Really Stupid Question About Db Levels

i’m trying to understand how something works here and i’m having trouble grasping the concept or even finding any basic technical information on it…

sound is measured in decibels… which are rated according to magnitude. like ~30db is a whisper, ~60 is somebody talking, ~90 is somebody yelling, ~120 is a loud concert or a jet engine, etc. i think i understand that concept

but then why are sound recordings measured in NEGATIVE decibels? why? and what does that mean? what strange unit of measurement is this? what does it correlate to in the real world? what significance does the number 0 have to the limits peaking out? etc?

anybody see where i’m coming from? i mean i’ve been messing around with this shit for like 8 years but i still don’t really understand how it works. any help would be appreciated. links to guides or anything like that.



actually dB is not a absolute measure… it shows how one figure relates to other…

i gotta run now… but will try to explain it later…

it’s measured in negative because this way there can be an easy reference point for distortion (zero), while the other values are based on a logarhytmic scale, meaning that the volume halves at each 3db of decay

log10(2) = 0.301 => 10^0.301 = 2

It also happened bacsuse of historical reasons: most preamps were built to have an output of 1V, which for some reason was roughly equated to 0dB.

It makes sense because in your preamp/premix it’s best to get all your sound right (with or without intended distortion) before you send it off to the slave amp (which could be your monitors or a PA). Many instruments need that preamp stage to bring them relative up to other instruments (like violyn, or a good condenser mic) - as It-Alien has said, everything is scaled up to that common point of 0dB.

Afaik the tracks in Renoise are set at -3dB so you’ve got headroom to fill up the mix a bit. The masterscope should give you a sense of how that all comes together (mathematically) using reductive vaules up to 0dB (the edges). You can go all the way down in negatives to -infinitydB, which makes more sense than reducing it down to zero, because you can always go lower than 0dB in sound!

There’s probably a ‘text book answer’ out there somewhere…

There you go.

that helps a bunch! thanks!

If this table is what i think it is, it could be very usefull for mixing / mastering these elements somewhat.
Which plugins use such reference tables?

10Hz 12,5Hz 16Hz 20Hz 25Hz 31,5Hz 40Hz 50Hz
-70,4dB -63,4dB -56,7dB -50,5dB -44,7dB -39,4dB -34,6dB -30,2dB

63Hz 80Hz 100Hz 125Hz 160Hz 200Hz 250Hz 315Hz
-26,2dB -22,5dB -19,1dB -16,1dB -13,4dB -10,9dB -8,6dB -6,6dB

400Hz 500Hz 630Hz 800Hz 1kHz 1,25kHz 1,6kHz 2kHz
-4,8dB -3,2dB -1,9dB -0,8dB 0dB +0,6dB +1,0dB +1,2dB

2,5kHz 3,15kHz 4kHz 5kHz 6,3kHz 8kHz 10kHz 12,5kHz
+1,3dB +1,2dB +1,0dB +0,5dB -0,1dB -1,1dB -2,5dB -4,3dB

16kHz 20kHz
-6,6dB -9,3dB

Waves VST has some spectrum ananlyzers with a visual field where all frequencies could be ideally placed. They might have based it on that table… or their own… dunno…

this is the so called “A-Weighting”: it is a table which represents the minimum volume at which that frequency, mixed to noise, can be percieved by human ear.

These values are probably weighted on 1Khz frequency, which has 0dB volume level. The other frequencies also, are multiple triplets.