Frequency Analysis: Sub 20Hz Frequencies Shown Across All Files

ericeholtz

New member
Hello,
I am running Adobe Audition CS6 in Windows 7 Professional 64 bit on an HP dv7-7000 Quad Core Edition laptop. All audio files show a bump below 20Hz in both Audition's Frequency Analysis, and NUGEN's Visualizer . This occurs across all files, no matter the source, i.e., personal recordings, ripped Cds, etc. Note: FFT Size is 8192
Here is an image using a ripped CD file:

frequency-analysis_CD_Audition_01.jpg

What could be causing the noise below 20Hz? This occurs on any file type, no matter whether I'm using the laptop's sound card or a USB audio interface.

Any comments and suggestions are greatly appreciated.

Thank you,
Eric
 
Hi Eric, first of all I suggest you download the free version of Right Mark Audio Analyser and repeat your tests in that RTA (need to be 16bit .wavs) .

However, the picture you posted gives me grief! If I right click on it (as I do elsewhere to get a larger picture) the post just "darkens out" and locks up. Another HR/IE11 SNAFU?) if I save to Pictures I cannot expand it because the resolution is not there, cannot read the scales.

Lastly, you might try generating white and pink noise (Audacity can do this) and see how the analyser present those?

Dave.
 
If you can't trace out the cause of the 'noise', it can be EQ'd out at some point in the mixing/mastering process. How did you discover this? Can it be heard?
 
If you can't trace out the cause of the 'noise', it can be EQ'd out at some point in the mixing/mastering process. How did you discover this? Can it be heard?

Did you manage to get better resolution on the graphs Pinks? OP might be OCD as I was when my son was buggering up the "studio" routing. We had computers, MD recorder, Sony S cassette and a Teac A3440 hooked up and a swift RTA of the noise floor often revealed hums at neg 50 or so and a bit of cable dressing could improve that by 10dB or even more.....If'n yer don't look!

Dave.
 
I have no idea why there is an issue with the image when right-clicked. The same image appears correctly on the Audition forum.
Here is another image, this one with Spectral Display on max. resolution. The source file is a sampled analog lead from Logic Pro:
frequency-analysis_Logic-lead_Audition_01.jpg
The image can be viewed larger if left-clicked. I've no idea why it is so small when right-clicked.
I am concerned about the information around 5Hz. This is a sampled synth, so there should not be anything near that region.
Note: This occurs when there is nothing connected to my computer, and when running on battery power.

Eric
 
Does this same anomaly appear with other sources/synths/plugins?

I'm leaning toward inherent noise in the capture of that synth sound. You haven't said whether it's audible (or felt) on playback. It's at a relatively low db (-45 or lower), which with other things active in the mix would be barely discernible.
 
Guys? The OP pretty specifically says it happens in all different kinds of files...

Consider for a moment that 120BPM is 2Hz. That means your 8th notes are 4Hz. ;)
 
@Pinky - No, the noise is neither audible or felt during playback. I simply do not know what could be causing such low frequency information. As I said before, it appears across all file types, no matter the source, or the bit depth and sample rate.
In addition, here is an analysis of a track from Joe Jackson's "Big World" CD. It was recorded live, direct-to-digital:
View attachment 98935
It too clearly shows all of the sub 30Hz information, with a prominent spike around 5Hz.
Just for heck of it, I opened up the Steely Dan track in Windows Media Player and took a screenshot of its visualization. It too shows all of the low freq. information that was revealed in Audition:
frequency-analysis_CD_WMP_04-1.jpg

The bottom line is that whether or not such low frequency information is detrimental, I simply would like to know what could possibly be causing such noise. I don't recall ever experiencing such on my other computer with Audition and Windows 7, or on this computer until just within the past few months.
I was really wondering if this could be computer hardware or operating system related.
 
Last edited:
Eric, one of the resident audio nerds clued us in with their above response...

If you want to fall asleep to the sound of your mind exploding, here's the 'easy' explanation :p :

https://en.wikipedia.org/wiki/Window_function

Hell's effin' teeth Pinks! I wish I had studied maths better when I was capable of taking this stuff in! TOTAL G-Gook to me. I have now remembered something, might be moot?

I used to drive valve amps with white and pink noise (in an attempt to bugger them. Hardly ever did) I used 4 different AC voltmeters to measure the noise and they all gave slightly different values despite being within 1/2 a dB of each other on sine waves.

Dave.
 
@Pinky - Holy cow, that's some light, bedtime reading right there. I'll have to put it right next to Mr. Lavry's "Sampling Theory For Digital Audio" which I've been going over...and over...and over. So, the take away from all this is that an FFT analysis window becomes increasingly unreliable in the lowest frequencies, while it excels in measuring the highest ones.

Hey, while I'm here, I am beginning work on re-transferring several analog mutitrack reels of my music group using Audition. Which forum would the most appropriate to start a thread? I mean, it's analog, but it's also digital recording with computers.
 
"Hey, while I'm here, I am beginning work on re-transferring several analog mutitrack reels of my music group using Audition. Which forum would the most appropriate to start a thread? I mean, it's analog, but it's also digital recording with computers."

The logic is undeniable to me? The SOURCE is analogue, as most are but the destination is digital so, bother the bits off Massive &Co!

Dave.
 
Hey, while I'm here, I am beginning work on re-transferring several analog mutitrack reels of my music group using Audition. Which forum would the most appropriate to start a thread? I mean, it's analog, but it's also digital recording with computers.

There was a thread in the nebie (?) area about transferring old 4 tracks tapes to digital, so maybe in that section it's appropriate.
 
Hey, while I'm here, I am beginning work on re-transferring several analog mutitrack reels of my music group using Audition. Which forum would the most appropriate to start a thread? I mean, it's analog, but it's also digital recording with computers.

All the stuff about optimizing playback from the deck fits in the analog forum. Questions about everything from the converters on can go in digital recording or something.

Definitely transfer all four tracks in one pass. Don't try to do pairs and match them up later.
 
Before I start a post regarding my analog multitrack project, I have a question about making sure the bit depth and sample rate in my DAW, audio interface, and Windows Sound Properties are all in sync. The Tascam US2000 interface shows 24 bit/88.2kHz, but the Line-In Properties in Windows 7 only shows a 16 bit setting:

Au-Preferences_Tascam-CP_Line-In-Properties.jpg

Is this a big deal, as my recorded test files are shown by MediaInfo to actually be 24 bit/88.2k? Is Windows down sampling from 24 bit to 16 bit for monitoring? Could any of this have affected my FFT analyses of low frequency signals?
I apologize if this should have been posted in a new thread.
 
I believe the windows settings can be ignored, as long as the DAW/software you're using is configured to only use the Tascam drivers (Tascam ASIO versus "windows/WDM" drivers).
 
Thanks Pinky. Adobe Audition CS6 is indeed configured to be using the Tascam ASIO drivers.
I believe all of my concerns regarding the title of this thread have been answered. My next set of concerns will be addressed in my next thread.
 
Back
Top