@ColinThomson said:

Wow. So that would be (for March) 74 separate lines on the manuscript paper? Is that how it is notated, or is it different?

The finalescore has just 61 lines and this is of course more than you'll find in the "optimized" pocketscore. because often 2,3 or 4 divided parts are notated together in the same line, but has to be seperated to make profit of the performance legato.

The Cubase Project has still some other extralines, because high strings and low strings do have their seperate Slave-PC's in my setup Since Appassionata chamberstrings and Solostrings are - due to the licening - not divided I had to dublicate those stringlines that uses samples from different slave-PCs.

best Steffen

Hello all, I haven't had time to listen to the three pieces yet but well done Steffen for the considerable effort you put into creating them. It's good to see the work of less well-known composers getting an airing here.

Just wanted to react to two comments:

Guy: "It sounds TOO perfect, TOO worked, TOO evenly balanced etc... to me it needs to loosen a bit but keeping it in a disciplinary way."

Dave: "I would even suggest it is possible to get a more subtle performance in a notation programme than playing everything in from a keyboard."

If you use a score to create sequences, every note will be quantised, i.e. perfectly in time. Conversely, if you play a line into a sequencer in real time, most of the notes will be slightly ahead or slightly late of the click (metronome beat). The difference in timing of individual notes can be as little as 10-20 milliseconds, but when listening to a phrase, the difference in FEEL between a quantised perfomance and a live one is enormous.

As we know, most sequenced pop music relies heavily on quantisation for its rhythm tracks, but usually live guitar or percussion (etc.) is added, which has the effect of 'spreading' the beat. if you quantise absolutely everything, your track will end up sounding like Kraftwerk (a German synth band whose music is designed to sound robotic and mechanical).

In my experience, orchestral music sounds terribly unrealistic if you quantise all the tracks. Real musicians just don't play that way, so if its realism you're after, you'll have to play in the lines to introduce some real human feel. Having said that, it's great that notation programs can scan scores, thus allowing us to quickly hear individual parts without the slog of playing them in by hand.

@Conquer said:

the difference in FEEL between a quantised perfomance and a live one is enormous.

I know this traditional argument very well and Kraftwerk who definitly are looking for a robotic sound just seem to be the perfect example. but it seems to me not that easy neither in classical music nor in popular or contemporary music.

If it really would be like this a large Part of nowadays contemporary pop music which relies on modern studiosoftware should sound stiff as Kraftwerk conciously intends to do. This is obviously not the case.

For classical music the importance of minor tempochanges or "rubato" or at least a kind of "microrubato" on the side of performance depends highly on the attitude of a certain composer to notation. Of course you can kill any chopinpiece, by playing it as midifile how your scanner has read it. Nearly the opposit is the case for the most impressionistic music of Debussy, Ravel etc, which you can often kill soon you dare to loosen the underlying metric and timeconcept.

For the most pieces and composer you just have to decide the strength of your metric relations depending on each certain composition and musical situation. In short you have to understand how a composition is thought and couldn't just leave the sequencer or your "human" playing imperfection all decisions how the music must be played.

Imho, not to have any clear Idea how and where those subtle changes are appropriate and nessesary doesnt grant you any thing neither feel nor musical sense. But if you have a detailed enough Idea how it should sound you are not restricted on a certain tool and its certain deficiences but can freely use what ever is available to make the music come to live.
best
Steffen

If I may interject, although I definitely agree with Conquer's comment my criticism was not aimed at the quantization factor in this case, I have no complaint about that.
Individually all the instruments seems great but because (I think) you focussed a lot on making each line perfect, too perfect, we lost something natural in the overall mix, it's very hard to explain and I'm sure I'm not doing a very good job at it, maybe later I could find better words... [:D]

>if you have a detailed enough idea how it should sound you are not restricted on a certain tool and its certain deficiences, but can freely use what ever is available to make the music come to live.

