Thanks for the great replies all!
That Parrus .177 is about six months old, with less than 1000 shots. I wanted to use something that I could be reasonably sure it didn't have other issues. Same with the KOI, brand new.
So it's noise level is straight-out-of-the-box plus a light cleaning. I figure the wide variance (not as bad WITH suppression it seems) is mostly the gun-noise. Tuned would be better, but it's the only threaded rifle I have.
That's why I shot with the meter just over a foot from the muzzle. Far enough away to avoid air-blasts, but close enough to dominate any other noises.
I will try later how it sounds at further distances. Right now editing a bunch of other guns I tested levels on (non-threaded, so no suppression other than build-in).
Oh, I did 11 shots each and dropped the lowest/highest one of each. I haven't done the math on this video yet to get averages.