What is "Attack"?

No, this thread doesn't belong in the Coffee Room.
Just another one of my periodic confessional queries about the meaning of a term understood apparently by everyone but me!

Something about playing pianos or how a piano plays, I gather...

Ariel

Ariel,

It's what I have to do to take on a particularly difficult piece. It's at time like this when I'm beating out a rhythm, or playing particularly loud beating out that rhythm, that my husband exits to his shop. Not really, but it plays good for the forum.

Actually, I find that with certain pieces of music where it requires that my "attack" on the music requires a certain kind of, or a quick response from, the piano, some pianos can deliver, and some can't. The fact that it's my playing certainly has nothing to do with it.

Honestly, someone else can explain it much better.

In discussions of piano sound, I've heard the *beginning* of the sound referred to as the attack.

The sound that a musical instrument makes can be divided into three parts: The attack is the initial sound produced - in the piano it is the sound first hear from the string(s) plus the "percussive" of the hammer hit. In an accordian, for example, the attack is slow as air pressure increases until the full tone (the second part often called the fundamental) becomes steady.

The other end is the "decay" - the sound made as the sound tapers off. In a piano the decay is rather long (if the damper pedal is depressed). In say a trumpet, it it rather short.

Yeah, I think I finally get it.

Attack, like “shock and awe” which is all over quickly followed by

Decay, like a few dead soldiers a week from sniping until all the killing stops.

Yeah, I get it.
:rolleyes:

David Burton writes: :D Only what comes between Attack and Decay? Taps?

Ariel

Sigh... and Ariel thought this thread doesn't belong in the Coffee Room.

It's attack, sustain, and decay.

Steve told you the truth Ariel

This is referred to as the "Sound Envelope"

Attack - initial sound produced, in a piano, this is the initial contact between hammer and string

Sustain - the actual souding of the note, as long as that may be.

Decay - the note trailing off.

All three of these are different ways to measure things, and describe different parts of the sound. When speaking of voicing, we're typically talking about the attack of the note, or the burst of sound that you first here.
Sustain comes into play when you're trying to really listen to the singing quality of a note. Play a note, and keep the key depressed, how long can you continue to hear it? How is the sustain in different parts of the piano?
How quickly does the note decay once it begins to?
All of these things can help you to describe what you're hearing in a piano, and when in the life of that note, you're hearing it.

hope that helps a bit

Steve and KlavierBauer isn't entirely correct;
In the general meaning of a sound envelope:

- Attack is the time the sound takes till it reaches it's full volume.
- Decay is how fast the sound decays with a note pressed.
- Sustain(level) - in the general music world, it actually means the volume level that an instrument keeps while the note is played. From what I gathered from piano dealers, they usually mean decay when they talk about sustain.
- Release, the time it takes for the sound to trail off, after a note got released.

So, a piano has a short attack, long decay, basically no sustain level (since a played note eventually becomes silent), and a short release (although my 80 year old piano seems to have a longer one )

a pan flute also has a short attack, a short decay, a slightly less sustain level than full volume, and a short release (can of course change depending on how it's played)

a violin has a longer attack (unless it's pizzicato), full sustain level, no decay, no release.

I hope that helps.

If you define attack only as time for a sound to reach it's full volume, then strictly speaking the piano has no attack, only decay. While this is how attack is defined in the world of synthesis, in the acoustic instrument world it is commonly used used to mean the broad transient response of an instrument when a note is first sounded. On the piano the transient includes all of the frequency transients as well as an initial rapid decay that is essentially an amplitude transient. If you look at a plot of amplitude over time you will see an initial decay that has a high rate (~ 8dB/sec) and brief duration, followed by a second decay that has a lower rate (~ 2dB/sec) and is much longer in duration. The first decay along with the frequency transients are typically called the "attack". Even though the piano does not sustain, it is common to refer to the second decay as the "sustain". This makes sense to me because the second rate is low enough (or should be) that it gives the illusion that the piano is sustaining when it really isn't.

Sustain and attack, then, are two defining characterstics of piano tone. There is a third, which is the duration of the attack. Some pianos have a shorter initial decay, causing the amplitude to drop relatively little before the second decay rate kicks in. Other pianos have a longer initial decay, allowing the amplitude to drop more before the second decay rate kicks in. The first piano might sound more "bell" like to our ears, while the second will have a crisper, more distinct attack. Of course there are other defining characteristics, including the filter characteristics of the bridge/soundboard assembly, the vibration characteristics of the strings, and the physical characteristics of the hammers. These get rather complex in a hurry.

Here is a link to a picture of the amplitude envelope as measured with an oscilloscope. As you can see, there is no rise in amplitude, the sound essentially starts at it's maximum amplitude and decays from there. You can also see the two decay rates.

http://www.speech.kth.se/music/5_lectures/weinreic/motion.html

Ryan

thanks for the clarification Pete and Ryan.
I guess I assumed we were talkig about the sound envelope as it pertains to pianos.
The term sustain is pretty much used universally in the piano world to describe this second decay rate. To the ear it does sustsain, but is more a measurement of time than of volume. People are listening to how long it take a piano note to decay, and the term for this in the piano world is sustain (since to our ears, that's what it's doing ... fighting the tendancy to decay).

