I was very interested in this since I've played horn on this in brass choirs and really like the music.

I hope you don't mind my criticizing. I don't mean to be too negative since I admire you for doing this.

First, in general the trumpet dynamic sounds mf, but the horns are ff. They do not match timbrally even when playing against each other.

Secondly, there is almost no phrasing. It cannot be played by live brass like this, because they have to take breaths. You need to look at every line and figure out where a human being would breathe. On this piece it is VERY often, because it's loud.

Third, the notes are too connected - they need more separation when they are not legato. There are some marcatos here that sound like they are played on an organ. But this could be rememdied by having the accented notes more separated. To create a huge contrast between the legato, and the more marcato notes helps a lot with realism because especially in brass it is normal to shorten notes that you want to give emphasis to.

Fourth there should be more inaccuracy - it sounds very MIDIfied and mathematically perfect. And along with this there should be some small decresc or cresc in the lines to bridge together shorter phrases. J Bacal did an absolutely fantastic job of subtle dynamics on this on his pieces. It isn't just written-in-the-score dynamics, but the implied ones that players and/or conductors always put in as part of expressive playing.

Anyway I hope you're not offended by my criticizing because I like what you've done here and am always interested in someone trying an ambitious performance like this.

This became incredibly hard to listen to almost right away. It is way too mathematically perfect. There is no relishing of notes.

I think people who sequence orchestral music really need to know what it's like to "perform" music.

I say, get up in front of 100 people and try singing/humming/whistling the melody of what you just sequenced, and you'll learn soon enough, that you need to shape these lines to keep people's interest.

This sequence was excruciatingly difficult to listen to. It makes you tense, because it doesn't ever relish any of the beauty of the lines.

Instead, try adding spaces, pauses, and such for dramatic effect. Make it sound conducted. Plus, in my opinion, Copland intended for this piece to be about 15% slower than this. Get into the "space". Feel the reverb. Slap that trumpet up on your keyboard and instead of listening to the notes, listen to the reverb. Every hall has it's own performance characteristics. Players drag notes, conductors sharpen phrases, so that it resonates best with the hall.

This is too much "sequence and music" and not enough "sound and compassion". Music is love and sharing. Forget the notes. Feel the spirit, don't hear the music.

If you want, you can email me, and I'll hum to you what a great performance sounds like. I used to play this work while I was 1st chair in the orchestra, and I have many recordings of the work.

I'd love to hear your next revision!

Sorry to be sharp ... I intend to work with you more on the next revision when you post it. I know you can do it!

Evan Evans

Hehhe! Well, there isn't much I can add to the previous two posts, except to say I agreed to listen to this, simply because I did it many months ago myself, using First Ed. and no perf tools.

On paper, Fanfare for the Common Man looks pretty straightforward, but to get the nuance that Copland wishes through punching notes into a sequencer using a mouse has got to be fairly difficult I would imagine. I wouldn't try that personally.

I have the same problem with brass parts in that I make them too perfect, thus not allowing the brass parts to breath (which if this were real players, they would soon run out of breath).

The only thing I would add, and noticed more than anything else, is the speed of it. It's way too quick with nothing like enough spaces between the phrases. The percussion, and this is just taste, could way more aggressive. But that's just me. I'm not saying that we should do mock-ups exactly like our favourite recordings btw.

Another quick thought:

It would be great to have more "light and shade" in the performance. It needs to build more; the climax of this piece is a triumphant celebration of brass timbre. There is nothing to celebrate if the timbre (and dynamic) is similar throughout.

How do you know something is loud, if you've never heard anything quiet...

