Arguments on properly recording/sampling acoustic pianos (+more)

Also sweep signals are mostly used for when we need to apply corrective EQ for speakers in a acoustically non‑treated room.
The sweep signal from linear (as much as possible) loudspeaker is blasted across the room and a flat resp. mic is capturing it at the position of listening (where the audio engineer will be sitting). This way we not only can capture the BRIR but also "see" which freq. from the sweep get more resonant or less in order to inverse them with the 'corrective EQ'. The reverb BRIR is not the goal here.
For pianos we only need the BRIR reverb response. They can be balanced in freq. and volume later. That is highly audio technical stuff part of the proper preparation.

@Claude AI Bot check the final few responses and do a strict red team analysis. Not sure what the argument is since it’s well known how to do impulse responses. None of this seems to point to a better way to record a piano, even theoretically.

dikrek None of this seems to point to a better way to record a piano

There is no "better" way. You need a better vision and equipment. Sampling a piano is different than just recording a piano (for a song\mix\style). Sampling is much more technically involved.

PASHKULI you have neither a better vision nor equipment, almost all your claims have been systematically annihilated in this forum by both humans and AI.

So far nothing you’ve said has led anyone to believe you have a better approach for recording a piano.

dikrek

Yes. That is because I have to first build the piano I would love to sample\record.
Until then, I can only modify existing pianos. No one will "sacrifice" their pianos. Therefore I have to by one, but then it has to be in a special room (close to anechoic, not necessarily a real anechoic, expensive logistics and time).
I am just talking from experience recording other instruments and working with mics (I had a quite expensive austrian made stereo pair 12 years ago and proper equipment).

Look at acoustic guitars. They are recorded with two mics (typically) one aimed at the body (resonator soundboard) and the other at about the middle of the fretboard (strings too). And for ambience there can be a room stereo pair. But most recordings include either piezo signal or pickups (the latter especially for such guitars with metal strings). Then you mix those signals as a form of art with some scientific knowledge needed about phase.

Sampling though → a completely different beast. Not only because the player moves, but also because there is a need for direct sound where you can not hold the player static. Well, the guitar can be on a stand and sampled that way but it is awkward or the operator. Piano is static, that is great. But it is huge compared to a guitar, cello. So, it is what it is. Piano is a piece of cake compared to a guitar, callo.

I am just an amateur hobbyist (of Music, instruments for Music).

PASHKULI You don’t need a piano to test your claim. Use a steel-string acoustic guitar with magnetic pickup + piezo. Record those and a fixed microphone simultaneously. Build your best pickup/piezo/IR approximation using one set of performances, then freeze it. Test it blind against the microphone on completely new notes, chords and dynamics.

If the frozen model really is “99% perceptually indistinguishable”, people shouldn’t be able to ABX it. If they can, the claim fails on an instrument vastly simpler than a piano.

Test generated via a collaboration between ChatGPT and Claude. Seems reasonable and straightforward.

dikrek

All great, I am all in for it. What is needed to setup, execute and complete that request:

• 1 × a steel string electro–acoustic guitar (€300 to €500) so most of them have built‑in adapter or rented
• 6 × nu pickups with audio converter (€250 + €150)
• pair of microphones for guitar (€380 + €450) or rented
• IR microphone such as Earthworks M30BX (€780) or rented
• at least some decent and ultra‑quiet small studio room (may vary for the duration of experiment but one day rent will be enough, about €100 per day\few hours)
• audio interface + laptop to store the data (I got fairly good ones, old but will do)
• some additional acoustic dampers just in case needed (€100)
• logistics (transport, test try of studio room beforehand till one that satisfies me is found; might take a few days, so another €100)

Let's see the total: ≈€3000 needed (if microphones and guitar are rented probably will drop to about €2000)

Ah, we might need to add some software needed, especially the one for the mixing and IR loading (a bit custom).

Keep in mind that this is only possible for open strings (of the guitar). Any other fretted note won't be applicable as both hands are involved in the sound → something piano can not do. Therefore:

dikrek the claim fails on an instrument vastly simpler than a piano

Piano is much simpler with regards to Music performance (dynamics, vibrato, muting, bending). Just because it is bigger does not make it more complicated (unless for logistics).

PASHKULI You’ve turned a simple falsification test into a €3,000 laboratory procurement exercise.

You don’t need six Nu pickups, a stereo pair, an Earthworks measurement mic or an anechoic room. One steel-string guitar, one magnetic pickup, its piezo, one ordinary mic, an interface and a DAW are enough. Record all three simultaneously. The room doesn’t even need to disappear — the fixed mic signal is simply the target you’re claiming the pickups can reproduce.

And open strings are enough. If a frozen pickup+piezo model can’t reproduce one fixed microphone position for six open strings across different attacks and dynamics, the larger claim already fails. Fretting is irrelevant because the three signals are recorded simultaneously; nobody is trying to reconstruct the performance from separate takes.

But the bigger confusion is this:

simpler to play ≠ simpler physical system.

A piano key may be an easier gesture than fretting/plucking a guitar note. Acoustically, that one key can excite 1–3 coupled strings, two transverse polarisations, bridge motion, many soundboard modes, duplexes, sympathetic strings and pedal-dependent coupling.

