AlexanderSchiborr
Senior Member
Last edited:
Dance of the Reed Flutes - Tschaikowsky - Close enough to live?
- Thread starter AlexanderSchiborr
- Start date
re-peat
Senior Member
I don’t know how many flutes you loaded for this arrangement, Alexander, but it sounds like a lot more than what the score prescribes. Not saying there are a lot more, just saying that it sounds as if there are a lot more.
Also, those startled shrieks from the flutes at 0’18” and 1’00 are — unintentionally, I assume — *very* funny. Very, very funny indeed. I am sorry, but they are. Those two moments literally had me laughing out loud (never had that happen before, actually) and I instantly rewinded the music to hear them again. That’s the sound nymphs make when caught by surprise while bathing in the river, isn’t it?
And another thing: with those deep, throbbing pizzicati, you forfeit every possible chance, right from the start, of ever ending up with an elegant, airy, light-footed dance. These pizzicati are way too heavy, dark and booming, to my ears, for this music. More the sort of sound you expect to hear when Colonel Hathi and his troupe of marching elephants are parading through the jungle. Not the merrily skipping sound above which the flutes can dance like butterflies with carefree abandon.
There are also a couple of very slight performance issues — notes not quite in sync with the rhythm — and I don’t know whether that’s the samples (inconsistent attack times of the staccato’s, which one comes across a lot in sample libraries) or the programming.
Despite all of the above, I did enjoy it. Lovely music, lovingly recreated.
But if it were mine, I would do some more work on it.
_
Also, those startled shrieks from the flutes at 0’18” and 1’00 are — unintentionally, I assume — *very* funny. Very, very funny indeed. I am sorry, but they are. Those two moments literally had me laughing out loud (never had that happen before, actually) and I instantly rewinded the music to hear them again. That’s the sound nymphs make when caught by surprise while bathing in the river, isn’t it?
And another thing: with those deep, throbbing pizzicati, you forfeit every possible chance, right from the start, of ever ending up with an elegant, airy, light-footed dance. These pizzicati are way too heavy, dark and booming, to my ears, for this music. More the sort of sound you expect to hear when Colonel Hathi and his troupe of marching elephants are parading through the jungle. Not the merrily skipping sound above which the flutes can dance like butterflies with carefree abandon.
There are also a couple of very slight performance issues — notes not quite in sync with the rhythm — and I don’t know whether that’s the samples (inconsistent attack times of the staccato’s, which one comes across a lot in sample libraries) or the programming.
Despite all of the above, I did enjoy it. Lovely music, lovingly recreated.
But if it were mine, I would do some more work on it.
_
Last edited:
AlexanderSchiborr
Senior Member
Well, thank you piet for the feedback. And yes it is a work in progress, still I thought to share it, just to get also some feedback prior finalizing in order to collect informations what I can improve in that track still. I used that reference here in particular this one:
I used actually 3 flutes here. But let me see what I can improve. And that funny moment at 18 seconds is where I tried to mimic the performance of reference, I really to re-create everything considering what samples can do or don´t, most of the time they don´t do it, so there is a lot of programming time and customizing patches ongoing.. Now this is not easy as you know also what sampled Woodwinds sound like. A forte, but yet very musical expressive long note with a molto vibrato performance..is not what sampled flutes do freely and even not with some violence.
I appreciate your feedback, thanks a lot.
Last edited:
muk
Senior Member
As all of your work this mockup shows the exremely high level of your technical and musical prowess. If I am allowed to nitpick: I agree with all of Piet's comments. The flutes don't sound right to me either. Maybe you can achieve a better blend there. Then, for my taste that counterline in the bassoon could be a smidgen more in the foreground. A bit too much reverb on the bassoon for me as well. The cor anglais-samples are letting you down in my opinion. To my ears it sounds clunky and not like a musician performing a musical phrase. I have a feeling that it isn't due to your mockup, but the samples themselves.
Finally, there are places where I think the tempo variations are not right. For example from 0:26 o 0:29 the first beats of the bar are slightly faster, immediately followed by a slight rallentando. That doesn't sound musical to me - it sounds like an overzealous attempt at speeding up, just to be immediately hypercorrected. There are a few more spots like that. 0:43-044 is another one. The flutes are too fast here, and then too slow in an abrupt manner. I assume that you tempo-mapped to the video you linked to. If so, I guess that in these places the mapping is not precise enough. It doesn't need much variation - just a sixteenth note off, and it doesn't feel right anymore.
In any case you have set yourself a difficult task, and solved it beautifully. Even if my comments may sound very critical, it's a very accomplished mockup.