William-
Thank you so much for taking the time to give a very detailed response - I really do appreciate it. I am not at all offended by this, I fully expected to be grilled by the people in this forum with very very high standards - this is good.
First, I agree about the matching of timbres - but I don't know how to fix this. I used the Opus 1 articulations: 'marc_4s or marc_3s' for all the marcato instruments, if available. Is this something I can fix, or an inconsistency with the samples?
Second, I agree that there needs to be more phrasing. I used some 17/16 bars, but need more breaths and pauses. I will try adding sixteenth note rests where breaths should be. I am a trumpet player, and have played this many times in brass ensembles in college and after college.
Third, agreed also. I will change some of the tenuto quarter notes to dotted-sixteenth notes and see if that helps. Great point!
Fourth, I also agree here there should be more innacuracy - or at least some. [:D] Is there a way to 'randomize' or humanize the performance if I export the midi file to cubase sx2?
I guess I am at the point where I know what I need to do, just not 'how' to, and need some tips on that. Preferably in Sibelius, but if not then easy things to do with the sequence once it's in Cubase sx2.
thanks William,
Mike

Evan-
Thanks... I guess for your post. I have been in Wind Ensembles and brass ensembles continuously throughout college and professionally for more than 16 years, and have conducted wind ensembles for the last 5 years, so I would not assume - as you do - that I don't know how to shape a line or that I don't know what 'a great performance sounds like'. I will pass on your offer to 'hum me a great performance' - you must be a virtuoso hummer! [:)] Anyways, aside from those very arrogant remarks, let's get to the substance of what you wrote:
I agree that I should add spaces, pauses and such for effect. I tried to do this with odd-metered bars, but I need more - perhaps one for each breath. About the tempo, that is a great point - I take the piece at mm. =66, but he marks mm. = 52, which I think is a tad slow and becomes a real 'chop buster' for the trumpets at that tempo. But, with samples that doesn't matter. [:D]
But, that was not my reason for going faster, it was becuase the perf_legatos would not sustain long enough to finish the line, and in one section, even had to creatively 'replace' a couple performance legato sustained notes with regular sustains, which is very hard to make un-noticable (I'm not sure if you noticed this or not.)
About your 'spirit of music' deal, I agree - but I need to do the best I can with sibelius 3 and giga. Is there anything I can do to help this in sibelius 3? I don't know if you are familiar with it, but in general is there anything that can be done to the original sequence, or would you just play the parts in live?
I will make another revision, and I hope you will still help me even though I did not appreciate a couple of your remarks. I really appreciate you taking the time to assist me in this learning process.
thanks,
Mike

Paul R.-
Can you post 'your' version of the Copland? I would love to hear it! About your comment on the 'punching notes into a sequencer' - like I said in the first post, I will be using all my samples with sibelius 3, so I kind of have to make the most out of that combo. I can use cubase sx2 for some editing, but I don't want to spend a whole bunch of time on it because I don't really have that much time for that.
How can I fix the breathing problem? Just add some rest where they should breathe?
I agree about the speed, but the samples would not last that long - see previous post.
Also, the percussion is at the highest velocity, and as agressive as they can get as far as I can see. I tried using London Orchestral Persussion, but the VSL has a way better sound, in my opinion - at least on the instruments used in this piece. How would I attain a more 'aggressive' sound (I did try to, BTW)?
thanks,
mvanbebber

@nicks@aubergine.co.uk said:

Another quick thought:

It would be great to have more "light and shade" in the performance. It needs to build more; the climax of this piece is a triumphant celebration of brass timbre. There is nothing to celebrate if the timbre (and dynamic) is similar throughout.

How do you know something is loud, if you've never heard anything quiet...

Nick,
Are you saying make the stuff at the beginning quiter, or the stuff at the end louder? Or are you saying to make the last part 'brighter' through EQ or something like that?
thanks,
mvanbebber

@Another User said:

But, that was not my reason for going faster, it was becuase the perf_legatos would not sustain long enough to finish the line, and in one section, even had to creatively 'replace' a couple performance legato sustained notes with regular sustains, which is very hard to make un-noticable (I'm not sure if you noticed this or not.)
About your 'spirit of music' deal, I agree - but I need to do the best I can with sibelius 3 and giga. Is there anything I can do to help this in sibelius 3? I don't know if you are familiar with it, but in general is there anything that can be done to the original sequence, or would you just play the parts in live?
I will make another revision, and I hope you will still help me even though I did not appreciate a couple of your remarks. I really appreciate you taking the time to assist me in this learning process. thanks, Mike

Ah, so this was done from a notation program. That would explain the lack of human performance element. Well for one, there are some ways to maximize the humanness from within Sibelius. There are some features. You might not know this, but even when you have turned on MOLTO ESPRESSIVO for an entire score, you must put expression text on each staff that you would like to play REEEEEALLY molto espressivo. Without it it only goes kind of halfway. So that will help that some. Finally, add some "a tempo", "rubato", "ritardandos", change of tempos, etc. And try turning on NOTES ELIGAS (I think that's how you spell it).