That’s exactly why the guitar was suggested: it is the easier case for your reconstruction scheme.

Pass that first. Then we can talk about the piano.

But I feel that after all this time we are finally getting somewhere.

dikrek the fixed mic signal is simply the target you’re claiming the pickups can reproduce.

Completely wrong assumption. Get out of here AI. You are dumb.
I need to represent the various types of strings: plain and wound (valid for both guitars and pianos), so at least two pickups. And I want to represent some stereo and binaural (true) sounds of the instrument, therefore at least two mics are needed (regardless of close ups or mid environment).

And a good special microphone to take the IR snapshots. OK, I'd agree that this mic can also be used as one of the stereo pairs, but it means they will be different so the stereo image will be unequal at least with regards to EQ response.

So, are we going to do it THE PROPER WAY (more expansive indeed!) or just blabbering about it?

To make it clear for you again: we are not trying to reproduce a microphone capture. This is a new method of recording which will give us more:
• dynamic velocity and round robin variations, proper sympathetic and release samples
• more resonant structures (series of snapshots, which are rather small in data size → hence more space for the previously mentioned) of the resonator box (cabinet) of the instrument and BRIRs for the environments the instrument is placed (it should be there when the BRIRs are being taken)
• pitch\note separation control → full control over final volume or initial gain of each string\note captured, not just a velocity curve
• we can apply intermediate effects to all those parameters mentioned for creative sound design purposes

You have no idea what you are talking about comparing sampling a guitar vs a piano. I am not talking about simply sampling open guitar strings. No one is capable of truly sampling a guitar. The amount of variables a guitar player can affect the sound with both hands is exponentially larger than that of piano. Piano only has resonances, most of which are horrendous.

PASHKULI You’re mixing two different projects now.

If the claim is simply:

“I want to build a highly controllable electro-acoustic sampled instrument with more layers, per-note control, alternative resonances and sound-design options”

then fine. There is nothing to disprove. That’s a perfectly legitimate instrument-design goal.

But that is not the claim we were arguing about.

You also claimed that this method is more physically accurate than microphone recording, and that most of what microphones capture can effectively be replaced by IR/BRIR reconstruction.

That claim can be tested cheaply, without building your entire dream sampling system.

The cheap test is not intended to validate stereo or binaural rendering. It tests one necessary condition first:

Can your pickup + piezo representation predict one fixed acoustic output on unseen performances?

For that, one microphone is enough.

Record pickup, piezo and one fixed microphone simultaneously. Use one plain and one wound string if you want both string types represented. Build your best mapping on one set of strikes and dynamics, freeze it, then test it on different strikes and dynamics.

If it cannot reliably reproduce even that one fixed acoustic perspective, then the earlier “more accurate / effectively indistinguishable from microphones” claim is already in serious trouble.

If it passes, great. Then move on to stereo. Then, if you want to test actual binaural rendering, use proper left/right ear measurements.

That is where terminology matters: two ordinary microphones give you stereo, not automatically “true binaural”, and one measurement microphone cannot capture a BRIR. A BRIR is the pair of transfer paths to the two ears.

So no, you do not need six Nu pickups, a stereo pair, a dummy head and a full BRIR library just to test the first necessary condition.

And yes, I agree that guitar has more player-controlled gestures than piano: fretting, bending, vibrato, plucking position, muting, etc.

But that is performance complexity.

It says nothing about whether the piano’s coupled strings, bridge and soundboard form a simpler vibroacoustic system.

So pick the claim:

new, more controllable instrument? Perfectly defensible.

more physically accurate reconstruction of the acoustic instrument? Demonstrate it.

dikrek can effectively be replaced by IR/BRIR reconstruction

Yes, but it has to be done correctly. It works well for physical spaces, speakers (other gear too such as amplifiers).

dikrek Can your pickup + piezo representation predict one fixed acoustic output?

It needs IRs too. So, those has to be included as part of the mix. Typically those IRs will be simplified BRIRs as SMIRs (source mono impulse responses), namely the IRs from the body chamber of the acoustic instrument. For instrument such as acoustic guitars, cellos and especially for violins this is not an easy task because the microphone has to be a special one, namely the mentioned Earthworks M30BX (€780) as it has a thin nozzle which can be inserted in the openings in the bodies of those instruments (and others too).

Why this is important is self evident, hence we will be limited of such mic placement by the geometry openings in the resonant body of the instrument (thankfully for a piano there is no such limitation):

dikrek more physically accurate reconstruction of the acoustic instrument? Demonstrate it.

A microphone captures sound pressure in the moment with everything inside the captured environment. Most microphones have their specific freq. response EQ "character". Some specialized microphones have near "flat" EQ response. There is no "right way" to place a microphone (or stereo pair of those) but generaly they are separated in three groups:

• close up mics → capturing the actual sound source direct acoustic signal with minimal early reflections and hopefully no late reflections
• mid field mics → usually to capture overall body of the music instrument or human voice (chest), early reflections if present
• ambient mics → to capture room acoustics, predominantly for late reflection but they also capture early reflections too and direct signal is rather weaker (distance dependent)

All those microphones must be phase aligned as precise as possible (even the ambient ones despite their captured phase starts tens or hundreds ms later than the direct close up mics), especially because those are in stereo pairs.