Bach: Goldberg Variations

[nattawa]

好！也推荐hapsichord的GV. 现代钢琴因为声音sustain很好反而演奏中不方便加入巴洛克式的装饰，harpsichord就没有这样的限制，音乐可以演奏得听上去特别“复杂”和有味道，有可能更接近作曲家当时想象的声音。今天常听的GV都是钢琴版，偶尔听听harpsichord有耳目一新的享受。

[nattawa]

作者: Woody-Allen

看看Angela Hewitt关于现代钢琴弹巴赫的好处：
http://www.chinasmile.net/forums/sho...d.php?t=493595 ("http://www.chinasmile.net/forums/showthread.php?t=493595&quot...

我看她不是在讲“好处”，而是在讲不能把钢琴当harpsichord用，和如何在演奏巴赫时利用现代钢琴的优势。这很象彩色摄影跟黑白摄影，各自有一套不同手法。

[nattawa]

作者: Woody-Allen

她在第一讲里很清楚地示范了在羽管键琴上无法分辨的两个voices在钢琴上能够同过音色的不通清晰地分辨除来。这当然算是钢琴的“好处”了。

她的意思是说在harpsichord上无法象现代钢琴那样用控制dynamic的方法控制声部平衡, 并不是说多声部在harpsichord上无法分辨.

[nattawa]

作者: Woody-Allen

You think you can identify the two voices probably because you have heard it on modern piano before......

She did say very clearly and explicitly t...

I didn't think I could. I actually can. So can you and anyone else.

What Hewitt was talking about by saying "you simply cannot do that" was exactly referring to the dynamic control, or lack of it on the harpsichord part. Never did she say anything about or imply that harpsichord being unable to bring out polyphony textures in that video clip. She would never do so because it is simply not the truth, no matter how much she favors her Fazioli. Now anyone who doubts a harpsichord would have to question himself why Bach would ever compose voices one can't make out from an instrument it's meant to be played on, and this can be a tough one to answer.

It is a fact dynamic control on a modern piano opened vast possibilities for music interpretation, even for works not composed with dynamics in mind, this however never by any sense makes a harpsichord a wrong instrument for polyphony compositions.

[nattawa]

作者: Woody-Allen

(1) Most of the keyboard pieces that he composed were meant for organ, not for harpsichor...

Organs have no or little dynamic control either, therefore, not polyphony friendly, would you not say?

作者: Woody-Allen

(2) Pieces like GV and those of WTC are not for performance to an audience, but for his kids or students to practice and perfect the art of counterpoint.

I'd have to pull my hair out trying to imagine one could practice and perfect an art when he could hardly hear himself. Don't tell me Beethoven did it please.....

作者: Woody-Allen

....... how you could distinguish them.

Very simple, looking for them, as opposed to having a pianist shove a voice up in my face when it is a modern piano. When there is "noise" coming up one would know a new voice has kicked in. It's sort of searching for the composer's secrete, hard but fun.

[nattawa]

They have multiple stops

Stops will not help, will they? Harpsichords have them too. Hewitt would tell you "you simply cannot do that on an organ either" if you'd ever try that little canone on an organ and manage another person pushing or pulling stops for you while you have 2 voices on the right hand and a 3rd on the left (impossible scene, isn't it?). You'd find the stops change sound of both right hand voices at the same time as they happen to be proceeding within the same octave, and what was "indistinguishable" would remain as "indistinguishable".

I found it more straightforward to say Bach composed keyboard works because they could be heard, understood, and enjoyed on the instruments available at his time, despite lacking dynamic control, than saying any other ways around.

Knowing that a new voice has entered is simple, of course. But when both voices are developed...

So it's not impossible to ID voices on a harpsichord, is it? That's the point I was trying to make.

Following the voices' development fully by simply listening, once? I've never even thought of having that gift. I believe most people can not either, even if it's a piano. That's probably why people come back to listen to same pieces again and again, and make discoveries every time. Perhaps this is the way music is meant to be enjoyed.

Modern pianos did not make impossibles possible. What they did to Bach's voices was hopping and waving, "hey, hey, I'm here", so that you don't have to try as hard to look for.

[nattawa]

作者: Woody-Allen

Hewitt would never say that.

