Understanding FFT in Audio Measurements

[Lambda]

One of the biggest things often stated is my room noise floor is “x” so all modern devices are transparent

That's oversimplification/wrong.
Imagine your romm nois is 50dB but only at 20-50Hz that's not hindering you from hearing a 2kHz sine wave at 10dB.

 

[Dj7675]

That's oversimplification/wrong.
Imagine your romm nois is 50dB but only at 20-50Hz that's not hindering you from hearing a 2kHz sine wave at 10dB.

Sorry I wasn’t clear. That is my point... the linked article explains it really well of why that thought is wrong. It would be really great to have that explanation in video form is the point I was attempting to make.

 

[tarikuz]

Well done! Easy to understand very nice movie.

 

[eliash]

FFTs are the fundamental to many measurements I show in my reviews. You can search for it online but I am not sure you will find any easy to understand explanation of it that applies to the way we use them in audio measurements. So I thought I do a walk-through using the Audio Precision of what it means, and how its parameters can drastically change what it displays.

As I asked the video, please let me know if you like these kinds of technical tutorials. To be sure they are not as much "fun" to watch as audio debunking ones. So like to know if you want to see more of them or not. And whether you could follow this one which is probably one of the most difficult concepts we have in audio measurements.

While I have not done a full text write up on this, the concept is included in my tutorial on digital audio measurements: https://www.audiosciencereview.com/...derstanding-digital-audio-measurements.10523/

Great stuff, thanks!
...Just playing around with my new (cheap) Rigol digital oscilloscope, which can do FFT too, observing similar things...seems limited to some -60dB of noise floor though. Will try out to check at least vinyl cartridge distortion and see if it matches...

 

[rodrigaj]

To answer your question - yes, please continue with the technical videos. Easy to understand if one is willing to put forth a small amount of effort.

 

[pufferfish]

The 3blue1brown explanation of the Fourier transform is quite good (not fft algorithm) if anyone is interested

 

[GeorgeBynum]

FFTs are the fundamental to many measurements ...

I use data acquisition in motion control systems. Looking at forces and positions is routine. My DAS (Data Acquisition System) software offers FFT for the physical world as yours does for the electronic world. I found your presentation accurate and well done.

A suggestion I would make is to have mentioned that the square wave odd harmonic contribution amplitude is 1/harmonic number ... the 3rd harmonic contribution is 1/3 amplitude (about -5dB), 5th is 1/5 (about -7dB), etc. That might allow better understanding of the diminishing levels on the frequency domain display. And I would have used a triangle wave as well discussing harmonic levels as well.

I wonder if audio systems are ever tested with square wave inputs and the output frequency domain compared to the ideal for a square wave. That would seem to "test" with all frequencies from the fundamental to equipment limits.

I don't recall your mention of your education background. My 1st year EE courses (1969) discussed Fourier (the fast wasn't discussed then, although in the literature in 1965, some 4 years before my EE Fourier introduction) and laPlace and their places in circuit analysis. The "unit impulse" containing all frequencies at a uniform amplitude (white noise) would seem as valuable in an audio system as in theory. Would not a square wave provide similar analysis capability via FFT.

 

[Billy Budapest]

Well done video, i think it is even better then debunking video's.

Agreed. I think that a debunking video every once in a while is good, but IMHO you don’t want debunking to become a “religion” with “followers” as ardent as the audiophiles who profess belief in snake oil solutions. I don’t think you would want to be known as the next James Randi.

 

[eddantes]

Thank you for this video and the one in the comments. I can't imagine spending Sunday afternoon in a better way.

 

[Wes]

Yes, I like these kinds of technical tutorials.

Also, I predict an anti- thread will be started at Audiophool Stylism about this.

 

[ezra_s]

I have to re-watch this, trying to watch it on sunday after lunch was too much for me, in any case I want to appreciate the efford you are making to explain these things to people like me, thank you thank you.. A video a day since you appeared on youtube for the first time also, amazing!

 

[sabristol]

FFTs are the fundamental to many measurements I show in my reviews. You can search for it online but I am not sure you will find any easy to understand explanation of it that applies to the way we use them in audio measurements. So I thought I do a walk-through using the Audio Precision of what it means, and how its parameters can drastically change what it displays.

As I asked the video, please let me know if you like these kinds of technical tutorials. To be sure they are not as much "fun" to watch as audio debunking ones. So like to know if you want to see more of them or not. And whether you could follow this one which is probably one of the most difficult concepts we have in audio measurements.

While I have not done a full text write up on this, the concept is included in my tutorial on digital audio measurements: https://www.audiosciencereview.com/...derstanding-digital-audio-measurements.10523/

Thanks for this it's an excellent video @amirm.

