One thing to note: using loopback ports inserts one additional audio buffer in your processing, so latency will increase by one buffer if you loop through the master rack.
That’s the reason I don’t loop through the master rack for my main signal chain. Instead, I use a linked rack in all of my songs which takes care of the mastering (limiting, final EQ). I have also set this rack’s state to not be controlled by the containing song, so whenever I make a change e.g. to my master EQ, this sticks across all songs, as long as the rack stays in memory (I don’t usually save gig-specific EQ or volume changes) - which is not a problem with pre-loaded setlists.
Not really, but the first thing you need to learn (if you haven’t done it yet) is how to use linked racks (see guides and videos). Also make sure you can see environment ports in your racks (settings).
Also, learn about pre-loading setlists (again, guides / videos).
Next, create a “master rack” that contains the signal chain for all the finishing touches you want to make to your sound. In my case, it contains an EQ and a latency-free limiter (LoudMax). I then route the output of the last plugin in my master rack directly to my output port (Main Speakers). That’s why I want to see environment ports in my racks.
Now all you need to do is add this linked rack to all your songs, and route the output of your plugins to this master rack’s input instead of to your speaker output. Now everything will be processed through this finalizing signal chain.
Using a linked rack and setlist pre-loading, this rack will stay in memory and won’t need to load with every song.
Last: make sure that the setting “Let the parent song contril this rack’s selected state and gain” is turned off. That way, you can have different states of this rack for different situations or bands that you can select at the beginning of a gig, and not have individual songs change the song state (which you don’t want for a “global” rack).