The Acoustic Guitar Forum

The Acoustic Guitar Forum (https://www.acousticguitarforum.com/forums/index.php)
-   RECORD (https://www.acousticguitarforum.com/forums/forumdisplay.php?f=43)
-   -   Microphone modeling the new frontier (https://www.acousticguitarforum.com/forums/showthread.php?t=479520)

KevWind 08-12-2017 07:04 AM

Microphone modeling the new frontier
 
Having had the good fortune to have actually heard and used some great high end professional studio mics, I admit have been somewhat skeptical of "Modeling"...It seems however that despite any misgivings I personally may harbor, the relentless march of technology keeps moving forward :)

So this is an interesting video about the new Townsend Labs microphone "system"
While not cheap at $1500, "IF" it does in fact get you anywhere near the sounds of the legendary mics it is modeling, it could be a real bargain

This video was released when it was still in the pre production stage and seeking "crowd funding" . Since it is listed to be in stock soon at both Sweetwater and Vintage King it appears that the production run is underway

FrankHudson 08-12-2017 08:19 AM

I tried to ask this in a different thread once, but it probably got lost in the shuffle. When I hear (old guy, not the greatest hearing; listen to a lot of music, but am not a pro-level recording engineer) a set of microphone tests I hear mostly different EQ profiles. I understand that there are also likely different Attack, Sustain, Decay, Release envelopes too, but I don't hear them as clearly, and outside of audiophile recordings that don't work on earbuds in the wild or inside a car, modern mastering and mixing will greatly blur that ADSR response in real world use.

In terms of the EQ curves, a good, well-educated ear would seem to me to be able to make a lot of mics sound like a lot of different mics with only the ubiquitous parametric EQ plugins. Of course you can't add treble or bass that info that isn't there, but much of what we record and spend the most time on getting "right" (acoustic guitar, voice) doesn't have that broad a necessary range that we need to capture. I believe there used to be a popular plugin which I tried out in the 20th Century called "Microphone Modeler" that I did this for you using a set of profiles, which I suspect were EQ profiles.

The ADSR envelope is a little tougher, but I suspect software can do quite a bit in that realm too, with Attack being the tough one.

This is not a knock against the idea of a hardware device that has these software functions built in. It's an explanation of why it might work quite well.

I await the more knowledgeable responses than mine. But if this doesn't or cannot work well, I'd like to hear why.

Brent Hahn 08-12-2017 08:28 AM

Quote:

Originally Posted by FrankHudson (Post 5441951)
In terms of the EQ curves, a good, well-educated ear would seem to me to be able to make a lot of mics sound like a lot of different mics with only the ubiquitous parametric EQ plugins.

You'd think so. Trouble is, the more educated your ears get, the more you realize it doesn't work that way.

KevWind 08-12-2017 08:52 AM

Quote:

Originally Posted by Brent Hahn (Post 5441964)
You'd think so. Trouble is, the more educated your ears get, the more you realize it doesn't work that way.

I would tend to agree if for no other reason than the EQ is still only boosting or cutting what is actually there, and what is actually there is different (perhaps very subtle at times but still different ) on different types of mics. Not to mention with vintage mics often individual mics of the same model sound different.

Also I think there is in fact more going on in the sum total of microphone sound, than just strict EQ response.

I would think it would take significantly more sophisticated algorithms to mimic the different, electronics , body size and resonance , diaphragms , etc. of different mic designs, than what is going to be found in even the most feature rich EQ plugins. Just my opinion.

Al Acuff 08-12-2017 10:16 AM

Phase coherent midrange and good off axis frequency response probably can't be modeled and emulated. Don't sell your U47!

FrankHudson 08-12-2017 11:04 AM

Quote:

Originally Posted by Brent Hahn (Post 5441964)
You'd think so. Trouble is, the more educated your ears get, the more you realize it doesn't work that way.

Not trying to be argumentative, real question:

How does it work then? Or do you mean to say something like "It's a mystery that is not understood."

FrankHudson 08-12-2017 11:32 AM

Quote:

Originally Posted by KevWind (Post 5441983)
I would tend to agree if for no other reason than the EQ is still only boosting or cutting what is actually there, and what is actually there is different (perhaps very subtle at times but still different ) on different types of mics. Not to mention with vintage mics often individual mics of the same model sound different.

Also I think there is in fact more going on in the sum total of microphone sound, than just strict EQ response.

I would think it would take significantly more sophisticated algorithms to mimic the different, electronics , body size and resonance , diaphragms , etc. of different mic designs, than what is going to be found in even the most feature rich EQ plugins. Just my opinion.

Again, my question is a sincere one, and with all with all sincere questions, it must come from ignorance. Your list above are all causes for there to be different sound qualities produced, but what would be their manifestations on the ear if not in different EQ profile and ADSR? Is there some overtone series being generated that was not there in the original sound source? Something else I'm missing?

