Melodeon.net Forums

Please login or register.

Login with username, password and session length
Advanced search  

News:

Welcome to the new melodeon.net forum

Pages: [1]   Go Down

Author Topic: Working with frequency response curves  (Read 1273 times)

0 Members and 1 Guest are viewing this topic.

gmatkin

  • Gavin Atkin
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 708
    • Julieandhersqueeze.com
Working with frequency response curves
« on: October 26, 2018, 02:54:05 PM »

I've been doing some experiments, and have learned how useful it can be to compensate out some of the peaks and troughs you find in manufacturers' published frequency response curves. Next time you're working with a mic at home and find you like a mic you have on one source and not another, check the curves, draw the opposite graph in Audacity or whatever you use and your view might be altered...

I don't find this thought is often found on the Internet, so I mention it as one amateur to a bunch of others.

Here's a YouTube that may or may not reveal the benefit of this EQ process, depending on YouTube and your equipment. https://youtu.be/j0MP_Vur2wM

Gavin
« Last Edit: October 26, 2018, 03:40:36 PM by gmatkin »
Logged

Dick Rees

  • Neo-Luddite
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1240
Re: Working with frequency response curves
« Reply #1 on: October 26, 2018, 05:06:21 PM »

Gavin...

This is an interesting project.  I've worn many hats over the years:  musician, live sound/broadcast/recording technician and avid listener.  The common frequency response charts you see from manufacturers are heavily "smoothed" and give only a fairly gross indication of the characteristics.  Increase the resolution and things get squidgy fairly quickly.

If you know anyone who owns an Allen & Heath sound console such as a Qu series or above, they offer a real-time, scrolling spectrographic frequency display which gives a wonderful visual display across the frequency spectrum as you play, allowing you to directly observe the changes as you apply parametric EQ to the mic signal. 

I've used Audacity and find it a very nice tool...and FREE is good.  I do find the real-time spectrographic display more to my liking, though.   I think your example holds up well enough on YouTube through my iPad into Sennheiser HD-25 headphones.  It would be very interesting to re-do the video with a spectrographic display as well.  That would tend to offset any loss or skewing of audio in the YouTubing.

Keep up the good work.
Logged
"You're making the wrong mistakes."
...Thelonius Monk

"I never made one of my discoveries through the process of rational thinking."
...Albert Einstein

Cogito ergo bibo.
...Bacchus

gmatkin

  • Gavin Atkin
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 708
    • Julieandhersqueeze.com
Re: Working with frequency response curves
« Reply #2 on: October 26, 2018, 09:05:40 PM »

Thanks Dick. A real time analysis would be a boon, though I'm a bit overwhelmed by Audacity's snapshot analyses of selected samples.

As you'll realise I'm thinking about how to do the best job I can - and the rest of us can - on the cheap.

I realise about the smoothing. From looking at specs and response curves, it'spretty obvious that more standardisation in the specs could make comparisons easier. Though of course it isn't just about the numbers.

I suppose I should now do something similar with a Rode M3, as they are cheap, I have a couple and I'm inclined to think a good result could be got from them.

Gavin

Winston Smith

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 3769
  • AKA Edward Jennings
    • "Our Luxor B&B" Luxor life, slice by slice.
Re: Working with frequency response curves
« Reply #3 on: October 26, 2018, 10:04:24 PM »

Beam me up, Scotty!
Logged
At last, broken and resigned to accept conformity.
Oh, how I LOVE Big Brother!

gmatkin

  • Gavin Atkin
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 708
    • Julieandhersqueeze.com
Re: Working with frequency response curves
« Reply #4 on: October 26, 2018, 11:16:53 PM »

This /is/ life, Jim!

Winston Smith

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 3769
  • AKA Edward Jennings
    • "Our Luxor B&B" Luxor life, slice by slice.
Re: Working with frequency response curves
« Reply #5 on: October 26, 2018, 11:26:53 PM »

But not as we know it! (Well, some of us, anyway.) I'll stop now.
Logged
At last, broken and resigned to accept conformity.
Oh, how I LOVE Big Brother!

Fred

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 719
  • G/C Pariselle, D/G Loffet & maybe too many Hohners
Re: Working with frequency response curves
« Reply #6 on: October 26, 2018, 11:34:31 PM »

Well, what shall I say. I don't hear any difference. Like none at all. Is it me or is it youtube? Nobody knows for sure.
Logged

Dick Rees

  • Neo-Luddite
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1240
Re: Working with frequency response curves
« Reply #7 on: October 26, 2018, 11:52:17 PM »

There is a slight bit of a difference in the instrumental passes, but for purposes of judging the changes I would prefer to hear the original track before the processed track.  Or even better, taking the EQ in and out several times in the same track to A/B any effect.  As is it's hard to rule out any expectional or conformation bias.

I wouldn't expect to hear any change in the voice at all as the critical voice frequencies fall into the 2k to 5k range and the HF peakiness on that mic is 1 to 2 octaves above that, give or take.

Still interesting and worth the time spent.  Any chance you kept notes on the frequencies adjusted and to what degree?  Inquiring minds...
« Last Edit: October 27, 2018, 12:04:25 AM by Dick Rees »
Logged
"You're making the wrong mistakes."
...Thelonius Monk

"I never made one of my discoveries through the process of rational thinking."
...Albert Einstein

Cogito ergo bibo.
...Bacchus

gmatkin

  • Gavin Atkin
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 708
    • Julieandhersqueeze.com
Re: Working with frequency response curves
« Reply #8 on: October 28, 2018, 06:57:10 PM »

Neither my darkish voice nor my melodeon are the most demanding things to mic.

If it's any help, listening to the YouTube I still feel the first sample is more natural, both on the Samson SR850 headphones I've got and the Alesis nearfield monitors that arrived yesterday (and have no been run in).

Here's my curve. It's pretty approximate...




Gavin

Dick Rees

  • Neo-Luddite
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 1240
Re: Working with frequency response curves
« Reply #9 on: October 28, 2018, 08:19:26 PM »

Neither my darkish voice nor my melodeon are the most demanding things to mic.

If it's any help, listening to the YouTube I still feel the first sample is more natural, both on the Samson SR850 headphones I've got and the Alesis nearfield monitors that arrived yesterday (and have no been run in).

Here's my curve. It's pretty approximate...




Gavin

The graph shows filter center frequencies.  The actual resulting curves will depend on filter width.  The summed effect will look markedly different.
Logged
"You're making the wrong mistakes."
...Thelonius Monk

"I never made one of my discoveries through the process of rational thinking."
...Albert Einstein

Cogito ergo bibo.
...Bacchus

gmatkin

  • Gavin Atkin
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 708
    • Julieandhersqueeze.com
Re: Working with frequency response curves
« Reply #10 on: October 29, 2018, 12:17:08 AM »

Quite so - and any remainder is likely less than 3dB, and in any case I like the result better. Your mileage may vary and you might do it better and more meticulously - as I likely will if I'm going to use such a curve regularly. And I think I likely will, though possibly not with this mic, as I've got SDCs that I like better on my melodeons and another LDC that I like better for vocals.

To be fair, I chose this mic for this exercise because it was inexpensive and also comes with a fairly detailed response curve. They're not all like this...

Gavin
« Last Edit: October 29, 2018, 09:33:43 AM by gmatkin »
Logged
Pages: [1]   Go Up
 


Melodeon.net - (c) Theo Gibb; Clive Williams 2010. The access and use of this website and forum featuring these terms and conditions constitutes your acceptance of these terms and conditions.
SimplePortal 2.3.5 © 2008-2012, SimplePortal