Anyway, thanks for the clarification on the REAL sound envelope!

I was suspecting that the terminoligy would be different when talking about pianos. Thanks for the info.
I'd like to add though, that from a technical standpoint, the attack of a piano isn't exactly 0.
I measured an attack rate of ~5ms on a middle C piano key (forget exactly which one).
Just because the hammer is hitting the string doesn't mean, the sound is there right away, it has to develop in the pianobody. Plus, the hardness of the hammer plays a little role also.
Technically we'er talking milliseconds, but musically speaking I agree, that a piano has 0 attack

Pete, you're right, there actually is a rise in amplitude. I didn't know what the number was, but I assumed it was small enough that it was safe to ignore Interestingly, while I was doing some experiments with synthesis I found that the ear can hear differences in attack times even in the sub-10ms range.

Ryan

There is always an attack. Even so-called square waves exist no more than in theory. Usually there is some overshoot, as well, since it takes more energy to start something moving than to keep it moving. That's all part of the attack.

This overshoot also increases the pitch a bit, too, since the tension is higher. This was one of the things that bothered me about Harry Partch's music. His kitharas were strung at such low tension that the pitch dropped off considerably after they were plucked, certainly more than the comma between a pure and equal-tempered fifth. That defeated the purpose of his advocacy of perfect intervals, as fas as I was concerned.

Still, different people hear things in different ways. I have two friends who have hyper-acute hearing sensitivity. Some things they hear are apparent once they are pointed out, so I have to believe that they hear the things that are not so apparent. (One is a world-famous maker of electric basses, the other is a trumpeter that Wynton Marsalis feels that he learned a lot from.)

an analog square wave created by a transistor gate actually doesn't have any attack time or overshoot, since as soon as the gate opens, electricity is flowing.
The overshoot happens in the digital world, when trying to simulate a square wave with a sum of sine waves, in order to avoid aliasing.

That's not actually true. There is always some time needed to get from zero voltage to whatever, whether the gate is digital or analog. You just need very sensitive instruments to measure it.

I'm not 100% sure, so I'll stop arguing
but then we're talking nanoseconds. I can't count that fast anymore

Don't mind me, I was a math student in college until I decided to do something useful. Which reminds me of a story.

A researcher pins a $100 bill on the wall, and tells a mathematician, a physicist, and an engineer that they can have the money if they can cross the room to get it. They just have to follow this one rule. When they have walked half the distance to the wall, they have to stop and wait for a second. The mathematician thinks about this and walks away. The physicist walks halfway across the room, waits a second, then walks half the remaining distance and thinks about it, and walks away. The engineer walks halfway across the room, waits a second, walks half the remaining distance and waits a second, walks half the remaining distance and waits a second, walks half the remaining distance and waits a second, walks half the remaining distance and waits a second, then walks half the remaining distance and thinks, "That's close enough!" grabs the $100 bill and walks away.

ok, you want jokes?

A doctor, a civil engineer, and a computer scientist were arguing about what was the oldest profession in the world. The doctor remarked "Well, in the Bible it says that God created Eve from a rib taken from Adam. This clearly required surgery so I can rightly claim that mine is the oldest profession in the world." The civil engineer interrupted and said "But even earlier in the book of Genesis, it states that God created the order of the heavens and the earth from out of the chaos. This was the first and certainly the most spectacular application of civil engineering. Therefore, friend doctor, you are wrong; mine is the oldest profession in the world." The computer scientist leaned back in his chair, smiled, and said confidently, "Ah, but who do you think created the chaos?"

This is a much belated expression of thanks to all the expert posters who replied in October to my thread question "What is 'Attack'?"

After bcarey, and Steve and KlavierBauer weighed in I became so overwhelmed with the hyper-expertise and detail expressed by ryan, pete blues and BDB that I didn't know how to reply at all - even to ask intelligent questions.

I had therefore planned to write a witty rejoinder expressing both my gratitude and my confusion, incorporating some of the terminology - especially from ryan's linked article - see below...Only I couldn't put that together either.

Ryan, you and BDB (BDB, oh thou math major, you REALLY pushed me over the edge)- you obviously know so much it's become impossible for mere mortals to approach your celestial heights!

...But, anyhow, it's been hanging over me that you all put a lot of effort into your answers and I never even expressed my appreciation. And I really DID appreciate them, although I'm not positive I have a real grasp of the concept of "attack" - at least not in sine waves.

When sustain turns into decay, is a subject of some confusion too, but I guess it's somewhere between licking the sound envelope and ripping it open impatiently...

You guys have impressive expertise. BDB and pete...I liked your jokes at the end!

