Date   

Re: Considering OCIO --> ASWF

David Wortley <davewo...@...>
 

Having CI for Windows Builds would be amazing to have! 


On Wed, Sep 26, 2018 at 9:49 AM Francois Lord <flord...@...> wrote:
I'm all for it. OCIO has become an essential part of the VFX world and
making sure its development will continue in a healthy fashion is
important, in my opinion.

I'm very happy to hear this.

Francois


On 2018-09-21 03:22 PM, Larry Gritz wrote:
> Hey, everybody. We're thinking about applying to turn OpenColorIO governance over to the Academy Software Foundation (ASWF), and they are very enthusiastically seeking our application. Before we get that underway, I want to make sure there's a chance here to air any reservations or objections so that we can move forward with a high degree of community consensus, and to answer any questions you may have about the process.
>
> ASWF got a lot of publicity at SIGGRAPH, so I'm going to presume you all know its significance and relevance to us and the basic value proposition. But do speak up if you want more explanation, and I'm happy to expand on it.
>
> I think there are several reasons why turning over governance to ASWF is good for this project:
>
> (1) It emphasizes that OCIO really is a work of and for the VFX community as a whole and shouldn't be thought of as an "Imageworks project", especially in light of the new infusion of contributions from Autodesk and elsewhere. Having it be underneath the foundation's umbrella should be a help to adoption and contributions, versus continuing to be thought of as belonging to one studio.
>
> (2) There are a variety of licensing, legal, and other issues that will be nice to turn over to a neutral party that likely can manage it better, and can do so with close coordination and uniformity with other projects in the ecosystem.
>
> (3) It lets us use the shared infrastructure and engineering support from LF for continuous integration, artifact archival (i.e. compiled release binaries people can directly download and install!), and release management tasks, that are being coordinated and dedicated to ASWF projects. (That's where a lot of the money from the corporate members is going.)
>
> (4) The corporate members not only fork over the $$, they also have to pledge a certain amount of engineering effort to open source. The first couple years (when there is a paucity of ASWF projects), it can go to any project that seems important in the ecosystem, but eventually it will be tightened and fulfilling the engineering obligations of the members will be expected to be dedicated as much as possible to ASWF projects. So by being an official ASWF project, OCIO is more likely to be the recipient of that engineering effort.
>
> (5) OCIO has, to be honest, had some periods when it lacked a strong and consistent leadership, design direction, and implementation contributions. It's pretty healthy right now, but being part of this organization I think is also insurance for the future, to help us with organization, resources, and coordination if and when needed or in times of crisis.
>
> If we go ahead with this, on a day-to-day basis, not much would change. The open source license we use (BSD) would remain the same. The copyrights would continue to be held by the individual authors or their companies. There might be a new CLA (or equivalent), perhaps a common one for all ASWF projects, and maintained by them rather than by Imageworks. Maybe the official canonical repo would eventually move from imageworks to the AcademySoftwareFoundation account on GitHub? But the same developers would work in the same way, the same people would have commit privileges, our process for developing and setting the design and engineering direction would remain as before, all by the same people. There's no micromanagement from above; the healthy projects run their own communities, unless they need help. There might be a more formal notion of a technical steering committee for the project, with the organization helping to facilitate the stakeholders staying in contact. An accepted project gets to elect a representative to the overall ASWF Technical Advisory Committee. I think that covers all the bases of what we'd see on the ground.
>
> I want to emphasize that we haven't made the formal application yet, the ASWF has not yet accepted the project (though since they have solicited our application, our odds are good), and also Sony Imageworks (the current owner/sponsor) has not yet formally committed to the change in governance. But before proceeding further on any of those fronts, I want to make sure this is aligned with the goals and sentiments of the OCIO stakeholders.
>
> Questions or discussion? Pros/cons as you see it?
>
> --
> Larry Gritz
> l...@... / l...@...
>
>
>
>


--
You received this message because you are subscribed to the Google Groups "OpenColorIO Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ocio-dev+u...@....
For more options, visit https://groups.google.com/d/optout.


Re: Build error on CentOS 7.5

Sachin Shrestha <noizf...@...>
 

Thanks Larry, I realised that too after posting this and managed to build it with 1.8. The ociotools got compiled properly too.

Thanks,
Sachin

On Wed, 26 Sep 2018 at 23:09, Larry Gritz <l...@...> wrote:
Yeah, I'm sorry, in OIIO master branch (the *next* release, in development), there has been an API change that OCIO has not yet caught up with. That's my fault, it's on my list to fix right away.

In the mean time, OCIO requires OIIO 1.8 (the current official stable release) or earlier, if you are building ocioconverr and the other command-line utilities. (OCIO's core library doesn't have any OIIO dependency at all.)

-- lg


On Sep 26, 2018, at 12:21 AM, noizf...@... wrote:

Hi,

I'm using a CentOS 7.5 (3.10.0-693.2.2.el7.x86_64) docker instance to build ocio with devtoolset-6 and my build fails with the following error:

/oss/imageworks/OpenColorIO/src/apps/ocioconvert/main.cpp: In function 'int main(int, const char**)':
/oss/imageworks/OpenColorIO/src/apps/ocioconvert/main.cpp:122:66: error: cannot convert 'OpenImageIO_v1_9::ImageInput::unique_ptr {aka std::unique_ptr<OpenImageIO_v1_9::ImageInput>}' to 'OpenImageIO_v1_9::ImageInput*' in initialization
         OIIO
::ImageInput* f = OIIO::ImageInput::create(inputimage);
                                                                 
^
/oss/imageworks/OpenColorIO/src/apps/ocioconvert/main.cpp:311:69: error: cannot convert 'OpenImageIO_v1_9::ImageOutput::unique_ptr {aka std::unique_ptr<OpenImageIO_v1_9::ImageOutput>}' to 'OpenImageIO_v1_9::ImageOutput*' in initialization
         OIIO
::ImageOutput* f = OIIO::ImageOutput::create(outputimage);

I also get these warnings initially during the build but not sure if they eventually cause the build to error out?

