Soliloqueen's k87(k67) and k47 capsules

[Roman Beilharz]

Not sure what you mean by this, you do realize you are hearing with your brain, not ears? Brain compensates for this. Hadn't we had linear equipment this thread and capsules discussed here wouldn't exist.

Yes but no. Our hearing differs a lot, regardless of brain compensation. Plus our HRTFs are like fingerprints: highly individual - I know we are not talking about localization here, but still - it factors in. Average English native speakers are more sensitive to frequencies above 7kHz than, say, an average German like me, which as a achieved higher sensitivity for consonants in the midrange. Hence culturally achieved and regional differing preferences when it comes to e.g. mastering. So our brain compensates, but not for everything, unfortunately.

Time domain errors and phase cohrence issues show in frequency response. Røde capsules have no special issues of their own regarding phase or time domain, they are as flaud as k67. Maybe you want to share measurements of these errors? How do you measure and correct for c100's ultrasonic range? How do phase, and time domain issues behave at c100's crossover region? C100's smaller capsule is of different construction than large one, how does the polar pattern look time domain wise at crossover from the two capsules? Sure that crossover frequency is 20K?

That's what I've read somewhere, but can't hickup the source right now. The SDC is taking over up until 50k, because the mic is also designed to work for HR recordings (useful e.g. for extreme down-pitching without lots of artifacts).

View attachment 131308
You don't think this area shows serious time domain/phase issues?

This could absolutely be the case, but as I am using the C100 in cardioid most of the time and I rather actually use my mics more often than measure them, those issues don't bother me in my everyday recording business. But what really matters to me is natural and believable sounding results - and fast transients. The NT1 sounds narrow and unnatural to me even over my smartphone speaker. But maybe it's just me.

I have to clarify: I am not so much focused on time domain errors in terms of multipattern characteristics, but rather on the speed/timing/ringing of the capsule/circuit in cardioid and maybe omni. I have not yet found a good way to measure and visualize fast attacks and sustains = a quick and natural transient response of mics with my equipment. But the differences are easily audible - at least with drums and percussion. I could post snippets to underline this, but I would really love to back off and cut all that C100 talk now to make room for Adrienne's capsules again. Let's open another thread if you guys would like to discuss this any further. OK?

Ro

 

[kingkorg]

The only reason i entered the "discussion" is because these aspects are relevant when it comes to the original post due to the diverse aspects of capsules, and to defend frequency response measurements as somehow inferior just becouse people don't know how to interpret them.

Weirdly, i know more about how brain works and how it misbehaves, than i'll ever know about microphones. Even across diferent nationalities. It's what i do for living. I'm sure we wouldn't agree in this area either, and absolutely not worth discussing here. I don't recommend any kind of DIY work with brain. Not even meditation
I do measurements because i don't trust my brain

 

[Roman Beilharz]

King, this is actually getting really interesting and yes, I am absolutely on your side thinking FR measurements are worth considering! It's just not the whole truth IMO, when it comes to carefully listening and actually using those devices for a living. Timing in relation to frequency is also more relevant for me when it comes to, say, speaker or headphone systems. Fast and transient-wise accurate (quick amplitude rise/little post-ringing) systems with a not-so-ideal FR sound better to me than slow or ringing/lagging systems with an ideal FR. So I am wondering, how this phenomenon impacts mic capsules as well: metric partials/harmonics and speed of the diaphragm kicking in at given frequencies/SPL/capsule tensions...

And I know you are way more experienced to interpret FR graphs properly than me . I carefully measure every step of my builds as well as good as I am capable of. But at the end of the day, what matters is what it sounds like to me and my customers in a given situation (instrument/style/preamp/musician/room, you name it), because luckily I can trust my own ears/brain when fed with audio over my speakers in my carefully treated room to a wide extent (at least in the mornings after a good coffee).

SO: I WILL ORDER A PAIR OF ALL ARIENNE CAPSULES AND SEE FOR MYSELF.

 

[Dylan W]

Yes but no. Our hearing differs a lot, regardless of brain compensation. Plus our HRTFs are like fingerprints: highly individual - I know we are not talking about localization here, but still - it factors in. Average English native speakers are more sensitive to frequencies above 7kHz than, say, an average German like me, which as a achieved higher sensitivity for consonants in the midrange. Hence culturally achieved and regional differing preferences when it comes to e.g. mastering. So our brain compensates, but not for everything, unfortunately.