Ariel
***********************************************



The typical decay of a piano tone is shown in Fig. 1, which displays the sound pressure level as a function of time. We see that the string is struck by the hammer at about t = 2 seconds, and the damper is released, stopping the vibration, at about t = 17 seconds. The vertical scale is in decibels, so that the ordinate of the graph is proportional to the logarithm of the pressure amplitude. In such a plot, the drop in level would appear as a straight line if the decay of the sound were of a type called exponential, which is what a physicist would expect from a linear system such as the string and the soundboard. Instead, it is clear that the curve breaks into two portions of quite different decay rates. The initial portion, called "prompt sound," drops (in this case) at a rate of about 8 dB/sec; the final one, called "aftersound," at less than one-quarter that rate. As we shall see, the prompt sound is simply related to the theoretical decay rate determined by the string's coupling to the soundboard; whereas the aftersound, which gives the piano its perceived sustaining power, represents the "miracle." [/b]

emphasis mine!
*******************************************
pete: And as for you, pete - you cut that out!

Ariel,

What is Attack?

"TomK."

Need you ask more?

She could ask you to sit on one.

With respect to all of the learned opinions and observations that have gone before — Unless I've missed something (As I might well have done. I’ve just returned from a trip and it’s late.) there is at least one point that seems to have been missed.

When a piano hammer strikes a string (or string set) it causes a creates an abrupt and violent physical displacement in the taut wire(s). This displacement will ultimately stabilize into some more-or-less coherent wave motion, or oscillation, as it develops into the characteristic wave motion that ultimately drives the bridge and creates the relatively steady-state sound of the piano. “Relatively” because it is not really steady-state, the piano strings are very lossy oscillators and the vibrating energy is decaying at some usually variable rate. Still, it is sustaining enough for us to identify pitch and volume.

However, before the string settles into this more-or-less coherent oscillating motion there is a brief period of utter chaos. There is a broadband wavefront that radiates out away from the hammer impact point and only becomes a coherent wave motion over some finite period of time — a few milliseconds down toward the bass, less up toward the treble (this time period is not clearly defined and depends on many variables such as the length of the scale, its tension, etc.). This chaotic wave motion reaches the front string termination well before it reaches the bridge (there is less distance to cover) and well before it becomes anywhere close to coherent. On reaching the front string termination, be it an agraffe or a V-bar, it causes the plate to vibrate.

Now, the plate may not actually be a very efficient soundboard, but it does move enough to create at least some sound energy. Some of this motion, along with some residual energy that is coupled to the rim and other case parts, creates most of the sound we call “hammer knock” when it gets too loud but which is otherwise so closely linked to the tone production of the piano that when it is missing (i.e., deliberately removed by careful editing) our ears often fail to recognize the resultant sound as coming from a piano.

Going the other way, this chaotic wavefront also travels toward the bridge(s). Along the way it is transforming itself into a coherent waveform, but this transformation may not fully occur until well after the initial wavefront reaches the bridge. So, although the string’s energy is very rapidly settling down into the aforementioned more-or-less steady-state wave motion, there can also be a brief period of relatively broadband noise generated by the soundboard. Again, how long all this takes is a function of scale length and tension.

All of which brings me to my answer to the original question: that period of time during which the motion of the string(s) is primarily chaotic, both at the front string termination and at the bridge (where it is coupled to the soundboard assembly), and the sound generated during this period of time, is what I call the “attack” period of the piano waveform envelope.

ddf

I totally agree with Del, (and appreciate his precise description !)

To me attack is percussion noise, mostly, that stabilize more or less fast in tone.

thats the most important part of the tone to be tuned, as the way the attack provide energy to the tone is how the tone will be projecting.

That is also the sensation that the pianist feel under its finger (he does not feel the sustain or dwell)
So you can have a piano perfectly tuned but with something missing too the pianist : the attack which have not be treated.
thats normal because the ear of the tuner close when predicting the noise.

What's the difference between sustain and decay? Are they both sections of the graph were the volume of tone is decreasing....?

I just reread the explanation in the most recent edition of the Piano Book(4+ years ago, around p. 41). Fine put them in this order: attack, decay, sustain. He defines decay as the initial and quick drop off from the peak of the attack portion of the graph and sustain as the longer part of the graph where the volume more slowly dies to zero.

1. Are the terms decay and sustain usually used as per the Fine explanation or is there disagreement on how they're used? There seems to be a difference of opinion about the order/meaning of decay and sustain compared to earlier posts.

2. In good my view good sustain involves two things:
(a)maybe most importantly, the portion of the graph where the tone drops at a slower rate starts a volume that's not too low compared to the maximum attack volume. In other words, if a tone remains audible for a long time but only at a very low volume compared to the attack maximum, that's not useful or good sustain.

(b)How long it takes before a tone dies out to the useless(not meaning inaudible)stage.

Is this a reasonable explanation of good sustain?

(If I had a graph it would make things easier to ask/explain)