Jump to content

That Wheatstone (Etc.) Sound


Recommended Posts

 

I guess the reeds benefit from regular exercise...

 

In my concertinas (and I suspect, most others), some of the reeds get played much more frequently than some of the others. But you rarely hear of a vintage concertina having an uneven sound, which would likely be a problem if each reed's sound was dependent on how much "exercise" it got.

 

Good point David, and I guess you're right at large.

 

However, from my experience there may be some reeds which are sounding somewhat acute due to not really having been played . As to my own concertina we don't speak about the really seldom used reeds, but for instance the low F# had really mellowed over, say, two years of frequent use.

 

Maybe the previous owner just played melodies in a higher register, or even stuck just to C-maj and A-min? I don't know, but there has been an improvement with actually two reeds (push and pull likewise) which obviously weren't replacements and therefore likely about 100 years old... without changes to the envirement (pads or travel or whatever).

 

There are some more reeds which I believe to have improved, but none of them belongs to the natural "white key" center rows...

 

Best regards - Wolf

Edited by blue eyed sailor
Link to comment
Share on other sites

Sound does get technical pretty quickly, doesn't it?

 

But perhaps some useful -- though by no means perfect -- information can be gleaned from what's already out there in the world.

 

For example, I've read that the action on a Lachenal can be a bit noisy and have seen listings that note steps taken to quiet the clackiness.

 

But viewing this (Collin Botts on a Lachenal of unspecified vintage, admittedly in a big hall)

 

http://youtu.be/HpCBdONhtjk

 

 

Eeek, remind me to add stainless steel pads on agate action boards to the pad leakage tests I'm conducting elsewhere!

 

Terry

Link to comment
Share on other sites

And, now that I'm over the shock of hearing that clacking, I should add that I think that first recording demonstrates why we would need to set some rules. That recording really tells us very little about the sound of that instrument, other than that it desperately needs some attention to the pads (or whatever is causing the clacks). It tells us more about the hall than about the concertina. Great demonstration, in fact. Foot tapping included.

 

And, yes, sound does get technical pretty quickly, but I don't think that needs to become an obstacle. We'd just need a "Concertina Comparison Expert Group", like the Joint Photographic Expert Group (JPEG) or the Motion Picture Expert Group (MPEG), to discuss the aims of the project, and work back to the requirements. Then set out a simple set of instructions that anyone could follow.

 

The clacking concertina does remind us that one of the instructions would have to be to fix any problems with the concertina before recording it. That suggests that the right people to contribute sounds to the comparison would be concertina repair people, or their customers!

 

Terry

Link to comment
Share on other sites

 

...we'd ideally need to get all samples adjusted to the same average level...

No.

Differences in dynamic range are also a characteristic of concertinas. Equalizing the loudness between samples from a pinhole Æola and a metal-ended Jeffries honker would misrepresent an important difference between them.

 

All things being the same, listeners always favour the louder signal.

Again, no.

 

I, for one, don't "always favour the louder signal"... not when listening, and certainly not when accompanying a song. And when comparing instruments I would want to be able to hear what its dynamics are over a normal range of bellows pressure, not artificially made louder (or softer).

 

 

The problem is that if you don't set them all to an equal level, you take pot-luck on whatever level the recordist opted for, and that is very misleading. You can experience pot-luck by listening to a range of recordings presented on YouTube, which, for some bizarre reason, leaves setting the overall level up to the discretion of the maker. So you can have recordings where the sound is hardly audible, to those where it is totally distorted. Compare that with recordings on Spotify, where the professional recordists have maximised their recording's impact by use of compression, limiting and normalising. We don't want that either, but it illustrates the opposite extreme. That's why I suggested using the old fashioned VU meter so we can at least supply recordings that have the same average sound level and are thus comparable.

 

The only way I can think of that you could preserve accurate absolute levels would be to ask recordists to take a sound level meter measurement at the mic position during the recording, and use that data in post production. Or to also record a calibration signal. I suspect very few people in the field would be equipped to do that.

 

Setting the various recordings to the same average level won't affect the dynamic range of the recording (compression and limiting would). But it does remind us that we wouldn't want to use a recorder with automatic gain control, as that will mess with the dynamics. This would include many (most?) video cameras that have no manual control on sound level.