/oss/imageworks/OpenColorIO/build/ext/yaml-cpp/src/ptr_stack.h:35:8: warning: 'template<class> class std::auto_ptr' is deprecated [-Wdeprecated-declarations]
   std
::auto_ptr<T> t(m_data.back());

I'm following the build instructions on this page. I have already built oiio and its installed in the default system location and ocio is picking that up correctly during the build process. But I came across this post re the cyclic dependency issue between ocio and oiio and I'm wondering if I need to follow the instructions on that thread in Linux as well since I intend to build all the ocio apps?

Thanks,
Sachin

--
You received this message because you are subscribed to the Google Groups "OpenColorIO Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ocio-dev+u...@....
For more options, visit https://groups.google.com/d/optout.

--
Larry Gritz




--
You received this message because you are subscribed to the Google Groups "OpenColorIO Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ocio-dev+u...@....
For more options, visit https://groups.google.com/d/optout.


Re: Build error on CentOS 7.5

Larry Gritz <l...@...>
 

Yeah, I'm sorry, in OIIO master branch (the *next* release, in development), there has been an API change that OCIO has not yet caught up with. That's my fault, it's on my list to fix right away.

In the mean time, OCIO requires OIIO 1.8 (the current official stable release) or earlier, if you are building ocioconverr and the other command-line utilities. (OCIO's core library doesn't have any OIIO dependency at all.)

-- lg


On Sep 26, 2018, at 12:21 AM, noizf...@... wrote:

Hi,

I'm using a CentOS 7.5 (3.10.0-693.2.2.el7.x86_64) docker instance to build ocio with devtoolset-6 and my build fails with the following error:

/oss/imageworks/OpenColorIO/src/apps/ocioconvert/main.cpp: In function 'int main(int, const char**)':
/oss/imageworks/OpenColorIO/src/apps/ocioconvert/main.cpp:122:66: error: cannot convert 'OpenImageIO_v1_9::ImageInput::unique_ptr {aka std::unique_ptr<OpenImageIO_v1_9::ImageInput>}' to 'OpenImageIO_v1_9::ImageInput*' in initialization
         OIIO
::ImageInput* f = OIIO::ImageInput::create(inputimage);
                                                                 
^
/oss/imageworks/OpenColorIO/src/apps/ocioconvert/main.cpp:311:69: error: cannot convert 'OpenImageIO_v1_9::ImageOutput::unique_ptr {aka std::unique_ptr<OpenImageIO_v1_9::ImageOutput>}' to 'OpenImageIO_v1_9::ImageOutput*' in initialization
         OIIO
::ImageOutput* f = OIIO::ImageOutput::create(outputimage);

I also get these warnings initially during the build but not sure if they eventually cause the build to error out?

/oss/imageworks/OpenColorIO/build/ext/yaml-cpp/src/ptr_stack.h:35:8: warning: 'template<class> class std::auto_ptr' is deprecated [-Wdeprecated-declarations]
   std
::auto_ptr<T> t(m_data.back());

I'm following the build instructions on this page. I have already built oiio and its installed in the default system location and ocio is picking that up correctly during the build process. But I came across this post re the cyclic dependency issue between ocio and oiio and I'm wondering if I need to follow the instructions on that thread in Linux as well since I intend to build all the ocio apps?

Thanks,
Sachin

--
You received this message because you are subscribed to the Google Groups "OpenColorIO Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ocio-dev+u...@....
For more options, visit https://groups.google.com/d/optout.

--
Larry Gritz





Build error on CentOS 7.5

noizf...@...
 

Hi,

I'm using a CentOS 7.5 (3.10.0-693.2.2.el7.x86_64) docker instance to build ocio with devtoolset-6 and my build fails with the following error:

/oss/imageworks/OpenColorIO/src/apps/ocioconvert/main.cpp: In function 'int main(int, const char**)':
/oss/imageworks/OpenColorIO/src/apps/ocioconvert/main.cpp:122:66: error: cannot convert 'OpenImageIO_v1_9::ImageInput::unique_ptr {aka std::unique_ptr<OpenImageIO_v1_9::ImageInput>}' to 'OpenImageIO_v1_9::ImageInput*' in initialization
         OIIO
::ImageInput* f = OIIO::ImageInput::create(inputimage);
                                                                 
^
/oss/imageworks/OpenColorIO/src/apps/ocioconvert/main.cpp:311:69: error: cannot convert 'OpenImageIO_v1_9::ImageOutput::unique_ptr {aka std::unique_ptr<OpenImageIO_v1_9::ImageOutput>}' to 'OpenImageIO_v1_9::ImageOutput*' in initialization
         OIIO
::ImageOutput* f = OIIO::ImageOutput::create(outputimage);

I also get these warnings initially during the build but not sure if they eventually cause the build to error out?

/oss/imageworks/OpenColorIO/build/ext/yaml-cpp/src/ptr_stack.h:35:8: warning: 'template<class> class std::auto_ptr' is deprecated [-Wdeprecated-declarations]
   std
::auto_ptr<T> t(m_data.back());

I'm following the build instructions on this page. I have already built oiio and its installed in the default system location and ocio is picking that up correctly during the build process. But I came across this post re the cyclic dependency issue between ocio and oiio and I'm wondering if I need to follow the instructions on that thread in Linux as well since I intend to build all the ocio apps?

Thanks,
Sachin


Re: Considering OCIO --> ASWF

Sebastian Sylwan <syl...@...>
 

Thumbs firmly up from me as well. 

FWIW, OCIO was one of the case studies that the VES Tech committee brought to the ASWF discussion as deep dive examples, so the ASWF structure was in a way biased to address what was identified as useful to these projects (as many of you already know).  

S


On Mon, Sep 24, 2018 at 4:47 AM Kevin Wheatley <kevin.j....@...> wrote:
On Sun, Sep 23, 2018 at 11:58 AM, Steve Agland <sag...@...> wrote:
> Thumbs up from me. Another reason I see is that in my mind the Academy's ACES config is now the de facto "reference implementation" for OCIO and it makes sense to have them living closer together.
>

Whilst I agree with the reasoning, I think it is better to be clear
that there is no hard commitment for the ACES work to be moved under
ASWF, there was discussion about this at the ACES next BoF but I got
the feeling that people thought the ASWF was emphasizing  the Software
side of things for the moment and ACES covers larger concerns, and
thus was a 'not yet'.

That said the last ASWF TAC meeting did bring up discussion over none
software artefacts.

Also to add to Larry's point about who is/would be driving things, you
can find out yourself by simply participating in the public calls (I
just listen in).

