It is currently Mon Apr 06, 2020 9:47 am

All times are UTC - 6 hours




Post new topic Reply to topic  [ 145 posts ]  Go to page Previous  1 ... 6, 7, 8, 9, 10
Author Message
 
 Post subject: Re: Rockstro
PostPosted: Sat Feb 29, 2020 1:59 am 
Offline
User avatar

Joined: Sat Nov 03, 2007 7:19 pm
Posts: 531
Location: Hood River, Oregon, USA
Yes, I've read your reports and postings about "thread strangulation". I think the conclusions you draw are not fully supported
by your data. I don't doubt that tenon compression is a real phenomenon. People have known, and written, about that for centuries.
Lots of methods for recognizing, dealing with, and avoiding that issue are well known.

You keep posting that Rudall tuning graph, claiming that it shows some tuning anomalies, and then you allude to the fact that they
are due to thread strangulation. I think your posts are misleading for the following reasons.

First, the most obvious issue with that tuning graph is that the flute in question has its slide closed to the point that it is being
played at something like British high pitch (A=452+), but the flute was obviously not designed and tuned to be in tune with itself
at that pitch. This is why the overall trend in the tuning climbs when going from left to right in the graph. Graphing the tuning of
a flute when its slide is not set to the point where the flute is best in tune with itself has the effect of
artificially amplifying some of the alleged tuning anomalies. Had the tuning slide been pulled out to the point where
the flute was best in tune with itself, the line would not have such an obvious overall slope, and most of the points would be
closer to the horizontal axis.

Second, the line would still have peaks and troughs, but these are at least as much an artifact of the choice of Equal Temperament
and Meantone as the reference as they are to the flute's tuning. This is not evidence of thread strangulation, it is mostly
evidence that you probably have not selected the same tuning target as the maker. You could, for the sake of example, have chosen
to plot the line against Just Intonation based on C. If you did that it would look a lot smoother and more level.

These flutes were designed to play music in a variety of key signatures, and across three octaves, in a time period when the
A=440 tuning standard did not exist. The makers had a wide choice of options for how to target the tuning of their flutes, which key
signatures to prioritize, etc. They also had to balance a lot of competing needs. There is lots of discussion of this in the literature.
This page at FluteHistory.com might be a good place to start.

Anyway, the point is that showing a single octave of tuning relative to a modern, and not particularly meaningful tuning standard for
those flutes, just obfuscates a lot of important issues.

Finally, the tuning of the keyed notes is greatly affected by issues such as the thickness of the pads, the amount they protrude into
the tone hole, the height setting of the key when open, and the thickness of the cork under the key touch.

I know that you already know most of this, so I find it disappointing that you would choose to use a graph like this to try to convince
people that this graph is some kind of scientific proof that your thread strangulation theories are valid, and that they explain the tuning
characteristics of antique flutes. It is especially disappointing to see you claim this after having argued persistently in this same thread
that bore perturbations at the tenons, in the form of chambers, couldn't possibly have any effect on tuning.


Top
 Profile  
Reply with quote  
 
 Post subject: Re: Rockstro
PostPosted: Sat Feb 29, 2020 6:44 am 
Offline
User avatar

Joined: Sun Dec 12, 2004 4:12 pm
Posts: 2179
Location: Malua Bay, on the NSW Nature Coast
I'm not arguing that Rockstro and Ellis' data is an indication of thread strangulation, although I think it is a distinct possibility. I'm simply pointing out that tuning on flutes in that era was seriously wonky. Unacceptably wonky. And not easily explained by resort to period or earlier tuning schemes.

And the choice of slide extension is of course Rockstro and Ellis' - you can hardly blame me for a decision taken in the second half 19th century! I do think that Rockstro's doggedness that English pitch had been always high is made fun of by his own collected data. It's yet another affirmation of Rockstro's inability to tell the truth. After Welch's forensic dismemberment, why would we believe anything this man said?