That's what I've read somewhere, but can't hickup the source right now. The SDC is taking over up until 50k, because the mic is also designed to work for HR recordings (useful e.g. for extreme down-pitching without lots of artifacts).


This could absolutely be the case, but as I am using the C100 in cardioid most of the time and I rather actually use my mics more often than measure them, those issues don't bother me in my everyday recording business. But what really matters to me is natural and believable sounding results - and fast transients. The NT1 sounds narrow and unnatural to me even over my smartphone speaker. But maybe it's just me.

I have to clarify: I am not so much focused on time domain errors in terms of multipattern characteristics, but rather on the speed/timing/ringing of the capsule/circuit in cardioid and maybe omni. I have not yet found a good way to measure and visualize fast attacks and sustains = a quick and natural transient response of mics with my equipment. But the differences are easily audible - at least with drums and percussion. I could post snippets to underline this, but I would really love to back off and cut all that C100 talk now to make room for Adrienne's capsules again. Let's open another thread if you guys would like to discuss this any further. OK?

Ro

Not to the derail the thread further, but I agree with what Roman is saying here. Maybe the language being used isn’t exactly correct, but perception of clarity in the time domain (ie transient response) is very important. I understand that in a minimum phase sense freq response and transient response are linked, but that doesn’t explain why two systems with same frequency response can have very different perceived clarity on transients. There are other ways to explain this. I believe distortion modes would be one, which in capsules could easily be related to the mechanical properties of the capsule, type of Mylar, and so on. Very narrow resonant peaks may be another, which *are* obviously tied to frequency response but would not necessarily show up in a smoothed graph.

 

[kingkorg]

Not to the derail the thread further, but I agree with what Roman is saying here. Maybe the language being used isn’t exactly correct, but perception of clarity in the time domain (ie transient response) is very important. I understand that in a minimum phase sense freq response and transient response are linked, but that doesn’t explain why two systems with same frequency response can have very different perceived clarity on transients. There are other ways to explain this. I believe distortion modes would be one, which in capsules could easily be related to the mechanical properties of the capsule, type of Mylar, and so on. Very narrow resonant peaks may be another, which *are* obviously tied to frequency response but would not necessarily show up in a smoothed graph.

You nailed it. I am all about eliminating frequency response as a variant. One of the reasons i started measuring mics is when i manage to match the FR of two mics i want to quantify what else makes them different. And a lot has to do with stuff in time domain. However it is important to use the right terminology, and probably impossible to talk about topic like this without very precise technical stuff in order not to perpetuate mythology that let the whole comunity astray in the first place.

Most important aspect of any mic, that literally never comes up is how three dimensional sound is rendered to a simple mono audio file. Three dimensions of space simply colapse to one. This collapse of three-dimensional space is incredibly complex, and probably impossible to quantify, measure, present. A lot of things happen that contribute to how the end result feels, and very little has to do with frequency response.

 

[soliloqueen]

A frequency response is only a processed, highly limited cross-section of the super complex 4 dimensional response of a microphone. There's no more "the" frequency response of a microphone then there is "the" picture of a lake. It's useful for visualizing certain problems and aspects but it's not prescriptive

 

[garp]

A lot of things happen that contribute to how the end result feels, and very little has to do with frequency response.

I have been fighting for so long with a lot of ingeneers totally blind (deaf ?) of this concept.

 

[kingkorg]

There's no more "the" frequency response of a microphone then there is "the" picture of a lake.

What a beautiful analogy!

I have been fighting for so long with a lot of ingeneers totally blind (deaf ?) of this concept.

Exactly, but now when i read just the part you quoted, i get how easy it would be to take it out the context.

What's important is to take the FR out the equation, by compensating for it, and do it the right way. If not compensated for, there's no point moving on to discuss other, more nuanced, and complex stuff like 3D, or even 4D image rendering, if we are swayed by FR difference. Not to mention taking anecdotal vs empirical route. It's no different to shooting out two mics, when one of them is 6db louder. It will surely 'feel' better.

 

[garp]

For me FR is a picture of something constantly moving ... therefore just informative.

 

[rogs]

....... Most important aspect of any mic, that literally never comes up is how three dimensional sound is rendered to a simple mono audio file. Three dimensions of space simply colapse to one. This collapse of three-dimensional space is incredibly complex, and probably impossible to quantify, measure, present. A lot of things happen that contribute to how the end result feels, and very little has to do with frequency response.

