Topic by matthew

  • Avatar
  • matthew
  • Member
  • Offline
  • From: Seattle, USA
  • Registered: 18-01-2014
  • Posts: 55

Topic: Questions about Dynamic Range

Bonsoir,

My name is Matthew, and I live in Seattle (USA).

I must first say that I appreciate the democratizing nature of software such as this, allowing measly apartment-dwelling liberal arts grads like myself to have an aural experience not unlike that of the richer folk who tap upon their fancy grands. Glenn Gould would have loved Pianoteq.

I have one particular question for any savvy users. At times, on rainy evenings, I like to make believe that I am the young Haydn, recently kicked out of St. Stephen's, as he played upon his old, rickety, "worm-eaten" clavier in his garret apartment of Vienna. On such occasions, I like to light a candle or two and choose, of course, the clavichord preset. I have noticed, however, that the dynamic range is quite narrow for this instrument (20 dB). This narrow range makes me think I might be missing something. And when I increase it to, say, 40 or 50 dB, it just doesn't sound right.

When going through the other, more modern pianos (at times a wave of Mendelssohnian melancholy washes over me), I notice that the dynamic range for the "recording" presents is much narrower than that of the "player."

My questions are: What is the best dynamic range for practice of classical piano? Which range should I stick to if I want my practice to be representative of a real grand piano (if one day I actually see or touch one of these mythical things)? Will practicing with a preset that uses a narrow dynamic range eventually hurt the expressiveness of my eternally feeble playing?

I imagine that it's best to stick to the "player" presents for practice, which have a dynamic range of 40-45. If it matters at all, I use headphones (Haydn, alas, did not). Any feedback?

Merci beaucoup à tous!

Reply by Luc Henrion

  • Avatar
  • Luc Henrion
  • Member
  • Offline
  • Registered: 02-05-2007
  • Posts: 933

Re: Questions about Dynamic Range

Well, a clavichord is not able to deliver what you'd call a thunderous sound in reality!
I agree, though, that a grand piano can, and if your headphones or your monitors are able to follow, why not give a test to a broader dynamic range? There is one thing more: in the real world, the further away you are from the instrument, the narrower the dynamic range. That's why when you put mics inside the piano you have to compress heavily, while a more classical approach (about 2 or 3 meters from the pîano) doesn't require such a compression, or no compression at all.
My 2 cents

Luc Henrion's Website

Reply by cpcohen

  • Avatar
  • cpcohen
  • Member
  • Offline
  • Registered: 15-04-2013
  • Posts: 79

Re: Questions about Dynamic Range

Luc Henrion wrote:

. . .  There is one thing more: in the real world, the further away you are from the instrument, the narrower the dynamic range. That's why when you put mics inside the piano you have to compress heavily, while a more classical approach (about 2 or 3 meters from the pîano) doesn't require such a compression, or no compression at all.
My 2 cents

Two questions:

. . . Does anyone have good data for the dynamic range (at various pitches) of an acoustic piano?

That is, the range (in dB) from "pppp" to "FFFF" ?

. . . Can anyone explain this effect Luc mentions -- that the dynamic range is wider,
. . .    the closer you are to the instrument?

THanks ==

.    Charles

Reply by delt

  • Avatar
  • delt
  • Member
  • Offline
  • From: Montréal, Canada
  • Registered: 11-06-2013
  • Posts: 401

Re: Questions about Dynamic Range

Also, i personally find that all types of headphones have a "compessing" quality to the sound, in general. When using headphones you tend to hear very soft sounds much more clearly, than you would if the same sound came from a certain distance - as in, from speakers, or from a live instrument.

http://soundcloud.com/delt01
Pianoteq 5 STD+blüthner, Renoise 3 • Roland FP-4F + M-Audio Keystation 88es
Intel i5@3.4GHz, 16GB • Linux Mint xfce 64bit

Reply by kalessin

  • Avatar
  • kalessin
  • Member
  • Offline
  • From: Germany
  • Registered: 06-12-2013
  • Posts: 306

Re: Questions about Dynamic Range

cpcohen wrote:

Two questions:

. . . Does anyone have good data for the dynamic range (at various pitches) of an acoustic piano?

