ProEXR v2.0 help video for Octane C4D

Maxon Cinema 4D (Export script developed by abstrax, Integrated Plugin developed by aoktar)

Moderators: ChrisHekman, aoktar

Post Reply
Meshmashuk
Licensed Customer
Posts: 5
Joined: Mon Jan 25, 2016 2:13 pm

A friend of mine was confused about getting Octane multilayer renders into After Effects using ProEXR2. So I made a help tutorial which someone else might find handy.
Good luck!
https://vimeo.com/303997000
DinoMuhic
OctaneRender Team
Posts: 194
Joined: Wed Mar 25, 2015 10:24 pm

Hey thanks for making that video but I have to disagree on some points here.

While you do make some valid points I would reconsider on some of your choices.

1. You are absolutely right when you say to use the c4d export settings and not the octane settings.
2. However I would not set the general tonemapping to Tonemapped for the passes. Reason is that everything will be clipped!
3. it's true that for the correct position pass you need to set it to tonemapped, but that is a problem in octane (either plugin or core). The tonemapping for the position pass should always be internally tonemapped but setting it like that for all other passes loses all the floating point data, which a compositor usually wants.
4. To get a correct z-pass set the tonamapping to linear, then you need to work in 32bit in AE and can down-expose the z-pass to get it into a usable range.
5. Generally always work in 32bit inside AE if you deal with EXRs since you otherwise lose all the floating point data above 1.
6. You set up the EXR to be 16bit inside the cinema export settings. This is not advisable, what you should do instead is leave it at 32bit but choose 16bit half float in the further collapsed settings. This will compress your floating point data (so everything from 0-1+) into a 16bit range and reduce filesize significantly while still enable you to work with overbrights.
7. Then you can use B44 compression (60-70% reduction compared to normal RLE) or even DWAA compression (90% reduction) to get a decently small filesize.

And this is the main reason why I have to disagree with your workflow, because the white ball in your example scene is overexposed. With your settings it would not be possible to correct this overbrightness correctly in post. However with a floating point unclipped workflow you would retain all dynamic data of your image for further enhancing in post while still having a small filesize.

So to sum up:

1. Use C4D export settings
2. In the octane passes settings enable multilayer, save beauty and set tonemapping and color profile to linear / linear.
3. In the C4D export settings choose 32bit exr, open up the extra settings and choose 16bit halffloat with a B44A/DWAB compression.
4. In AE choose 32bit workflow and apply a linear->srgb color profile correction (either with the ICIO plugin by fnord or using the inbuilt color profile converter) to the passes needing a conversion (all except z-pass and position pass in your example).
5. Render the position pass seperately tonemapped to get it correctly out of octane.
DinoMuhic
OctaneRender Team
Posts: 194
Joined: Wed Mar 25, 2015 10:24 pm

Here is my version.
Render out my one, then open them both up in AE using ProEXR 2.

Make sure the AE project is set to 32bit.

Now open up the beauty passes of both (Comp "Denoised beauty source") and in the viewport expose down to something like -7
As you can see your beauty pass will not expose down correctly.

The only problem with my method is that the position pass is wrong but that is due to octane being just coded wrong in this regard.
Attachments
Octane_ProEXR2_AOV_001_DinoMuhic.zip
(99.25 KiB) Downloaded 151 times
Meshmashuk
Licensed Customer
Posts: 5
Joined: Mon Jan 25, 2016 2:13 pm

Hi Dino

I only partially agree with your post. I personally always prefer to work fully linear at 32bit as you suggest. However there are many individual artists who have limited storage space and also limited need for deep colour workflows. In fact the last project I worked in 16bit was for an Oscar winning VFX house who regularly drop to 16bit workflows when there is no practical benefit in 32bit. There’s sometimes a difference between theoretically correct practises and the real world. Down stopping z passes and separate EXR outputs for Position is a faff if you just want a quick colour perspective effect and post DOF.

My friend specifically wanted help to reduce file sizes and for much motion graphics work I strongly recommend a 16bit workflow as a way do to that. Many artists do very little post work - I wish they did that but that’s the practical reality. I suspect my video is more useful to them than for yourself who clearly didn’t need to watch it anyway!
Post Reply

Return to “Maxon Cinema 4D”