Finally, there are places where I think the tempo variations are not right. For example from 0:26 o 0:29 the first beats of the bar are slightly faster, immediately followed by a slight rallentando. That doesn't sound musical to me - it sounds like an overzealous attempt at speeding up, just to be immediately hypercorrected. There are a few more spots like that. 0:43-044 is another one. The flutes are too fast here, and then too slow in an abrupt manner. I assume that you tempo-mapped to the video you linked to. If so, I guess that in these places the mapping is not precise enough. It doesn't need much variation - just a sixteenth note off, and it doesn't feel right anymore.
In any case you have set yourself a difficult task, and solved it beautifully. Even if my comments may sound very critical, it's a very accomplished mockup.
AlexanderSchiborr
Senior Member
Sure you are totally allowed, it was the first reason I posted it because I felt to get some good tips during the stage of creation from you guys as this helps quite a lot. You know when you sit over hours and hours for a couple of days with that piece and doing constanly a/b your ears get that fatigue that I simply can´t nail everything there. Trying to inject some life into sample Woodwinds is imo a very difficult task, so thats a good thing you guys point me to some spots so that I can work on them in order finalizing the piece.
Tapping the tempo is quite a challenge because it needs to be tapped by each quarter bar, sometimes even on the 8th notes which makes it also very challenging.
Then I am battling with the slight attack fluctuation imperfection what the samples have which is quite a very insane programming routine because you need to move sometimes each triad manually so that the whole chord and balance of the chord is right. I tried to accomodate more realism here so that prior beginn I created custom patches which are beyond the berlin Wind standard patches to surpass their sloppyness and expression limitations which is not evident when you write lets say normal music but mocking up this stuff exactly makes necessary to nail every aspect of the performance of the players.
The dying duck: I know that the english horn isn´t quite succesful which has to do with expression (which believe it or not) was treated with care but the slight off tuning (not in a pleasant way on certain notes) which is just a reason of bad sample source which I will try to work around by pitching single notes or combining different english horns for that one line...is there anything in sample world out there which sounds good? I don´t know. I guess pulling of musical soloing with an english horn is very specific for the sample world, so I am not sure.
Thanks again for the tips.
Guy Bacos
Senior Member
Hi Alex,
Nice work and I agree with Muk about your usual high level of production, especially with your great compositions. Classical pieces, as we know, are always a bigger challenge, but we have to be honest if that's what you want to go for. To my ears the flutes sounds semi staccato combined with a bit too much reverb makes them sound a bit sloppy, sounds almost like there's some sustain, needs to be spicier. I would go for staccatissimo to make up for this, we need to hear those notes more distinct.
At 8s and 18s, the accent sounds harsh, I know those notes are accentuated in the live version, but if you could make it sound less harsh, I think would make it smoother.
I'm looking forward to hearing updates.
Nice work and I agree with Muk about your usual high level of production, especially with your great compositions. Classical pieces, as we know, are always a bigger challenge, but we have to be honest if that's what you want to go for. To my ears the flutes sounds semi staccato combined with a bit too much reverb makes them sound a bit sloppy, sounds almost like there's some sustain, needs to be spicier. I would go for staccatissimo to make up for this, we need to hear those notes more distinct.
At 8s and 18s, the accent sounds harsh, I know those notes are accentuated in the live version, but if you could make it sound less harsh, I think would make it smoother.
I'm looking forward to hearing updates.
AlexanderSchiborr
Senior Member
Hey Guy,
Thanks for chiming in and your interest taking time to comment. I actually think that this is somehow a thin line. These short articulations are all kinds of different ones when you listen closely. There are definitely spots where the flutes do those staccatissimo but also slight longer notes which is the point: You simply can´t just use a staccatissimo patch here throughout the whole parts, because real flutes do all this sorts of slight different note lenghts variations, even though they are not so obviously first, but when you listen very closely you can hear how they vary in note lengths and attacks. What I can try is using for certain parts is a staccatissimo patch though and try out if I can get there a bit more of what you are saying. Thanks for the pointer. Yes the 8s and 18s mark can be a bit tamed down by dynamic Eq with a notch at 5K, agreed here.
Guy Bacos
Senior Member
Just to be more specific, real flutes would just need to play it staccato but because of the nature of the samples I'm hearing, or due to your reverb setting, I would shorten the staccatos enough so the notes don't sound connected to each other which is really what bothers me the most, this overlap, makes it a bit fake considering this is a passage we know soooooo well. 
Last edited:
re-peat
Senior Member
Alexander,
Intrigued as I was by that odd phenomenon of 3 Flutes sounding like a herd of 30 — in a manner of speaking of course — I thought I’d have a quick-ish stab at it myself, also with instruments from the Berlin Legacy Woodwinds, and https://users.telenet.be/re-peat/Nutcracker_Mirlitons_3Flutes_Test.mp3 (this) is what I came up with.
