Are you using shared nodes?

I see a ton of potential but end up finding them tedious to use.

For example it'd be great to:
Do a base grade on the master shot > make all the nodes shared > copy grade to the closeup > grade the closeup and make these new nodes shared > copy to all instances of the closeup. Then move onto the next angle starting with the base grade again and repeat.

The problem is I have to go through all 3, 6, 10 or however many nodes are in the base grade to make them shared. Then I want to go through and label them as being part of the base grade, because otherwise I'll get confused about which shared nodes are base and which are angle-specific. Then I have to do that again for every new angle. For node-heavy projects, this is just too many clicks and I find myself resorting to using groups as usual.

Am I missing something? Is there a better way?

I don't think multiple shared nodes are a good idea, but that's just me. I can see it for ONE node, but if I had to have several nodes, I'd use a Post-Clip Group grade. At least Resolve is flexible enough that they give you several different ways to accomplish the same thing.

I using it when I assume that there will be a little corrections (from client) and then pre-set one or two shared nodes.

My only problem with groups is that Clip is sandwiched in between Group Pre-Clip and Group Post-Clip. So any clip adjustment is going to be either downstream or upstream of the group node tree; I can't just insert a node arbitrarily in the middle.

I suppose I could split up the node tree between Pre-Clip and Post-Clip if I wanted to...

I use Shared Nodes frequently, and find them a very welcome feature addition to Resolve. But I use them one at a time. On a White Balance node, for example.

Having said that, my work is not terribly complex or high end, so any given clip will have only 1 to 3 nodes, rarely more.

My only problem with groups is that Clip is sandwiched in between Group Pre-Clip and Group Post-Clip. So any clip adjustment is going to be either downstream or upstream of the group node tree; I can't just insert a node arbitrarily in the middle.

One option you might consider -- particularly with longform projects -- is a Fixed Node Tree, where you have X number of labeled nodes that provide the same function in every shot in the entire project. 90% of the time, these nodes might be bypassed, basically doing nothing, but you always have the option of rippling a change just to those specific nodes in the middle of the tree and adding that precise correction.

Here's one example of a Fixed Node Tree, as used by veteran LA colorist Mark Todd Osborne:



I use variations of this idea in my feature and longform TV work, specifically so I can make one similar change in the middle of a node tree for a selection of shots. It also helps to have certain kinds of windows (deactivated) ready to go, should you need them. Note the Vignette that Osborne has at the end of one row.

I also like the idea of having all the nodes pre-labeled. That way, if you happen to come back to the project 3 months later, 6 months later, or even a year later, you can pull up the session and immediately know what was done and what function each node was doing, without having to "hunt" for it.