But even if we "pull the slide out" post factum, which we can do mathematically, we still have a major disconnect between foot and body to deal with. And between LH and RH. No tuning systems can address those disconnects. Something is up. Perhaps it's thread constriction. Perhaps Quantz was right - the C foot is a disaster. Whatever, something is up. And we need to understand what. Otherwise we are deluding ourselves. And others.

Blind faith in the genius of 19th century flutemaking is unsupported. If these guys were so good, and their flutes have not suffered subsequent damage, then their flutes should still be exemplary. But they are not. So which is it? Not so good, or subsequent distortion? Or both?

A very interesting detail is that the bore of the head was measured at 18.8mm, as was the starting bore of the LH section. We rarely see the top of the LH so large - check my chart of the first 70mm of bores about two posts up. Only two of those flutes look like they could fit that criterion. And remember the flute you advanced, #3801, whose LH bore starts at 18.3mm. Assumes ghastly Bing Crosby Christmas song voice: "It's beginning to look a lot like serial strangulation..."

Paddler: "You could, for the sake of example, have chosen to plot the line against Just Intonation based on C. If you did that it would look a lot smoother and more level."

Feel free to do so, but also explain why that would make sense. As you go on to say:"These flutes were designed to play music in a variety of key signatures, and across three octaves, in a time period when the A=440 tuning standard did not exist." So why would anyone in the mid 19th century want a Just Intonated flute based on any key or pitch? Remember Bach and his "Well-tempered Clavier"? We'd left Just Intonation a long way behind a lot earlier than Rudall and Rose.

Paddler: "Anyway, the point is that showing a single octave of tuning relative to a modern, and not particularly meaningful tuning standard for those flutes, just obfuscates a lot of important issues. Finally, the tuning of the keyed notes is greatly affected by issues such as the thickness of the pads, the amount they protrude into the tone hole, the height setting of the key when open, and the thickness of the cork under the key touch."

Take all that up with Rockstro and Ellis, not me! I won't be born for another 70 years! But we can ignore the keyed notes, just go on the open hole notes, and the harsh reality still presents itself. The tuning is wonky.

Paddler: "It is especially disappointing to see you claim this after having argued persistently in this same thread that bore perturbations at the tenons, in the form of chambers, couldn't possibly have any effect on tuning."

I don't think I've argued they couldn't have any effect on tuning. I do still have numerous reservations:
- I've not previously seen/heard any suggestion that "chambering" is associated with / located at tenons. Quite the contrary.
- The alleged "chambering" at the tenons looks more like thread compression than deliberate reaming to me.
- The alleged "chambering" hasn't been fully documented - we'd need at least maxima and minima bore dimensions, and a lot more data points
- The small amount of material removed doesn't look like it would achieve very much
- Nobody has attempted to identify what sort of changes to tuning we might expect from such manipulations.

We can do better than this, and we need to better than this.


Top
 Profile  
Reply with quote  
 
 Post subject: Re: Rockstro
PostPosted: Sat Feb 29, 2020 9:59 am 
Offline
User avatar

Joined: Sat Nov 03, 2007 7:19 pm
Posts: 531
Location: Hood River, Oregon, USA
Your notion of "wonky tuning" is just another way of saying that the tuning of the flute doesn't agree with your
preconceived, and unjustified, notion of what the tuning should be. Your inability to recognize this, combined
with your list of statements about chambering vs thread compression, just confirms that you have not read or
understood most of the material already referenced in this thread. That is disappointing.


Top
 Profile  
Reply with quote  
 
 Post subject: Re: Rockstro
PostPosted: Sat Feb 29, 2020 4:50 pm 
Offline
User avatar

Joined: Sun Dec 12, 2004 4:12 pm
Posts: 2179
Location: Malua Bay, on the NSW Nature Coast
Paddler, what do you think the tuning should look like?

I'm imagining that if one had been engaged to play flute at Lady Blacknall's afternoon tea party, it would be deuced convenient to have a reasonable alignment with the notes on her A430-tuned Broadwood.

