Showdown - conical / cylindrical

[Calum Stewart]

For quite some time now, I've been playing and working with a cylindrical bored flute. A flute by Rudall Carte & co, simple system, 8 keys, in cocus wood made in 1933 (in "D" or "C" if you are that way inclined). I was drawn to this flute in the first instance, not because of the fact it was cylindrical bored, but because of the powerful and varied sound properties / tuning / lively and barky nature when pushed, coupled with a large colour palette and a connection I felt with it. The fact that it was cylindrical rather than conical, didn't even come into the equation and didn't pose any issues with me. So I'm writing this, to kind of nail my colours to the mast, so to speak, as I feel there is a lot of poppycock, being drawn out time and time again on this subject of bore, and I aim to address this now. In addition to that, I'm going to make it all jolly exciting but inviting you all to listen and judge for yourself - Using sound samples. Huzzah I hear you cry!

I realise by doing this, I am fully prepared to be proven wrong. So this makes it so exciting for me too! So my goal in this particular test, is to prove that bore type (between a wooden pratten type model, and wooden cylindrical model) makes no difference whatsoever to the sound of a flute. I am of course referring to our beloved wooden flutes, simple system instruments - not modern silver, boehm keywork flutes. I could indeed include Rudall and Rose models here too, but I think my point can be proven quite clearly using a pratten model - as this type of flute is generally thought of as delivering "the" sound.

I hope to prove this, by recording sound samples of my Rudall Carte wooden flute (cylindrical bore), and sound samples of a conical bore flute based heavily on Pratten model. I cannot say the maker of this flute, neither if this flute is by a modern maker or original - because it doesn't come into the equation - Suffice to say, this conical flute is a very very fine example of typical concert pitch, large holed instrument, based on Pratten model. Both these flutes are in concert pitch 440.

Personally, I've always judged each flute on its own merit; With regard to maker, wood, model, and also in this instance bore type. Sometimes, Ive been very surprised by what Ive thought of as a good flute. Famous makers, who's flutes didn't quite cut the mustard for me, and also unknown makers or very unusual looking flutes / unlikely set ups (such as metal heads, or different bore systems, or key layout, etc ..) which totally kick ass.

I've never been hung up on trends personally, and don't tend to take notice of "flute fashion". In fact, quite the opposite - I believe strongly in the importance of individual sound, and finding your own way and having a connection with the timber in your hands. I believe there is far greater difference in sound, that comes from players own variables such as embouchure, than the differences we tend to discuss here in flute aesthetics. (At least what I'm attempting to prove as aesthetics).

So to finish off - I will make a poll (either a new thread, or as part of this one), in which I will ask the following 2 questions: 1) which flute do you think is conical and which cylinder? 2) which sounds better?

MODs is it possible to have a "blind" poll, so no one sees the ongoing results until the end? I think that is fair, so people really do rely on their own ears. It could be great too, if people took part in this, but didn't air their opinions until the end. That way everyone makes their own mind up, and no one is influenced by other members on the board.

I am quite prepared to be proven wrong, and really hope that everyone will take part in this poll. (Or forever hold your peace!). Both these flutes are great instruments and I will play them both with 100 percent. I won't give playing preference to one or the other, because actually I am quite willing to being proven wrong!

If anyone has any suggestions in which I can make this poll and recording test clearer / fairer, please do say.

I will record both flutes during the next week or so (when I can find time to set up my little portable home studio), on the same day, in the exact same conditions, using the exact same unbiased mic technique, playing the exact same tunes.

Game on...

[Nanohedron]

MODs is it possible to have a "blind" poll, so no one sees the ongoing results until the end? I think that is fair, so people really do rely on their own ears.

I don't think we have that option, unfortunately. But I'll double-check. Not a bad idea at all.

Good idea for a thread, too. You're so impassioned, I already believe you.

It could be great too, if people took part in this, but didn't air their opinions until the end. That way everyone makes their own mind up, and no one is influenced by other members on the board.

Good luck with that, I guess. I already had a comment or two myself, but ... will ... not ... post ... argh ...

You still have the option of turning this topic into a poll; after 72 hrs a mod will have to do that for you.

[Akiba]

Sounds great. I did something similar a while back playing the same tune on a Murray, Olwell and Casey Burns. Definitely there were tonal differences and of course great differences of opinion. I expect the same here.

The one factor I'm pondering lately is the flute's inherent "overtone" profile in the overall sound. I believe I read that Michael Grinter commented favorably on hearing another maker's flute, but that it did not have the same overtones as his flutes. I wonder how/if those overtones do make the Grinter sound vs. the Olwell sound--I think they might.

Thanks in advance for doing this, Calum. I look forward to hearing the clips.

[Nanohedron]

Okay, looks like one can't hide poll results here. We'll just have to watch the mounting numbers much as if watching a horserace, with all the attendant tension, sweep, and drama .

[I.D.10-t]

If you're logged in it doesn't immediately show the poll results, but one can still peek at them.

Hope the tune goes well into the second register.

[jemtheflute]

Having a little prior experience of this type of thing (both a 3 head, 2 body trial I posted myself a while back and also stuff Calum has sent me privately in the past when trying out flutes, for me to make guesses and a fool of myself!), I suspect that, as Calum well knows, all that'll really happen is he will show how much he sounds like himself whatever he is playing.... But we'll have some fun along the way.

