Thanks for the suggestion, but that doesn’t have the desired result because both SPH and LFO.T have no effect on a free running LFO. Trying this out did lead me to a cool use case that almost accomplishes what I want, though:
So, enabling LFO.T on first step AND parameter locking the LFO MODE to TRIG does consistently start the LFO at whatever SPH is set to, and the free running takes over from there. This is pretty cool, and leads to a consistent free running modulation, however it will reset along with the sequence, which makes perfect sense. What I’m trying to accomplish is an LFO that loops in say 3/16 against a typical 16 step sequence such that it lines up every three beats forever, but with this it will always repeat beats 1 and 4 (instead of 1.1, 1.4, 2.3, etc).
The offness (IMO) of this design stands out a lot when your LFO DEST is SAMPLE START. Imagine a sample with various very different textures. Now, every time you press play you get completely different start points, so you can’t really dial in a composition that can be recalled in a set or whatever. Again with the p-lock LFO TRIG approach you do get repeatable modulation, but if the LFO phase simply started where SPH was set to on pressing play, you’d have control over the composition and get polymetric variation - all without having to mess about with p-locks etc.
I do see the point about having a setting to disable / enable this, however if I had to pick one I’d prefer the phase starting a 0 like pretty much every free synced LFO in Live (the latest M4L one even calculates the offset based on the timeline so if you stop and continue mid-measure it will line up mathematically - blew my mind.