very readable, the skin on the blue/purple side has a lot of information

-sRGB (OpenDRT v0.0.90b2): gradients of the lights are getting too sharp, blue light on the skin is also a bit too sharp

-sRGB (JzDT): Similar to OpenDRT v0.0.90b2, only a bit brighter

  1. Cornell Box:

-sRGB (ACES): Very saturated and contrasty, AO in the corners a bit too intense

-sRGB (OpenDRT v0.0.83b2): Red wall is a bit flat.

-sRGB (OpenDRT v0.0.90b2): looks very similar to OpenDRT v0.0.83b2 except for the red wall which has more tones

-sRGB (JzDT): very similar to OpenDRT v0.0.90b2, just a little darker

  1. lightsabers:

-sRGB (ACES): Blue light is too saturated, red tones on the female character are flattened too much, overall image is has more contrast which makes it darker

-sRGB (OpenDRT v0.0.83b2): red light is very intense on the female character, not that many different tones visible

-sRGB (OpenDRT v0.0.90b2): Overall my favorite, there’s a lot of values on the female character that are missing on the others, the red light also appears more pink and the blue has a good balance of not too light and not too dark

-sRGB (JzDT): blue is a bit too light

Character art lead:

Be aware that they reviewed the images in the ACES JzDT thread under uncontrolled conditions and that, while both the tech-art lead and the env-art lead knew that OpenDRT and JzDT don’t have a baked in look and are meant to be graded afterwards, I don’t think that this fact is well understood by our char-art lead.

Late again! But here are the recording and notes from meeting #29.

1 Like