Oh, I'm sure she would. You apparently forgot what stops in an organ can do and cannot do, and in what way. Again, in the very canone Hewitt took example, stops would certainly not help set apart the voices were it on an organ, by the design of the mechanism of stops itself, i.e. stops only work by octave or group of octaves, they change sound of all notes within octaves that they are associated with by the same way.

I did not, and have no intention to say or imply a harpsichord is as good as an organ in handling polyphony music. Neither would I mind if one is much favored over another. What I was trying to point out is that you could have not brought up this multiple stop thing, as it serves you no support.

As of the fact that Bach composed a lot more for organ of all keyboards, why could not he have done so simply for the magnitude of the sound, let alone it would earn him the most money possible (and it did seem so)? I bet even the instrument maintenance cost wasn't out of his own pocket.

[nattawa]

作者: Woody-Allen

I am not an organist nor a mechanic, but I did hear about mutation and mixtures in organ stops. Explain to me why they are not useful for changing ton...

They do make changes. Only that all notes within the octave get the same change.

[nattawa]

作者: Woody-Allen

I don't think this applies to mutation stops.

Mutation stops are not much different if not identical. We probably didn't have to come this far and break down an organ. The simple fact is, no stops in an organ are designed to work by voices.

Coloring voices with stops can be done only when voices are written/played on different octaves, as each octave has its own stops. For voices played by a single hand, like the piece Hewitt demonstrated, it is impossible, the two are too closely placed and sharing many notes within the same octave.

[nattawa]

作者: Woody-Allen

A stop can be used for a rank of pipes or an individual pipe. Conversely, a rank of pipes can have multiple stops. I don't see why it has to be the ca...

Stops and their controls are different things.....

They say one picture betters 1k words....so let's build an organ for Woody.



We have a simple one-key organ prototype with our organist, Woody Allen, at the keyboard. Despite the smallish range the organ has 10 ranks of pipes per key. The 4 smallest pipes are mutation pipes, they are connected together so they will sound or mute together as one pipe.

Woody wants to be able to change the sound quality, or timbre, from time to time when he hits the key. So we put in 7 control valves (called stops or “sliders” in the picture) at the bottom of the pipe system so that each of the pipes can be turned on or off at Woody’s will by operating the valve controls prior to hitting the key.

So how many control handles, or stop knobs as they call them, do we need to install for Woody to operate the valves? Seven, obviously, one for each valve. Wrong!

Woody is not interested in the valves themselves, what he wants is the timbre of sound. Then what is that? It’s nothing but certain combination of the on/off state of the 7 valves. The total number of combination is a huge figure.

We easily pick 20 combination of Woody’s choice and built a complicated linkage system and installed 20 knobs on a side panel (not shown in the picture). Each knob has the name of the sound timbre printed on it so that Woody can easily identify and select one or more to control and explore.

Woody is happy but soon he wants a real instrument. We then build him a 61-key organ, again with 10 pipes per key, but 61 pipes per rank this time. When it comes to the design of the stops we have a problem. There will be too many stop knobs if we are to offer independent timber controls to individual keys, insanely 1220 knobs!

Woody quickly realizes it would be impossible to set up the instrument’s sound timbre quickly with so many knobs, plus timbre control down to the key level is hardly desired musically. So we decide to keep the 20-knob design unchanged, except that we extend the valve controls, so that the control that goes to one pipe in the one-key organ now goes to all pipes in its rank. The result, changing the knob setting changes the sound timbre of the entire instrument.

[nattawa]

作者: Woody-Allen

Haha! Nice try, but you made it much more complicated then necessary. How many phone numbers that we need to dial? Numerous. How many number keys we ...

How does your analogy work? The operating principles between an organ stop control and a telephone dialing are fundamentally different. An organ is to DECODE the pre-programmed control knobs, while a telephone is to ENCODE an address of device with numbers. Besides, why 10 numbers? Two would be enough, 0 and 1.

[nattawa]

作者: Woody-Allen

No need for PREPROGRAMED stops. At any time the organ player can pull more than one stops.

Ah! I see what you meant, number pad vs 1-tough speed-dial.

The 7 straight-control stops are among the 20 knobs in our organ, they are pre-program in its simplest form. The other 13 are speed-dial knob, so to speak. Are you sure Woody doesn't want speed-dial at all? We can easily remove them and save the panel space for future expansion.