John, good question! We haven't tested time lapse or boost recording specifically with any of the Smart CODECs we've tested (Axis, Hikvision, Panasonic).
Without testing, I'm not sure what effects dynamic GOP may have on boost recording. In theory, it shouldn't be an issue, because when you switch the framerate, the stream should respond with a new I-frame. But that's not guaranteed. I've seen them drop in some rare cases under normal circumstances, let alone regularly changing stream settings.
In general, though, I think that smart CODECs reduce or eliminate the need to do boost recording. In very still scenes, where the stream would stay low frame rate in a boost scenario, smart CODECs have their biggest impact. Savings of 70% are not uncommon in these scenes, with 90% at times in our tests. So trying to eke out an additional couple of percent from boost recording may not be worth the effort.
Thanks Ethan. I'll be running some trials to measure the storage efficiency of time lapse Smart Codec vs boost recording. Switching to time lapse recording of smart codec also introduces some wrinkles on the storage calculation.