castleofargh
Sound Science Forum Moderator
- Joined
- Jul 2, 2011
- Posts
- 10,448
- Likes
- 6,067
sorry, I should have specified that my personal example already considers peak level when I give my listening level. that's what I measure in the end with a full scale signal, not some average loudness based on crest factor or whatever other method that will better align with the subjective impression of overall loudness in a song.I thought 110dB (or 115, or 120, or whatever) is chosen to account for maximum dynamic range of music (plus whatever EQ is being applied) over the hypothetical maximum average listening level?
So, say I normally listen at 85dB, but on occasion I like to rock out at closer to 95dB, and the classical album I'm listening to has a dynamic range of 30dB. Keeping in mind that half of the dynamic range will fall below the average listening level, that means I need power for 15dB on top of 95dB average = 110dB max.
If I apply +8dB of EQ, then the total should be 118dB (though I don't honestly know how EQ affects average loudness).
Anyway, that's what I've read and how I think about it, though, practically speaking, I don't think I ever get anywhere close to those levels.
Maybe we're talking about the same thing? Or maybe I'm misled on the how and why of it...
so if on your hand you're thinking loudness as an average subjective value for an entire song or selection of songs(which is correct TBH), indeed you'll have to consider that this is not the loudest case scenario the amp will have to be able to "drive". the peaks in the song are.
about EQ, again probably my bad for keeping the basic assumptions in my head^_^. let's say I boost the bass by +5dB, I will usually also apply a -5dB gain setting in the EQ to the entire signal just to be safe about the possibility of the bass boost clipping the digital signal. so now if I want to listen to the same song and hear a bass boost instead of hearing that everything but the bass is quieter, I'll use my amp to compensate for those now missing 5dB. meaning it has to be able to do whatever it was previously doing+5dB.
is it clearer or did I still forget some super important detail?