# Topic: PianoConcerto Revised and Revisited

1. ## PianoConcerto Revised and Revisited

Concerto for Piano and Orchestra in D major

part 1 - Adagio, Allegro, Poco cantabile, Allegro
part 2 - Andante
part 3 - Vivace

I hope you won't lament and think "oh no, not again!" The last weeks I spent studying/experimenting a lot of papers about mastering, reverbs, tonal balance, orchestration and even frequency characteristics of strings (in particular violins), etc. These varied from pure acoustical stuff to mathematical calculations (I skipped the majority of those formulas) and X-Y grahps. I can recommend those things, it is very easy to find them on the Net, it only cost time to get to the right answer (among millions).

Thanks to Randy, who pointed me into the right direction using Sonar in a particular way, and the "new powerful" computer, I could set it up the way I thought right.

Still not having money for Altiverb, I looked for an alternative. There isn't any, so I dived into my CD collection to listen to older and more recent recordings of piano concertos, violin concertos and in particular listened to the sound of the strings, and the reverb time. On my system I tried to get that "same sound", even by copying parts of scores of those concertos. Finally I've found something that came rather close: the Auri Duculus method.

The story is not ended yet, but you are here in the Listening Room and NOT the Reading Room. To emphasize: I used in one go 6 occurences of Garritan Ambience (concerthall1), equalized the higher strings with three "octave based" different settings and in the final mix (with Audition) I applied another equalizing for enhancing the mid frequencies and to lower that harsh-sound a bit. Very modest compressing(1.5:1 above -10dB), only to get rid of the highest peaks (almost at 0 dB), giving me the possibility to amplify it to the (almost) max.

For those who already are familiar with this work, when you notice that it sounds different than before, you are right. I didn't add notes to the score, except for some corrections (at max. 20 notes). Everything was already there but never heard. Now I used other accents on groups of instruments and finally I can hear things I wrote down last year.

I hope you enjoy listening (again),

Raymond
(The piano is the Grand2 model1 Steinberg- where a +1-raise in velocity gives you another timbre. Strange!)

2. ## Re: PianoConcerto Revised and Revisited

Well you could at least have cut back on the mid-frequencies on those early reflections from the 2nd violins.

I'm kidding of course.

I never realized there was so much to reverb. I just pick a preset from the Garritan Ambience and slap it on the master bus. Or if I'm feeling sophisticated, have two buses, one for the near instruments and one for the far away instruments.

It's fine. The only thing I'd say -- since I take it you want us to listen real closely -- is that sometimes when a section gets louder, it sounds as though it's moved closer to us and not just played more loudly in the same location. For example, there are some pizzicato strings (I can't see the timing on your esnips player) where it sounds as though the string section jumped up and came closer to us and then went back to their places to play arco again.

But really, it sounds very good Raymond.

3. ## Re: PianoConcerto Revised and Revisited

Hello, Raymond!

Goosebumps - I actually got goosebumps (you know this expression? perhaps "goose flesh" as the English say?) while listening to the new recording of your Piano Concerto--!

I remember the previous version, and how you were aware that you wanted to be more successful with your aural realization of it--You wanted better reverb, and had hopes of unburying things that were getting lost.

I think you have More than succeeded. This is an entirely beautiful recording. What you've done with emulating the sound of natural ambience on the CDs in your collection, and the EQ adjustments, the touch of compression - all of that really worked. If I helped at all with the few technical things I passed on awhile back, well great, I'm certainly glad my humble contributions may have helped you.

Extremely rarely, there is some unnaturalness not in the recording but in the MIDI data, like around 6:40 of part one, there's a moment of machine-gun strings. It's OK and doesn't really matter. It's the trade off perhaps when recordings get so much closer to reality, as yours does, then small imperfections can stand out a bit more perhaps.

I know this has been a huge project for you for quite some time. I hope you have a deep feeling of satisfaction that it was time Very well spent.

Bravo to you, Raymond - And thanks for letting us hear the newly revived and exciting new version of your wonderful work!

