Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Eddie Eagle]

Is this sound technology or?
https://www.indiegogo.com/projects/sphere-precision-microphone-modeling-system--2#/

[klaus]

I can only address this issue indirectly, having had no first-hand experience with mic modeling so far.

Two comments:

1. It is inconceivable to me how the quality of the capsule, the crucial conversion of acoustic waves to electrical current, could be irrelevant in mic modeling, even if the electrical/digital processing downstream from the capsule were flawless (see #2): if the processing of a Chinese copy capsule leaves much to be desired, as we all know, how could that same capsule suddenly be a foundation to elicit C12-like quality from a mic modeling system? Garbage in = garbage out, no?

2. A bunch of highly respected and well-known audio engineers, producers and studio room tuners recently had a week-long retreat in the California Sierras. The subject: compare DACs. These are the boxes which convert digital information back to analog sound signals, to be played over hopefully superb home or studio playback systems.

The results: each of these boxes processed the sound audibly differently. With clear (and universally agreed upon) rankings of resolution, musicality and what have you*.

My conclusion: the quality of digital converters is all over the map, yet hugely relevant for the quality of the musical experience. So, even if the acoustic conversion (see #1) were absolutely perfect, the question of the digital processing in modelers would be paramount to the quality of the experience, in my mind.

I am speculating that, in low-res listening environments (project/home studios) the difference between a Chinese analog mic and a China-capsule mic modeler may not be all that much, but as soon as you get into a high-res environment, the differences may be stark and indisputable.

I will try to get one of these modeling systems into a high-res studio, compare its "vintage $20k" model to the real thing** and report back.

* a bonus result of these listening tests was the revelation that recordings which used digital mics, were not an improvement over classic analog set-ups. The physically limited size of converters inside digital mics may be the bottleneck in this particular application

** how many user (not manufacturer) reports about the audio quality of mic modelers come from people with frequent, extensive exposure to vintage classics kept in superb working quality? I guess it's a distinct minority. Otherwise the pricing of these modelers would address the top tier of users and rival that of the classics, rather than aim for the "vintage copy" market.

[Timjag]

To quote Iggy,

"A person should have a personality. You won’t get one dicking around on a computer"

That's as far as I can go with that.

[Eddie Eagle]

I like the idea of the modelers, but don't think the technology is properly honed and applied yet to compensate for all the detail in acoustics and sound of high end analog.

[radardoug]

This idea of modelling is ridiculous. Making an SM58 sound like a U47? Not going to happen!
The subtleties of the electronic design can not be accurately modelled, let alone the acoustic differences.

[Eddie Eagle]

Making an SM58 sound like a U47? Not going to happen!

I agree with that. This particular mfg seems to be in the biz of selling mics as a previous mfg, Slate, who introduced their product a few years ago. Seems like a mic for this purpose would have to start at the upper end of quality (not a 58) to be able to have software manipulate signal accurately with algorithms to mimic a 47 or other mic of such quality. It's a futuristic "idea" that will compete with actual hardware and will be interesting to see which succeeds or if both succeed.

[Avgatzeblouz]

As far as I'm concerned, when recording someone, I take great care of the mic placement and pattern. You always find a spot where the music happens more than on another. Those mic modeling systems take that out of the process, and make it less creative and fun. You just stick a mic close to a source and you record it. The involvment, artistic choices a good engineer has to make are gone.
Second, a real performer reacts to what the mic is giving to him/her. The emotionnal attraction of the mic IS part of the equation. And this concept is taking this out as well.

Globally, it can be much fun, but it is, again, lowering the quality of the artistic implication of a great engineer, therefore, the overall quality of the production.
These tools are not musical instruments as a real microphone, even an SM58 btw, can be.

[David Satz]

From a purely objective standpoint, several of the main characteristics of any microphone are interdependent, and post-processing can't change them without also affecting other, related characteristics. For example, it isn't possible to alter the bandwidth and frequency response of a system without also affecting its impulse response, or vice versa; they're fundamentally interdependent.

