I was doing some tests with “Render With Takes” and I'm getting mixed results I would expect. Here are some of the tests I tried. (Bold settings are the ones that differ from the Base Case.)

No parameters have been added to the A take. The hip take is where all the parameters are being set.

What is expected? What is a bug?

Mantra ROP with…

Base:
Current Take - A
Output Picture - $ACTIVETAKE.main.tif
Render With Take - (Current)
Output - A.main.tif
Result - Expected

Case 1:
Current Take - hip
Output Picture - $ACTIVETAKE.main.tif
Render With Take - A
Output - hip.main.tif
Result - Bug?

Case 2:
Current Take - hip
Output Picture - `run(“takeset”)`.main.tif
Render With Take - A
Output - A.main.tif
Result - Expected

Case 3:
(Note: Mantra ROP is wired into a Batch ROP. The Render Button on the Batch ROP is the one being pushed. The Render With Take is being set on the Batch ROP.)
Current Take - hip
Output Picture - `run(“takeset”)`.main.tif
Render With Take (Mantra ROP) - (Current)
Render With Take (Batch ROP) - A
Output - hip.main.tif
Result - Bug?

I'm basically looking for a way to set up my ROPs so that I can just set the paths to query what the current take is in my hip take, that way I don't have to micromanage all the different outputs for the different takes.

I can kinda understand why $ACTIVETAKE isn't 100%. But the `run(“takeset”)` kinda scares me in the Case 3. I would have thought the Batch ROP would have overrode whatever the Mantra ROP had set.

Please redirect this to support with your hip file. Thanks!

Case #2 is a bug. You should be getting A.main.tiff. I suspect the output picture parm is being evaluated before the take is being applied, or perhaps the parm value is being incorrectly cached.

Case #3 is correct. The “render with take” option only applies to that ROP. In fact, the Batch ROP probably shouldn't have that parm at all, since it doesn't do any rendering. Though I suppose you could set its single parm differently in takes :wink:

The only node whose “Render with Take” parm affects other nodes is the Subnet ROP - it sets its take for the duration of the rendering of the children, so any child with Render with Take set to ‘Current’ will use the parent subnet's take.

twod
Case #2 is a bug. You should be getting A.main.tiff. I suspect the output picture parm is being evaluated before the take is being applied, or perhaps the parm value is being incorrectly cached.

You mean Case #1?

2d
Case #3 is correct. The “render with take” option only applies to that ROP. In fact, the Batch ROP probably shouldn't have that parm at all, since it doesn't do any rendering. Though I suppose you could set its single parm differently in takes :wink:

The only node whose “Render with Take” parm affects other nodes is the Subnet ROP - it sets its take for the duration of the rendering of the children, so any child with Render with Take set to ‘Current’ will use the parent subnet's take.

Ahh.. Good to know. So many new ways of working….so little time to test.

Thanks Mark.

You mean Case #1?

Sorry, yeah – I was looking at the second block and not reading the title ops: