• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Is Amateur Piano Recording This Hard?

??? As I said, my teacher uses a grand piano. How is software playback going to help with that?
Tell your teacher he is going to have a digital piano for teaching online and a physical piano for teaching in person
The amount of cash required for a great recording outweighs the benefit here.
Cost: 500 to 1000 EUR inclusive VST / cabling

Else, - comparable sound? Equip the room acoustically optimized, and get a 1k+ microphone fixture, place Piano in optimal position in room
Good luck with that. That's much more of a headache and interpersonally a nightmare.
Cost? 6000 EUR - with DIY maybe close to 3000 EUR
 
Last edited:
I love this particular one:


It's so natural,... with the 8030c and closed eyes is very close to reality... perhaps it has to do with proximity of the mics... don't know really.

I guess to some degree there is no right or wrong piano sound, it depends on the instrument, the room and specially the listener position. As a player the sound is different than from the audience or from the side... I guess there is no objective piano sound.

I've had some acoustics and some digitals. The Casios GP have different versions of the sound, simulating what you here from the front, as an audience.. it's just a blunt approximation but it allows for quick comparison.

Personally, I prefer clarity or even dryness if you want over pomposity, but I guess this is too subjective as a description :)
 
  • Like
Reactions: DEF
Here is an example of piano recording which sounds nice but the wide stereo sounds a bit artificial to me:


The guy in that video has a link to the gear he uses for his recordings: https://kit.co/WattKeys/grand-piano-recording-kit

He uses a fairly cheap set of two microphones going into an ordinary audio interface (with at least two channels). If you don't like the wide stereo sound you don't have to pan the tracks as wide as he did in his mix, and/or use another stereo microphone technique than what is seen in the video.
 
There is nothing in the world that remotely recreates the sound of a piano when you are playing it. There is a wonderful reality and 3-D presentation as you play it that stereo can't even come close to recreating it. Those artists saying it can like the pianist in the video, either hear the difference but don't want to say, or don't hear it which is well, just as bad.

Agree. The “Steinway Lyngdorf” open baffle designs apparently are supposed to come close. But one of the tricks Lyngdorf used to convince Steinway was that the speakers needed to be “in the room” like a real piano as opposed to near the wall like most speakers real are placed.

What’s interesting is that Kawai is one of the most technical piano makers out there.
First with robotics. First with carbon fiber, etc. Their algorithms for the piano modeling on the digital side are also in-house.

1728000585689.png


Hopefully some of their engineers read ASR and will start putting more effort into their audio section in their digital pianos. :)

Tell your teacher he is going to have a digital piano for teaching online and a physical piano for teaching in person
The amount of cash required for a great recording outweighs the benefit here.
Cost: 500 to 1000 EUR inclusive VST / cabling

One problem is that a lot of the higher end player piano/silent systems/digital pianos track things like velocity with higher precision than MIDI 1.0 and there aren’t a lot of MIDI 2.0 keyboards with a nice feel. Then there’s a whole separate issue of pedal feel…. It’s a little bit similar to chasing SINAD though — even a basic digital piano can achieve great sound in the hands of a skilled pianist so I see your point. I went from the Kawai MP8 to the Korg which should be a downgrade in feel, but the more rigid frame of the Korg (as compared to even a premium stage piano stand; I used a Yamaha LG800) and superior pedal makes it more enjoyable to play.

Tagging @napilopez since he is an ASR/pianist too. He linked to

1728030004660.png
 
One problem is that a lot of the higher end player piano/silent systems/digital pianos track things like velocity with higher precision than MIDI 1.0 and there aren’t a lot of MIDI 2.0 keyboards with a nice feel. Then there’s a whole separate issue of pedal feel…. It’s a little bit similar to chasing SINAD though — even a basic digital piano can achieve great sound in the hands of a skilled pianist so I see your point. I went from the Kawai MP8 to the Korg which should be a downgrade in feel, but the more rigid frame of the Korg (as compared to even a premium stage piano stand; I used a Yamaha LG800) and superior pedal makes it more enjoyable to play.

I'm not a piano player, but s few years ago I bought a Roland FP-30 for my daughter and she loves this digital piano. I think the keys have the feel and weight of a real piano and the sound quality out of the in-built speakers is fairly good.

