Video

There is a new color transform in Blender 4.2 called "Khronos" which preserves important colors from textures, e.g. label data, for the color representation of the original sRGB data. ACES 2.0 looks very much improved and fixes the problem with collapsing HUE colors, but it still needs a stacked workaround to fix the blue/violet color shift problem.

There is just one Arnold Noice. But it looks like the Post denoiser in the Render View works differently because it doesn't use the denoiser AOVs. You need to denoise the sequence by Arnold > Utilities > Arnold Denoiser (noice) manually. Unfortunately the Arnold GPU isn't still supporting AOVs.

Concerning analog film material, this is correct. If you compare the color sweeps of ACES, Filmic, and AgX with that of an ARRI film camera, which shows how color and light are recorded, then you will understand why the AgX protocol shows none of the known problems with ACEC or Filmic. Desaturated colors can be color-corrected in post-production (grading).

A raw file needs to be view-transformed anyway. Every view transform needs more or less a final color grading.

The Fresnel effect is a reflection that depends on the viewing angle. In Blender 4.0, the Principled Shader has now been adapted so that the Fresnel effect is correctly adjusted physically via the IOR. The intensity can still be controlled via a specular value. In versions prior to 4.0, the IOR does not control the Fresnel, but the Specular value. The default value of 0.5 creates a Fresnel effect on every object. You simply can prevent this by using the IOR formula in 3.X and setting the IOR to 1.0 for this formula or you set the IOR to 1.0 in Blender 4.X.

Thanks for your answer @@3dvfxprofessor

The ACES transform protocols are broken. They are not able to create overall consistent and stable results.

Thanks! It's not perfect because the AFX tracker can't track micro-movements with motion blur. I need to take a look into NUKE and maybe 3D Equalizer. But overall it's good enough for my course concept. It's fast and fun. This version also uses half the size of the render layers. I scaled them up and used some filters like sharpen to match the iPhone backplate look as good as possible.

Thanks for the funny comment. Appreciated! 😊

WIth Blender. But you can use the AgX OCIO in Maya, NUKE or AFX, too.

aiStandardSurface with Roughness: 1.0.

www.arri.com/en/learn-help/learn-help-camera-system/image-science/reveal-color-science

@@3dvfxprofessor I ment the value of albedo/base color.

@@piotrus3333_CGI The standard value of 0.8 (white).

@@3dvfxprofessor thank you. all clear.

The protocols for view/color transform. Just watch the video. ACES is not able to produce a correct and natural-looking result. There are color shifts and collapsing HUE colors.

Check my other videos about this topic. Troy Sobotka has a lot of EXR images for testing on his GitHub page. Or just google "Obi-Wan face blue light" and you see good examples of what ACES is messing up. This won't happen with AgX.

@@3dvfxprofessor Yes I'm just in the midst of trying to familiarizing myself with these terms. Would you mind clearing something up for me? For instance at 0:51 we see a mostly blue light source hitting the sphere. We are getting a purplish hue on the sphere, and a mostly blue hue on the walls. As my understanding goes, this is from the Abney effect, stating that as more value is introduced to the light, the hue should shift and eventually just appear white at super high values. Two things about this 1. Does this have to do with hue shift? 2. Isn't the Abney effect a natural phenomenon and desirable? I don't see it very much in AgX compared to ACES. I see the shift to losing saturation but not much hue shifting. Is the ACES form of this too pronounced maybe? I'm not sure I can detect hue collapse yet, but I'm looking at your videos and others.

Hi@@BoboTheImp The HUE shift has nothing to do with the Abney effect. The Abney effect describes the perceived color shift by the stimulus that occurs when white light is added to the light of a monochromatic light source. What you see in the video is a result of the ACES color transform protocol which does it wrong. The Abney effect would occur when you have a blue color being desaturated towards white. The stimulus lets your eyes and brain see purple colors in between. When you use a colored pure blue light source shining onto a white background the background looks blueish. If you raise the light intensity the background will get brighter but not bluer than blue. That's what AgX is doing correctly. And this is the problem with the ACES View Transform. ACES tries to keep the purity of colors and doesn't desaturate the colors toward white. This ends up in a HUE collapse because there is no higher blue than the blue of the HUE spectrum. The color shift from blue to purple is a specific error in the protocol. The newer ACES 2.0 is doing something similar to AgX now in this scene. But it still looks oversaturated. Now folks would say that this is fine now but when you do further tests with a a different setup (check my latest videos) you will still get bad results while AgX is doing it perfectly. Troy Sobotka has published an archive of EXR images with extreme situations on GitHub for testing. The only protocol which is able to deal with all of them is AgX. Every other protocol fails. Especially ACES.

@@BoboTheImp, you can see the HUE collapse very clearly here: czcams.com/video/7-zFWruNJ7U/video.html

There is no hope. ACES is deeply broken. People argue why it's broken when it's widely used in the film industry? I answer: "You are brainwashed by the term INDUSTRY STANDARD!" There are a lot of examples with horrible results. No one cares because it can't even be fixed by color grading when the initial result is clipped or collapsed. Especially when it comes to blue and the "line of purples" colors.

In my description is also a link to a 10-bit MP4 without compression artifacts of CZcams.

What OCIO did you use for AgX in Maya, is it the original Sobokta one or a custom one?@@3dvfxprofessor

@@DomZeal You can use the one of Blender 4.0 from EaryChow on GitHub. I recommend to check out the OCIO collection of "Joegenco/PixelManager" on GitHub. The AgX-Resolve Default is really nice. Or the ARRI K1S1. Troy has a lot of EXR images for testing on his GitHub page.

​@@DomZeal The results in my movie were view transformed in Blender 4.0 by using the specific OCIO profiles. I cross-checked in NUKE. Same results.

You're right... still with ACES 1.2 in Blender / Nuke and 1.3 in Davinci Resolve to fix the clamping bug on raw footages.

The ACES transform protocol is broken and needs to be fixed. There is no workaround. ;-)

@@3dvfxprofessorso I can’t use ACES at the moment in blender or do I misunderstand?

@@jbdh6510 You can use ACES by replacing the specific OCIO file in the folder for color management.

The top is AgX. Bottom ACES.

@@3dvfxprofessorthanks for the video. With version of ACES and is gamut compression being applied?

+++ CORRECTION +++ "And it's not used at all in the industry." was wrong translated in my brain. I wanted to point out "Arnold is not generally/everywhere used in the industry". Thanks to @andrews4084 for the comment. There are not that extreme differences in Quality between Arnold and Blender when it comes to casual work. Arnold has more options in the standard PBR shader but the core path-tracing technology is the same. Arnold is one of the industry renderers for two reasons: its history and the capability to deal with high complexity. And it's not used at all in the industry. The bigger studios like WETA or Animal Logic have developed their own renderer as PIXAR did with Renderman or Dreamworks with MoonRay. Brecht van Lommel (one of the Cycles coders) has worked for Solid Angle on Arnold in the past. ;-)

Arnold is not necessarily better, Blender in this test is more realistic, and Arnold was less, but there are other tests where Arnold is clearly superior, such as rendering eyes realistically, it is best to combine the best of both and reach the best result

@@NoiecityHacking I'm not only talking about realism, Export settings, render passes, shaders, etc.

@@3dvfxprofessor You say Arnold is "not used at all in the industry". Many movies have been rendered entirely with Arnold (especially special effects-heavy movies).

@@jimmyblimmy Thanks for your comment! That was a translation error. I wanted to point out that "Arnold is not generally/everywhere used in the industry". WETA and Animal Logic for instance had to develop their own renderer for specific reasons.

When I remember right it was very straight forward and not complicated.

Don't see what all the hype is about, just want a single linear encoding system that encompasses all current colourspaces and downconverts to display identically across monitors, which we hoped ACES would be the one final solution for.

I think you don't understand it. obviously the colors are way better with AGX, aces is primarily developed for movies and a certain style. AGX is way WAY better@@locinolacolino1302

There is no Blender problem here. How do you come to the claim that Blender is extremely buggy? In what case? We are using Blender since 2016 for commercial productions. The problem is the color transform protocol of ACES. Troy Sobotka has been pointing out the fundamental problems for a very long time and they have been confirmed by Alex Fry. You will get the same color transform issues with EXRs coming from Arnold, Redshift, etc. in NUKE, Fusion, AFX ... using the specific OCIO profiles. Alex Fry answered my test on LinkedIn: "The issues you’re pointing to in the current RRT/ODT rendering are absolutely real, and absolutely a problem." If anyone has a deep understanding of this topic, it is currently Troy and Alex. Source: www.linkedin.com/feed/update/urn:li:activity:7033776158114508801/?comment

"The look is determined at grading phase at the end." That's right, but ACES has serious issues like color shifts when it comes to a proper color transformation. I recommend making some tests with AgX and you might understand the difference.