That is, the range (in dB) from "pppp" to "FFFF" ?

. . . Can anyone explain this effect Luc mentions -- that the dynamic range is wider,
. . .    the closer you are to the instrument?

THanks ==

.    Charles

It might be worth noting that there is a hard 'floor' when loudness is concerned: below a certain velocity, the piano will remain silent since the hammer just does not hit the string(s) anymore. Also, there are practical limits on the highest velocities that can be reached by a human's muscles and/or without damaging the instrument. So, 'pppp' is a tad optimistic in my opinion, as is 'ffff'; I usually think of a piano's range in terms of 'ppp' to 'fff', but that's just me.

That said, the dynamic range of a concert grand piano can reach about 50 decibels, at least as far as I know (actual pianists, anyone?). Parlour and baby grands offer a somewhat smaller dynamic range, and uprights AFAIK even less. About 40-45 decibels for a medium grand sounds about right to me.

Concerning distance: in an ideal recording environment without noise and/or reverberation, the dynamic range cannot be a function of distance, at least when my understanding of the basic math does not fail me. However, reverberation is always present, and it describes the amount of sound that is reflected in relation to the part that is absorbed. Absorption is energy transfer, and energy transfer is usually a function of intensity. That means in terms of acoustics: I wouldn't be surprised if the relative reflected amount gets larger the quieter the sound is, because absorption decreases. But this is just a plausibility consideration.

If that were true however, this would effectively mean that reflected sound is always de facto compressed. And when you move the mikes away from the piano, what you effectively do is change the ratio of 'direct' to 'reverberated' sound. So it might actually be the case that the effective dynamics decrease with increasing distance. I am not a sound technician, so I have never tried it. If it does, though, I suspect the effect will not be huge: noticeable but not gigantic.

Apart from that, a dynamic range of 50 or even 60 decibels is easily recordable even at 16 bits (and professional equipment usually records at 96/24 or 192/24) even without any compression. If such a recording without any compression whatsoever is very 'playback-friendly' is another matter: I think listening fidelity is a main motivation for slight compression even in classical recordings. One should also consider that the usual noise floor in a normal, 'quiet' environment is already rarely below 30-40 decibels, and everything above 80 is usually already uncomfortably loud. This is the reason that, were I a sound technician, I would probably aim for 30-35 decibels of dynamic range (modern pop and rock, metal recordings have significantly less than even that).

Last edited by kalessin (18-07-2014 08:23)
Pianoteq 6 Standard (Steinway D&B, Grotrian, Petrof, Steingraeber, Bechstein, Blüthner, K2, YC5, U4, Kremsegg 1&2, Karsten, Electric, Hohner)

Reply by stamkorg

  • Avatar
  • stamkorg
  • Member
  • Offline
  • Registered: 05-04-2014
  • Posts: 467

Re: Questions about Dynamic Range

I find that from a players perspective, a dynamic between 35-40 in Pianoteq is nice to play, IMO.
But I would like to know if the dynamic range should not be different on each key, with a greater dynamic range on the low notes, and shorter on the highs.
I don't know if it is the case in Pianoteq

Reply by Philippe Guillaume

  • Avatar
  • Philippe Guillaume
  • Admin
  • Offline
  • Registered: 25-04-2007
  • Posts: 5,405

Re: Questions about Dynamic Range

stamkorg wrote:

I find that from a players perspective, a dynamic between 35-40 in Pianoteq is nice to play, IMO.
But I would like to know if the dynamic range should not be different on each key, with a greater dynamic range on the low notes, and shorter on the highs.
I don't know if it is the case in Pianoteq

This paper:
http://www.google.fr/url?sa=t&rct=j...8958,d.d2k
shows the dynamics at several places from bass to treble for different piano brands (see page 11).

Philippe Guillaume's Website

Reply by cpcohen

  • Avatar
  • cpcohen
  • Member
  • Offline
  • Registered: 15-04-2013
  • Posts: 79

Re: Questions about Dynamic Range

Phillippe -

. . . Thank you!

Kalessin --

What _might_ be happening:

The amplitude of reflected sound will be linear (very nearly) with the amplitude of the piano's vibrations, as will the amplitude of direct sound.

