Questioning the plot spectrum (frequency) analysis

For Musicians to Share Their Music, Tips, Tricks and Reviews.
Forum rules
ImageIf you require help using Audacity, please post on the forum board relevant to your operating system:
Windows
Mac OS X
GNU/Linux and Unix-like
Black Dog Bluez
Posts: 626
Joined: Sat Jun 16, 2012 9:43 pm
Operating System: Windows 7

Questioning the plot spectrum (frequency) analysis

Post by Black Dog Bluez » Thu Apr 28, 2016 2:50 am

Reading the Audacity plot spectrum:

Question: Is there a general consensus for frequencies/peaks below 100 hertz, roughly the 0 to 70 hertz range, that these should be reduced, if too high, to a certain level?

It seems about 25 dB lower than the songs peak seems the norm (?).

steve
Site Admin
Posts: 80677
Joined: Sat Dec 01, 2007 11:43 am
Operating System: Linux *buntu

Re: Questioning the plot spectrum (frequency) analysis

Post by steve » Thu Apr 28, 2016 10:27 am

Black Dog Bluez wrote:Question: Is there a general consensus for frequencies/peaks below 100 hertz, roughly the 0 to 70 hertz range, that these should be reduced, if too high, to a certain level?
For vocals, peaks below about 100 Hz tend to make the sound "muddy" / "boomy" and can make it difficult to get the vocals high enough in the mix without distorting. For bass instruments (such as bass guitar, kick drum...), frequencies below 100 Hz are an important part of the sound and should normally be retained, though you probably don't want frequencies much below about 50 Hz.
A simple "high pass filter" ("low cut") is usually sufficient for reducing frequencies that are below the useful range. You don't generally need to go extreme with this - just 6 dB per octave is usually adequate, though there are no hard and fast rules.
9/10 questions are answered in the FREQUENTLY ASKED QUESTIONS (FAQ)

Black Dog Bluez
Posts: 626
Joined: Sat Jun 16, 2012 9:43 pm
Operating System: Windows 7

Re: Questioning the plot spectrum (frequency) analysis

Post by Black Dog Bluez » Sun May 01, 2016 11:34 pm

Thanks, Steve

Black Dog Bluez
Posts: 626
Joined: Sat Jun 16, 2012 9:43 pm
Operating System: Windows 7

Re: Questioning the plot spectrum (frequency) analysis

Post by Black Dog Bluez » Mon May 23, 2016 11:32 pm

What about the high end, any such behavior/suggestions up there? Say, like should everything above a certain hertz be cut? 8, 10, 20 kHz? ... thanks

steve
Site Admin
Posts: 80677
Joined: Sat Dec 01, 2007 11:43 am
Operating System: Linux *buntu

Re: Questioning the plot spectrum (frequency) analysis

Post by steve » Tue May 24, 2016 12:41 am

Why the giant writing?
9/10 questions are answered in the FREQUENTLY ASKED QUESTIONS (FAQ)

Robert J. H.
Posts: 3633
Joined: Thu May 31, 2012 8:33 am
Operating System: Windows 10

Re: Questioning the plot spectrum (frequency) analysis

Post by Robert J. H. » Tue May 24, 2016 8:39 am

Black Dog Bluez wrote:What about the high end, any such behavior/suggestions up there? Say, like should everything above a certain hertz be cut? 8, 10, 20 kHz? ... thanks
Essentially not, quite the contrary. The range above ~8 kHz adds clarity transparency, presence and air.
Most people add a high shelf boost of 1 to 3 dB, starting at 4 to 8 kHz.
Apart from low-cut (100 to 200 Hz) and high-shelf (at about 8000 kHz) there's often a cut in the mids (around 450 Hz) and some engineers throw these three EQ settings automatically on a track.

Kick and Bass need a slightly different set-up.
However, the treble range shouldn't be cut there either.

Robert

Black Dog Bluez
Posts: 626
Joined: Sat Jun 16, 2012 9:43 pm
Operating System: Windows 7

Re: Questioning the plot spectrum (frequency) analysis

Post by Black Dog Bluez » Fri May 27, 2016 12:12 am

Sorry about the big text/.. And thanks Robert, interesting.

I noticed on my MP3 player's EQ when listening to low quality podcasts that cutting the highest band of the 5 band EQ, and some of the 4th helps lower the screechiness.. though../??

