What is a good piano reproduction???

[RayDunzl]

Ran across this today...

 

[amirm]

Can see why she would be so popular! Man her fingers seem to move faster than the camera shutter!

 

[RayDunzl]

I don't think I've ever heard that piece before now.

"Prokofiev's blisteringly difficult Toccata in D minor, Op. 11."

http://www.el-atril.com/partituras/Prokoviev/Prokofiev_-_op11_toccata.pdf

 

[amirm]

Certainly seems blistering difficult!

 

[Keith_W]

I don't think of piano as a revealing test of a system because it is inherently coloured and limited dynamically - no matter how many notes are playing, they're all being generated by a single, thin wooden board, the very definition of resonance, colouration and intermodulation (which is why we like the sound). Solo piano is not difficult for an audio system to reproduce, and would tell me nothing about whether my system had similar (unwanted) 'wooden board' characteristics.

There is no way that piano is limited dynamically!!

Many years ago, a friend of mine was going to give a piano concert and wanted my opinion on which piano to choose. We went backstage to a large-ish room (about twice the size of my listening room). In it there were four concert grand pianos. She went to each one and played and asked me what I thought. Well, I couldn't think any more. My feet were jelly. I can tell you that when you hear one of these things close up, and in a small room ... it is a physical experience. It vibrates the air in your lungs.

Perhaps you might think it has limited dynamics because you are used to sitting 10-20m away.

In any case, piano has been my go-to test of a system's capabilities for years. The lowest key on the piano is A (about 27Hz), the highest is C (about 4200Hz). In addition, these notes generate plenty of harmonics. You need a system that is linear from top to bottom with excellent time coherence, otherwise piano will sound strange.

People who know me will know that I have been complaining about the bass of my speakers for years. That is, until I managed to time align them. What I was hearing was a piano where the bass seemed disconnected from the harmonics. In hindsight it's so obvious - what I was hearing was the harmonics arriving at my ear first, followed by the fundamental. The measured discrepancy between woofers and horns in my system is about 10ms - more than enough to smear piano notes.

Missing bass in a system does not harm piano listening all that much, because of the missing fundamental effect. But it is much better if it is there - it sounds so much more realistic.

Systems with superb dynamics do the best job with piano reproduction IMO. This is why I have never been a fan of electrostatic or panel speakers - very few of them (if any) are able to produce the slam of a sudden fortissimo.

And lastly - resolution. You will need this in spades if you want to hear the timbre of the notes as well as the acoustic of the concert hall.

I think piano is great for testing all aspects of your system. But you do need to know what good piano sound is in the first place.

 

[fas42]

Many years ago, a friend of mine was going to give a piano concert and wanted my opinion on which piano to choose. We went backstage to a large-ish room (about twice the size of my listening room). In it there were four concert grand pianos. She went to each one and played and asked me what I thought. Well, I couldn't think any more. My feet were jelly. I can tell you that when you hear one of these things close up, and in a small room ... it is a physical experience. It vibrates the air in your lungs.

Perhaps you might think it has limited dynamics because you are used to sitting 10-20m away.

Spot on! Most musical instruments produce immense intensity of sound when you're up close - the impact of the sound pulsates through your being, you're swamped by the richness of the notes, it's a remarkable feeling ...

Of course most systems fail terribly in this aspect, they fall at the hurdle of reproducing that intensity - which is key to getting all recordings to deliver of their best. A system fully sorted will deliver this experience, but it may require a great deal of concerted effort and attention to detail to make it happen.

Overall, it is worth doing ... because the rewards are so great ...

 

[fas42]

And, right on queue, a video report from the RMAF 2016 show, which shows a system which doesn't suffer from this problem - it's becoming easier to assemble the bits to make it happen, and then it just requires "debugging" the remaining issues ...

 

[amirm]

And, right on queue, a video report from the RMAF 2016 show, which shows a system which doesn't suffer from this problem

On what basis do you say that? I was in that room and this is my assesment: http://audiosciencereview.com/forum/index.php?threads/rmaf-2016-raidho-acoustics.859/

 

[RayDunzl]

The reviewer said something like "best bass ever" at the end - 11:45.

RTA of the video:

 

[fas42]

Amir, you said "The sound was quite dynamic but again, a tad bright." - that's exactly where a system should be on its journey to getting "special" sound. If a setup has a heavy overlay of sounding bassy, and drab, then there is a lot more that has to be done. The "brightness" is the indicator that the rig has the grunt to deliver BIG sound - if it sounds a bit off then you're listening to distortion artifacts which need to be eliminated.

 

[Keith_W]

The reviewer said something like "best bass ever" at the end - 11:45.