But the leading transient -- the hammer strike -- will be "smeared" in time, in the reflected sound.  And the further away you are from the piano, the greater the "smear". 

So when you're close to the piano, the peak will be shorter (and hence, louder) than it will be when you're 30 feet away, and a lot of what you're hearing is reflected sound.


I don't know if that's _right_, but maybe it's plausible.<g>

.       Charles

Reply by stamkorg

  • Avatar
  • stamkorg
  • Member
  • Offline
  • Registered: 05-04-2014
  • Posts: 467

Re: Questions about Dynamic Range

Thank you Philippe, it is interesting but a little hard for me.

But I understood that the dynamic range on the high notes is greater than on the lower. I thought it was the opposite

Reply by Philippe Guillaume

  • Avatar
  • Philippe Guillaume
  • Admin
  • Offline
  • Registered: 25-04-2007
  • Posts: 5,405

Re: Questions about Dynamic Range

stamkorg wrote:

Thank you Philippe, it is interesting but a little hard for me.

But I understood that the dynamic range on the high notes is greater than on the lower. I thought it was the opposite

Those figures are not easy to read, in particular because of the "noise". Let's take a look for example to the one labelled "Bösendorfer SE290" in figure 7, and lets consider all data without distinction: legato ("lg", in green) and staccato ("st", in red).

The lowest curve, corresponding to C1, shows more or less 58 dB at velocity 20 and 98 dB at 120. That makes 40 dB dynamics from velocity 20 to 120.

The upper curve, corresponding to G6 (and C5), shows more or less 72 dB at velocity 20 and 110 dB at 120. That makes 38 dB dynamics from velocity 20 to 120.

Hence 40 dB dynamcis in the bass range, 38 in the upper range. Because of the noisy curves, those 2dB difference are not very significant. The other figures show a similar behavior. Hence, the rough conclusion I would draw from those curves is that:
- the dynamics ir more or less the same for all notes,
- the treble notes are louder than the bass notes.

Philippe Guillaume's Website

Reply by Wheat Williams

  • Avatar
  • Wheat Williams
  • Member
  • Offline
  • From: Dayton, Ohio, USA
  • Registered: 23-01-2011
  • Posts: 132

Re: Questions about Dynamic Range

matthew wrote:

...the clavichord preset. I have noticed, however, that the dynamic range is quite narrow for this instrument (20 dB). This narrow range makes me think I might be missing something. And when I increase it to, say, 40 or 50 dB, it just doesn't sound right.

A clavichord is not a piano!

A real acoustic clavichord is a very quiet instrument with a weak tone and a very small dynamic range, due to the particular design of the key mechanism, which does not use hammers like on a piano. The sound of the clavichord is in no way comparable to a modern piano. If you amplify your Pianoteq clavichord with your speakers or in your headphones to make it as loud as a modern piano, you are not doing it correctly. By the same token, if you increase the dynamic range, it will certainly sound unnatural.

Furthermore, the most common examples of historical clavichords only had a range of about three and a half octaves (around 45 keys), far short of the 61 keys found on the Neupert clavichord (upon which the Pianoteq virtual clavichord is modeled), or 88 keys found on the modern piano.

Do a YouTube search on "clavichord" and you can hear some demonstrations of the real thing.

Here is one example.

http://youtu.be/ZWXh3Ec2Ypw

Last edited by Wheat Williams (23-07-2014 14:45)
Dayton, Ohio, United States of America
macOS 10.14.6 Mojave • Apple MacBook Pro (2017), no Touch Bar • 2.3 GHz Intel Core i5, 2 core • 8GB RAM

Wheat Williams's Website

Reply by matthew

  • Avatar
  • matthew
  • Member
  • Offline
  • From: Seattle, USA
  • Registered: 18-01-2014
  • Posts: 55

Re: Questions about Dynamic Range

Yes, thank you. I am indeed aware of the differences between a piano and a clavichord. I was simply concerned that too much long-term practice (like a couple hours daily) on an instrument with a narrow dynamic range would have detrimental effects on my technique upon switching back to the piano presets (or a real piano), especially since I'm using the same controller.

However, I believe that my questions from this old post have been answered, and that 40db in the piano presets seems like a safe range for practice representative of a real piano.