WRT an RIAA equalized feedback preamp with about the right gain at 1 KHz (I think around 35 dB) the HF boost comes for free, until the open loop gain bandwidth is used up. 20+ years ago I incorporated that tweak (I did not know of any of this) into a hybrid passive/active EQ solution with a switch. Generally I preferred the boosted version and I'm still not convinced its because it corrected for the cutting side.
On the digital side i could not make any sense of KBK's dissertation, its over my head. But I know that no recordings are made with microphones with the extended bandwidth necessary to need 100 KHz bandwidth and that the analog media never had that bandwidth. Getting the bandwidth in analog required a tradeoff in SNR that would not be acceptable for recording.
Digital fixed the record/reproduce limits with virtually perfect amplitude phase response if desired. But microphones and speakers are still not anywhere near that frequency range or phase perfection.
True about the microphones. Microphones that go flat to 20kHz are rare, ones that are also quiet enough for pro audio recording are much rarer. For example, the response of the Neumann U87 studio condenser mic, a favorite of recording engineers apparently -- https://www.neumann.com/zoom.php?zoomimg=./assets/diagrams/u87ai_diagrams.htm&zoomlabel=Diagram&w=878&h=295.
It's dropping fast at 17kHz and only makes it there by virtue of an on-axis peak. 96kHz sampling records pretty much just noise above 22kHz in music media.
It's dropping fast at 17kHz and only makes it there by virtue of an on-axis peak. 96kHz sampling records pretty much just noise above 22kHz in music media.
You forget you are upside down, I could say do you guys sleep all day.
Being upside down, good for the circulation.
Well, the curves you are showing is in its omnidirectional configuration. It can be tuned in Cardioid and 8 as well, with different response curves. Note too that it is, most of the time, used in cardioid and proximity. With the big size of the membranes, this increase the bass and low medium and gives a lot of warmth and presence. Voices don't need very high frequencies.
In the same time, we will use other brands of mics with extended bandwidths for acoustic guitars etc.
Last edited:
Having 30+ years of experience with a bunch of mics, the Neumann's are FAR from perfect... Being a centre terminated capsule they waste the main vibration mode of the diaphragm, and being large diaphragm they have significant polar response issues with lobing at high frequencies.
George Wuttke of Schoeps gives a convincing account of why small diaphragm mics are significantly better, less coloured etc. For most popular recording, the mic becomes part of the colouring used to get the sort of sound they want - typically like someone else's sound. The Neumann's are quite euphoric... They don't pretend it's true to life, but it's what they want.
George Wuttke of Schoeps gives a convincing account of why small diaphragm mics are significantly better, less coloured etc. For most popular recording, the mic becomes part of the colouring used to get the sort of sound they want - typically like someone else's sound. The Neumann's are quite euphoric... They don't pretend it's true to life, but it's what they want.
Last edited:
I had a U67 witch was a pure wonder. Never used-it for something else than voices. And, yes, i used a lot of Shoeps for guitars and cymbals... and never for voices ;-)
And, yes, we use mics to "create the sounds" we want. It is not a search for "real life", but for "make believe" and hyper realism. Like Photoshop, see what i mean.
people should be disappointed with the real voice of most the singers !
And, yes, we use mics to "create the sounds" we want. It is not a search for "real life", but for "make believe" and hyper realism. Like Photoshop, see what i mean.
people should be disappointed with the real voice of most the singers !
Last edited:
In a conventional active RIAA equalizer using the Lipshitz design equations and methodology, the top end (here I mean 20 kHz) gain is usually 0.5 to 1 dB higher than it should be. This is simply because, as Lipshitz explains, using an all active feedback network, you cannot get a perfect fit in the top octave. To correct for this, a 'post filter' pole is added that lies above 50 kHz and in some cases as high as 100 kHz. The location of this pole is dependent on the accuracy of the feedback network and the interaction of the pole zeroes in the audio band and is usually set empirically - I use LTspice but practically you could look at the square wave response or inverse network flatness.
I built a Lipshitz spread sheet which is up on my website and if you want accurate all active RIAA you need the post filter pole. Self discusses this as well.
Is this perhaps where the Von Neumann confusion comes from that we read about?
Note, passive EQ does not have this problem.
I built a Lipshitz spread sheet which is up on my website and if you want accurate all active RIAA you need the post filter pole. Self discusses this as well.
Is this perhaps where the Von Neumann confusion comes from that we read about?
Note, passive EQ does not have this problem.
Last edited:
The real challenge (as if there were not enough in the studio) is to get a system in a home that can be a chameleon of all that. One is art, is producing, the other is reproducing. Let's face it, the latter is rarely achieved to a great level. But it can be done.
Re the sound of microphones, what about ribbons? They have a very low frequency resonance, then working above that can be quite clean, whereas condensers have to work both below and above that (clearly the designer fiddles with it to get a certain result). Anyone recall those James Boyk piano recordings where some complained they sounded dull, but in fact used ribbons. So condensers generally sounds more toppy, but are not response wise. So if you have a resonance that's manipulated flat does not necessarily make it sound flat, it's still there.
Re the sound of microphones, what about ribbons? They have a very low frequency resonance, then working above that can be quite clean, whereas condensers have to work both below and above that (clearly the designer fiddles with it to get a certain result). Anyone recall those James Boyk piano recordings where some complained they sounded dull, but in fact used ribbons. So condensers generally sounds more toppy, but are not response wise. So if you have a resonance that's manipulated flat does not necessarily make it sound flat, it's still there.
- Status
- Not open for further replies.