Good point Steffen, I agree with you - there are occasions when quantisation can be musically very apt, and a mixture of quantised and non-quantised material can sound great regardless of musical style.

One further point about quantisation: instruments 'speak' at different rates depending on their attack speed - strings often have quite slow attacks, while percussion samples generally have a fast attack. If the whole score is quantised, the slow attack instruments will sound later than the fast ones and the performance will sound rhythmically disunited. To overcome this we can 'slide' the strings back in time till they feel right with the percussion, but as you've implied, such procedures take time and require careful judgement.

P.S. >If it really would be like this a large part of nowadays contemporary pop music which relies on modern studiosoftware should sound stiff as Kraftwerk conciously intends to do. This is obviously not the case.

Modern pop feels good because live performances can easily be 'spun in' over sequenced rhythm tracks using Pro Tools etc., and instances of bad timing in the live playing (and singing) can be subtly corrected without losing the overall feel. But in the '80s a lot of pop music DID sound stiff to me because of over-reliance on drum machines and quantisation.

P.P.S. 'Soft' quantisation - moving notes closer to their exact metronomic positions rather than putting them 100% in time - is often a good compromise when working under pressure, I find.

I don't know what you can or cannot do with Sibelius regarding this, but in Finale, you can de-quantize attacks and releases - - putting them exactly where you want them - - while varing tempo in small or large ways. If, like the Vienna Instruments, the virtual instrument you are working with allows it, you can also shape ADSR's dynamically, turn release samples on and off, employ MIDI Volume and Expression, use velocity crossfades, etc. Finale allows you to write and edit almost any kind of MIDI CC data you'd like. You can aslo use the Human Playback feature and write custom preferences for it to suit the sample library you are using.

Unfortunately, due to a contractual agreement with Native Instruments, Finale is currently limited to playing only "powered by NI" instruments directly - - i.e. as plugins within Finale. This, however, does not prevent you from hosting plugins in another program (e.g. Logic, DP etc.) running on the same computer with Finale and accessing them via the IAC drivers - or of course from hsoting plugins on slave computers. (I believe the previosuly mentioned contractual arrangement will come to an end next year.)

Very interesting discussion I seem to have started. I am a keyboard player (though not a very good one), and so it has always seemed like it would be cheating for me to just typing the notes and have the computer do the work. But I have been working on recreating Across the Stars by John Williams, and have inserted everything in Logic's notation window (I don't have Finale or Sibelius), and have been amazed at what it sounds like. Of course there is much tweaking left to do, but the parts that I think are close to right sound great, and not as stiff as I would have thought they would. The piece starts out with the Oboe in the foreground, and soft strings in the background. I started out by writing in the background material, and then playing the Oboe part. But (due to my lack of talent) it seems that it sounds better when I just put in the notes and let the computer play it.

Conquer, couldn't what you said about the different attack rates of instruments actually be a help in keeping it from sounding quantized? In a live performance, wouldn't the strings attack be slower when played soft, and faster when played forte? And of course, no matter what dynamic, in performance, the string attack will always be longer than most percussion attacks. The attack time, of course, can be adjusted, but I have not experimented enough with that to see if it loses realness in the process. Just some thoughts from me.

I understand the concern about loosing the human touch with notation programmes driving the performance. It is however perfectly possible to tweak every every aspect of the performace indead you can tweak every parameter of every note as much as you want. A good musician will create a good human performance.

I also find it fascinating to listen to notation programme performances of the most complex works such as Sorabji's feindish piano music. Where often even the best pianists in the world use a fog of pedalling to cope, the notation prgramme will let you listen to every single note correctly played in the correct place. It can be a revelation.

@Guy said:

I(I think) you focussed a lot on making each line perfect, too perfect, we lost something natural in the overall mix, it's very hard to explain and I'm sure I'm not doing a very good job at it, maybe later I could find better words... [:D]

Hi Guy,
to much of honor for me. My part in shaping each certain line is quite poor compared to what Berg has already done in his composition.