Kevin

--
You received this message because you are subscribed to the Google Groups "OpenColorIO Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ocio-dev+u...@....
For more options, visit https://groups.google.com/d/optout.


Re: Considering OCIO --> ASWF

Kevin Wheatley <kevin.j....@...>
 

On Sun, Sep 23, 2018 at 11:58 AM, Steve Agland <sag...@gmail.com> wrote:
Thumbs up from me. Another reason I see is that in my mind the Academy's ACES config is now the de facto "reference implementation" for OCIO and it makes sense to have them living closer together.
Whilst I agree with the reasoning, I think it is better to be clear
that there is no hard commitment for the ACES work to be moved under
ASWF, there was discussion about this at the ACES next BoF but I got
the feeling that people thought the ASWF was emphasizing the Software
side of things for the moment and ACES covers larger concerns, and
thus was a 'not yet'.

That said the last ASWF TAC meeting did bring up discussion over none
software artefacts.

Also to add to Larry's point about who is/would be driving things, you
can find out yourself by simply participating in the public calls (I
just listen in).

Kevin


Re: Considering OCIO --> ASWF

Sean Looper <sean....@...>
 

I didn't realize how much I wanted a mysterious council of alien overlords until you said we weren't getting one. ;)

That all sounds great, Larry. Thanks for the clarification. Within the ASWF structure I hope to see continued leadership from you and others who have helped propel OCIO forward. It sounds like a win for everyone involved if all goes to plan.


On Sun, Sep 23, 2018 at 5:23 PM Larry Gritz <l...@...> wrote:
On Sep 23, 2018, at 12:28 PM, Sean Looper <sean....@...> wrote:
 I would like to hear how AWSF plans to avoid design/development by committee and the tendency that such a strategy has to fall short of the mark. 

ASWF is not involved in the design or development per se.

It's *us* who avoids design by committee, if that's what we want.

In the process of being accepted, we (as a community) would need to establish a charter that lays out our procedures. They are happy to accept a benevolent dictator, or a committee, or whatever else we want that would seem to work. We need to decide how to proceed.

To clarify, here are some things ASWF can do for us:

* Be a neutral party to hold assets like the domain, trademarks, etc., as well as encourage usage and contributions by making it a shared industry project rather than having the appearance of "ownership" by one studio.

* Shield us from liability by holding each project in its own LLC, legally distinct from any of us, our companies, or other projects.

* Provide and share some infrastructure and expertise, including some of the painful nuts and bolts related to CI, porting, testing, release engineering.

* Help direct toward this project the engineering contribution obligations of the premier member companies.

* Help coordinate communication with other projects and copies in the ecosystem, and help us get organized if we again fall into some period of disorganization.

It's worth also pointing out that ASWF is not some mysterious council of alien overlords. It literally *is* us, with its governance and technical direction set by the same companies that have been involved in using and building OCIO all along -- Dreamworks, Autodesk, DNeg, Disney, Sony (er, any day now, really), etc.


--
Larry Gritz




--
You received this message because you are subscribed to the Google Groups "OpenColorIO Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ocio-dev+u...@....
For more options, visit https://groups.google.com/d/optout.


Re: Considering OCIO --> ASWF

Larry Gritz <l...@...>
 

On Sep 23, 2018, at 12:28 PM, Sean Looper <sean....@...> wrote:
 I would like to hear how AWSF plans to avoid design/development by committee and the tendency that such a strategy has to fall short of the mark. 

ASWF is not involved in the design or development per se.

It's *us* who avoids design by committee, if that's what we want.

In the process of being accepted, we (as a community) would need to establish a charter that lays out our procedures. They are happy to accept a benevolent dictator, or a committee, or whatever else we want that would seem to work. We need to decide how to proceed.

To clarify, here are some things ASWF can do for us:

* Be a neutral party to hold assets like the domain, trademarks, etc., as well as encourage usage and contributions by making it a shared industry project rather than having the appearance of "ownership" by one studio.

* Shield us from liability by holding each project in its own LLC, legally distinct from any of us, our companies, or other projects.

* Provide and share some infrastructure and expertise, including some of the painful nuts and bolts related to CI, porting, testing, release engineering.

* Help direct toward this project the engineering contribution obligations of the premier member companies.

* Help coordinate communication with other projects and copies in the ecosystem, and help us get organized if we again fall into some period of disorganization.

It's worth also pointing out that ASWF is not some mysterious council of alien overlords. It literally *is* us, with its governance and technical direction set by the same companies that have been involved in using and building OCIO all along -- Dreamworks, Autodesk, DNeg, Disney, Sony (er, any day now, really), etc.


--
Larry Gritz





Re: Considering OCIO --> ASWF

Sean Looper <sean....@...>
 

My only concerns are around the general benefits of having a single, production-focused sponsor (SPI in this case). As Larry alluded to, there are pros and cons to a single sponsor, the greatest con in my mind being the potential for fits and starts being tied solely to that sponsor. The major benefit, as I think we've seen with projects like OpenEXR, Alembic, USD and others is an authoritarian focus on a single, owner-centric goal. I don't necessarily need convincing of AWSF governance for OCIO given its sometimes aimless history since Jeremy moved on but I would like to hear how AWSF plans to avoid design/development by committee and the tendency that such a strategy has to fall short of the mark. 


On Sun, Sep 23, 2018 at 11:05 AM Haarm-Pieter Duiker <li...@...> wrote:
Two thumbs up. OCIO is a great candidate for the services and standardization that the AWSF will provide.

HP



On Fri, Sep 21, 2018 at 12:22 PM, Larry Gritz <l...@...> wrote:
Hey, everybody. We're thinking about applying to turn OpenColorIO governance over to the Academy Software Foundation (ASWF), and they are very enthusiastically seeking our application. Before we get that underway, I want to make sure there's a chance here to air any reservations or objections so that we can move forward with a high degree of community consensus, and to answer any questions you may have about the process.

ASWF got a lot of publicity at SIGGRAPH, so I'm going to presume you all know its significance and relevance to us and the basic value proposition. But do speak up if you want more explanation, and I'm happy to expand on it.