Also I just occurs to me that pickup patterns are another significant practical difference, and one that may not be able to be removed from this discussion, even though my original question is based on listening to microphone comparisons and hearing mostly EQ as the distinction.

In my ignorance and probably degraded hearing, I hear EQ as being much more than half of what I hear when I listen to a mic A vs. mic B comparison.

I realize that amplitude response (ADSR) curves are there too, particularly in pure comparison tests with a single sound is recorded to one track with no effects and compared, but that in a final mastered recording I suspect these differences become even smaller compared to the EQ differences.

I realize questions are easy and answers are hard, but given the cost of a good microphone locker I'm wondering if it's possible to fake it to the degree than few can tell.

Bob Womack 08-12-2017 12:15 PM

Here are a couple of the biggest characteristics of a mic that can't just be EQ'd in: dynamic response and headshell resonance.

You see, a dynamic mic has a diaphragm with a little tube attached to the center around which wire is wound. That tube is fitted into a circular slot in the mic's magnet form. When the diaphragm moves the coil of wire moves within the magnet, inducing a small current into the wire. That mass of tube and wire increase the mass of the diaphragm's center and thus induces a drag that makes the mic resist compliance to impulses and high frequency sounds. So you've got to figure in compliance and dynamic response BY FREQUENCY because the dynamic response changes with frequency.

Then there is headshell resonance. Sound feflections off the inside surface of the microphone body up at the end with the pickup combine with the direct sound to cause reinforcement and cancellation of small bands of frequencies, in larger numbers and smaller bands than you can take care of with an EQ. The precise combination of these bands contributes to the character of a mic's sound.

And then there are the grosser, smoother response curves that define the mic's published response.

And that is what these programs are attempting to model. I hope that helps.

Bob

KevWind 08-12-2017 12:27 PM

Quote:

Originally Posted by FrankHudson (Post 5442155)
Again, my question is a sincere one, and with all with all sincere questions, it must come from ignorance. Your list above are all causes for there to be different sound qualities produced, but what would be their manifestations on the ear if not in different EQ profile and ADSR? Is there some overtone series being generated that was not there in the original sound source? Something else I'm missing?

Also I just occurs to me that pickup patterns are another significant practical difference, and one that may not be able to be removed from this discussion, even though my original question is based on listening to microphone comparisons and hearing mostly EQ as the distinction.

In my ignorance and probably degraded hearing, I hear EQ as being much more than half of what I hear when I listen to a mic A vs. mic B comparison.

I realize that amplitude response (ADSR) curves are there too, particularly in pure comparison tests with a single sound is recorded to one track with no effects and compared, but that in a final mastered recording I suspect these differences become even smaller compared to the EQ differences.

I realize questions are easy and answers are hard, but given the cost of a good microphone locker I'm wondering if it's possible to fake it to the degree than few can tell.


I see Bob has once again brought some great info to the discussion (and he posted while I was slowly typing my reply)

First I never assumed your question was anything but an honest sincere question. I simply think that some of the assumptions you are basing it on are misplaced and or not totally accurate.

First to clarify so everybody is on the same page..... you are not hearing "EQ" differences between mic's, EQ is process you apply to the sound of a mic or other sound source , equalization is not inherent in the tonal quality of a mic. You are hearing the sum total tonal differences, which do include but are not simply limited to, frequency response differences.


While certainly EQ is used to alter frequency response, EQ is simply a process of increasing or decreasing volume/level at a specific frequency, or a specific band of frequencies. But again tone is more complex than simple volume differences

So if we step back for a minute and propose "If what you are saying were accurate" then with a good EQ you could then supposedly make any guitar sound like any other guitar. I think you would find this is not case.

While you might be able to get frequency responses closer with EQ between mics or guitars, I think there is still going to be tonal differences .. again just my opinion.

Brent Hahn 08-12-2017 01:11 PM

Quote:

Originally Posted by FrankHudson (Post 5442127)
Not trying to be argumentative, real question:

How does it work then? Or do you mean to say something like "It's a mystery that is not understood."

Well, it's more of a mystery to me than it probably is to a lot of other other people. But let me offer up one example:

The AKG 414. Over the years there have been at least half a dozen different varieties. Different capsules, some with transformers and some without, I don't know what all else. If you were to look at the frequency-response plots of them, they all look pretty much the same. But they don't sound the same at all. In fact, even within one model type, the 414EB, there were two variations, one with a brass capsule and the other with a teflon capsule. The brass-capsule 414 sounds very bright and airy, while the teflon is a lot darker.

Now here's where it gets weird. You can record a vocal with a modern 414 that sounds just as bright as the old brass-capsule 414, and then record the same thing with the darker, older teflon EB. Then bring them up in the same mix, maybe a loudish rock band. The "dark" vocal will punch right through, while the "bright" vocal will be subjectively harder to hear. Why is that? Because the old teflon mic has a transformer in it while the new one does not. To me, it falls into the category of "a mystery that is not understood." But I nevertheless know it to be true.

