How do you find db level of a song?

[Weigel21]

I'd like to imagine the -10dB tone would be good for most music and really, you'd only have to worry about frequencies recorded at higher level if and only if yu were to be playing the stereo at the levels used when the gain were set.

On top of this, it'd be next, if not completely impossible to eliminate distortion from many songs as they are recorded with distortion present already.

Not entirely sure how accurate it is, but Audacity has a option to show clipping in a selected range of a track. I haven't played with Audacity much, but I know that with the dozen or so songs I played around with in it, all of them showed areas of clipping when selecting the option "find clipping".

Still, I understand what you're getting at. With a fair amount of music, I'm willing to bet there is bass recorded at -5dB and maybe even higher (surely nothing recorded at 0dB I'd think though), but I'd like to believe the frequencies human ears are more sensitive to are recorded at more like -15dB.

Curiosity has me now, I'm going to go play with Audacity for a couple minutes.

[Weigel21]

Alright, so I don't really have a lot of music on my computer's HDD, I have most of my collection stored on an external HDD in 320kbps MP3s, but I do happen to have a handful of songs on my computer.

First, I selected the song I liked best out of what I had on the computer, it was "Beautiful World" by Utada. Since my music is stored in MP3 format, the extreme highs are what would have been cut to compress the file. Looks like the highest frequency in the song is 16,440Hz and it's recorded at -59.3dB. The frequency recorded at the highest level looks to be 46Hz and it's recorded at -7.1dB. Looks like a large portion of the frequencies are recorded t anything from -12dB to -40dB, higher frequencies being more attenuated.

Not sure if you're at all familiar with that song or Artist, but she's an American born (New York) Japanese Artist.

So lets see what else I have. Looks like the only song I have on my computer that's not J-Rock or J-Pop is "Hypnotize" by Young Jeezy. This song is known for some DEEP bass, has some nice bass hits around 27Hz. Anyways, highs look to be nonexistent above 13kHz, but it appears it's one of my lower bit files being 256kbps. Anyways, frequencies from 100Hz and up look to be recorded at -20dB all the way down to -42dB. The highest recorded level is -2dB and it looks like frequencies between 34-53Hz are all recorded at such. 27Hz looks to be recorded at -4dB.

[csshakka]

Alright, so I don't really have a lot of music on my computer's HDD, I have most of my collection stored on an external HDD in 320kbps MP3s, but I do happen to have a handful of songs on my computer.

First, I selected the song I liked best out of what I had on the computer, it was "Beautiful World" by Utada. Since my music is stored in MP3 format, the extreme highs are what would have been cut to compress the file. Looks like the highest frequency in the song is 16,440Hz and it's recorded at -59.3dB. The frequency recorded at the highest level looks to be 46Hz and it's recorded at -7.1dB. Looks like a large portion of the frequencies are recorded t anything from -12dB to -40dB, higher frequencies being more attenuated.

Not sure if you're at all familiar with that song or Artist, but she's an American born (New York) Japanese Artist.

So lets see what else I have. Looks like the only song I have on my computer that's not J-Rock or J-Pop is "Hypnotize" by Young Jeezy. This song is known for some DEEP bass, has some nice bass hits around 27Hz. Anyways, highs look to be nonexistent above 13kHz, but it appears it's one of my lower bit files being 256kbps. Anyways, frequencies from 100Hz and up look to be recorded at -20dB all the way down to -42dB. The highest recorded level is -2dB and it looks like frequencies between 34-53Hz are all recorded at such. 27Hz looks to be recorded at -4dB.

Thanks for messing around with it. How exactly are you analyzing each song? Selecting the whole piece then plot spectrum?

[_paralyzed_]

This stuff is so over my head I got a sore neck from looking up and trying to figure it out.

[Weigel21]

Yeah, I was selecting the entire length of the song, figured that would give a more accurate reading for what you seemed to be wondering about. I will admit though, I've not used Audacity a whole hell of a lot. I've had it for probably 1 and a half years on my computer now and as of now, probably only messed with it a couple dozen times.

[Weigel21]

Yeah, I was selecting the entire length of the song, figured that would give a more accurate reading for what you seemed to be wondering about. I will admit though, I've not used Audacity a whole hell of a lot. I've had it for probably 1 and a half years on my computer now and as of now, probably only messed with it a couple dozen times.

Now I know that not "all" of a particular frequency is recorded at the level listed and that those are what they each peak at, but without selecting a much smaller portion of the track and doing so numerous times until the entire track were analyzed, I couldn't say just how frequently the higher -2dB levels occurred.

[csshakka]

Yeah, I was selecting the entire length of the song, figured that would give a more accurate reading for what you seemed to be wondering about. I will admit though, I've not used Audacity a whole hell of a lot. I've had it for probably 1 and a half years on my computer now and as of now, probably only messed with it a couple dozen times.

Now I know that not "all" of a particular frequency is recorded at the level listed and that those are what they each peak at, but without selecting a much smaller portion of the track and doing so numerous times until the entire track were analyzed, I couldn't say just how frequently the higher -2dB levels occurred.

Yeah it makes sense looking at it this way. Another thing I just tried was using the wave stats plugin in audacity. Basically analyzes the track and gives RMS db level. Rather than just do the whole song, I did three tests. I used audacity's high/low pass filters to mimick my low, mid, and high ranges and then ran wave stats on each one to get a better idea what range, on average, was seeing.

[Weigel21]

Cool.