How to reproduce sound of piano

Try ATC active speakers and a JL 113 sub or SVS PB13 Ultra (with the ports
blocked) - I guarantee it will get you pretty darn close. Piano is not easy as it
is both percussive (110 db dynamics) and has precise timbre as well as a
presence (like a harp) from the resonance of the unplucked strings (needs
super low IMD distortion and very low Q speaker on low notes). It also
requires a speaker with wide even dispersion or it will not sound natural. You
definitely need a sub though for a convincing grand at your price point. The
speaker designer was a professional pianist and this seems evident in how
the design handles piano.

[quote] Are electrostatics way to go? [/quote]

For midrange clarity yes but for realistic dynamics and LF response - no. Let
me say you need a very clean waterfall plot. If you go dynamic then try
damped drivers (paper/pulp or soft/woven fabric - designs that do not ring -
as you want to hear the piano strings resonate not the sound of the speaker
driver)

People say panels like the Quad 57 sound fast...this is the effect of a clean waterfall...there is no lingering ringing coloration to the sound - everything has a "light" realistic quality about it and you just hear more of the details on the recording. In a conventional box speaker this requires drivers that have internally critically damped drivers within their operating bandwidth (this generally requires doped paper/pulp or woven fabrics - anything to "deaden" the sound of the diaphragm itself and prevent ringing)

Shadorne,
How true, but I hate to have my head mimic a soundstage.

Exactly - me too. I don't like the sound being in my head. Headphones eliminate the pinna and that means they can never sound realistic even if they have superb performance.

Anyone else feel that best headphones do better justice
to piano than most speakers?

Headphones have very low distortion levels and only a handful of speakers are
designed such that they could begin to compete with $1000 headphones. I
mentioned the numerous distortion and dynamic compression problems in
conventional speakers above - a $1000 headphone playing at tiny output levels
close to the ear has a much easier job than a speaker. Remember the vast
majority of speaker drivers cost less than $100!!! (basically there is no contest
between excellent headphones and most speakers)

Hello, Shadorne said a while back in this thread,"Let
me say you need a very clean waterfall plot."

What exactly do you mean by that? Also why do you want this? What does it mean ? And how do you get this? How do you identify this?

The best is to give a classic example - the Quad 57. Because the cumulative spectral decay plot (waterfall) is so clean in the mid band this became a famous speaker - despite many other shortcomings. Most people will say this is one of the best sounding midrange around.

The reason is that the subtle timbre of the piano is not MASKED by driver resonance.

IMHO, it was a sad day when the industry began widely adopting lightweight metallic drivers with small motors - most of these designs ring badly and require "notch" filters to try and limit the ringing. Most people do not realize that it takes only a ringing at one frequency within several octaves to completely mask all the other subtle sounds within that band. It is called "masking" and has been studied extensively - it happens when one sound at a certain frequencies "masks" other sounds across an entire band or range of normally audible frequencies and SPL levels - this scientifically proven effect is used to compress audio files in the MP3 standards (basicaly they remove sounds that they "know" you can't hear anyway and make the compressed audio file much smaller).

this generally requires doped paper/pulp or woven fabrics - anything to "deaden" the sound of the diaphragm itself and prevent ringing

I would add that the latest approach is to this problem is to use "Constrained Layer Damping" - instead of doping a speaker cone with a viscous damping fluid they are now sandwiching a viscous fluid between two rigid cones - kind of like a springing car suspension with "shock absorbers" to dampen it. This idea works well because the internal energy within the cone is lost as the constrained layer shears.

Basically if you can't dissipate the stored energy in a rigid cone you unavoidably end up with some resonance - after all this is exactly how cymbals are made and they ring a long long time...

Newbee and Dave,

You guys are on to something - how to make a grand piano sound real in a room that is not big enough for a grand piano. It seems we all missed the obvious...it just ain't possible. The room is a good 50% of what you hear...