[Terry McGee]

Sounds good, Calum. I'm all for exploding preconceptions, even if they may be mine! (Of course, I prefer to call them facts...)

Can you include a few long notes in isolation on each flute? A second or so on each note is enough, and say D and G in both octaves, or ideally the scale from Low D to 2nd octave B, to facilitate the picking of nits.

I'll also be interested to see what spectral analysis of each flute tells us. If you fear that that might give the game away, you could either send them separately to me (and to anyone else who asks), or perhaps record and provide them in a different sequence to the tunes, with a different identification system only known to you (eg flutes a and b for the tunes, and flutes x and y for the tones.) I'm happy to sit on the results until you give the all-clear. It will be interesting to see if there is easy correlation between human perception and clinical analysis.

In terms of methodology, mic position is very important. Too distant and you get too much room reverb, too close and position becomes critical. Probably good to test the recordings on some friends first, in case they can point out weaknesses you need to address. I'm inclined to think that slow tunes offer the listener more chance to judge tone than fast ones, but it might be that it isn't just tone that matters - it might be agility too. Perhaps a fast passage and a slow passage?

If you feel the recordings are hard to distinguish between, you could also try this approach. You provide three recordings, a, b and c. The question you ask is "does b sound more like a or c?"

I experienced this approach back in the early days of CODEC research for transmission of digitised sound. There were 6 CODECs under test and it was conducted by Expert Groups in each country, and a vast number of countries were involved. I foolishly agreed to be one of their listeners in the Australian group. They were conducted in a sound lab, and with both headphones and speakers. They threw everything at us - harpsichord, human voice, string quartets, popular music, country, rock, folk, narration, electronic music, sound effects, etc etc. I was there for over an hour. The operators had no control over what I heard, it came from a computerised random source, with randomly repeated tests to check for guessing and acoustic fatigue. The only given was that b was always the uncompressed original. I couldn't see the operators, I'm not sure if they could see me. It was scary, because at first, I couldn't tell the difference between any of the three samples. But a reassuring voice over the speakers said, don't worry, you're probably just hearing some of the CODECS that offer very minor compression, but also therefore have big bandwidth requirements. You'll soon hear something obvious. Sure enough, a few sets later, there was something really ghastly, and I was able to settle down to the task without the sense of rising panic!

I found incidentally, that the human speaking voice was a surprisingly telling test. Surprising to the extent that the bandwidth requirements of the human voice are not that great (or the telephone might not have been invented quite so long ago!). But of course, we humans are expert at pulling nuances out of the human speaking voice. Will flute players prove to be experts at pulling nuances out of flute recordings?

Terry

[Calum Stewart]

Perhaps we could use something like "survey monkey"? Which is external to the board, but seems to be "blind" until the poll is finished. Of course, to make sure no funny business is going on, perhaps a 3rd party could set that up?

I forgot to mention that both flutes are made of cocus. So it really is the bore which separates them.

Good information there Terry. I'd be happy to do some long notes, and perhaps more in depth studies afterwards. That would be very interesting when analyzing the final result. But at first, I'm going to do this playing music, slow and fast tunes in 2 octaves of the flute.

I also want to add, that I've been spending a lot of time with both flutes... I don't feel particularly at an advantage technically or "better" on one more than the other.

@Jem, of course I will sound like me (I hope!) but I still think there is a clear point to be made.

Does anyone have any particular tunes that they think will work best? Perhaps a couple of session favorites or really well known tunes? I think I'll do aout 2 or 3 minutes or audio for each flute. 1 or 2 times through 2 or 3 tunes.

It could be really great if all kinds of members of the board joined in with this. Professionals who lurk silently, beginners and amateurs of what ever level. The important thing is we get as many people as possible to joint in. So that I can prove you all wrong.... Or else you can all have a big laugh when it's me that is wrong (and view subsequent eBay post for my RC !!! )

[I.D.10-t]

I just wish you had a low D whistle thrown in for fun.

eskins use to have a web page where it seemed that he played the same tune on every piccolo, whistle, and flute that he laid his hands on. It was kind of nice having an apples to apples demonstration of the different instraments.

[Lars Larry Mór Mott]

How about 'Trim the velvet' for a tune suggestion?

[radcliff]

does your flute head Mancke fits a standard bohem flute too? I wonder if such a flute would sound so different in this kind of experiment...

[jemtheflute]

Calum, to keep the poll secret you could quite simply set up a standard form of answer in a post and ask that people copy and paste it into a PM or e-mail (latter better given PM storage limits here), edit in their responses and send it to you. Of course, you'd then have to analyse the resulting data and present it back, but I doubt you'll get more than dozens of participants at best. You could also simply copy and paste any narrative comments from responders into the public results thread too.

[benhall.1]

Calum, to keep the poll secret you could quite simply set up a standard form of answer in a post and ask that people copy and paste it into a PM or e-mail (latter better given PM storage limits here), edit in their responses and send it to you. Of course, you'd then have to analyse the resulting data and present it back, but I doubt you'll get more than dozens of participants at best. You could also simply copy and paste any narrative comments from responders into the public results thread too.

Yes, but awkward people ... can't think who ... might still post their opinion on the thread, and skew the results in that way.

[Denny]

yes, well, awkward people with opinions...

[MTGuru]

Yes, but awkward people ... can't think who ... might still post their opinion on the thread, and skew the results in that way.

There is something we could do about that. After all, we just love deleting stuff. It makes us feel evil and omnipotent. Bwaa-ha-ha!!!