KORG Poetry (Digital Piano/Bluetooth Speaker) Measurements

[GXAlan]

KORG Poetry Digital Piano/Soundbar
The KORG Poetry is a digital piano which has been designed as "The Piano For Lovers of Chopin." It uniquely captures an 1843 Pleyel vintage piano that was Chopin's choice of piano and also provides a modern Fazioli grand piano sample.

What's unique is that the audio system is essentially a large soundbar with two 10 cm full range drivers that utilize a whizzer cone (25W x2). Since the piano takes up physical space, Korg allows you to use it as a Bluetooth speaker which can be run in parallel with the piano tone generator.

The LP-380 utilizes the same (? previous generation) 10 cm full range driver as the Poetry. The LP-380 advertised 22W x2 while the Poetry (and C1 Air) are 25W x 2.

Measurements
For reference, on an 88-key piano, the fundamental goes from 27.5 Hz to 4.2 kHz
A 61-key keyboard (which captures a lot of the music you may hear) spans from 65 Hz to 2.1 kHz

I took a Windows laptop, connected to the unit via Bluetooth and ran a REW sweep with a UMIK-1 at the piano bench near the listening position and then far away (3m in the room, off axis). The piano bench may not be accurate in that you would expect the presence of the pianist to alter the actual sound.

It's pretty impressive to see this from a full-range driver with a whizzer cone that's only 10 cm in size. I'm not sure if the piano samples have more fine tuning than the Bluetooth input, but it sounds pretty good for a "lifestyle" product. Extreme DSP isn't an option for these pianos since even a traditional stage piano going to a regular ADC (with no room correction) adds too much latency.

The modern Fazioli sound:

The historic Pleyel sound:

 

[kyuu]

Neat. As a bit of pianist myself (currently own a Kawai CA49), I'd love to see measurements of the speaker systems across a wide line of these to see how much (or little) engineering actually goes into making their speaker systems decent. Impractical, I know, with the considerable cost and bulk of them, but one can dream.

 

[GXAlan]

The Kawai team (and I guess Yamaha team) are pretty aggressive about incorporating Hi-fi technologies into their digital pianos. I have a Kawai MP8 and if money/space wasn't an issue, I'd probably want to go for the NV10s for my hybrid, while sitting tight and saving up for a Shigeru Kawai.

Many of the new Kawai digital pianos have Onkyo branded audio, but I believe it's a real integration since they do use Onkyo's SpectraModule headphone amp which is a feature of their integrated amp, and the idea of dual DAC signal conversion is pretty neat in my opinion. That said, Kawai uses Bluetooth MIDI but not Bluetooth audio. This KORG Poetry has bluetooth audio (not sure what codec) but no Bluetooth MIDI.

This is marketing from Kawai (not the KORG that I tested)

 

[kyuu]

I'd definitely be interested to see if any of that marketing blurb actually panned out. I believe my CA49 is supposed to have benefited from that partnership with Onkyo. Though that partnership was short lived and their new models no longer advertise who provides their audio components.

If I had unlimited funds and my piano skills were good enough to warrant it, the Novus NV5S would be my ultimate piano want (barring a full-size SK grand of course). I'd be really interested to see measurements on that, given the use of transducers and a soundboard on that and the NV10S.

 

[williamr]

Very nice. Do you any idea how a body sitting at the piano would affect the "at the bench" measurements?

It would be amazing if we could somehow get a well known DP reviewer to start to make measurements. Someone like Stu Harrison from Merriam Music. I've always wondered how the sound from a Novus NV5S differs from the NV10S. I guess is that the sound engine and piano model differences dominate the aural differences between different brands.

 

[GXAlan]

Though that partnership was short lived and their new models no longer advertise who provides their audio components.

Looking at the CA901, it looks like their expertise was brought in house. They show an ESS DAC and still say they use a dual DAC architecture but now say that it is Kawai R&D. It’s possible that when Onkyo Japan went bankrupt the IP went with Kawai.

Very nice. Do you any idea how a body sitting at the piano would affect the "at the bench" measurements?

I probably could get binaural microphones and try, but the KORG is still an “entry level” standalone piano, so the performance is already better than I would have guessed.

On the other hand, the Bluetooth mode is designed for room filling music, so the far field makes more sense even though you are seeing the effects of my particular room and the bass reinforcement from the rear wall. The KORG is ported though, so the wall reinforcement may not be that much.

Likewise, we can see from the Canon S-50 measured here, which also uses a whizzer cone, one big limitation of the whizzer cone is dispersion but if you are relying on reflection…

I guess is that the sound engine and piano model differences dominate the aural differences between different brands.

For sure, the sound engine (how it’s recorded and layered) is going to make a bigger difference than the SINAD. Low noise is nice, because some people complained of the Korg G1 having some hiss at baseline and you have to ask yourself — how many harmonics do you need to capture for the very highest note of the piano when the fundamental is 4.2 kHz? If you have tweeters that go to 20 kHz, that’s just opportunity for noise. But if you are marketing a piano that can play an accompaniment, maybe you do want a true full range speaker.

Here’s the next question. In its primary use as a digital piano, a real piano will have string resonance and a lot of intermodulation effects. You can model this like Roland and rely on super precise speakers to capture it, but what if speaker’s IM distortion gives you some naturalness to piano.

Do you want to optimize for the pianist or the audience?

For digital pianos with a BT music capability, it would be pretty easy to take measurements. I suppose you could test self-noise on the headphone jack (easy) vs line out as well.

Then there’s the whole issue of keyboard feel. Even the hybrids from Kawai and Yamaha use “baby grand” mechanisms and is this like trying to get vinyl to sound like digital where you are taking an archaic technology and making it work in a smaller environment? Or are digital first designs like folded hammers or Casio’s GP-line good enough?

I posted my actual piano impressions at PianoWorld.

https://forums.pianoworld.com//ubbthreads.php/topics/3548405/re-korg-poetry.html#Post3548405

 

[williamr]

This brings up the questions of audio mimicry and how can quantify how well one system mimics another one at the audio level. In other words, how well does a DP sound like a "piano". Let's assume a "piano" means an acoustic piano even though I realize this doesn't have to be goal of the DP designer necessarily. This must be fiendishly difficult to mimic an acoustic given all the intermodulation, reverberation, and extended spatial sources of vibration, etc.

 

[kyuu]

Digital pianos nowadays have pretty sophisticated engines consisting of many (many) samples mixed with simulations (or in some cases, it's entirely simulated), and they sound very good. Better than any upright piano, depending on the engine and sound system, barring possibly the largest and most expensive models. IMO, for home use, unless you have the space and budget for a full on grand piano (and the desire/ability to keep it maintained and in tune), most people are better served both practically and acoustically by a digital piano nowadays.

 

[tomtoo]

Measures very good for widebands.