FrankHudson 08-12-2017 01:43 PM

Quote:

Originally Posted by Bob Womack (Post 5442201)
Here are a couple of the biggest characteristics of a mic that can't just be EQ'd in: dynamic response and headshell resonance.

You see, a dynamic mic has a diaphragm with a little tube attached to the center around which wire is wound. That tube is fitted into a circular slot in the mic's magnet form. When the diaphragm moves the coil of wire moves within the magnet, inducing a small current into the wire. That mass of tube and wire increase the mass of the diaphragm's center and thus induces a drag that makes the mic resist compliance to impulses and high frequency sounds. So you've got to figure in compliance and dynamic response BY FREQUENCY because the dynamic response changes with frequency.

Then there is headshell resonance. Sound feflections off the inside surface of the microphone body up at the end with the pickup combine with the direct sound to cause reinforcement and cancellation of small bands of frequencies, in larger numbers and smaller bands than you can take care of with an EQ. The precise combination of these bands contributes to the character of a mic's sound.

And then there are the grosser, smoother response curves that define the mic's published response.

And that is what these programs are attempting to model. I hope that helps.

Bob

Thanks! I knew the response curve (what I tend to call ADSR, just a hold over from fooling around with synthesizers) differs between capsules and types, and that A part (attack), that fast or slow rise, a faster rise would be the tougher thing to create by filter. Your point about the limits in terms of bands or parameters of what even a series of parametric EQ filters can do adds a new (knowledgeable/experienced) perspective. I'd guess it'd be a pain in the neck to do by hand, and if even attempted you'd want to do it once and save it, which is what I guessed the old microphone modeler software plug in did years ago.

Again, I appreciate it. Questions are easy/answers are harder.

FrankHudson 08-12-2017 02:01 PM

Quote:

Originally Posted by KevWind (Post 5442219)
I see Bob has once again brought some great info to the discussion (and he posted while I was slowly typing my reply)

First I never assumed you question was anything but an honest sincere question. I simply think that some of the assumptions you are basing it on are misplaced and or not totally accurate.

First to clarify so everybody is on the same page..... you are not hearing "EQ" differences between mic's, EQ is process you apply to the sound of a mic or other sound source , equalization is not inherent in the tonal quality of a mic. You are hearing the sum total tonal differences, which do include but are not simply limited to, frequency response differences.


While certainly EQ is used to alter frequency response, EQ is simply a process of increasing or decreasing volume/level at a specific frequency, or a specific band of frequencies. But again tone is more complex than simple volume differences

So if we step back for a minute and propose "If what you are saying were accurate" then with a good EQ you could then supposedly make any guitar sound like any other guitar. I think you would find this is not case.

While you might be able to get frequency responses closer with EQ between mics or guitars, I think there is still going to be tonal differences .. again just my opinion.

Thanks! I put these "sincere" qualifiers in because on internet forums sometimes folks can't hear attitude and assume snark and attack when instead someone just doesn't understand something (and I'm the later in this case).

Yes, in the old days of 1 Mhz 8 bit processors some (me) naively thought just a little more filtering and we'd be able to approximate any instrument using only the parameters of our synthesizers. Some instruments were fairly easy to get close to (tuned percussion, flutes and the like), some not so much (oh, acoustic guitar...).

I assume digital modeling is behind most "pretty close" guitar approximations, not filter based synthesis.

jim1960 08-12-2017 03:24 PM

I haven't heard the Townsend in person but a friend of mine did a few shootouts with the Slate VMS vs a variety of mics, including some very affordable mics from Advanced Audio. The Slate was at or near the bottom virtually (pun intended) every time. He owned it less than a year and finally sold it because it was never the best option for anything.

KevWind 08-12-2017 04:44 PM

Quote:

Originally Posted by jim1960 (Post 5442377)
I haven't heard the Townsend in person but a friend of mine did a few shootouts with the Slate VMS vs a variety of mics, including some very affordable mics from Advanced Audio. The Slate was at or near the bottom virtually (pun intended) every time. He owned it less than a year and finally sold it because it was never the best option for anything.

Yes the Slate has gotten mixed reviews.
Obviously the proof would be in actually hearing this Townsend , we can theorize, speculate, postulate, and pontificate, this that and the other, till the cows come home .. None of which holds a candle to being in the room and trying it.

jim1960 08-12-2017 05:22 PM

Quote:

Originally Posted by KevWind (Post 5442456)
Yes the Slate has gotten mixed reviews.
Obviously the proof would be in actually hearing this Townsend , we can theorize, speculate, postulate, and pontificate, this that and the other, till the cows come home .. None of which holds a candle to being in the room and trying it.

Absolutely. Unfortunately for the Slate system, I was in the room for one of those shootouts.


All times are GMT -6. The time now is 03:47 PM.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.
Copyright ©2000 - 2022, The Acoustic Guitar Forum

vB Ad Management by =RedTyger=