OTOY Forums

Hello guys, I am rendering a scene with Trails ( X particles). It's a very large resolution animation.

Do you think I can render in grayscale to save rendering time and then colorize them in After Effects?

Can I render with less bits to save render time?

Would that give me a faster rendering?

What do you think?

Thank you very much for your opinion/advice!!!

-Snoopy

As I understand it, no, the primary kernels are always going to render in full bit depth, then down-sample/encode to whatever file format you have chosen. You might be able to do a particles-only render in a standard kernel with most of the render settings turned way down and still get what you need.

Depending on how you need to use the particle pass, you might want to explore the Info Channel options and possible combinations of Info, instead of normal image data. Info Only renders can be wicked fast. That's what I'd look into, first.

Great suggestions from frankmci. While the renderer could output a greyscale (similarly to monochrome sensors or "B&W photographic films") result "at kernel level" (prior to any post processing, i.e. OCIO, LUTs), it is currently not possible. However, if it was supposedly possible, it would likely not excessively "speed-boost" that much.

Have you locally considered tweaking kernel settings and optimizing the scene instead?

Thank you guys!

frankmci » Thank you, I will look into the Info Channels and see what I get.

elsksa » Thank you, I have been tweaking the Kernel settings but that link is awesome, I am reading it right now, maybe there's some info that I can try.


Thank you very much for your ideas and suggestions!!!

-Snoopy

I am rendering inside C4D. Does it render the scene faster if I send ite to the Standalone version?

-Snoopy

You are welcome.

Snoopy wrote:I am rendering inside C4D. Does it render the scene faster if I send ite to the Standalone version?
-Snoopy

Universally and in some ways, yes. Not having a 3D DCC "host" is always a "relief" for the renderer.

However, Standalone calls for a different, conscientious, workflow/pipeline. I've written a bit about it (with more to come), if you are interested. There is merit to both Octane as a plug-in and Standalone. The choice is yours.

Wow that's impressive.

Extremely useful!


Thank you for doing that!

-Snoopy

Snoopy wrote:I am rendering inside C4D. Does it render the scene faster if I send ite to the Standalone version?
-Snoopy

In my experience, not significantly. Once the scene assets have been converted and loaded to the GPU, render times are essentially identical. There are instances, such as textures that need to be updated every frame, or CPU based animated shaders, where it can have more of in impact, but even then, it is minimal. For me, the less I need to jump around between applications, the less likely I am to make mistakes, especially when stressed. Since a mistake can cost many hours, unexpectedly, and at the worst possible time, I'd much rather spend a few extra seconds or minutes when I expect them.

frankmci wrote:

Snoopy wrote:I am rendering inside C4D. Does it render the scene faster if I send ite to the Standalone version?
-Snoopy

For me, the less I need to jump around between applications, the less likely I am to make mistakes, especially when stressed. Since a mistake can cost many hours, unexpectedly, and at the worst possible time, I'd much rather spend a few extra seconds or minutes when I expect them.

To which of their own. Both have advantages and inconveniences.

One of the major benefit of Standalone is not only stability, as it removes the 3D DCC host "odds" of its own failure (or plugin related issues) but also provides high flexibility as assets are carefully and independently exported/loaded. Ensuring a sort-of QC stage in the middle that doesn't slow down the pace as much as one could think.

Such approach (analogous to Katana, Gaffer, Clarisse...) provides a practical scene-control & asset updates, among other benefits, some of which I've listed.

Octane Standalone on Linux is also a great(er) choice. Similarly to Standalone vs Plug-in, a slight but welcomed positive difference.

elsksa wrote: To which of their own. Both have advantages and inconveniences.

One of the major benefit of Standalone is not only stability, as it removes the 3D DCC host "odds" of its own failure (or plugin related issues) but also provides high flexibility as assets are carefully and independently exported/loaded. Ensuring a sort-of QC stage in the middle that doesn't slow down the pace as much as one could think.

Yes, definitely, whichever workflow fits for you. Personally, I like to use C4D's XRefs to manage assets across a project as they change. Stability isn't so much of an issue for us, as we tend to stay a comfortable distance behind the cutting edge, sticking with a proven solid tool set until there is a compelling reason to update. Even then, we generally don't update to the latest and greatest, for the sake of reliability. Render speed just isn't as important as it used to be, with mindblowingly powerful GPUs, network rendering, and cheap cloud rendering, even with 4K production. New features have to be really compelling to justify a jump and a prolonged period of stability and compatibility testing across the entire team.