 

Just to explain further on the favouring louder signals, there is a good technical reason why this happens. The Fletcher-Munson equal loudness curves illustrate that, as we reduce level, our perception of the bass and treble falls off faster than the mid frequencies. So take just one sound and present it as two choices, one louder than the other, and listeners will be attracted to the louder. Provided of course, they are both not already too loud! The louder seems to be richer, the quieter seems to be thinner.

 

fm1.gif

 

Now, I might be scaring you with all this techno-babble, so I should conclude by saying that I don't think any of this is hard to deal with. Failing to deal with it is the problem. A jumble of recordings made with a wild range of microphone types at any angle and distance in environments ranging from anechoic to cathedral at any level from inaudible to deathmetal and perhaps including dynamic processing will achieve nothing that YouTube hasn't already achieved. Set and follow some simple rules and it will all work out fine.

 

Terry

Link to comment
Share on other sites

...we'd ideally need to get all samples adjusted to the same average level...

No.

Differences in dynamic range are also a characteristic of concertinas. Equalizing the loudness between samples from a pinhole Æola and a metal-ended Jeffries honker would misrepresent an important difference between them.

 

All things being the same, listeners always favour the louder signal.

Again, no.

 

I, for one, don't "always favour the louder signal"... not when listening, and certainly not when accompanying a song. And when comparing instruments I would want to be able to hear what its dynamics are over a normal range of bellows pressure, not artificially made louder (or softer).

 

The problem is that if you don't set them all to an equal level, you take pot-luck on whatever level the recordist opted for, and that is very misleading.

 

I didn't say it would be easy. In fact, like definitions of "fair", I suspect it's impossible to get an answer that's "right" for all situations.

 

Just to explain further on the favouring louder signals, there is a good technical reason why this happens. The Fletcher-Munson equal loudness curves illustrate that, as we reduce level, our perception of the bass and treble falls off faster than the mid frequencies.

Exactly. So if you artificially increase the loudness, the listener will hear a "richer" sound from a quiet instrument than they will when encountering that same instrument "in person"? It'll leave them baffled. (Pun intended. ;))

 

So take just one sound and present it as two choices, one louder than the other, and listeners will be attracted to the louder. Provided of course, they are both not already too loud! The louder seems to be richer, the quieter seems to be thinner.

Some listeners will prefer "richer" to "thinner", but not necessarily everyone. Just as not everyone -- and not every tradition -- prefers lots of vibrato from a violin/fiddle. And if "thinner" is what the instrument gives in real life, then I would say that providing a "richer" recording is misrepresentation, whether that difference came from artificial reverb or just adjusting volume levels.

 

I suspect that it's possible to get truly precise comparisons only if all the recordings are made under precisely controlled conditions, and that almost certainly means the same studio or other location, at a minimum. (Just out of curiosity, can variations in barometric pressure make a difference?)

 

But then there's the difference between precision and accuracy. There's no point in reporting (or recording) ten digits of precision if reproducibility/accuracy is no better than three digits. In the end, I think we'll need to accept that no matter how good the recordings, they will probably never sound to our ears exactly as they will "in the flesh".

Link to comment
Share on other sites

I'd expect, providing we were to take reasonable steps to make the recording processes similar, we would easily discern the essential differences. Whereas, if we take no precautions, we will definitely end up with a useless jumble.

 

The degree of caution needed is proportional to the similarity between instruments that we might expect to encounter. If the sounds of the most disparate instruments are only very subtly different (when two good examples are played side-by-side in a good listening environment), we will need to be very careful in recording to ensure we capture that subtle difference. If the difference is "night & day", it will be obvious in relatively casual recordings.

 

Perhaps we should ask those who are fortunate enough to have experienced that difference. Are we talking subtle differences or a broad range of sounds? Do we need the wide-angle lens or the scanning electron microscope?

 

Terry

Link to comment
Share on other sites

The degree of caution needed is proportional to the similarity between instruments that we might expect to encounter. If the sounds of the most disparate instruments are only very subtly different (when two good examples are played side-by-side in a good listening environment), we will need to be very careful in recording to ensure we capture that subtle difference. If the difference is "night & day", it will be obvious in relatively casual recordings.

 

Perhaps we should ask those who are fortunate enough to have experienced that difference. Are we talking subtle differences or a broad range of sounds? Do we need the wide-angle lens or the scanning electron microscope?

 

The former, I'm positive!

 

(If you're not just accustomed to a very average sounding - good - instrument, for instance a best-period wooden-ended Aeola, you might encounter real difficulties to find an instrument which might replace the own one, be it in the long term or just for a gig or session)

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...