In the specs it says it has MIDI format: Conforms to GM2, GS. Does that mean it has MIDI 1.0 or MIDI 2.0?
Anyway, we used the program Addictive Keys when we recorded a song and I thought the result was good, and definitely good enough for the little recording project we did.
 
I'm not a piano player, but s few years ago I bought a Roland FP-30 for my daughter and she loves this digital piano. I think the keys have the feel and weight of a real piano and the sound quality out of the in-built speakers is fairly good.

In the specs it says it has MIDI format: Conforms to GM2, GS. Does that mean it has MIDI 1.0 or MIDI 2.0?
Anyway, we used the program Addictive Keys when we recorded a song and I thought the result was good, and definitely good enough for the little recording project we did.
Hello goat76.
I own a Roland FP30.
it supports MIDI 1.0 - from 0 to 127.
it does NOT support MIDI 2.0

What others said about MIDI 2.0 or even KAWAI XP MIDI standard is not making a difference.
Or Yamaha Disklavier
We are incapable of consistently playing 127 different levels of play consistently.

Even if we were capable, the sensing mechanisms are far too worse on it, even on high-end digital grands for 8000 or more EUR.
And some are just jumping in velocity. E.g when I tested I couldn't get the Avant Grand to values in between of 83 to 127.
It immediately jumped to 127 after a certain threshold.

Moreover, GOOD systems randomize the tonality by quite a bit for every keypress.
PianoTeq does this. It humanizes / randomizes. You can turn that off, if you wanted.
--- vastly oversimplified ---
For MIDI 2.0 - you have double point precision. It is not quite correct, but I say it for ease of understanding.
So you could have 60.34 etc for velocity etc up to 127.00
----

You see, I downgraded from Roland RD-2000 to FP-30 because the keyboard action was the thing I was after.
And there is not a difference that would make it up for me.

If there IS an upgrade, it would be along the lines of Yamaha P525.
Why? Just for the action. I hook up an VST anyways, it outperforms it.

Sidestepping this - it would be great if we had different sized Pianos.
I can get one upright custom made for 20 Grand. This is expensive.
If only we had digitals...
Well, David Steinbuhler did that in the past, now sells the uprights.
Look, in the past he provided: https://up.tail.ws/images/Custom MP11-SE.jpg

And digitals are often having latency themselves.
 
Last edited:
I suppose recording a piano is mostly about timbre and catching subtleties in reverb, attack etc. There is no imaging to catch, as the piano is basically a large piece of furniture that emits diffuse sound from each of its surfaces. At least from the audience's vantage point.

Commercial recordings seem to take a number of different approaches with regard to imaging. In many jazz recordings the piano is assigned spot on the virtual sound stage like all the other instruments. Classical recordings often spread it out across the entire width of the soundstage (not something I like). Sometimes they move it more to either side, and sometimes they offer a left-to-right gradient, e.g. bass to the right and treble to the left (audience perspective), or the other way around (player perspective).

In an actual live performance with instruments spread and spaced across a 45-60 degree angle (like a chamber music piano quartet from a favourable distance) the piano will be the hardest to pinpoint.
 
I suppose recording a piano is mostly about timbre and catching subtleties in reverb, attack etc. There is no imaging to catch, as the piano is basically a large piece of furniture that emits diffuse sound from each of its surfaces. At least from the audience's vantage point.

Commercial recordings seem to take a number of different approaches with regard to imaging. In many jazz recordings the piano is assigned spot on the virtual sound stage like all the other instruments. Classical recordings often spread it out across the entire width of the soundstage (not something I like). Sometimes they move it more to either side, and sometimes they offer a left-to-right gradient, e.g. bass to the right and treble to the left (audience perspective), or the other way around (player perspective).

In an actual live performance with instruments spread and spaced across a 45-60 degree angle (like a chamber music piano quartet from a favourable distance) the piano will be the hardest to pinpoint.
Each hammer can be seen as one speaker - more or less
 
Those artists saying it can like the pianist in the video, either hear the difference but don't want to say, or don't hear it which is well, just as bad.

Very interestingly, I found an interview explaining more about the session. Limitation is the machine translation, but it makes more sense.