Randy B.
(rbowser)

4. ## Re: PianoConcerto Revised and Revisited

Diligamus wrote:It's fine. The only thing I'd say -- since I take it you want us to listen real closely -- is that sometimes when a section gets louder, it sounds as though it's moved closer to us and not just played more loudly in the same location. For example, there are some pizzicato strings (I can't see the timing on your esnips player) where it sounds as though the string section jumped up and came closer to us and then went back to their places to play arco again.
I noticed that too and couldn't find a solution for it. For me the premise is that a reverb should sustain its original value irrespective of the volume (loudness) of the sound. Maybe I am wrong about this, but when the sound is low in volume the reverb dies out sooner (not audible) than with louder sounds. Maybe it is just the Garritan Ambience. Today I will experiment with some reverb settings to find out (and NOT presenting it again on this forum).

Randy wrote:Extremely rarely, there is some unnaturalness not in the recording but in the MIDI data, like around 6:40 of part one, there's a moment of machine-gun strings. It's OK and doesn't really matter. It's the trade off perhaps when recordings get so much closer to reality, as yours does, then small imperfections can stand out a bit more perhaps.
Randy I will have a look at it today. Maybe you are right. One thing I didn't solve were those piccolo trills, they weren't fully played (stuttered). On occasions Sonar can't get all the notes recalled from the libraries in time. I noticed that with the Grand. I had to amend the duration of the notes drastically to get all notes played. Don't know why and how. Latency playes a role in this, that is what I noticed, but the reason is going beyond my comprehension.

Both, thank you for the comments. Apart from those remarks, worth looking into it, I will start a new project. Now this "one year project" is finished (and I am fed up with it, really, but enjoyed doing it. It opened my ears).

And that Auri Duculus plugin is just a "Silly/Ridiculous Aural plugin" [haha] or crazy ears, mine. Auri=ears in latin , Ridiculus=latinized Ridiculous.
Raymond

5. ## Re: PianoConcerto Revised and Revisited

Nice composition, Raymond!

Long

6. ## Re: PianoConcerto Revised and Revisited

Originally Posted by Raymond62
On occasions Sonar can't get all the notes recalled from the libraries in time. I noticed that with the Grand. I had to amend the duration of the notes drastically to get all notes played. Don't know why and how. Latency playes a role in this, that is what I noticed, but the reason is going beyond my comprehension.
I've had these kinds of problems too, and the reason is also beyond my comprehension. Fixing it has generally involved changing the buffering under Options -> Audio. But I cannot discern any general rules. Sometime increasing the size works; sometime decreasing the size. Sometimes increasing the number of buffers; sometimes decreasing the number of buffers. And sometimes a setting that works for one project will have to be changed for another project.

7. ## Re: PianoConcerto Revised and Revisited

Hello again, Raymond

Even though a couple of issues have been mentioned on this thread, I hope you realize that your time on this intricate piece has been time very well spent - and that the recording is very good, and more successful than many things we hear in The Listening Room.

You and Diligamus have been talking about reverb and how it can seem to behave differently throughout a piece, depending on the varying volumes:

"...For me the premise is that a reverb should sustain its original value irrespective of the volume (loudness) of the sound. Maybe I am wrong about this, but when the sound is low in volume the reverb dies out sooner (not audible) than with louder sounds..."

But you're describing the natural nature of sound reverberating in a space. If an aural event is loud, there's more sound to bounce around in the venue, causing the multitude of echoes which is reverb. If something is quiet, there's not as much sound to bounce around - it will naturally cause less reverberation.

If you're unhappy with moments in a mix where the effect seems to be that instruments are temporarily moving closer, like the example given of strings going back and forth between Arco and Pizz - Don't Forget that you can automate your reverb sends in Sonar. You simply arm the Send knob, push the "record automation" button up in the transport control, and record moving the knob up to a level you want for a passage, and down again. That kind of correction is always available during the course of a piece.

I'm not understanding the other issue of some notes being cut off:

"... I had to amend the duration of the notes drastically to get all notes played..."

