Ran ipamorelin this past summer mostly for sleep, cycling it off now for spring to get a clean read on natural GH pulse. I never ran DSIP, epitalon, or selank, so take the back half of this as mechanism, not experience. The thing nobody splits out in these stack threads: these compounds aren’t competing on the same metric. DSIP and selank get sold for the front of the night, falling asleep and the anxious-mind variety of insomnia. That’s sleep onset latency. Real if it works for you, but it’s a feeling, and it mostly won’t show up on a wearable as a number you can argue with. ipa and sermorelin work a different lever. They nudge the GH pulse, which lands on slow-wave sleep. That part is measurable. My deep sleep minutes went up enough on ipa that it wasn’t noise, and honestly the drop-off since I came off is the cleaner read. so the question “which sleep peptide is worth it” is kind of malformed. worth it for what part of the night? If you can’t fall asleep, a GH secretagogue isn’t your tool, it does nothing for onset. If you fall asleep fine but wake up feeling like you banked no depth, that’s the slow-wave bucket, and that’s where ipa actually has a mechanism to point at. fwiw the variable I’d track harder than compound choice is dose timing. Pre-bed conditions vary more than people admit, and a shot 90 min before bed reads different than one at lights-out. I log injection time off the watch complication so the WHEN is in the data, not a guess two weeks later. And none of this beats a fixed wake time plus magnesium. Every honest write-up I’ve seen, mine included, the peptide is the small lever and the boring inputs are the big one. If you stacked all three at once you changed too many variables to know what moved anything.
edit: clarifying
the “drop-off since I came off is the cleaner read” is the part I’d trust most in your whole writeup, because that’s a delta against your own prior, not a single-night absolute. a wearable’s deep-sleep number is noisy as hell night to night, but the direction of change when you pull the lever is harder to fudge. one thing I’d add from the TRT side: testosterone itself can suppress slow-wave in some guys, so if anyone’s running a GH secretagogue on top of trt and not seeing deep minutes move, that’s a confound worth ruling out before blaming the peptide. muscle and recovery run on slow-wave specifically, not just hours logged, same bucket you’re pointing at. logging injection time off the watch is the right instinct. I keep a short free-text note on each pin in the tracker i use, time and how the night before felt, and the 8-week picture reads way cleaner than memory two weeks later. fixed wake time still doing the heavy lifting though, no argument there.
the “testosterone itself can suppress slow-wave in some guys” line is the one I’d put an asterisk on. Not saying it’s wrong, the direct hormonal angle shows up in the literature and it’s a fair confound to name. But three years on TRT, total T parked in the 700s, and I’ve got no clean read that the T itself is eating my deep minutes. What I’d rule out first, before blaming the hormone directly, is sleep-disordered breathing. TRT can nudge AHI up in some guys, and fragmented breathing trashes slow-wave way harder and more reliably than any direct suppression mechanism does. So if someone’s running a GH secretagogue on top of trt and the deep minutes won’t budge, my order of suspects is apnea/AHI first, hormone-direct second. A home sleep study settles it faster than guessing. The other thing your free-text-per-pin habit gets you, and most people miss this, is the cross-correlation once you’ve got enough nights logged. I run my injection times and a sleep note through a med tracker that surfaces correlations across entries, and it flagged that my worse deep-sleep nights clustered behind late dinners, not behind any peptide variable at all. Boring input, biggest lever, again. agreed on the delta-against-prior point being the trustworthy part, that’s the whole reason the come-off read is cleaner than any single night on. fixed wake time still doing the heavy lifting, no argument from me either. fwiw I’d want to see your 8-week note set sorted by night-before-feel vs measured deep minutes, bet the two diverge more than you’d expect.
Putting apnea ahead of the hormone-direct mechanism is the right order, and the reason is exactly the one you’d care about: AHI is something a home study actually measures and a CPAP actually fixes, where “TRT is eating my deep minutes” is a story you can’t isolate from your own n=1. The late-dinner cluster your tracker surfaced is the same lesson from the other direction, the boring input moved the slow-wave number while every compound variable sat there doing nothing you could prove.