Evan Evans

Hello Mike,

I'm glad you weren't irritated by my remarks since this is a very difficult thing to do - really expressive sound, without a lot of notes, is actually harder by far to get a good MIDI sound with than something really fast with lots of notes. I encountered the same difficulty on a violin piece - the Allegro, which had fast notes everywhere, was much easier to do than the Adagio, which is childishly simple to play in notes but all espressivo. I still am not satisfied with the performance I did of that one.

I think you'll find that the inaccuracy and the breath marks will work wonders. Exaggerate them beyond what you think is right - it will be right.

I agree it is harder to do this with a notation program than with a sequencer - I play everything in by keyboard, so it is automatically very inaccurate [[;)]] However, if you are feeling the musical line, the inaccuracy can be expressive - for example, one instrument like a first trumpet getting to the high note just a little faster than the other trumpets and hanging onto it longer - as 1st trumpet players usually do being the prima donnas they are.

So you have some extra work, ironically, tweaking this from a notation program.

The slight cresc or decres are very important - like near the end there are some opportunities for these to build up to the climactic chords that are so spectacular there. You know this, and have heard it, being a player and conductor, but it is difficult to translate into midi - I know from personal experience. Anyway, it is not easy and good luck on a great project!

William,

Why did you delete your post regarding you playing the Fanfare and how it could breathe with the reverb more?

Evan Evans

Hi Mike,


if you really want to stick with Sibelius, then do not only follow the hints Evan gave but also add phrasing with tempo changes. I usually compose first in Sibelius and then use the sib file as a tempo reference to play in every part by hand (in cubase). And the score usually has lots slight tempo changes then. So when general tempo is 60, then there might 59 or 61 or sometimes 57 and so on to keep the tempo going with the phrasing. And added beats, pauses or fermatas certainly don´t hurt, too.
Then I use the sib file for the click track to play in the part by hands and select the appropriate articulation (can sometimes change note by note).

I understand your dream of connecting a virtual orchestra to a notation program. But if you want to have a mockup with a musical feel you won´t get away with a LOT of tweaking and effort.

Like our famous old comedian from Munich, Karl Valentin, once said: "Art is beautiful, but takes a lot of trouble..." (hm, the voice of this great guy is of course quite important to get the smile. Well, anyway.... [:D] )

Bests,
- Mathis

@mvanbebber said:

Can you post 'your' version of the Copland? I would love to hear it! About your comment on the 'punching notes into a sequencer' - like I said in the first post, I will be using all my samples with sibelius 3, so I kind of have to make the most out of that combo. I can use cubase sx2 for some editing, but I don't want to spend a whole bunch of time on it because I don't really have that much time for that.
How can I fix the breathing problem? Just add some rest where they should breathe?
mvanbebber

Yes you could hear it, but I have no way of posting it. And I haven't used the legato function.

Can't you play the notes into a sequencer via a keyboard, for inaccuracy? That's what Bill was alluding to. If every note of say a brass 'choir' chord starts more or less the same time, it sounds organ-ish.

In real life, say you have three trumpets, three french horns, two or three trombones and a tuba and their respectives notes all start on the same beat: what do you think that would look like if you were able to somehow show it as midi?
And these same instruments: Do they stop playing the notes of this chord all at the same time in real life, even if the notes on the score say they should?
Do they get louder at the same time, even if they're supposed to? With some brass samples, you may wish to start the note quite a way forward of where it's supoosed to start btw. Give it time to catch up with itself.

Thats what the others meant by breathing. Reverb of course. Lots of reverb for a piece like Fanfare. Loud piece of music. That's why Copland called it Fanfare.........