I think there are several reasons why turning over governance to ASWF is good for this project:

(1) It emphasizes that OCIO really is a work of and for the VFX community as a whole and shouldn't be thought of as an "Imageworks project", especially in light of the new infusion of contributions from Autodesk and elsewhere. Having it be underneath the foundation's umbrella should be a help to adoption and contributions, versus continuing to be thought of as belonging to one studio.

(2) There are a variety of licensing, legal, and other issues that will be nice to turn over to a neutral party that likely can manage it better, and can do so with close coordination and uniformity with other projects in the ecosystem.

(3) It lets us use the shared infrastructure and engineering support from LF for continuous integration, artifact archival (i.e. compiled release binaries people can directly download and install!), and release management tasks, that are being coordinated and dedicated to ASWF projects. (That's where a lot of the money from the corporate members is going.)

(4) The corporate members not only fork over the $$, they also have to pledge a certain amount of engineering effort to open source. The first couple years (when there is a paucity of ASWF projects), it can go to any project that seems important in the ecosystem, but eventually it will be tightened and fulfilling the engineering obligations of the members will be expected to be dedicated as much as possible to ASWF projects. So by being an official ASWF project, OCIO is more likely to be the recipient of that engineering effort.

(5) OCIO has, to be honest, had some periods when it lacked a strong and consistent leadership, design direction, and implementation contributions. It's pretty healthy right now, but being part of this organization I think is also insurance for the future, to help us with organization, resources, and coordination if and when needed or in times of crisis.

If we go ahead with this, on a day-to-day basis, not much would change. The open source license we use (BSD) would remain the same. The copyrights would continue to be held by the individual authors or their companies. There might be a new CLA (or equivalent), perhaps a common one for all ASWF projects, and maintained by them rather than by Imageworks. Maybe the official canonical repo would eventually move from imageworks to the AcademySoftwareFoundation account on GitHub? But the same developers would work in the same way, the same people would have commit privileges, our process for developing and setting the design and engineering direction would remain as before, all by the same people. There's no micromanagement from above; the healthy projects run their own communities, unless they need help. There might be a more formal notion of a technical steering committee for the project, with the organization helping to facilitate the stakeholders staying in contact. An accepted project gets to elect a representative to the overall ASWF Technical Advisory Committee. I think that covers all the bases of what we'd see on the ground.

I want to emphasize that we haven't made the formal application yet, the ASWF has not yet accepted the project (though since they have solicited our application, our odds are good), and also Sony Imageworks (the current owner/sponsor) has not yet formally committed to the change in governance. But before proceeding further on any of those fronts, I want to make sure this is aligned with the goals and sentiments of the OCIO stakeholders.

Questions or discussion? Pros/cons as you see it?

--
Larry Gritz
l...@... / l...@...




--
You received this message because you are subscribed to the Google Groups "OpenColorIO Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ocio-dev+u...@....
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "OpenColorIO Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ocio-dev+u...@....
For more options, visit https://groups.google.com/d/optout.


Re: Considering OCIO --> ASWF

Haarm-Pieter Duiker <li...@...>
 

Two thumbs up. OCIO is a great candidate for the services and standardization that the AWSF will provide.

HP



On Fri, Sep 21, 2018 at 12:22 PM, Larry Gritz <l...@...> wrote:
Hey, everybody. We're thinking about applying to turn OpenColorIO governance over to the Academy Software Foundation (ASWF), and they are very enthusiastically seeking our application. Before we get that underway, I want to make sure there's a chance here to air any reservations or objections so that we can move forward with a high degree of community consensus, and to answer any questions you may have about the process.

ASWF got a lot of publicity at SIGGRAPH, so I'm going to presume you all know its significance and relevance to us and the basic value proposition. But do speak up if you want more explanation, and I'm happy to expand on it.

I think there are several reasons why turning over governance to ASWF is good for this project:

(1) It emphasizes that OCIO really is a work of and for the VFX community as a whole and shouldn't be thought of as an "Imageworks project", especially in light of the new infusion of contributions from Autodesk and elsewhere. Having it be underneath the foundation's umbrella should be a help to adoption and contributions, versus continuing to be thought of as belonging to one studio.

(2) There are a variety of licensing, legal, and other issues that will be nice to turn over to a neutral party that likely can manage it better, and can do so with close coordination and uniformity with other projects in the ecosystem.

(3) It lets us use the shared infrastructure and engineering support from LF for continuous integration, artifact archival (i.e. compiled release binaries people can directly download and install!), and release management tasks, that are being coordinated and dedicated to ASWF projects. (That's where a lot of the money from the corporate members is going.)

(4) The corporate members not only fork over the $$, they also have to pledge a certain amount of engineering effort to open source. The first couple years (when there is a paucity of ASWF projects), it can go to any project that seems important in the ecosystem, but eventually it will be tightened and fulfilling the engineering obligations of the members will be expected to be dedicated as much as possible to ASWF projects. So by being an official ASWF project, OCIO is more likely to be the recipient of that engineering effort.

(5) OCIO has, to be honest, had some periods when it lacked a strong and consistent leadership, design direction, and implementation contributions. It's pretty healthy right now, but being part of this organization I think is also insurance for the future, to help us with organization, resources, and coordination if and when needed or in times of crisis.

If we go ahead with this, on a day-to-day basis, not much would change. The open source license we use (BSD) would remain the same. The copyrights would continue to be held by the individual authors or their companies. There might be a new CLA (or equivalent), perhaps a common one for all ASWF projects, and maintained by them rather than by Imageworks. Maybe the official canonical repo would eventually move from imageworks to the AcademySoftwareFoundation account on GitHub? But the same developers would work in the same way, the same people would have commit privileges, our process for developing and setting the design and engineering direction would remain as before, all by the same people. There's no micromanagement from above; the healthy projects run their own communities, unless they need help. There might be a more formal notion of a technical steering committee for the project, with the organization helping to facilitate the stakeholders staying in contact. An accepted project gets to elect a representative to the overall ASWF Technical Advisory Committee. I think that covers all the bases of what we'd see on the ground.

I want to emphasize that we haven't made the formal application yet, the ASWF has not yet accepted the project (though since they have solicited our application, our odds are good), and also Sony Imageworks (the current owner/sponsor) has not yet formally committed to the change in governance. But before proceeding further on any of those fronts, I want to make sure this is aligned with the goals and sentiments of the OCIO stakeholders.