I first realised this many years ago....
I live in a small(ish) seaside resort on the south coast of the UK. By a quirk of fate the 'town band' - first created over 100 years ago - has turned into a world class orchestra - the Bournemouth Symphony Orchestra
I have had the chance to hear the orchestra on many occasions over the years, in varous local venues.

What I can categorically state is that the Orchestra - live - sounds nothing like any recorded version of any of their music . Same for any orchestra.
The best microphones, the best mixers and audio recorders do not even come close to reproducing all the aspects of sound that make a live concert so special..... The 3 dimensional sound space, the subtle interactions at ultrasonic frequencies, the totally uncompressed dynamic range ... the list goes on.
Here we are discussing the benefits - or otherwise - of artificial nuances in electronic items that can only make a falry crude stab at capturing the subtleties of the real acoustic world...... Even the very best don't really come that close.

Good fun experimenting though !

 

[mrgrooves666]

SO: I WILL ORDER A PAIR OF ALL ARIENNE CAPSULES AND SEE FOR MYSELF.

That is probably your best bet, indeed

 

[kingkorg]

I first realised this many years ago....
I live in a small(ish) seaside resort on the south coast of the UK. By a quirk of fate the 'town band' - first created over 100 years ago - has turned into a world class orchestra - the Bournemouth Symphony Orchestra
I have had the chance to hear the orchestra on many occasions over the years, in varous local venues.

What I can categorically state is that the Orchestra - live - sounds nothing like any recorded version of any of their music . Same for any orchestra.
The best microphones, the best mixers and audio recorders do not even come close to reproducing all the aspects of sound that make a live concert so special..... The 3 dimensional sound space, the subtle interactions at ultrasonic frequencies, the totally uncompressed dynamic range ... the list goes on.
Here we are discussing the benefits - or otherwise - of artificial nuances in electronic items that can only make a falry crude stab at capturing the subtleties of the real acoustic world...... Even the very best don't really come that close.

Good fun experimenting though !

Couldn't agree more. My last experience of this sort was in a small tavern in Lisbon, Portugal.

Great bunch of musicians were performing Fado, moving around our table with nothing but acoustic instruments. I probably seamed crazy to everyone, but i spent almost the whole time with closed eyes, concentrated on the sound of the instruments in the room as they were moving around me. I also tilted and turned my head to "mix" what i was hearing, enjoying the alteration of binaural field. I was thinking the whole time, how in the world could a microphone capture this? No way! Kinda what Soliloqueen said, you can only take a snapshot of this lake of sound.

I almost got to meet the amazing @Whoops on this trip, but the stars didn't align.

Infinitely charismatic, coolest ever Dave Rat made a simple, fun project to try and experiment with this kind of stuff. Has a couple more vids on this concept. To anyone interested in audio, and not following Dave, subscribe immediately!

 

[ElSmurf]

To keep the sidetrack alive, I always wonder how much sound engineers in the forties and fifties cared about any of these abstract technical considerations. And yet they were able to make some incredible sounding recordings, some of which even hold their own today.

The way I see it, this focus on technicality is a must if you're actually designing capsules - or anything else for that matter. The more information you can gather through measurements, and the more accurate these measurements are, the better informed your design decisions will be. Then afterwards, us lowly DIYers can cram the result of that meticulous process in our little DIY mics and marvel at the result. My own journey in DIY mic land has convinced me that with the right parts, you can make very good and cost-effective microphones, especially when compared to the top tier overpriced commercial stuff out there. And people like @soliloqueen make this possible for us.

 

[Roman Beilharz]

YES, EISmurf!! Plus in regards to the OT - the development of fine mic capsules, their sound properties and and capabilities - I quickly want to add some information that I came up with digging deeper into the rabbit hole last night:

1. A lengthy, highly controversial but still interesting/informative thread about whether or not condenser microphones are minimum-phase devices and thus all of their frequency and time domain properties can be completely grabbed, analysed and back-and-forth transformed or not:
https://gearspace.com/board/remote-...aphram-fast-small-diaphrgam-condensers-9.html

2. A paper by Neumann developer Stephan Peus regarding Transient Response of microphones


My conclusion for now in respect to question 1: Regardless whether we use a sweep, shot or a DIRAC for the measurement - I'd now vote for YES, it's all in the FR/FFT data, but it's really hard to visualize the timing of the first 100-or-so microseconds of an impulse over the audible range in a meaningful way. Plus intermodulation distortion of complex musical signals will further degrade the clarity.