As the the proud owner of a QuantAsylum QA-401 / QA4-451, which may not be in the same league (price or spec) as the AR equipment but do an excellent job, I found this video a really good reminder. The noise floor changing due to FFT size was new to me and left me with the question "how should I actually go about measuring an amplifiers noise floor then?".

Yes I would like to see more on testing tutorials. A focus on specific parameters: THD, THD+N, IMD, output impedance, S/N ratio, FR etc could be the way to go.

Thanks again.

 

[BDWoody]

As the the proud owner of a QuantAsylum QA-401 / QA4-451, which may not be in the same league (price or spec) as the AR equipment but do an excellent job, I found this video a really good reminder. The noise floor changing due to FFT size was new to me and left me with the question "how should I actually go about measuring an amplifiers noise floor then?".

Yes I would like to see more on testing tutorials. A focus on specific parameters: THD, THD+N, IMD, output impedance, S/N ratio, FR etc could be the way to go.

I've got the same setup, along with the 480, and I have to say I am a long way from being able to make it look easy. The more I poke around with it, the more things make sense, but I still find myself doing stupid things like forgetting to properly account for the +/- gain being applied, etc. I got an autoranger to go in front of it, and I'm trying to figure out what to use for a dummy load so I can get more into playing with my amps, but I've been having fun with it as I learn what it can do.

A video on how to do some simple measurements with a sound card might give a lot of people something to dig into. The basics aren't that hard. It's everything else that comes after!

 

[hansik]

FFTs are the fundamental to many measurements I show in my reviews. You can search for it online but I am not sure you will find any easy to understand explanation of it that applies to the way we use them in audio measurements. So I thought I do a walk-through using the Audio Precision of what it means, and how its parameters can drastically change what it displays.

As I asked the video, please let me know if you like these kinds of technical tutorials. To be sure they are not as much "fun" to watch as audio debunking ones. So like to know if you want to see more of them or not. And whether you could follow this one which is probably one of the most difficult concepts we have in audio measurements.

While I have not done a full text write up on this, the concept is included in my tutorial on digital audio measurements: https://www.audiosciencereview.com/...derstanding-digital-audio-measurements.10523/

Thank you Amir, it´s a pleasure to listen to your clear explanation of things previously tought too difficult for me.

 

[TheTalbotHound]

The 3blue1brown explanation of the Fourier transform is quite good (not fft algorithm) if anyone is interested

someone else covered the DFT/FFT in a similar style if you want that too

 

[Helicopter]

I had to pay attention and rewind a couple times, which is nice. In the league with the audiobooks I most enjoy on long drives or, in the before times, on my commute.

 

[bogart]

Really enjoyed this video. I liked the methodical explanation and demonstration of how different settings lead to different results all from the same input signal. It’s really helpful to see it live and in real time.

I still find myself struggling to map time and frequency domains back and forth mentally. It’s helpful to be reminded that math tells us we don’t need to

 

[scott wurcer]

Let me contribute a slightly different presentation that I find wonderful because it is based on polynomial multiplication which is more basic algebra and understood by more folks in general.

 

[anchan]

I also have a similar setup- an older QuantAsylum QA-400.
I usually run 32K points, and Hann windowing because.... actually I have no idea. Seems like a happy median setting. No idea how to compute the FFT gain. And also, does the FFT gain affect just the bottom edge of the noise floor, or the noise peaks? Maybe Matt from QuantAsylum can make some videos like this for his product? hint hint
I have to make an unsolicited and unaffiliated plug for him btw- it's a great product and tremendous value.

Thanks for this it's an excellent video @amirm.

As the the proud owner of a QuantAsylum QA-401 / QA4-451, which may not be in the same league (price or spec) as the AR equipment but do an excellent job, I found this video a really good reminder. The noise floor changing due to FFT size was new to me and left me with the question "how should I actually go about measuring an amplifiers noise floor then?".

Yes I would like to see more on testing tutorials. A focus on specific parameters: THD, THD+N, IMD, output impedance, S/N ratio, FR etc could be the way to go.

Thanks again.

 

[Blumlein 88]

I also have a similar setup- an older QuantAsylum QA-400.
I usually run 32K points, and Hann windowing because.... actually I have no idea. Seems like a happy median setting. No idea how to compute the FFT gain. And also, does the FFT gain affect just the bottom edge of the noise floor, or the noise peaks? Maybe Matt from QuantAsylum can make some videos like this for his product? hint hint
I have to make an unsolicited and unaffiliated plug for him btw- it's a great product and tremendous value.

Your FFT gain is 42 db. It is in effect if you have wide band noise wider than the FFT bins. If you have a resonant or narrow frequency range it wouldn't apply.