Black Dog Bluez
Posts: 626
Joined: Sat Jun 16, 2012 9:43 pm
Operating System: Windows 7

Re: Questioning the plot spectrum (frequency) analysis

Post by Black Dog Bluez » Thu Jun 09, 2016 3:01 am

Here's one.. (see picture attached) Got this song from Youtube with a download helper that I downloaded as MP4 and edited with Audacity to normalize (etc.) and convert to MP3. It came in 1 or 2 decibels in the red and has this incredible 40 hertz peak (maxed) BTW It's a controversial rap song by Eric Dubay. So I experimented bringing the peak down with EQ and it totally took away from the song's essential bass element.. 40 hertz peaked! Wow, anyone know how they do that? Is this raps modus operandi /secret..(?).

BTW I am not a rap entusiast.. as it is usually a trite disregard for any kind of benevolent reality.. I do not tolerate well. If not another covert culture creation programming scheme, secretly being manufactured for the ignorant, to keep ignorant, by the crafty malevolent media conglomerates, and otherwise powers that be.

Anyone, please share frequency knowledge, particularly the effect on man's psyche..

From what I've heard so far, something to do with man's 'frequency' being around 8 hertz with brain waves 18-20 hertz (?).

There's also great info out there about how musical instruments were recalibrated about a hundred years back from 432 hertz tuning to current 440 hertz.. And this is thought to be for a nefarious reason.. research that one. "Out of tune" may not be out of tune!
Attachments
Capture_RAP_song_FREQ.PNG
Capture_RAP_song_FREQ.PNG (111.15 KiB) Viewed 940 times

steve
Site Admin
Posts: 80677
Joined: Sat Dec 01, 2007 11:43 am
Operating System: Linux *buntu

Re: Questioning the plot spectrum (frequency) analysis

Post by steve » Thu Jun 09, 2016 7:42 am

Black Dog Bluez wrote:40 hertz peaked! Wow, anyone know how they do that?
I'd guess they've got a synth bass drum sound that has a 40 Hz sine as a strong element.
An anecdote: Back in the days when Dub reggae was the big thing on the urban / underground live music scene, huge home-made speakers were a common sight (the size of wardrobes). These speakers were not well designed, they were just big, so although they produced huge amounts of bass, it was a strange kind of bass with strong resonance at certain frequencies. This was sometimes referred to as "one note bass". It became such a distinctive part of the live sound, that in more recent times, some producers have emulated this "flaw" deliberately.
Black Dog Bluez wrote:Anyone, please share frequency knowledge, particularly the effect on man's psyche..
If you're interested in the science behind the myths, look up "Neural oscillation". Pretty sure that Wikipedia has an article on the subject... yes, here: https://en.wikipedia.org/wiki/Neural_oscillation (also see the "talk" page. That's where you'll find discussion about what has been written)
Black Dog Bluez wrote:There's also great info out there about how musical instruments were recalibrated about a hundred years back from 432 hertz tuning to current 440 hertz.
Most of it is bull. Due to the amount of misinformation that was being posted about this subject, I did some research about it a while back. A very brief post that I wrote around that time: http://forum.audacityteam.org/viewtopic ... 78#p267178
9/10 questions are answered in the FREQUENTLY ASKED QUESTIONS (FAQ)

Robert J. H.
Posts: 3633
Joined: Thu May 31, 2012 8:33 am
Operating System: Windows 10

Re: Questioning the plot spectrum (frequency) analysis

Post by Robert J. H. » Thu Jun 09, 2016 11:17 am

Many (classical) composers associated certain key signatures with certain emotions.
If a part in a Beethoven symphony is called "Majestic" then it will most likely be in a majestic, triumphant key (such as D Major or A Major).
See for example:
http://www.biteyourownelbow.com/keychar.htm

A lot of music purists claim that due to the adoption of 440 Hz tuning the original intention of older compositions is somewhat lost. Mozart for instance based his work often on A4=428.

Some compositions in the orchestral context are indeed harder to sing/play with A4=440. Most noticeably flutes and brass instruments have not a linear harmonic spectrum over the their range and a tuning up can therefore make the sound thinner.
This does certainly also apply to string instruments. A violin manufactured by Stradivarius was probably differently tuned in the beginning and the resonator modes are therefore slightly off (although the art is to actually disguise these modes in order to have a almost linear sound)
I'm sure Steve knows how awful cheap fiddles can sound at certain keys where the resonance is either exaggerated or nearly cancelled by the strong corpus modes.

There's also a reason that we guitarists prefer the "CADGE" keys... or do you have a lot of songs in Eb major? ;)

Robert

Post Reply