In regards to 2. it seems to be common sense in microphone development, that it was feasible to construct (and there are quite some examples of) microphones that have decent FR but lousy transient response and vice-versa. There seems to be a difference between the impact of the first pressure wave of sound and the phase when the diaphragm is moving already - following the sound pressure. So time smearing is definitely something to be considered at least over the audible range when it comes to how a mic performs in the real world.

The "unlucky" HF response of the C100 (I am really sorry) looks similar to some U47 builds and sounds great: 80% of the energy that makes the sound of a close on-axis-mic in cardioid (which is what most people use most of the recording time, at least in "pop" music - whatever this is) is being emitted below 10k anyway. And the visible sibilance in the graphs does never come along as harsh, it rather sounds "intimate" and shiny. Could that be a time domain property or is this just low whatever-distortion?

- now back to the king and all the others

 

[garp]

Despite their poor hf reponse, the Melodium's 42b have the most impressive 3d "effect" of all the other mics in my locker (i have some).

 

[kingkorg]

YES, EISmurf!! Plus in regards to the OT - the development of fine mic capsules, their sound properties and and capabilities - I quickly want to add some information that I came up with digging deeper into the rabbit hole last night:

1. A lengthy, highly controversial but still interesting/informative thread about whether or not condenser microphones are minimum-phase devices and thus all of their frequency and time domain properties can be completely grabbed, analysed and back-and-forth transformed or not:
https://gearspace.com/board/remote-...aphram-fast-small-diaphrgam-condensers-9.html

2. A paper by Neumann developer Stephan Peus regarding Transient Response of microphones


My conclusion for now in respect to question 1: Regardless whether we use a sweep, shot or a DIRAC for the measurement - I'd now vote for YES, it's all in the FR/FFT data, but it's really hard to visualize the timing of the first 100-or-so microseconds of an impulse over the audible range in a meaningful way. Plus intermodulation distortion of complex musical signals will further degrade the clarity.

In regards to 2. it seems to be common sense in microphone development, that it was feasible to construct (and there are quite some examples of) microphones that have decent FR but lousy transient response and vice-versa. There seems to be a difference between the impact of the first pressure wave of sound and the phase when the diaphragm is moving already - following the sound pressure. So time smearing is definitely something to be considered at least over the audible range when it comes to how a mic performs in the real world.

The "unlucky" HF response of the C100 (I am really sorry) looks similar to some U47 builds and sounds great: 80% of the energy that makes the sound of a close on-axis-mic in cardioid (which is what most people use most of the recording time, at least in "pop" music - whatever this is) is being emitted below 10k anyway. And the visible sibilance in the graphs does never come along as harsh, it rather sounds "intimate" and shiny. Could that be a time domain property or is this just low whatever-distortion?

- now back to the king and all the others

I really don't get what you are trying to say, imply, or prove. You keep flooding the thread with unrelated stuff, trying to make some kind of a point, finishing every post with let's move on.

Before you can take part in this kind of discussion, you need to get the theory, and terminology straight first. Instead of quoting GS posts, start a new thread on a topic, and post some science paper quotes. There are plenty of these on all the subjects you mentioned. Just the first post on that GS thread made my stomach turn, and what followed wasn't any better either.

The document you provided does nothing to prove your point regarding rise time, transient response, and i pretty much agree with everything stated there. There are couple of issues tho. It's not scientific, it doesn't specify what "pretty frequency response" is, and why the "pretty frequency mic" sound bad, but he does agree with me that fast transient response doesn't seem to sound necessarily good.

It doesn't help either that he doesn't take off-axis impulse response into the consideration. Or even off axis frequency response. He doesn't even mention what pattern characteristics DUT has.

If you are at a point where you think u47 and c100 have anything in common besides both being a microphone, with all the best intentions, being aware how edgy i can sound, please do some reading first, or at least start another thread where people can get to choose if they will participate or not.