Basically the actual recording engineer emphasizes that he recorded using carefully placed B&K 4009 mics with no modification whatsoever of the digital recording such as a low pass, etc. He does mention that PCM and DSD is least important element of the recording but mic placement and room selection. :)

From what GPT says, the B&K 4009 is a precision stereo matched version of the B&K 4003, which is a premium version of the B&K 4006 that Stereophile uses to measure speakers. The 4003 uses 130V instead of 48V and as a result handles and extra 10 dB of SPL without clipping and is more accurate in the lower frequencies as well



——

■Recording without any unnecessary signal processing

-- How was the recording done?

Ono: It is said that there are more than 200 ways to record piano. I have roughly eight different theories depending on the genre and the client. This time, I recorded Chopin in a classical, but not concert hall, salon-sized listening room, and the sound source was "high-resolution 2ch only, no processing", not intended for broadcast or CD release. In most cases, multi-microphone recording is done, and the balance is adjusted later, and the applause from the audience is mixed in to create a "good" final result. Even if the subtle nuances of the actual recording are different from "as is", the purpose of post-production is to create a "good" final result. If you are not going to process it "as is" like this time, there is no hesitation, just two B&K microphones. This was also the method used when we conducted an experiment in distributing DSD live streaming of the Chopin Competition from Warsaw in 2015. I don't think there should be any directing in the recording or tone at the competition. The touch and tone must not change from the actual image.
 Similarly, for this recording, I decided on the microphone position and did not apply low cut or any signal processing. Since the concept was "as is," I used a pair of microphones, the B&K 4009, which is also the standard for measurement. The famous 4006 is a 48-volt phantom specification, but the 4008 is 130 volts, and the matching pair is the 4009. It is a microphone that picks up the sound as it is heard on the spot. In fact, I brought another pair to the site to compare them. The latter spreads a sense of air a little farther. If it were an entertainment recording, I would have recommended the former, but I chose the B&K for the "as is" sound. Of course, I did not mix in off-mics later, and the stereo format was also decided, so I set up the microphones in a position where the direct sound of the piano and the reflection and reverberation of the room were just right, checked it with my own ears, and then had Murakami check the test recording immediately. For me, the microphone position was OK when Murakami said, "The sound is pretty close." If there is anything that bothers you, we will work together to adjust the reverberation of the room itself and the piano. Tuners are very important in piano recording. This time, the sound of the SK-EX was already completed when we arrived at the site. It's Murakami's home ground. The other important factor is how well pianist Ayano Maruya concentrates on playing. Rather than trial and error, you need to psychologically bring out the best performance from the performer to make a good recording. If you organize and rank the priorities for capturing the best sound quality, they are ① the pianist's performance, ② the piano itself, ③ the space, ④ the microphone position, ⑤ the microphone, ⑥ the microphone amplifier, ⑦ the recorder, ⑧ the cable, etc., and ⑨ the specifications and media format.
 In high-resolution articles, 11.2MHz DSD is often said to be the highest specification, but what is more important than the specifications is the instrument, the space, and the microphone position. That is why a professional balance engineer is essential for important recordings that will be listened to by many people. It's the same logic as not everyone can take photos like a professional photographer just because they have an expensive camera. How can you bring out the best smile in a photo?

—-
GOING back to your recording question…

Knowing that this B&K 4009 is special for piano recording, a Google search took me to this article


And toward the bottom, you have one recording engineer who gives very clear opinions

1) Mic 4-8 ft away from piano
2) Place about midway of the longest string from audience perspective and point the omnidirectional mics in the direction of the hammers, keeping them parallel
3) 18 inches to 5 ft apart. Width of the mic space captures desired width of the image of the final recording
4) Hard pan the left and right.
 
The above could be starting points.
There is no sure recipe unfortunately, loads of variables, mic type, piano type and age, player style, room…

I record pianos quite often (mostly upright) and it is indeed not easy at all, there’s a lot of dynamics and resonances to capture and the range is wide.

The fact that in the original video they are ok with it’s a different matter, that would keep us here for months, but YouTube in general shouldn’t be taken too seriously for high quality audio.
 
Back
Top Bottom