The polar response of a microphone can't be influenced by post-processing, yet it's one of the most fundamental ways in which different types of microphones behave and sound differently; you can't "model" a cardioid into an omni or vice versa, for example. But polar response has more subtle aspects, even among microphones designated as having the same pattern (e.g. cardioid), because the actual pattern always varies at different frequencies--sometimes a little, sometimes a lot. No two microphones that use differently constructed capsules ever have the same exact pattern as each other at all frequencies.

Say for example that microphone X, measured at 12 kHz, has a 3 dB rolloff at 90 degrees as compared to its on-axis response, while microphone Y's sensitivity at the same frequency and angle is 7 dB down relative to 0 degrees. That's a plausible degree of difference between two microphones, and it would have audible consequences in real-world use. If you EQ microphone X's signal so as to make its on-axis frequency response as close as physically possible to microphone Y's, the off-axis response of the two microphones will still be quite different. It's a fundamental characteristic which these systems can't, and don't, address.

All in all, microphone modelers operate on about the same ethical grounds as a magician at a child's birthday party. Everyone knows (or should know) that magicians can't really do what they try to make it appear they are doing. Thus you can't sue a magician on grounds of fakery, any more than you can sue a fiction writer for making up stories; that's the essence of what you are presumably there to experience. You might feel that the show is good enough for the occasion or you might not, but either way it's still a show and not "actual magic."

[Eddie Eagle]

You might feel that the show is good enough for the occasion or you might not, but either way it's still a show and not "actual magic."

Great analogy and breakdown, hence, the reason for the thread. Where is the line in human perception and physical attributes for technology and how do they blur to create aural illusion?

I'd love to find out what Klaus finds in the high end studio application.

[Jim Williams]

Imitation is the sincerest form of flattery.

I recall many years ago the first modeling agency (Antares) created a model of an Audio Upgrades modified AKG 414.

That's when I decided to use an SM58 for everything.

[klaus]

I don't understand your point. Please explain.

[Kai]

...That's when I decided to use an SM58 for everything.

That's obviously meant as a joke.
I tried the Antares microphone modeller, compared it with the real things:
It failed even on changing a Neumann U87 into a AKG C414 sound and vice versa.
It didn't even come close.

If you look at dynamic mic's, the frequency response of several specimen of the same model is all over the place.
E.g. I measured 10 Shure SM57s bought new at the same time. Guess what - not two of those where close enough to justify the use as a stereo pair.
So which one do you want to simulate or where is your starting point ?!

Looking at the Sphere L22 website the thing is quite tempting, specially at their price point.
The concept of separately recording front and back of a condenser capsule isn't new and does work for changing characteristics.
The main question that arises for me is: what is the quality of the microphone itself, cause this is always the biggest part of the equation.

[Timjag]

I guess it doesn't matter. Sure, to us it's a big deal but I work with kids - 19-26, the current craze is Hip Hop in particular Trap. These kids do everything on their MBPs - in abelton live or Logic and most of it sounds HORRENDOUS!! I spend most of my time with dynamic EQs carving off the top end which they've loaded up with just about every high end style plug in EQ you can imagine.

I'm telling you now these kids will be the market for this mic - they will only stretch their budget to 1 LDC and a midrange "fancypants" mic pre. The idea that one mic does all will be incredibly enticing to them.

Here's the thing though this music is coming on like a freight train with the 17-20 year olds because it's loud and rude and their parents HATE it! (Sound familiar) and it's exciting to be around - I'm just trying to guide the more musically enabled ones to some type of sonic excellence. (Which is hard - I swear they're deafening themselves with that music....hmmm that sounds familiar too    )

So I guess it's fair to say this mic already has a market. These kids would like to be in the nice studios with all the great gear and great players - but the fact that they can get something sort of like that(in their minds at least) at home means they're good to go.

We're dealing with a generation that would rather lose their car than their broadband access - it's a changing world.

[radiovinhet]

Mic modeler really sucks... but what do you think about Liquid Channel from Focusrite?

[klaus]

Mic modeler really sucks...

This forum would benefit from your detailed user report, rather than a blanket thumbs-down

but what do you think about Liquid Channel from Focusrite?

Not sure I understand the logic of dissing a mic modeler (see above) but not dissing the same principle applied -digital modeling of analog audio signal processors- to mic pres?