If you enjoy your c100, please by all means continue using it, i don't think there's anything about it that will ruin your recording. If you are trying to proove it's technical specifications are beyond mediocre you are wasting your time, because the DATA!!! says otherwise. Your C100 is not a minimum phase device. It has two capsules on top of each other. It can not be a minimum phase device. Whatever you like about it, it surely isn't phase related.

Take a break, and then come back and try to read and interpret what you just said here:

80% of the energy that makes the sound of a close on-axis-mic in cardioid (which is what most people use most of the recording time, at least in "pop" music - whatever this is) is being emitted below 10k anyway. And the visible sibilance in the graphs does never come along as harsh, it rather sounds "intimate" and shiny. Could that be a time domain property or is this just low whatever-distortion?

And keep in mind you continued with this after the sentence with microseconds, fast transients and hearing what you imply is impossible to visualize, whatever that might mean.

 

[elskardio]

@elskardio
I don't think anything crazy will happen with these cardioid @soliloqueen offers, i just think she's very pedantic and goes for highest possible standards.

That’s what I thought. Thanks for the reply

 

[Roman Beilharz]

Wow, kingkorg. My brain luckily is able to compensate for condescending tones as well.

I never wanted to prove anything. Maybe you didn't catch that drift, but I came here to find out more about sonic characteristics/differences between Arienne's "vintage" K47 and upcoming K67 and hoped for her to maybe chime in. I didn't get any meaningful information apart from the FR graph of the first batch, at least something. Then Khron said "it depends on what you want to achieve" for my 67-style build. I gave a description of what I want to achieve, when micolas asked whether I'd know mics that would match that description. Only then I gave my examples C100 and Josephson C617Set and I wish I hadn't done that, because it was the reason for various OT posts including your helpful and factual, but still OT information about the C100.

My apologies to the community that I have taken so much screen space contemplating transient response, I never intended to hijack this thread. I am sorry.

I still have not heard a single meaningful statement regarding my original question about the sonic differences of the capsules;, but in the meantime I have decided to find out for myself, as I am actually recording with my microphones rather than analysing them in the first place. So thanks for the slap on the butt, I will keep up doing what I am really good at - creating amazing music and music recordings - and leave the rest up to you guys.

Peace!

 

[kingkorg]

Wow, kingkorg. My brain luckily is able to compensate for condescending tones as well.

I never wanted to prove anything. Maybe you didn't catch that drift, but I came here to find out more about sonic characteristics/differences between Arienne's "vintage" K47 and upcoming K67 and hoped for her to maybe chime in. I didn't get any meaningful information apart from the FR graph of the first batch, at least something. Then Khron said "it depends on what you want to achieve" for my 67-style build. I gave a description of what I want to achieve, when micolas asked whether I'd know mics that would match that description. Only then I gave my examples C100 and Josephson C617Set and I wish I hadn't done that, because it was the reason for various OT posts including your helpful and factual, but still OT information about the C100.

My apologies to the community that I have taken so much screen space contemplating transient response, I never intended to hijack this thread. I am sorry.

I still have not heard a single meaningful statement regarding my original question about the sonic differences of the capsules;, but in the meantime I have decided to find out for myself, as I am actually recording with my microphones rather than analysing them in the first place. So thanks for the slap on the butt, I will keep up doing what I am really good at - creating amazing music and music recordings - and leave the rest up to you guys.

Peace!

In my experience Soliloqueen is very responsive to DM, both here and through her site.

I'll try to answer your question, although, without trying to be condescending, I have to say it is confusing.

As @Khron suggested, k47 and k87(k67) can't be compared. Totally different animals, and because the major differences in construction, minute differences like transient response are dominated by other, dominating differences. That is also why you can't compare any of these with Josephson, or Sony as they are way too different.

You could test if you had a lab, team, and perform a battery of tests. I don't think anyone has done this, or ever will.

As far as i know all the k87 capsules Ari built have isolated backplates, k87 is the same as k67 which doesn't have isolation.

You can't build u67 with any kind of k47 boosted or not, unless you are trying to achieve something unique.

If you are interested at all in my personal opinion... Before i retired from being a professional musician, every time i focused too much on the technical stuff the art suffered. I occasionally record and produce stuff still. Every time i picked a mic because it made sense to use that particular one based on what i knew about it's technical properties i made an error and suffered later on.

 

[McIrish]

I see that Ari has the new K47 capsules available on her site now. Very tempting. I have a flat 47 already but this might be worth checking out as well.