One of the main issues with RW Cameras: currently there is no way to link a viewport to a Design Layer camera.
WHAT CURRENTLY HAPPENS:
You place a RW camera on a design layer and adjust it to the preferred POV. If you have the camera Activated when you create a Sheet Layer viewport, the camera is removed from the Design Layer and becomes embedded in the new viewport. If you don’t have the camera Activated, the RW Camera remains on the Design Layer and the viewport is created with the same POV (but with no embedded camera).
IN EITHER CASE, the viewport’s POV is never linked to a camera on the Design Layer. This is currently impossible.
Examples of why this doesn’t work in actual workflows:
SCENARIO 1: MULTIPLE VIEWPORTS WITH SAME POV
Let’s say you have a set design and you create a viewport to render the design. You then duplicate this viewport 15 more times to show different design variations via class visibilities and class overrides. Perhaps each viewport shows different textures or furniture options or lighting. Or perhaps it's a theatrical play with different scenes.
Then for whatever reason the POV for these renders needs to change (which is not uncommon). There is no easy way to change the POV once and have all 16 viewports match. The only way to do this currently is very manual: First you adjust the POV for one viewport, then you move all of the existing viewports off to side so that you can duplicate this viewport 15 times like you did before and place each one where it goes, then you re-apply all of the class overrides by using the eyedropper tool and sampling from the old viewports, then you delete the old viewports, then you update all the new ones and hope you didn't mess something up. (I should note that all my rendering viewports are on one sheet layer in my workflow).
This is a VERY tedious process, and especially annoying when the POV keeps on changing.
THE PROPOSED SOLUTION:
You place a RW camera on a Design Layer and adjust it to the preferred POV. When you create a viewport there should be the following options:
Leave camera on the Design Layer and LINK the viewport’s POV to this camera.
Leave camera on the Design Layer and create a duplicate camera to be embedded into the viewport.
Embed existing camera in the viewport (just like the current functionality).
The key here is option 1. This leaves the camera right where it is in the Design Layer, and would allow you to have multiple viewports linked to this one camera. In the scenario above, when you want to change the POV for all of your renderings, you simply move the camera’s position in the Design Layer and then update all of your existing viewports. Done.
SCENARIO 2: THE ENTIRE SET CHANGES POSITION.
Let’s say you have a design for an outdoor music concert stage and you’ve set up 8 renderings showing the stage from different points of view in the audience. Now for whatever reason the entire stage is changing position on the site and needs to rotate to the left. You need the renderings to following along with the new position. While you're not completely screwed, the current way of dealing with a rotation like this is quite laborious (which I won’t type out here). It’s possible, but realistically takes about 20 minutes to change all the viewports and make sure you haven’t messed anything up.
With the proposed functionality, since the cameras still live on the Design Layer, you'd simply rotate the camaras along with the other geometry, then update your viewports. Nice and simple.
SCENARIO 3: CAMERA PLOT
In general it would just be nice to be able see the points of view for your renderings while working in a design layer. Say for example you’re creating a camera plot for a tv show and want to see the camera positions for all the shots simultaneously, and you want to be able to move them around and have the linked viewports adjust accordingly. This would be possible with this new functionality: what you see in your Design Layers will always match what you get in your Sheet Layers.
(All of the above scenarios have arisen for me in the last few months, and each time I quickly realize the shortcomings of the camera tool.)
The OIP for a viewport would indicate that the viewport is linked to a Design Layer camera, and it would display the camera name. There would be a button for a pop-up that would allow you to choose a different camera to link the viewport to. There would also be a button to unlink the viewport from the DL camera (leaving you with the same POV but no longer connected to anything). Existing viewports that are currently unlinked should be linkable to an existing camera via the viewport's OIP.
If you double click a viewport with a linked camera and choose Edit Camera, there would be a pop-up warning you that the viewport's camera is a Design Layer camera, and adjusting it will also affect other viewports linked to the same camera. There would be a secondary option to “create a duplicate camera" so that the changes you make only apply to the one viewport, leaving the original camera intact.
The OIP for a DL Camera with linked viewports would indicate what viewports are linked to it. This dialogue box would list all linked viewports, with an "unlink" button next to each one and a button to "Navigate to Viewport". When you try to adjust the camera, VW would warn you that the adjustments will affect all linked viewports. If you try to delete the camera, VW will warn you that all linked viewports will then become independent unlinked viewports.
As suggested elsewhere on these forums, another functionality that would be nice (and tie in well to the above considerations) is the ability to lock the POSITION of a RW camera, but have it still be selectable. Currently if you Lock a camera, you can’t even activate it to see its POV. And there’s no other way to prevent accidental changes to the camera (especially because if you have the camera Active and use the flyover tool, you’re actually moving the camera, which often trips up novice users).
I know this was a long post, but I think this is one of the most important changes VW could make in order to improve the rendering workflow. It would also be immensely helpful in the TV/film/theatre industries.
I'd love to hear other people's thoughts!