Questions or discussion? Pros/cons as you see it?

--
Larry Gritz
l...@... / l...@...




--
You received this message because you are subscribed to the Google Groups "OpenColorIO Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ocio-dev+unsubscribe@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


Considering OCIO --> ASWF

Steve Agland <sag...@...>
 

Thumbs up from me. Another reason I see is that in my mind the Academy's ACES config is now the de facto "reference implementation" for OCIO and it makes sense to have them living closer together.

Steve


Re: Considering OCIO --> ASWF

Francois Lord <flord...@...>
 

I'm all for it. OCIO has become an essential part of the VFX world and
making sure its development will continue in a healthy fashion is
important, in my opinion.

I'm very happy to hear this.

Francois

On 2018-09-21 03:22 PM, Larry Gritz wrote:
Hey, everybody. We're thinking about applying to turn OpenColorIO governance over to the Academy Software Foundation (ASWF), and they are very enthusiastically seeking our application. Before we get that underway, I want to make sure there's a chance here to air any reservations or objections so that we can move forward with a high degree of community consensus, and to answer any questions you may have about the process.

ASWF got a lot of publicity at SIGGRAPH, so I'm going to presume you all know its significance and relevance to us and the basic value proposition. But do speak up if you want more explanation, and I'm happy to expand on it.

I think there are several reasons why turning over governance to ASWF is good for this project:

(1) It emphasizes that OCIO really is a work of and for the VFX community as a whole and shouldn't be thought of as an "Imageworks project", especially in light of the new infusion of contributions from Autodesk and elsewhere. Having it be underneath the foundation's umbrella should be a help to adoption and contributions, versus continuing to be thought of as belonging to one studio.

(2) There are a variety of licensing, legal, and other issues that will be nice to turn over to a neutral party that likely can manage it better, and can do so with close coordination and uniformity with other projects in the ecosystem.

(3) It lets us use the shared infrastructure and engineering support from LF for continuous integration, artifact archival (i.e. compiled release binaries people can directly download and install!), and release management tasks, that are being coordinated and dedicated to ASWF projects. (That's where a lot of the money from the corporate members is going.)

(4) The corporate members not only fork over the $$, they also have to pledge a certain amount of engineering effort to open source. The first couple years (when there is a paucity of ASWF projects), it can go to any project that seems important in the ecosystem, but eventually it will be tightened and fulfilling the engineering obligations of the members will be expected to be dedicated as much as possible to ASWF projects. So by being an official ASWF project, OCIO is more likely to be the recipient of that engineering effort.

(5) OCIO has, to be honest, had some periods when it lacked a strong and consistent leadership, design direction, and implementation contributions. It's pretty healthy right now, but being part of this organization I think is also insurance for the future, to help us with organization, resources, and coordination if and when needed or in times of crisis.

If we go ahead with this, on a day-to-day basis, not much would change. The open source license we use (BSD) would remain the same. The copyrights would continue to be held by the individual authors or their companies. There might be a new CLA (or equivalent), perhaps a common one for all ASWF projects, and maintained by them rather than by Imageworks. Maybe the official canonical repo would eventually move from imageworks to the AcademySoftwareFoundation account on GitHub? But the same developers would work in the same way, the same people would have commit privileges, our process for developing and setting the design and engineering direction would remain as before, all by the same people. There's no micromanagement from above; the healthy projects run their own communities, unless they need help. There might be a more formal notion of a technical steering committee for the project, with the organization helping to facilitate the stakeholders staying in contact. An accepted project gets to elect a representative to the overall ASWF Technical Advisory Committee. I think that covers all the bases of what we'd see on the ground.

I want to emphasize that we haven't made the formal application yet, the ASWF has not yet accepted the project (though since they have solicited our application, our odds are good), and also Sony Imageworks (the current owner/sponsor) has not yet formally committed to the change in governance. But before proceeding further on any of those fronts, I want to make sure this is aligned with the goals and sentiments of the OCIO stakeholders.

Questions or discussion? Pros/cons as you see it?

--
Larry Gritz
l...@larrygritz.com / l...@imageworks.com




Considering OCIO --> ASWF

Larry Gritz <l...@...>
 

Hey, everybody. We're thinking about applying to turn OpenColorIO governance over to the Academy Software Foundation (ASWF), and they are very enthusiastically seeking our application. Before we get that underway, I want to make sure there's a chance here to air any reservations or objections so that we can move forward with a high degree of community consensus, and to answer any questions you may have about the process.

ASWF got a lot of publicity at SIGGRAPH, so I'm going to presume you all know its significance and relevance to us and the basic value proposition. But do speak up if you want more explanation, and I'm happy to expand on it.

I think there are several reasons why turning over governance to ASWF is good for this project:

(1) It emphasizes that OCIO really is a work of and for the VFX community as a whole and shouldn't be thought of as an "Imageworks project", especially in light of the new infusion of contributions from Autodesk and elsewhere. Having it be underneath the foundation's umbrella should be a help to adoption and contributions, versus continuing to be thought of as belonging to one studio.

(2) There are a variety of licensing, legal, and other issues that will be nice to turn over to a neutral party that likely can manage it better, and can do so with close coordination and uniformity with other projects in the ecosystem.

(3) It lets us use the shared infrastructure and engineering support from LF for continuous integration, artifact archival (i.e. compiled release binaries people can directly download and install!), and release management tasks, that are being coordinated and dedicated to ASWF projects. (That's where a lot of the money from the corporate members is going.)

(4) The corporate members not only fork over the $$, they also have to pledge a certain amount of engineering effort to open source. The first couple years (when there is a paucity of ASWF projects), it can go to any project that seems important in the ecosystem, but eventually it will be tightened and fulfilling the engineering obligations of the members will be expected to be dedicated as much as possible to ASWF projects. So by being an official ASWF project, OCIO is more likely to be the recipient of that engineering effort.

(5) OCIO has, to be honest, had some periods when it lacked a strong and consistent leadership, design direction, and implementation contributions. It's pretty healthy right now, but being part of this organization I think is also insurance for the future, to help us with organization, resources, and coordination if and when needed or in times of crisis.