You have to be talking about live playback versus the recording, the rendering - ?

When I'm still working on a project, before it's been recorded, I have to accept the fact that my computer is under-powered to play back everything the way I'd want. When I'm using The Strad, if there's very much more happening in the project at the same time, its notes will be cut off as I listen back - BUT ONLY during live play back. Once I'm ready to try a mix, I export all the tracks, effects, automation etc. but NOT WITH QUICK BOUNCE and every detail is recorded as I intended. No notes are ever cut off that way.

Is that what you're talking about? You must be. -Yes, playing with buffer sizes can adjust things, but at least with the set up I'm working on, there's simply no way I can have my 60 MIDI tracks (average number) playing successfully in real time - I have to wait to hear an un-stuttered, un-truncated version for when I do my mix down. I'm very used to it, and it doesn't feel like a handicap at all.

Randy B.
(rbowser)

8. ## Re: PianoConcerto Revised and Revisited

Raymond, you said you worked a year on this project. I followed your struggles from a distance, as an interested listener and I can tell you that you really grew in the, I think, unexpected difficult job of composing and performing serious music. From disappointment to dispair you muddled through, but finaly you made it. Bravo. Your experiences till now are a good startingpoint for your next project.
Regards,
Gerard.

9. ## Re: PianoConcerto Revised and Revisited

GerardR, thank you for the comment. It was quite a challenge and there is material enough in it to make it "sing". One thing in life I can't stand is why can't I do it with the same material as another. Why can others give it the sound of the Concertgebouw Orchestra (to be honest nobody can, but you know what I mean) while I can't. So I went back to the "drawing table" and studied a lot, facing the problem "how to translate that text into computer processes". But for now, I am on the right track to make better recordings.

But still waiting for Altiverb (or some donations to get it - my bank account numbers will follow next time, I reply )

Randy, that stuttering sound is not only when playing back in real time with Sonar, but also when rendering. Some very fast passages with the Grand came out in chunks (after rendering). After having altered the durations af the successive notes this phenomenon was gone. For the piccolo I used in phrases very fast trills (half step and at other places whole steps). What ever I do with note length (durations), pause between notes or anything else two of those trills didn't get out as written, while the same trill one octave lower for the fluters and somewhere else sounded excellent.

Both with rendering and realtime listening. My system is quite powerful now, I have with this project:

• 36 midi instruments tracks
• 36 midi synth tracks (I always load the VST with all synth channels)
• 16 audio tracks which send reverb info to reverb buses (6)
• 6 send busses for reverbs
• 6 audio tracks to control the mix of dry and reverb
• 2 audio buses called Orchestra and Piano for a final adjustment before going to Master
• and I have 6 occurences of Garritan ambience
• 4 equalizing effects
To explain:
16 audio buses are
1. Flutes incl. Piccolo, Oboes, Clarinets, Bassoons sending to reverb bus WoodWinds;
2. French Horns, Trumpets, Tbones, Tuba sending to reverb bus Brass
3. Timpani and other "hit them hard, guys things" send to reverb Percussion
4. Violins1, Violins2 and Violas have their own reverb bus
5. Cellos and Cbasses a dedicated reverb bus
6. and at last the Piano
The output from the dry instruments and the reverbs come together in other audio buses: Mixed WoodWinds, Mixed Brass, Mixed Percussion, Mixed HighStrings, Mixed LowStrings, MixedPiano and each of them is directed to FinalOrchestra and FinalPiano combined to Master Output.

Now I can control with those sliders input volumes all over the process at any stage. By the way, I didn't touch the reverb level in automation (see the premise I made) I think I will experiment with it.

Clear?

Raymond

10. ## Re: PianoConcerto Revised and Revisited

I just checked my Sonar rendering settings. It has Fast Bounce checked.
I will uncheck this and see what happens. Anyway, whatever the outcome, thank you for commenting this piece of music again.

Raymond

#### Posting Permissions

• You may not post new threads
• You may not post replies
• You may not post attachments
• You may not edit your posts
•