And then if one is going on to play first flute with the Philo that same evening, it would be helpful to be in reasonable alignment with the strings up at A 452.

This would seem to suggest preferably two flutes (we don't seem to have very much evidence for that but we should keep looking), or a very long tuning slide and a body scaling set somewhere between those two extremes. We certainly find the very long tuning slides. And Rose's 1832 Patent Head reinforces that they expect one flute to cover the range.

But the body scaling? What might we expect? I said somewhere between the two extremes. 430 + 452 = 882, divide by 2 gives us A441, eerily close to our current pitch. But what do we find? The same as Rockstro and Ellis found 140 years ago. Wonky.

But why? Laziness? Incompetence? Shrinkage? Thread compression? An intentional mystery flute-only temperament nobody has so far put forward? One or more of the above? Other, please specify?

I think we've probably teased out the historical evidence about as far as we can, and still haven't been able to answer this most basic question. Time to turn to computer modelling to see what it can tell us. Even if it can eliminate any of the questions in front of us, we've taken a step forward.


Top
 Profile  
Reply with quote  
 
 Post subject: Re: Rockstro
PostPosted: Sat Feb 29, 2020 6:36 pm 
Offline
User avatar

Joined: Sat Nov 03, 2007 7:19 pm
Posts: 531
Location: Hood River, Oregon, USA
Well, you are the one making claims about what the flute should sound like, not me! I am simply pointing out that
there are a very many possible tuning targets, and that the ones you have chosen in order to declare the
flute's tuning as "wonky" are simply not convincing, especially since historical documents (and current ones)
consistently suggest that flute tuning should be considered more like that of non-fretted and non-keyboard
instrument (think violins, voice etc) which tend to be more focused on just intonation. When playing with
other instruments whose tuning is necessarily constrained, then it is up to the player to adapt, just as a singer
or fiddle player would.

Superficially, the difference between 12 TET and just intonation looks quite similar to the tuning graph you keep
posting. So according to your criteria, just intonation is wonky tuning! Just think for a moment how ludicrous
that statement is.

If you scroll down the page at this link you can find the difference (in cents) between 12 TET and Just Intonation for various intervals.
You will see plenty that deviate substantially.

Similarly, there are published research papers that describe ways to measure the consonance and dissonance
of various chords and intervals in different tuning standards. Just for example, below is a graph from a paper by
Meihui and Satoshi that shows the relative dissonance (using their calculated dissonance index) of Equal
Temperament tuning vs Just Intonation in various keys. The line for Equal Temperament is level because
it produces equal dissonance in all keys, showing, by definition, that an instrument tuned in Equal Temperament
is not perfectly in tune! Clearly, you can construct tuning graphs that make tuning in one temperament look wonky
compared to tuning in a different temperament, but that really doesn't prove anything!

Image

Anyway, I've had several members contact me saying that they feel I'm just wasting my time here. And I agree,
so this time I really am going to wrap up my contributions to this thread. I hope it has been valuable time spent
for anyone who is still reading.


Top
 Profile  
Reply with quote  
 
 Post subject: Re: Rockstro
PostPosted: Sun Mar 01, 2020 1:48 am 
Offline
User avatar

Joined: Sun Dec 12, 2004 4:12 pm
Posts: 2179
Location: Malua Bay, on the NSW Nature Coast
I never equated Just Intonation with wonky tuning. The pattern shown in the JI vs ET graph and the pattern from Rockstro and Ellis' analysis (and what we see regularly) might seem to have superficial similarities, but that's where it ends. Let's assign some Highs, Mediums and Lows to the notes from C to B in each and compare the patterns:

LMMLHLHLLHLH (the JI pattern) Letter count: 4H, 2M, 6L
MLLLMMMMMHMH (Rockstro & Ellis) Letter count: 2H, 7M, 3L

You'll notice that even if you "slide one of the patterns along and wrap it around" (to centre the JI on different pitches) they can never match. Amusingly, even the fact that the widths of the two 12-letter words formed are different tells us immediately we have differing numbers of H, M and L letters! I've provided a letter count for each to confirm that observation.