If we go ahead with this, on a day-to-day basis, not much would change. The open source license we use (BSD) would remain the same. The copyrights would continue to be held by the individual authors or their companies. There might be a new CLA (or equivalent), perhaps a common one for all ASWF projects, and maintained by them rather than by Imageworks. Maybe the official canonical repo would eventually move from imageworks to the AcademySoftwareFoundation account on GitHub? But the same developers would work in the same way, the same people would have commit privileges, our process for developing and setting the design and engineering direction would remain as before, all by the same people. There's no micromanagement from above; the healthy projects run their own communities, unless they need help. There might be a more formal notion of a technical steering committee for the project, with the organization helping to facilitate the stakeholders staying in contact. An accepted project gets to elect a representative to the overall ASWF Technical Advisory Committee. I think that covers all the bases of what we'd see on the ground.

I want to emphasize that we haven't made the formal application yet, the ASWF has not yet accepted the project (though since they have solicited our application, our odds are good), and also Sony Imageworks (the current owner/sponsor) has not yet formally committed to the change in governance. But before proceeding further on any of those fronts, I want to make sure this is aligned with the goals and sentiments of the OCIO stakeholders.

Questions or discussion? Pros/cons as you see it?

--
Larry Gritz
l...@larrygritz.com / l...@imageworks.com


ACES 1.1 config?

Shanky <shank...@...>
 

Any plans to update ACES config to version 1.1 (https://acescentral.com/t/aces-1-1-now-available/1420) in near future?

Thanks


Re: SIGGRAPH 2018 - OpenColorIO Birds of a Feather

George Siddiqui <geo...@...>
 

A big thank you from BlueBolt to all involved in the exciting developments announced last week at the BOF.

Please could someone with slack workspace admin rights send me an invite for: https://opencolorio.slack.com to: geo...@...

Please may I be added to the ocio-v2 google group if it's up and running?

Best wishes,
George


On Thursday, 16 August 2018 20:58:20 UTC+1, Michael Dolan wrote:
Hello OCIO community,

On Tuesday (August 14th) we held the OpenColorIO Birds of a Feather meeting at SIGGRAPH 2018 in Vancouver. Thanks to everyone who was able to attend, we had a great turnout!

The agenda focused on the latest pull requests from the Autodesk OCIO v2 team: Doug Walker, Patrick Hodoul, and Bernard Lefebvre. Doug ran a live demo in Flame showing Lut3D inversion, tetrahedral interpolation on the GPU, ICC and CLF support, and new Range and CDL ops. Patrick then gave an explanation of the new GPU renderer, greatly expanded unit tests, and additions to ocioconvert and ociodisplay for validating OCIO output.

I followed Doug and Patrick's presentation with a case study on how we implemented the new GPU renderer at Imageworks, and are seeing parity with the same color transforms in Nuke's implementation of the OCIO CPU renderer.

We also heard an update from Foundry's Tony Micilotta on the future of OCIO integration in Nuke.

Sean Cooper (DNEG) and Mark Boorer (ILM) concluded the meeting by moderating a discussion on the OCIO v2 proposal for rigorous built-in transforms and the potential for an external "OpenColorMath" C++ library.

The presentation slides are attached for your reference. Conversation will continue in this mail list and the OCIO Slack group, so feel free to add your thoughts or ask any questions you may have. A recorded version of Doug's demo will be made available sometime after SIGGRAPH, so stay tuned. 

Thanks!
Michael Dolan



On Thursday, 16 August 2018 17:55:38 UTC+1, Douglas Walker wrote:
Sean, you're very welcome to share.  The v2 project is being run entirely in the open and we want the community to be informed and to participate.

I'm on my way back to Montreal ATM and would be happy to share the working group slides on my return, although they were pretty minimal and mostly just intended as conversation starters, so the audio is probably a better record of the event.  Also, Patrick, Bernard, and I will be hosting an online meeting in two weeks for those that could not be in Vancouver and we will do a recap of the SIGGRAPH sessions there as well.

I second your shout out to Patrick, Bernard, and Mike, awesome work guys! 

Sean, it was great to see you in Vancouver.  It's very cool to see the Imageworks / Autodesk collaboration you helped us start last year delivering some nice results!

Doug


On Aug 16, 2018, at 4:07 AM, Sean Cooper <se...@...> wrote:

Thanks Mike!

I'm particularly interested to discuss the ramifications of built in colour science transforms.

I suspect that well defined and static colourspaces won't be too controversial, but dynamic colourspaces like ACES become more problematic. Where forcing recompiling and linking of the library for every ACES release isn't ideal. We're open to ideas on how to effectively solve this problem set and benefit us all with a uniform and reliable implementation of colour transforms.

And a special shout out to Patrick, Bernard, and Mike for the huge help and verification of our moving development. The community is very grateful for your contributions and care.

P.S. I recorded the conversation of this year's v2 working group in addition to Doug's notes. I'll append notes or provide the full audio log if some one requests it, pending Doug et. al. approval.

On Wed, 15 Aug 2018, 7:25 pm Michael Dolan, <mic...@...> wrote:
Hello OCIO community,

On Tuesday (August 14th) we held the OpenColorIO Birds of a Feather meeting at SIGGRAPH 2018 in Vancouver. Thanks to everyone who was able to attend, we had a great turnout!

The agenda focused on the latest pull requests from the Autodesk OCIO v2 team: Doug Walker, Patrick Hodoul, and Bernard Lefebvre. Doug ran a live demo in Flame showing Lut3D inversion, tetrahedral interpolation on the GPU, ICC and CLF support, and new Range and CDL ops. Patrick then gave an explanation of the new GPU renderer, greatly expanded unit tests, and additions to ocioconvert and ociodisplay for validating OCIO output.

I followed Doug and Patrick's presentation with a case study on how we implemented the new GPU renderer at Imageworks, and are seeing parity with the same color transforms in Nuke's implementation of the OCIO CPU renderer.

We also heard an update from Foundry's Tony Micilotta on the future of OCIO integration in Nuke.

Sean Cooper (DNEG) and Mark Boorer (ILM) concluded the meeting by moderating a discussion on the OCIO v2 proposal for rigorous built-in transforms and the potential for an external "OpenColorMath" C++ library.

The presentation slides are attached for your reference. Conversation will continue in this mail list and the OCIO Slack group, so feel free to add your thoughts or ask any questions you may have. A recorded version of Doug's demo will be made available sometime after SIGGRAPH, so stay tuned. 