Try and imagine where the brass players would breath before they play the next note. Play little games with yourself. Like Ist trumpet is a big guy with big lungs - 2nd trumpet is a little broncial today - 3rd trumpet tries to outdo Ist trumpet, but is a heavy smoker - and so on. Even give them names! [[;)]] Write their 'names' on the score. Tell them off when they make a mistake. [:O]ops: Get a little insane.

The higher and louder a brass player plays, generally, the more 'rasp' there is to the note. Also, good reverb, like Altiverb or Waves helps with 'space'.

Regarding the percussion - if you are at the max on velocity - turn up the volume.

Lastly - why do we do other composers mock-ups? A friend of mine once told me when I was having to do a mock-up of some composer or other - why do we torture ourselves like this?

Good point, don't you think.

There's a book coming out - or is out by now by Paul Gil (something, I can't remember sorry) which looks good and may be of interest to you.

these are all good suggestions by Paul. Everything that can be done to make each individual instrument play a line by itself instead of just fill in notes of a chord is desirable.

By the way I have encountered that same problem with the ff timpani or bass drum. They are very rarely as agressively loud as they need to be, probably because when recording samples it seems a little crude and uncivilized to really beat the hell out of the instruments, though in a performance it is done all the time.

@William said:

By the way I have encountered that same problem with the ff timpani or bass drum. They are very rarely as agressively loud as they need to be, probably because when recording samples it seems a little crude and uncivilized to really beat the hell out of the instruments, though in a performance it is done all the time.

distort them. really. crank the marshall to 11.

@evanevans said:

William,

Why did you delete your post regarding you playing the Fanfare and how it could breathe with the reverb more?

Evan Evans

I didn't mean to delete it, I don't know what happened.
-mike

@PaulR said:

Lastly - why do we do other composers mock-ups? A friend of mine once told me when I was having to do a mock-up of some composer or other - why do we torture ourselves like this?

Good point, don't you think.

There's a book coming out - or is out by now by Paul Gil (something, I can't remember sorry) which looks good and may be of interest to you.

Paul-
Good question, but I have a good answer for you on why we do others' mockups. Here are just a few reasons:
1.) You can be sure the piece is well orchestrated and suited to each instrument it is written for. The piece is already considered to be great, therefore any problems with the mockup are your fault, not that of the composer.
2.) You can compare your version to previosly recorded 'real' versions considered 'great performances' and see how yours compares.
3.) You can learn to compose and orchestrate by literally copying other great composers in a much more thourough way than just simple transcription or score study. You are forced to realize the subtleties in each piece of music.
thanks,
Mike

Point 3 is a reasonable one.

Well Mathis I've done all sorts of things to "fake" a louder dynamic - simple volume, layering with bass drum, filters, etc. - but it would be good to have the actual samples of fff timpani for example because the timbre is so different. I remember trying to use some very good timpani samples at the end of a piece that had a fff timpani solo, and they sounded terrible even though they were excellent quality in general. So I got an old Roland S-50 sample floppy disc I had lying around the studio - it was not even multi-sample! - that happened to be an fff that sounded extremely "savage" and was perfect. Fortunately there were only two notes in the solo, so the stretching was not a problem.


Point 3 of this last post is a very good reason, because it's true you get a more "intimate" view of the music than any other kind of score study. The only thing better is orchestral playing or actually conducting it and hearing where all the problems are and how everything is put together. So MIDI is a tremendous learning tool in this sense especially when combined with such detailed samples as VSL.

@William said:

Point 3 of this last post is a very good reason, because it's true you get a more "intimate" view of the music than any other kind of score study. The only thing better is orchestral playing or actually conducting it and hearing where all the problems are and how everything is put together. So MIDI is a tremendous learning tool in this sense especially when combined with such detailed samples as VSL.

This is exactly right. Unless someone knows what an orchestra actually sounds like they will never be able to programme orchestral music successfully, even if they have the best sample library in the world. This is why many composers use orchestrators, as they can get a reasonable result with MIDI that has no chance of sounding good in the real world. Making a "real" orchestration sound good with MIDI is much more difficult, and therefore a very good learning process.

DG