What might disturb you may have to do something with the certain position of this Composition in the music-historical context.

For me it is just the Point where a thrilling "dark side" of the upcomming musical Expressionism arises. (For me Berg in relation to Mahler seems to be somewhat comparable to Egon Schiele in relation to to Gustav Klimt)

To me it is a kind of esthetical narcism of going beyond total perfection towards something consciously artificial.

What means in this case going beyond Mahler and Debussy by just consequently develop their musical means further. Leaving behind their roots in their "natural" melodic or emotional judgement of musical invention heading for something new and irritating.

Thats why still very much people intuitive reacts disturbed on that kind of music. I understand that, but on the other hand I love that as part of the certain genius of Berg.

best
Steffen

Hi all, thanks for responses.

>couldn't what you said about the different attack rates of instruments actually be a help in keeping it from sounding quantized?

In theory yes, but in practise usually no, as it tends to make instruments sound a bit out of time with each other!

>In a live performance, wouldn't the strings attack be slower when played soft, and faster when played forte?

In general yes, though you can have fast-bowed quiet notes.

>the notation programme will let you listen to every single note correctly played in the correct place. It can be a revelation.

Absolutely - and getting back to Steffen's defence of the score as the traditional vessel through which orchestral musical ideas are expressed, clearly a perfect rendition of every nuance of a score can be a desirable thing. Computer-rendered scores are a great time-saver, if Steffen had played in every note by hand we'd still be waiting to hear the results!

>It is however perfectly possible to tweak every every aspect of the performace, indeed you can tweak every parameter of every note as much as you want. A good musician will create a good human performance.

Yes - however, I've found it impossible to 'create' a performance by tweaking a mechanically-rendered line; I have to first get it sounding half-decent by playing it in live, after which I can tweak the timing, note lengths, dynamics etc. (Maybe if I was a better programmer I could do the whole thing on screen without touching a keyboard?)

The Berg mock up is a nice addition here, I'm now thinking of doing a Charles Ives mock-up sometime in the near future, I love his orchestration, I think that would also be interesting to have on the site.

@Guy said:

The Berg mock up is a nice addition here, I'm now thinking of doing a Charles Ives mock-up sometime in the near future, I love his orchestration, I think that would also be interesting to have on the site.

Great, try Ives "unanswered question" or the simultaneous marches of "Three Places in New England" [[;)]]

best
Steffen

@Guy said:

The Berg mock up is a nice addition here, I'm now thinking of doing a Charles Ives mock-up sometime in the near future, I love his orchestration, I think that would also be interesting to have on the site.

Great, try Ives "unanswered question" or the simultaneous marches of "Three Places in New England" [[;)]]

best
Steffen

No! No! don't do the unsanswered question it's so over played but Yes! Yes! please do "Putnam's Camp" from "Three Places in New England" or the even more wild "The Fourth of July" from "The Holidays Symphony".

Yeah, I was thinking something in the genre of "Putnam's Camp".

Beautiful job, Steffen-- and a great compositional choice. There's not much else I can say that hasn't already been said except to express my thanks for choosing Berg and taking such great care with these works. What a nice treat this is!

I have severe problems with this music, compositionally, but it is a great performance.

@William said:

I have severe problems with this music, compositionally, but it is a great performance.

William, just do like 95% of the world does, pretend you really dig this music. [:D]

Got it. [:D]

Hi William - can you say what it is about the composition which is problematic for you? It's interesting to know what the 'turn-off' factors' are in a piece of music.

First let me thank all for their vivid reactions on my very first serious VSL-programming.

I am fascinated, that Berg nearly a century after writing this score still has the power to rise musical controversy, which for decades seemed to be totally overcome.

Obviously there is something timeless challenging in the very certain way musical skills, thought and passion has intermingled in Bergs Music.

I like that, but do not expect that evrybody has to.
best
Steffen