Thanks!
Michael Dolan


--
You received this message because you are subscribed to the Google Groups "OpenColorIO Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ocio-dev+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "OpenColorIO Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ocio-dev+u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.


Re: SIGGRAPH 2018 - OpenColorIO Birds of a Feather

Douglas Walker <doug_...@...>
 

Sean, you're very welcome to share.  The v2 project is being run entirely in the open and we want the community to be informed and to participate.

I'm on my way back to Montreal ATM and would be happy to share the working group slides on my return, although they were pretty minimal and mostly just intended as conversation starters, so the audio is probably a better record of the event.  Also, Patrick, Bernard, and I will be hosting an online meeting in two weeks for those that could not be in Vancouver and we will do a recap of the SIGGRAPH sessions there as well.

I second your shout out to Patrick, Bernard, and Mike, awesome work guys! 

Sean, it was great to see you in Vancouver.  It's very cool to see the Imageworks / Autodesk collaboration you helped us start last year delivering some nice results!

Doug


On Aug 16, 2018, at 4:07 AM, Sean Cooper <se...@...> wrote:

Thanks Mike!

I'm particularly interested to discuss the ramifications of built in colour science transforms.

I suspect that well defined and static colourspaces won't be too controversial, but dynamic colourspaces like ACES become more problematic. Where forcing recompiling and linking of the library for every ACES release isn't ideal. We're open to ideas on how to effectively solve this problem set and benefit us all with a uniform and reliable implementation of colour transforms.

And a special shout out to Patrick, Bernard, and Mike for the huge help and verification of our moving development. The community is very grateful for your contributions and care.

P.S. I recorded the conversation of this year's v2 working group in addition to Doug's notes. I'll append notes or provide the full audio log if some one requests it, pending Doug et. al. approval.

On Wed, 15 Aug 2018, 7:25 pm Michael Dolan, <mich...@...> wrote:
Hello OCIO community,

On Tuesday (August 14th) we held the OpenColorIO Birds of a Feather meeting at SIGGRAPH 2018 in Vancouver. Thanks to everyone who was able to attend, we had a great turnout!

The agenda focused on the latest pull requests from the Autodesk OCIO v2 team: Doug Walker, Patrick Hodoul, and Bernard Lefebvre. Doug ran a live demo in Flame showing Lut3D inversion, tetrahedral interpolation on the GPU, ICC and CLF support, and new Range and CDL ops. Patrick then gave an explanation of the new GPU renderer, greatly expanded unit tests, and additions to ocioconvert and ociodisplay for validating OCIO output.

I followed Doug and Patrick's presentation with a case study on how we implemented the new GPU renderer at Imageworks, and are seeing parity with the same color transforms in Nuke's implementation of the OCIO CPU renderer.

We also heard an update from Foundry's Tony Micilotta on the future of OCIO integration in Nuke.

Sean Cooper (DNEG) and Mark Boorer (ILM) concluded the meeting by moderating a discussion on the OCIO v2 proposal for rigorous built-in transforms and the potential for an external "OpenColorMath" C++ library.

The presentation slides are attached for your reference. Conversation will continue in this mail list and the OCIO Slack group, so feel free to add your thoughts or ask any questions you may have. A recorded version of Doug's demo will be made available sometime after SIGGRAPH, so stay tuned. 

Thanks!
Michael Dolan


--
You received this message because you are subscribed to the Google Groups "OpenColorIO Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ocio-dev+u...@....
For more options, visit https://groups.google.com/d/optout.

--
You received this message because you are subscribed to the Google Groups "OpenColorIO Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ocio-dev+u...@....
For more options, visit https://groups.google.com/d/optout.


Re: SIGGRAPH 2018 - OpenColorIO Birds of a Feather

Sean Cooper <se...@...>
 

Thanks Mike!

I'm particularly interested to discuss the ramifications of built in colour science transforms.

I suspect that well defined and static colourspaces won't be too controversial, but dynamic colourspaces like ACES become more problematic. Where forcing recompiling and linking of the library for every ACES release isn't ideal. We're open to ideas on how to effectively solve this problem set and benefit us all with a uniform and reliable implementation of colour transforms.

And a special shout out to Patrick, Bernard, and Mike for the huge help and verification of our moving development. The community is very grateful for your contributions and care.

P.S. I recorded the conversation of this year's v2 working group in addition to Doug's notes. I'll append notes or provide the full audio log if some one requests it, pending Doug et. al. approval.

On Wed, 15 Aug 2018, 7:25 pm Michael Dolan, <mich...@...> wrote:
Hello OCIO community,

On Tuesday (August 14th) we held the OpenColorIO Birds of a Feather meeting at SIGGRAPH 2018 in Vancouver. Thanks to everyone who was able to attend, we had a great turnout!

The agenda focused on the latest pull requests from the Autodesk OCIO v2 team: Doug Walker, Patrick Hodoul, and Bernard Lefebvre. Doug ran a live demo in Flame showing Lut3D inversion, tetrahedral interpolation on the GPU, ICC and CLF support, and new Range and CDL ops. Patrick then gave an explanation of the new GPU renderer, greatly expanded unit tests, and additions to ocioconvert and ociodisplay for validating OCIO output.

I followed Doug and Patrick's presentation with a case study on how we implemented the new GPU renderer at Imageworks, and are seeing parity with the same color transforms in Nuke's implementation of the OCIO CPU renderer.

We also heard an update from Foundry's Tony Micilotta on the future of OCIO integration in Nuke.

Sean Cooper (DNEG) and Mark Boorer (ILM) concluded the meeting by moderating a discussion on the OCIO v2 proposal for rigorous built-in transforms and the potential for an external "OpenColorMath" C++ library.

The presentation slides are attached for your reference. Conversation will continue in this mail list and the OCIO Slack group, so feel free to add your thoughts or ask any questions you may have. A recorded version of Doug's demo will be made available sometime after SIGGRAPH, so stay tuned. 

Thanks!
Michael Dolan

--
You received this message because you are subscribed to the Google Groups "OpenColorIO Developers" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ocio-dev+u...@....
For more options, visit https://groups.google.com/d/optout.


