I couldn’t agree more re dialogue. I always wipe existing levelling and ask for all clip levels to be set to 0 gain. I often find editors turn clips up which sound fine on playback but sometimes I may have to render their audio and that can send clips with a lot of clip gain into digital clipping.
To be honest, I usually ignore all supplied automation entirely and do it all myself, referring to the guide when needed. As you say, that’s one of the things we’re paid for. But sometimes, we’re not paid much at all and the budget/schedule is super tight and it helps things to keep the existing automation and refine it, which is another thing we’re paid for.
The point you make re undipped mixes and stems is a great point. It’s doesn’t seem relevant for the drama mixing I’m doing but for other narration based stuff I agree entirely. I always mix with my stems passing through actual tracks, do an initial flat mix without narration, then introduce the narration and dip the playback levels of the individual stem tracks. If time is against me I’ll keep the narration up the whole time and dip the track stems playback on the run, but then I don’t really get to hear the mix minus in its full glory.
I also prefade send those stem tracks to an undipped mix minus as well as the normal track output to the full mix. Once I’m happy with the mix I can print all stems and mixes simultaneously in real time if I want, which I usually do to hear the mix one last time. I often find I’m stopping and doing tiny tweaks and drop-ins on the run at this printing stage.
Anyway, I’m waffling on now. Thanks for your thoughts and time on this MattiasNYC.