Jump to content
Why become a member? ×

Guide to microphone positioning


charic
 Share

Recommended Posts

The problem with the FFT thing is, as an FFT is spread over time it looses accuracy. I tried spectral graphing too, but that was too innacurate. What I need to do is create a program which:

-Takes samples at 10ms periods
-Calculates the amount of samples taken
-Calculates FFT values for each sample
-Adds all values together and divides by the number of samples

You can see why this would be an extremely difficult program to create. However I've managed to gather some useful information this way :)

Enjoy the read :huh:

Link to comment
Share on other sites

hi, just finished reading it. i was going to make a post about ten minutes ago half way through, but i was only going to suggest everything you put in your '5.2 future work' section. like i said earlier, i'm sure you could make such an application using Max MSP, it's a pretty cheap and very well supported objective programming environment which i've seen apps written similar to the one you describe before.

In the context of relating mic positioning to EQ, i would have thought that timed spectral graphs would give far more objective, quantifiable and relevant data - was the inaccuracy you logged due to using white noise as a test source? from experience, swept sine tones across your test frequency band give more than adequately accurate results for relating the frequency response of a (calibrated) mic position to a similar response from an equaliser.

I would have liked to have seen the response graphs of the microphones used to give a context for testing and to be used as a calibratable baseline for assessment, then the differences in position could be overlaid and analysed in the way you describe in an environment such as Matlab.

overall, a very interesting read which has given me some great ideas for improving my mic technique - i especially liked your section on drum overheads - some surprising results which i'm itching to try in a session.

Don't think i'm being patronising by the above statement, they are just suggestions for future work which i'm sure you'll pursue due to your enthusiasm evident throughout the project. I look forward to hearing your responses to the above points, it's great to be reading something 'real' again (i'm teaching A Level music tech at a sixth form at the mo!)

Link to comment
Share on other sites

The tracks were a good 30 seconds long, the source was actual instruments rather than white noise. The inaccuracy was due to the fact that the timed spectral analysis is due to the very slight differences. Its plausible just not in the timeframe. Bare in mind i got 63 recordings from this work aswell lol. The microphones positions which were compared were all the same microphones so as to be a fair test so the response graphs seemed unneccesary.

I do wish to carry on this work, i just need the correct situation again. And your input is very much appreciated :)

Link to comment
Share on other sites

[quote name='charic' post='198620' date='May 14 2008, 11:43 AM']The tracks were a good 30 seconds long, the source was actual instruments rather than white noise. The inaccuracy was due to the fact that the timed spectral analysis is due to the very slight differences. Its plausible just not in the timeframe. Bare in mind i got 63 recordings from this work aswell lol. The microphones positions which were compared were all the same microphones so as to be a fair test so the response graphs seemed unneccesary.

I do wish to carry on this work, i just need the correct situation again. And your input is very much appreciated :)[/quote]

no problem. didn't realise the tracks were so long! i meant using swept sine tones in calibrating your test equipment, and using the frequency response plots from the microphones - either from factory testing or your own, and also in plotting the response of amps and other noise-making equipment, and not as a source for comparison, as you're working on real instruments; however, to compare the response of a mic position with the response of an equaliser set-up to deliver the same results would be an interesting addition to further works - and then the subjective application of that data to real instruments would provide an empirical baseline for your results. i agree 63 recordings is a lot, i bet you went nuts analysing it all!!

J

Link to comment
Share on other sites

Nuts indeed! I ended up with over 200 graphs to look through and compare!

Yeah it was a lot of work but more quenching personal interest than anything. Coincidently I'm considering teaching music tech at college. I learnt a lot from this and i really would like to carry on testing to gain a full set of results aswell as a program to calculate the frequency change for microphone placements. Which I'm starting to work out the basics of now :)

Glad you enjoyed the read and the overhead microphone technique sounds brilliant!

Link to comment
Share on other sites

  • 2 weeks later...

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

 Share

  • Recently Browsing   0 members

    • No registered users viewing this page.
×
×
  • Create New...