SIGGRAPH 2018 - OpenColorIO Birds of a Feather

Michael Dolan <mich...@...>
 

Hello OCIO community,

On Tuesday (August 14th) we held the OpenColorIO Birds of a Feather meeting at SIGGRAPH 2018 in Vancouver. Thanks to everyone who was able to attend, we had a great turnout!

The agenda focused on the latest pull requests from the Autodesk OCIO v2 team: Doug Walker, Patrick Hodoul, and Bernard Lefebvre. Doug ran a live demo in Flame showing Lut3D inversion, tetrahedral interpolation on the GPU, ICC and CLF support, and new Range and CDL ops. Patrick then gave an explanation of the new GPU renderer, greatly expanded unit tests, and additions to ocioconvert and ociodisplay for validating OCIO output.

I followed Doug and Patrick's presentation with a case study on how we implemented the new GPU renderer at Imageworks, and are seeing parity with the same color transforms in Nuke's implementation of the OCIO CPU renderer.

We also heard an update from Foundry's Tony Micilotta on the future of OCIO integration in Nuke.

Sean Cooper (DNEG) and Mark Boorer (ILM) concluded the meeting by moderating a discussion on the OCIO v2 proposal for rigorous built-in transforms and the potential for an external "OpenColorMath" C++ library.

The presentation slides are attached for your reference. Conversation will continue in this mail list and the OCIO Slack group, so feel free to add your thoughts or ask any questions you may have. A recorded version of Doug's demo will be made available sometime after SIGGRAPH, so stay tuned. 

Thanks!
Michael Dolan


Re: Digest for ocio...@googlegroups.com - 2 updates in 1 topic

Joseph Goldstone <jgold...@...>
 

Pretty sure there was a killer typo in Guarav’s illustration of tetrahedral interpolation though. There was supposed to be a second edition of that book (“Digital Color Imaging Handbook”, not “Digital Color Handbook”), and Amazon still shows a mock-up page for it, but it never came out.

To ARRI Partner Program members who are curious about this we recommend one of Kang’s books. You could look at pp. 70-72 of “Color Technology for Electronic Imaging Devices":
https://books.google.com/books?id=vzQH3qA_RKkC&pg=PA72&lpg=PA72&dq=kang+tetrahedral+interpolation&source=bl&ots=DC1eh_t_Xh&sig=3DrpvVzNk2RaTz3fZV7c32Wm2Hw&hl=en&sa=X&ved=0ahUKEwjql8_iz6vcAhUNKXwKHfneA8cQ6AEIVzAG#v=onepage&q=kang tetrahedral interpolation&f=false

But there’s also the Truelight documentation, and Richard Kirk does a nicer, I think, job of presentation than Kang (perhaps because Richard only outlines one scheme, whereas Henry Kang seems to outline every conceivable interpolation scheme ever invented). If you go to this page:
and download the "Truelight Software Library” document from 2006, § 6.6 (“Colour Cube Transforms”), pp. 55-57, does a nice job and includes straightforward C code. I haven’t personally tested the code by cutting and pasting but I suspect it works. You want to be sure that the tetrahedra are cut in such a way that the (0,0,0) to (1,1,1) chord of the cube runs along tetrahedral edges — this is kind of the natural way to do it, and in fact is how the (probably little used) 3D LUT processors in the original DLP Projector electronics packages did it, back in the day — a handy thing back when image processing components were twenty years behind today’s components in terms of performance.

On Jul 18, 2018, at 1:35 PM, ocio...@... wrote:

Topic digest 
View all topics
michael...@...: Jul 17 11:04PM -0700 

I guess this means that the lut processor doesn't use (256 256 256) 
nomenclature when it outputs a color, so it doesn't have to perform a 
conversion to (256 256 256) by multiplying by 63. That would make sense, 
since it would be faster. And as for interpolation, I think we were both 
referring to the same 2x2x2 cubes. You said ({6|7}, {12|13}, {18|19}), 
while I used cartesian coordinates. Your way of expressing it makes sense 
and fills in the blanks for me.
 
So I think I've got it, and thank you very much! How did you learn this 
stuff? I can't find even one book that explains the math of cube luts. Can 
you recommend any books or journal articles? My search has been 
frustrating, as I said before, and I feel very fortunate that you've helped 
me out. Anything you recommend I will definitely read.
 
Try it yourself, do a search and see if you can find any technical 
documentation. Maybe I just don't know the right keywords. But I have a 
feeling that other people who are curious about luts are going to find this 
conversation in the future, and it will be a good resource for them too.
 
 
 
On Tuesday, July 17, 2018 at 12:11:42 AM UTC-4, Dennis Adams wrote:
Jim Houston <jim.h...@...>: Jul 18 12:07AM -0700 

> On Jul 17, 2018, at 11:04 PM, michael...@... wrote:
 
> I can't find even one book that explains the math of cube luts.
 
One source is Chapter 11 of Digital Color Handbook by Guarav Sharma, CRC Press 2003 —
“Efficient Color Transformationion Implementation”
 
Jim Houston
You received this digest because you're subscribed to updates for this group. You can change your settings on the group membership page.
To unsubscribe from this group and stop receiving emails from it send an email to ocio-dev+unsu...@....

This message is confidential. It may also be privileged or otherwise protected by work product immunity or other legal rules. If you have received it by mistake, please let us know by e-mail reply and delete it from your system; you may not copy this message or disclose its contents to anyone. Please send us by fax any message containing deadlines as incoming e-mails are not screened for response deadlines. The integrity and security of this message cannot be guaranteed on the Internet.



This message is confidential. It may also be privileged or otherwise protected by work product immunity or other legal rules. If you have received it by mistake, please let us know by e-mail reply and delete it from your system; you may not copy this message or disclose its contents to anyone. Please send us by fax any message containing deadlines as incoming e-mails are not screened for response deadlines. The integrity and security of this message cannot be guaranteed on the Internet.




This email has been scanned for email related threats and delivered safely by Mimecast.
For more information please visit http://www.mimecast.com


Re: how to write .cube luts?

Jim Houston <jim.h...@...>
 


On Jul 17, 2018, at 11:04 PM, michael...@... wrote:

I can't find even one book that explains the math of cube luts.

One source is Chapter 11 of Digital Color Handbook by Guarav Sharma, CRC Press 2003 —
“Efficient Color Transformationion Implementation”

Jim Houston

421 - 440 of 2155