My point was that if you unplug the speaker, it will not play any sound at all. It will not play a sine sweep, MLS or stepped sine. All the microphone will record is the ambient noise. Read up the posts the lead to my earlier comment.
Just going quicly through the transcript, and stopped before halfway.
This is one very complex approach to justify the use of an usb mic.
I use Arta, in (pseudo) 2 channel, is very simple , accurate, reliable and reproducable. And of course no usb mic.
This is one very complex approach to justify the use of an usb mic.
I use Arta, in (pseudo) 2 channel, is very simple , accurate, reliable and reproducable. And of course no usb mic.
I bet just because It's easy, and cheap! Like processed food, pop into micro-owen and two minutes later belly full. Not healthy, but easy, quick win although long term bad.
Of course. But the recording part will still do the analysis!
When you have a filtered stepped sweep you will have a completely different result of the recorded noise as with an unfiltered stepped sweep. (extreme example) The ambient noise get's modified from the analysis part of the measurement! Can you follow?
Maybe just try it!
😄😄😄😄😄
It's not easier vs a audio interface + mic.
Unless you call connecting one additional XLR cable very difficult?
In that case, humankind is lost....
It's also barely cheaper, if at all.
USB mic is path of least resistance for beginner. e.g. its offered with miniDSP. only after some learning is it realized that a reference channel is a good idea
^^ Hah, yeah never underestimate power of quick win! Of course if someone is serious with measurements they'll get proper equipment, eventually. I think many just don't know enough for starters and buy the cheapest and easiest, if marketing material shows image of mic and a loudspeaker it must work, right? 🙂
Last edited:
@JanRSmit
/off topic
for now it may appear like a word linking/deleting synthesis machine. So you can ask it to translate that transcript to any other language, or ask it to write a song in the style of Tom Holkenberg’s on the topic of lcondenser microphones made by Australian RØDE, and it will do it.
Or go have a whinge about your boss or spouse and see what it says…
We ain’t seen nothing yet…
GPT5 is just months away and embodiment (humanoid robots with GPT built in) is coming…
/end
Or getting/making a additional little thing to be able to measure impedances.
Upgrade the microphone, or even getting your microphone cablibrated.
Or when it fails, everything fails
etc etc
I still don't get it. What "analysis" are you referring to? What do you mean by "filtered" sweep? What analysis or filtering do you apply to your noise recording?
But hey, in the end it does not matter so much to me. I will be able to continue with my life even if I don't get it ;-)
@mbrennwa I was afraid it's not so easy to get for some people so I spent my lunchtime and did some measurements:
https://www.diyaudio.com/community/...fferent-measurement-algorithms-graphs.410662/
Opened a new topic cause that's pretty specific. Hope this clears things.
And yes - it matters when you do acoustic measurements and want to measure THD 😎. Differences are huge. You need to KNOW what your software is doing to know if your measurements are valid!
https://www.diyaudio.com/community/...fferent-measurement-algorithms-graphs.410662/
Opened a new topic cause that's pretty specific. Hope this clears things.
And yes - it matters when you do acoustic measurements and want to measure THD 😎. Differences are huge. You need to KNOW what your software is doing to know if your measurements are valid!
The type of FFT window is a good example of this as well.
The majority of programs (and scopes) use a Hann or Hamming window, which is great for frequency resolutions, but not so great for amplitude resolution.
So for finding the right amplitude of the harmonics, something like a flattop window is a better method.
There are entire data acquisition books written on this btw.
It's only also very important not to fall into the trap of pixel peeping.
One first has to start with the question why so much resolution is needed to begin with and if it will have any significant change on the end result.
You need the resolution for exact low frequency measurements. But it creates big amounts of data and a totally useless amount of high frequency precision.
Of course not needed when you do windowing and only have a few ms to analyse anyways ...
Of course not needed when you do windowing and only have a few ms to analyse anyways ...
No thanks, I’ll keep my dual channel interface and few microphones. That is just as cheap and more versatile. Try to measure impedance or amplifier distortion with an USB mic.
pro tip for measuring speaker resonances: connect the driver to the analyser (mic) input. tap the box various placed and record the electrical signal generated by the driver. Thanks to the principle of reciprocity, this reveals the transmission from the driver to the panel being knocked on. The driver is a super low noise sensor. this method picks up acoustical box resonances as well as panel and driver resonances. you can even see the bass tuning and cone breakups.
cheers
Lars @purifi
cheers
Lars @purifi
The only drawback of this measurement is it is not absolute, that it requires some form of calibration to determine the magnitude of the panel resonances or other spurious outputs in relation to the direct driver outputs. In other words, how important is it to eliminate a specific resonance to the performance of the speaker in question.