We can safely rule out JI on historical grounds too - it was long abandoned in favour of Well-tempering and Meantone back in the Baroque (Bach's "Well-tempered Clavier" reminds us). Our flutes date from the Romantic era, where freedom to modulate without encountering the "Wolf Note" was the goal. Meantone was giving way to ET by the middle of the century. We can see why Rockstro and Ellis tested for both in the data graphed - they were the only reigning options. Note that they drew no conclusions from their study. I'm with them.

The other dead giveaway that we are not dealing with an intonation or tempering is that the same intervals in the various octaves differ. As I have joked in the past, our flutes are just Bad-tempered. The question we have yet to deal with is "why"?

I'm sorry you and your supporters feel you're banging your head on a brick wall, but I think it was important to test for alternative theories, even if we ended up not being persuaded.

The scary question remains - can we get any further?


Top
 Profile  
Reply with quote  
 
 Post subject: Re: Rockstro
PostPosted: Sun Mar 01, 2020 4:19 am 
Offline
User avatar

Joined: Sun Dec 12, 2004 4:12 pm
Posts: 2179
Location: Malua Bay, on the NSW Nature Coast
Above I'd said: "Note that they drew no conclusions from their study. I'm with them.".

But I had overlooked an interesting comment in the preamble that speaks to us:

"It has been stated that the large-holed flutes of Rudall and Rose were unrivalled in tone, and even if their intonation was not as perfect as might have been wished, still they were better in that respect than those of any other makers."

Again, it's a hedged statement, as almost all of Rockstro is. "It has been stated....". Spit it out, man! Is it a statement of fact or an anonymous allegation?

But the central tenet, "even if their intonation was not as perfect as might have been wished" goes directly to what Rockstro and Ellis found, and what we find when we examine these flutes today. We have to keep in mind that Rockstro had moved on from the 8-key flute to his own version of Boehm's cylindrical flute with none of the dramatic intonation problems previous flutes suffered.


Top
 Profile  
Reply with quote  
 
 Post subject: Re: Rockstro
PostPosted: Mon Mar 02, 2020 12:25 pm 
Offline
User avatar

Joined: Tue May 23, 2006 6:47 am
Posts: 6879
Location: N.E. Wales, G.B.
Sebastian Mayfield has very generously shared the data he took a couple of years ago from my longest-owned Rudall & Rose, #4683 (made c1843) and also of Rudall Carte #6845 (made c1881-2, which I sold to a chap in Australia a few years ago - a photo-story of its restoration with useful information can be seen here: https://m.facebook.com/jemtheflute/albu ... =bookmarks) as spreadsheets.

Sebastian comments: "Here are a couple of files with bore data for your Rudalls, which you're welcome to share as you see fit. Unfortunately they're the other way round to make it easier for reamer-making but the graph should help a bit, and the data can always be reversed to give Z-zero at the large end if that makes sense."

I should add that the measurements were taken using graduated nylon discs of known diameters inserted into the bore on a depth-measuring device. Each data point obviously indicates the limit of careful, unforced insertion of each disc. No account or representation of any ovalling is recorded.


I've uploaded both Excel documents to my Box resources:
#4683 - https://app.box.com/s/axfnkp9bsnbz0c3ugypjx0ott2ejwg5c
#6845 - https://app.box.com/s/u8y56yl5fxvp0vfoo0f68maa2116or37

Sebastian also provided a comparative plot of the two bores as an image, also now in the Box folder: https://app.box.com/s/zowo7lgdlp9oo5c3ab51oqkohq0glyvw

_________________
I respect people's privilege to hold their beliefs, whatever those may be (within reason), but respect the beliefs themselves? You gotta be kidding!

