Future of CTL

No need for further debate - it’s all pedantic anyway! ACES does not discriminate - it wants to be used by everything and everyone.

I think SPIRV or some other meta language that can be easily translated to popular shader languages is the best way forward. The big issue is that ACES is growing, and if CTL becomes more popular as a result, we are really going to need a way for all developers to easily and correctly translate CTL into their preferred graphics framework.

It is worth considering that relying on a third party tool with an uncertain future like SPIRV-Cross could end up backfiring, but I’m not aware of any alternatives.

:wink:

Anyway back on the topic, Microsoft maintains ShaderConductor that converts HLSL to DXIL and SPIR-V (via SPIR-V-Cross):

SPIR-V-Cross is actively developed and maintained. I’m curious as what makes you think its future is uncertain? If anything its future looks brighter than that of CTL! :slight_smile:

Hi,

@Thomas_Mansencal just pointed me to this thead. I’ve been playing with my own implementation of a CTL interpreter in the hopes of integrating it with OpenColorIO and our upcoming OpenColorMath library.
At the moment I have a bytecode VM in C with no external dependencies, an interactive REPL, and some rudimentry support for JIT compiling directly to native x64. I also plan to emit GLSL.

My motivations for re-implementing CTL were to remove some dependency requirements of the current implementation (eg, IlmBase), provide C linkage, and as a bit of a personal challenge.

I’d love to be a part of any future discussions, and hope to have something to show soon.

5 Likes

Hi Mark,
Thanks for your first post and welcome!
Steve T
ACESCentral Admin