RTA of the video:

View attachment 3775

Do you think this is a reflection of the speaker's performance, or a reflection of the recording equipment used, or an artefact of Youtube compression? I listened to that video through the same headphone setup I use for music (Beyerdynamic T5P and Schiit DAC) and even the voices at the beginning sounded tinny and metallic to me.

 

[RayDunzl]

Do you think

I try to.

Sometimes I just present a data point.

 

[fas42]

Do you think this is a reflection of the speaker's performance, or a reflection of the recording equipment used, or an artefact of Youtube compression? I listened to that video through the same headphone setup I use for music (Beyerdynamic T5P and Schiit DAC) and even the voices at the beginning sounded tinny and metallic to me.

Keith, part of the problem with YouTube at the moment is that they've recently changed the processing options - unless you send the video in the right format it will be encoded poorly, 126Kps at best. Optimum is 192Kbps, at 720p setting, for viewing - which unfortunately AVshowreports hasn't cottoned on to ...

One has to be able to "listen past" these aspects, when viewing.

 

[amirm]

Do you think this is a reflection of the speaker's performance, or a reflection of the recording equipment used, or an artefact of Youtube compression? I listened to that video through the same headphone setup I use for music (Beyerdynamic T5P and Schiit DAC) and even the voices at the beginning sounded tinny and metallic to me.

Audio compression doesn't change frequency response (with the exception of MP3 above 17 Khz). So the likely suspect is the camcorder microphone.

 

[Blumlein 88]

Amir, you said "The sound was quite dynamic but again, a tad bright." - that's exactly where a system should be on its journey to getting "special" sound. If a setup has a heavy overlay of sounding bassy, and drab, then there is a lot more that has to be done. The "brightness" is the indicator that the rig has the grunt to deliver BIG sound - if it sounds a bit off then you're listening to distortion artifacts which need to be eliminated.

Quite the conjuring here. The brightness is an indicator of where .......... come on. As Amir and Keith pointed out, the voices are tinny at the beginning. The camcorder or whatever mic is effecting the sound. And even with Amir saying he heard it and thought it a tad bright as someone listening to this video you don't know how much of this is the recording and how much is inherent to the speakers. You only know the recording is partly responsible.

 

[fas42]

Quite the conjuring here. The brightness is an indicator of where .......... come on. As Amir and Keith pointed out, the voices are tinny at the beginning. The camcorder or whatever mic is effecting the sound. And even with Amir saying he heard it and thought it a tad bright as someone listening to this video you don't know how much of this is the recording and how much is inherent to the speakers. You only know the recording is partly responsible.

Dennis, we're talking about two entirely things there. Amir, in the room in person, heard "brightness" from that system - and then we also have this video, encoded relatively poorly. I'm not the lightest bit interested in how "bright" the clip sounds, only in how well the dynamics were portrayed - this is where experience comes in, I'm listening to different things from what you are ... the telltale signs are what matters, not "the overall impression" ...

 

[Blumlein 88]

https://dl.orangedox.com/i0D1EHQHOynolfTmpc

You can download this. Grabbed the mp3 of the audio off that video. Looked at Ray's spectrum of it. Figured the mic gave up around 300 hz. So I boosted 300-75 hz by 12 db/octave. Rolled out of the boost to shelve it at 50 hz and below. This was just a guess from the graph and how vidcam mics work. Sounds more reasonable as a first guess. I spent maybe 60 seconds doing this.

The other thing apparent looking at this in Audacity is the video cam employed AGC of course. Frank can automatically compensate for unknown video cam mics running unknown types of AGC to access the dynamic portrayal of the entire rig. No reason to think AGC would mess with the telltale signs of dynamic capabilities.

 

[fas42]

Figured the mic gave up around 300 hz. So I boosted 300-75 hz by 12 db/octave. Rolled out of the boost to shelve it at 50 hz and below. This was just a guess from the graph and how vidcam mics work. Sounds more reasonable as a first guess. I spent maybe 60 seconds doing this.

Why would a mic "give up" at 300Hz?

 

[Blumlein 88]

Why would a mic "give up" at 300Hz?

Not a mic designer. However, many small condensor mics in cell phones and videocams are reasonably flat in response above some point. That point by design of the circuit or microphone is quite often around 200-300 hz. Maybe they are just rolling off the input to not get rumble and noise in the signal. Ray's FFT looked like it was 300 hz for this one. Download and hear the file I made available. It is in the ballpark. And I spent no real time on it.

EDIT to add:

Having looked up some info, even inexpensive condenser mics in laptops, cellphones etc likely have good response to 10 hz or so. So I am going with the guess they limit the low end on purpose to prevent low frequency noise pickup, handling noise, bumps with other objects etc. It would also prevent over loading the rest of the circuit which is more likely to happen with low frequencies.

 

[c1ferrari]

Ran across this today...

I was unaware of these NPR Music Field Recordings -- effusive thanks for turning me on to them!