My YouTube channel
My FB photo albums
Low Bb flute: 2 reels (audio)
Flute & Music Resources - helpsheet downloads


Top
 Profile  
Reply with quote  
 
 Post subject: Re: Rockstro
PostPosted: Mon Mar 02, 2020 10:54 pm 
Offline
User avatar

Joined: Sun Dec 12, 2004 4:12 pm
Posts: 2179
Location: Malua Bay, on the NSW Nature Coast
Ah, thanks Jem and Sebastian, very interesting.

Amazingly, despite having a 9 line image address, it looks like we can import the two bore graphs here:

Image

First thing to note is how similar they are, especially considering the difference in ages. You'd think that 40 years difference at a time when pitch was on the rise would manifest dramatically. The foot of the RC is just a smidge shorter; the lengths of the LH and RH sections identical.

Note that the two foot bores are almost totally identical, excepting the RR bore continues to reduce and then flare slightly, while the RC stops reducing earlier and becomes cylindrical for the last 20mm or so. Now, it's interesting to consider that a foot joint has no tenons, and should be therefore immune to any tenon constriction. Seems to be the case here.

The LH section has a tenon on both ends, and we often see constriction on both ends. No dramatic signs of constriction at the top (left) ends, although interesting that the RC starts around 18.8mm, about the head diameter, whereas the RR starts around 18.3mm. And both show a slight point-of-inflection within the length of the tenon.

Interesting both show that kink around 130mm. Same reamer? Or a point where one reamer stops and another continues?

Both in my view demonstrate bore constriction at the lower LH tenon. Paddler I think would argue that that was signs of backreaming to form a chamber.

The start of the two RH sections are remarkably similar, which again makes sense - there is a socket, not a tenon, so no expectation of compression.

But the lower ends of the RH sections demonstrate very different results. This is the tenon end, and I would argue the RR exhibits classic and significant tenon compression. The RC exhibits the opposite - a clear flaring of the bore which then terminates abruptly in the vertical cliff-face at the base of the foot socket. That one is hard to explain away. It would be good to examine that tenon closely to see if we can identify why. It could be signs of back-reaming but it could also be the result of damage. Our modelling should tell us what the effect of such a change would be.

A note incidentally on measurement technique, particularly as it relates to ovality. Jem tells us that Sebastian uses the "how far can I insert a disc" method. This gives us the minimum bore diameter at that length along the flute. I mostly use the "how far can I insert a telescopic T-gauge set to a specific width" method. That normally gives us the maximum bore diameter at the specific point along the flute. But it can also be used to sense the minimum diameter - having noted the maximum, you pull it out a bit and rotate it to find the minimum. It is rather time consuming, but we'll need to explore the issue of ovality as part of this investigation. All these methods have their place, but it's good to remember they can give differing results when ovality is involved.

Thanks Jem and Sebastian for access to that data. More grist to the mill.


Top
 Profile  
Reply with quote  
 
 Post subject: Re: Rockstro
PostPosted: Sun Mar 08, 2020 8:05 pm 
Offline
User avatar

Joined: Tue May 23, 2006 6:47 am
Posts: 6879
Location: N.E. Wales, G.B.
Sebastian has now kindly come up with the tone-hole measurements he took for the two flutes he measured. He only took external tube diameter measurements for one of them. I have added his two documents to the folder in my Box resources: https://app.box.com/s/qiyn6pvgsltmqyzaa9dtn9cs3gk3g5y3

_________________
I respect people's privilege to hold their beliefs, whatever those may be (within reason), but respect the beliefs themselves? You gotta be kidding!

My YouTube channel
My FB photo albums
Low Bb flute: 2 reels (audio)
Flute & Music Resources - helpsheet downloads


Top
 Profile  
Reply with quote  
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 145 posts ]  Go to page Previous  1 ... 6, 7, 8, 9, 10

All times are UTC - 6 hours


Who is online

Users browsing this forum: Conical bore and 4 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum

Search for:
Jump to:  
Powered by phpBB® Forum Software © phpBB Group
[ Time : 0.147s | 12 Queries | GZIP : On ]
(dh)