I listened to your audio files.
The "soft-hard-clipping.wav" one:
I hear a difference at the half of the length. I think, the last part is the hard one. If I apply the two kinds of clipping, I don't get such high difference....
I can only hear the difference using my headphones. The reason for it may be that my speakers are more for domestic use than professional.
I have multible sound systems at home:
- The stereo system isn't low-end, but also isn't high-end. Somewhere between but stuck somewhere in the late 90's. A classic one with stacked Amplifier, Tuner, CD...
- The 2.1 speakers are homemade and sounds a bit better than most of these systems but aren't high-end. These don't have the common gap between the bass.
- The build-in speakers from the television device. These sounds boomy and more like a bloated sound cloud around the screen. I'm not sure, if these are playing mono or stereo.

Somewhat bulky sound which I can't really explain...
No way... All systems ate these differences at all.
BTW: The last kind of devices is, what I target with the tracks where I need this limiter. In most cases, my homemade videos are being watched on televisions with ugly audio playback.
But it also needs to sound good using better systems.
If I don't filter them, there is nothing but noise, klicks (comes from fiddling with the camcorder while recording) and mysterious mumbling somewhere far away.
If I really need to mix something together that must sound good on every audio system, I use headphones for that but I also listen on the two audio systems after finished.
I wrote much....
All in all, I think, a misunderstanding comes up:
- some of you may use audacity in a more professional way using more professional and expensive equipment that provides bigger differences here. You spend a lot of time in mixing and mastering.
- I just want audio tracks for my homemade family videos that even works good using the average internal speakers of television devices without adjusting every second of sound there.