I must add immediately that I cheated as far as Flute 2 is concerned. After some dissatisfying experimenting with the Berlin Flute 2, I decided to replace it with the SWAM Flute and I edited the instrument (inside SWAM) to produce a noticeably thinner sound than usual.
Flute 3 (which is the Berlin Flute 3) was also made considerably thinner but in this case with the aid of an EQ (creating a rather steep valley of some -9dB around 450-500Hz)
The 1st Flute is the Berlin Flute 1, unaltered straight out of Kontakt, no EQ, no reverb on it, no processing whatsoever.
I also took Guy’s perceptive words into consideration and tried to give the performance a more pronounced staccato feel. But I think I may have gone a bit too far with that. It sounds just a fraction too staccato at the moment, for my taste.
Anyway, performance issues aside, I do think the thinning out of the 2nd and 3rd Flutes does help a bit in keeping the threesome sound like a threesome, no?
(The bassoon is the one from Donnie Christian. Very, very old library, but for certain things still unsurpassed, I find. The pizzicati are Spitfires. I didn’t attempt to make a correct orchestral mix, it ended up sounding more like a chamber orchestra version of the 'Dance of the Mirlitons'. But that’s not important.)
_
Intrigued as I was by that odd phenomenon of 3 Flutes sounding like a herd of 30 — in a manner of speaking of course — I thought I’d have a quick-ish stab at it myself, also with instruments from the Berlin Legacy Woodwinds, and https://users.telenet.be/re-peat/Nutcracker_Mirlitons_3Flutes_Test.mp3 (this) is what I came up with.
I must add immediately that I cheated as far as Flute 2 is concerned. After some dissatisfying experimenting with the Berlin Flute 2, I decided to replace it with the SWAM Flute and I edited the instrument (inside SWAM) to produce a noticeably thinner sound than usual.
Flute 3 (which is the Berlin Flute 3) was also made considerably thinner but in this case with the aid of an EQ (creating a rather steep valley of some -9dB around 450-500Hz)
The 1st Flute is the Berlin Flute 1, unaltered straight out of Kontakt, no EQ, no reverb on it, no processing whatsoever.
I also took Guy’s perceptive words into consideration and tried to give the performance a more pronounced staccato feel. But I think I may have gone a bit too far with that. It sounds just a fraction too staccato at the moment, for my taste.
Anyway, performance issues aside, I do think the thinning out of the 2nd and 3rd Flutes does help a bit in keeping the threesome sound like a threesome, no?
(The bassoon is the one from Donnie Christian. Very, very old library, but for certain things still unsurpassed, I find. The pizzicati are Spitfires. I didn’t attempt to make a correct orchestral mix, it ended up sounding more like a chamber orchestra version of the 'Dance of the Mirlitons'. But that’s not important.)
_
synergy543
Senior Member
I hesitate to comment as I have tremendous respect for the creators and commentators of both valiant attempts but there's something about the resonance that is excruciatingly piercing in both examples and I don't hear this in ANY of the real instrument performances. I don't know the answer although my guess is that there is something in the filtering of the air, or the depth that "softens" the resonances of a real instrument that is not happening with the samples nor the room emulators that are being used in these examples. Its almost as if they need some slight dynamic compression? I first noticed this effect at a live outdoor concert with prominent brass instruments and I had seats up front. I was expecting harsh resonances from the instruments being so close, but I was surprised at how smooth the instruments sounded even though they were quite loud. There was a very surprising "dynamic smoothness" from these real instruments in the open air. Well, as I said, I don't have the answer but maybe its something to think about? On second thought, maybe try experimenting with the new dynamic capability of Fabfilter Q3 to emulate this softening effect on the sharp resonances? I haven't tried it, but it might help.
Last edited:
re-peat
Senior Member
Thanks, Greg. You mean these flutes sound more like oversized-piccolos-with-a-hint-of-dentists’-drill-in-their-high-register, instead of regular nice concert flutes? Perhaps you’re right. It did escape my attention last night I must admit, but in my defense, I made this excercise between 01h and 03h in the morning to relax a bit after a few depressing hours discussing other sample libraries.
A big part of the problem begins with the samples themselves of course. Play in the C5 octave with just about any sampled flute and you’re very likely in trouble, soundwise. And it only gets more troublesome at higher dynamics. As you say: excruciatingly piercing.
Also, there’s only so much you can do with an EQ to addres the issue, especially where it concerns the flute, because tame that piercing shrillness too drastically, and you’ve also lost much of the essential airy and fluttery quality of the instrument. It’s no problem to make dull flutes that don’t sound piercing at all, but then you have dull flutes.
And if we limit ourselves to the Berlin Legacy Flute1, there very little that can be done, without completely ruining the flutyness, about that high G5 at 0’18”, I’m afraid. That’s a pretty merciless note no matter what you do. (And I'm still very dissatisfied with that moment.)