"to claim ACES color space is flawed". The ACES color space isn't flawed. It's the color transform protocol to compress eg. ACEScg color space to eg. Rec.709.

you're the reason why vfx houses are crushing down. Conservativism at all costs. AGx is newer technology, better optimized for today's needs than ACES. More flexible cause allows indipendent studios to create they're own pipeline adjusted to their color needs. Keep getting old while the world goes round, you'll be extremely happy and well employed I guess ahah

Least Hateful blender hater

Blender's standard color space Linear Rec.709.

@@3dvfxprofessor it looks real 😍

@@MustafaKhan-fs3np The main secret is an HDRI I've shot with the Insta360 One R and the most effort went into the compositing and the window's reflection of the facade. 😀

@@3dvfxprofessor will love to see the process of creating this video 😍❤️

@@MustafaKhan-fs3np Color Transform protocol was AgX (now in Blender 4.0).

Thanks for the comment. Yes, this scene is not representative of complexity and all possibilities. It's one possible comparison and very outdated now. Blender and Cycles have been improved since then. I am working on an interior project actually with over 140K meshes. Cycles is dealing with this very well. There are now many examples that show that there are hardly any differences in the engines. You have to additionally differentiate here which tasks you have to solve. Big productions like WETA, for example, use their own solutions because commercial renderers can't meet the demands.

I am coming from Softimage with Mental Ray, Arnold and Redshift. We are using at my studio Cycles for commercial projects since 2016. I don't see a problem here that Cycles can't solve with as high a quality as Arnold or Redshift. In the end, it is also a matter of experience with such productions and their customers.

@@3dvfxprofessor Yes, sure. I also think you can do pretty much the same with cycles if you know how to do it. I use it alongside the others as well. But the issue is that we don't get the nice results we would like to have on the first try, and we have to rely on compositing for many color management tasks, for example. That, at this moment, is far frustrating. In the last live Pablo mentioned the issue of migrating to ACES and how challenging it was to make Filmic the default due to the compatibility issues. But times go on, and ACES is why superior. Every render engine out there uses it. And there are many other issues that I hope will be addressed in the near future, Cycles has a way to go yet to fully catch up and then overcome the rest. By the way, I found that Gleb mentioned your work on his new lighting course. Well deserved ;).

@@yonlaycabrera4431 Thanks again for your time and thoughts. Appreciated! What? Gleb mentioned me? I need to check this out. Do you have a link with maybe a time code?🙂 I am in contact with Troy Sobotka and I was shocked that ACES has extreme weaknesses in the transformation of HUE colors. I made some tests and published them on my channel. The extremes are not noticeable in everyday production, but AgX is currently superior to ACES. Troy is currently working on an improvement to get as close as possible to the color sweep of a film camera. He rightly criticizes ACES for being the industry standard and the developers' lack of interest in solving the problems.

@@yonlaycabrera4431 I can recommend this video series about color management. This is the best and most well-described content I found about the confusing topic on the web so far. czcams.com/video/68xlyvRewKk/video.html

It's pretty simple. Just copy the node setup.

You're right today. But this wasn't the case in older Blender versions. czcams.com/video/WVjBevFZmes/video.html

Thanks for your interest. I am actually looking into the old tests. It looks like I started a Blender scene but finally used Maya with Arnold. I need to sort this out and will also link the scenes here.

Here's the scene: bit.ly/3G16B7q

It was animated light.

@@3dvfxprofessor ha, thanks for the update!

The exposure animation is done in Blender's compositor.

@@3dvfxprofessor I see, thanks. And the other hue experiment was also done after the render with compositing? Or did you render that one?

@@BlackPantherElite This one was done in the post using Blender with ACES, Filmic, and AgX. I have updated the information below the video with a link to the Maya project with the RS-Proxies (you can't animate anything here): drive.google.com/file/d/1nbIH84pM5odqmNtTOC1A4zjV-FXJIrjZ/view?usp=share_link

@@BlackPantherElite The other ones were rendered in Blender by animating the HUE colors and the intensity of an area light.

@@3dvfxprofessor Thanks for the answer! That is interesting, is there a specific reason you did this one via compositing and the hue one via a full animated rendering? My question being, how do you think would the rendered animated light compare to a hue change of the EXR via compositing?

Yes, because a giant build this model. ;-)