Bedroom mining overnight... is dB rating even the right thing to compare?

So i've been going back and forth on whether to put a mining heater in the bedroom for winter. My apartment runs cold and i already spend money on a space heater, so the math kinda makes sense on paper.
But every comparison i see focuses on the lowest dB number on the spec sheet. And honestly i'm not sure that tells you much about whether you can actually sleep next to the thing. A steady fan at 38dB can totally disappear into the background — my air purifier is around that and i forget it's on. But a fan that ramps up and down, even if it's technically "quieter" at its lowest setting, would absolutely wake me up. The change in sound is what gets you, not the volume itself.
Which got me thinking... would you actually run a Bitcoin heater in a bedroom overnight? Like has anyone here done it and not just for a single night but consistently?
I've been looking at the Avalon lineup since they seem designed for this use case. The Mini 3 is more interesting to me — 500W or 800W mode, and apparently it can idle at a target room temp. That's actually close to what a normal space heater does. Spec says 33-55dB across modes with a Night Mode, which sounds promising but 55dB is not quiet lol. I'd really want to hear it from the actual bed position before committing.
The other thing nobody talks about — the first overnight run probably shouldn't be the first time you test it. Like, run a full heating cycle while you're awake first. Check the plug isn't getting warm, make sure your blankets or curtains aren't anywhere near the airflow, basic stuff. These things are computers pulling real wattage, not a sealed oil radiator you can throw a towel over.
Anyway, for anyone who's actually slept in the same room as one of these (or any home miner really) — what was worse, steady fan noise or the sound of it cycling on and off? I feel like that's the real question nobody's answering in the spec sheet comparisons.