Anyway, I put the Waves F6 (also an excellent dynamic EQ) to work in this https://users.telenet.be/re-peat/Nutcracker_Mirlitons_3Flutes_Test_v2.mp3 (new mix), rather aggressively reducing several narrow bands between and around 2000Hz and 3000Hz (which, to my ears at least, is where much of the piercing seems to happening) whenever it exceeded a fairly lowly set treshold. Better?
_
A big part of the problem begins with the samples themselves of course. Play in the C5 octave with just about any sampled flute and you’re very likely in trouble, soundwise. And it only gets more troublesome at higher dynamics. As you say: excruciatingly piercing.
Also, there’s only so much you can do with an EQ to addres the issue, especially where it concerns the flute, because tame that piercing shrillness too drastically, and you’ve also lost much of the essential airy and fluttery quality of the instrument. It’s no problem to make dull flutes that don’t sound piercing at all, but then you have dull flutes.
And if we limit ourselves to the Berlin Legacy Flute1, there very little that can be done, without completely ruining the flutyness, about that high G5 at 0’18”, I’m afraid. That’s a pretty merciless note no matter what you do. (And I'm still very dissatisfied with that moment.)
Anyway, I put the Waves F6 (also an excellent dynamic EQ) to work in this https://users.telenet.be/re-peat/Nutcracker_Mirlitons_3Flutes_Test_v2.mp3 (new mix), rather aggressively reducing several narrow bands between and around 2000Hz and 3000Hz (which, to my ears at least, is where much of the piercing seems to happening) whenever it exceeded a fairly lowly set treshold. Better?
_
AlexanderSchiborr
Senior Member
I am uploading an update later today with a complete run through..
synergy543
Senior Member
Yes Piet, that is a great improvement and even at a fairly high monitoring volume too! As you say, it does affect some of the airyness of the flute tone so you could even pull back on the EQ level possibly. Especially if you were to give a bit more depth to the instruments. Nothing wrong with your chamber style version though this piece would typically be a but further away on a stage or in a pit. And I think the additional reverb might possibly help smooth things out even further.I'm curious to know how much dynamic effect you added (although the dynamics don't significantly here do they). This idea clearly works spectacularly so I'll have to experiment around with it at different volumes as it appears like a great way to tame a difficult problem. Thanks for making these demos as they're very enlightening.
AlexanderSchiborr
Senior Member
Here is an update, both versions, probably the best to dl them both and listen on either good speakers of headphones. I tried to a few things out coming from the several comments from you guys. And yes reducing too much of the airness in the center pinch zone takes out some life, I experiemented with that too, but you need to be carefully, otherwise you end up missing the certain overtone balance when the flutes play in their lower register also.
(broken link removed)
more improvements the next days then..
(broken link removed)
more improvements the next days then..
Last edited:
AlexanderSchiborr
Senior Member
I like the shorts, not quite sure about some spots like at 17 - 19 seconds though, also that high note..can you dial in more expression vibrato with more a kind diverse vibrato rate? Also the spot before (first time)I think the triadic balance is not quite on spot with flute 2, and 3 need more impact so that the triadic harmony is more obvious with all the 3 flutes.
Thanks for sharing.
Guy Bacos
Senior Member
I'm curious to know how much dynamic effect you added (although the dynamics don't significantly here do they). This idea clearly works spectacularly so I'll have to experiment around with it at different volumes as it appears like a great way to tame a difficult problem. Thanks for making these demos as they're very enlightening.
Yes, it is an improvement but at the same time I would of sworn these were alto flutes.
synergy543
Senior Member
Well Guy, there is supposed to be an alto in the score for Fl.3, so you heard correctly!
@Alexander - Nice improvements. Sometimes though I think the vibrato you use is too strong and in the later part (particularly with the E.Hn) you could still shape the arc of the phrasing a bit more. This last part is hard to explain, I think you either sense it or not but I think its something performers develop over time. Its the subject that is most often explained in masterclasses by performers such as Vengerov (below - though the best masterclasse ex. was taken down by YT). Of course, with samples we only have volume to work with so you can't as easily shape tone (though through selection of dynamic layers or filtering, you slightly can). So you work with what you've got, but I think there's a little room for some more expression at a few spots (think as a conductor of the arcs he might want to make here.)
Guy Bacos
Senior Member
No, I meant it sounds like 3 alto flutes playing.
synergy543
Senior Member
I kinda knew you were gonna say that.
However, its just one in the score. Who knows though, maybe Piet was trying getting as far away from the piccs as possible.
Although I think more likely the real reason is because the sound is so upfront in his example (but I could be wrong). Still, it wouldn't be a bad thing, as we're just illusionist trying to create a believable prestige.
