Head-Fi.org › Forums › Misc.-Category Forums › Music › Live classical concerts?
New Posts  All Forums:Forum Nav:

Live classical concerts? - Page 2

post #16 of 24

Sacramento, San Francisco, LA, NY, St. Petersburg, Mariisnki, Utah, Cleveland, Vienna, Berlin, Israel, Seattle, Boston, National--either Philharmonic or Symphony Orchestras...probably a few that I've forgotten!

post #17 of 24

I've been very lucky to see those great orchestras live.

 

Some of them even several times.

 

(I've not attended them in this order)

 

Český Národní Symfonický Orchestr,

London Symphony Orchestra,

London Philharmonic Orchestra,

Scottish Chamber Orchestra,

New York Philharmonic,

Wiener Philharmoniker,

Orchestre de Paris,

Concertgebouworkest Amsterdam,

Gewandhausorchester Leipzig,

Berliner Philharmoniker,

Staatskapelle Dresden,

SWR Sinfonieorchester Baden-Baden,

Heidelberger Sinfoniker,

Heidelberger Philharmoniker

NDR Sinfonieorchester,

Göttinger Symphony Orchester,

Symphonieorchester des Bayerischen Rundfunks,

Münchner Philharmoniker,

Bamberger Symphoniker,

and some "unknown/smaller/youth" ones more...

post #18 of 24
Haven't been to too many sadly.
The great Pittsburgh Symphony(home symphony smily_headphones1.gif )
Chicago Symphony
Cleaveland Orchestra
Various highschool/middle school orchestra

Really really want to one day see the Berlin Phiharmonic!!!!
post #19 of 24

Speaking of live classical concerts (and how the hall influences what we hear)...check out this article.

 

"How Concert Halls Enhance Crescendos"

 

This is a quick, but really fascinating read. It's made doubly cool by the fact that both the Author and the Researcher referenced in the article belong to the Perceptual Audio Group (which I created and for which I am the Moderator) on Linked In.

 

Nice to see them getting the props they deserve.

 

http://www.huffingtonpost.com/trevor-cox/how-concert-halls-enhance-crescendos_b_5483292.html

 

Mark

post #20 of 24
Thanks, very interesting, Mark. Odd that, despite all the praise for the sound of the Musikverein over the years, modern concert hall designers prefer more avant garde shapes rather than replicating what sounds best.
post #21 of 24
Quote:
Originally Posted by eyeresist View Post

Thanks, very interesting, Mark. Odd that, despite all the praise for the sound of the Musikverein over the years, modern concert hall designers prefer more avant garde shapes rather than replicating what sounds best.

You're most welcome.

 

It is interesting in a way. I mean, to think that some of the great halls and amphitheaters were designed hundreds if not millenia ago, and none with the help of modern technology. It's been said that some of the amphitheaters in Greece (and, I believe, Sicily and other parts of Italy) have acoustics that allow a person seated well back from stage to hear the sound of a match being struck.

 

Still, for those not familiar with the modern approach (most rely upon a technique known as ray-tracing (pretty easy to visualize - think of a laser pointer pointed towards a mirror as opposed to a wall - one reflects much more light than the other - ray tracing sort of works this way)), you should look into the CATT and ODEON pieces of code ( http://www.odeon.dk/archaeological-acoustics-open-theatre# ) as well as their auralisation demos. This is some pretty powerful stuff that allows, often with a very high degree of precision and realism, to simulate halls before they are built. The thing is, like any tool, the simulation (and resulting auralisation) really comes down to just how well the space has been modeled. Gettting all of the material properties right is absolutely paramount. After all, computers only do what we tell them to.

 

Here's an example of an auralisation demo that some may find interesting:

 

http://www.odeon.dk/sound-musikverein

 

Apart from the field of modeling, people like Angelo Farina (University of Parma) have spent years collecting impulse responses of various well-known concert halls. Why do this? Well, those who work in audio recording can make use of these impulse responses using a technique known as convolution (reverb). Convolution reverb differs from traditional simulations in that you are using the actual impulse response of a venue to create a more mathematically-correct reverb. Again though, just how you actually get the impulse response is itself a pretty heady process. In essence though, the idea is to use some sort of stimulus in a space with microphones located around the venue. Sometimes these are stereo pairs, sometimes mono, and sometimes, binaural mannequin head microphones. Many different forms of stimuli have been used (and there are many papers written on the subject), but the most common ones re things like a starter's pistol, a firecracker, a balloon burst, or a dodecahedron speaker, into which is sent either Gaussian noise, or a chirp (a rapid sweep).

 

Why is the impulse so important?

 

Mathematically speaking, an impulse is (more or less) defined as a signal that is 'infinitely high' and 'extremely short-lived' (a Dirac delta function, or in related sense, the Kroneker delta) - you can think of a single hand-clap (or a single finger-snap) as a kind-of-sort-of-mostly-in-a-way-but-not-really an impulse response. Again, a true impulse in the time domain is basically "nothing" then the impulse, then "nothing" once more, so that hand-clap and finger-snap aren't really true impulses, but...you get the idea.

 

Now, if you take the transform of the impulse response from the time domain into the frequency domain, you will end up with a perfectly flat spectrum; all frequencies appear at equal power. However, this is often difficult to realize in practice - making something that fits such a definition is all but impossible. However, getting close is often enough to properly model the space. So by hitting a device's input with an impulse, you are imparting all frequencies at the same magnitude - you can see how this could be useful. Incidentally, this is the very same reason why during structural testing (buildings, planes, brdiges, what have you), the stimulus is often a blow from a special type of hammer that happens to be equipped with a load cell (which measures force). Why do this? Because if the impact is 'clean' you will impart all frequencies to the device. As such, if resonance frequencies exist, they will be excited by the blow from the hammer (and of course, things like the damping of the structure (how easily vibrations decay) can be measured.

 

Have you ever noticed, perhaps while cleaning out an almost empty garage or a shed, just how loud a broomstick can sound when it meets the floor? You can try this for yourself by placing a foot on the broom and pulling it back at the top (like a spring) and then releasing it. When it hits the concrete, something similar to an impulse is created. Because there is very broad-band energy produced, your ears hear (almost all at once) this very broad-ranging sound, and consequently, it seems quite loud to your ears (if you think about the hearing mechanism and such, and read up on things like temporal masking, you can start to get an appreciation for just how complex is our hearing system).

 

Anyway, the impulse response technique isn't all that new, but it is extremely powerful. In fact, this very same approach is used in audio gear to essentially measure the properties of that system - only in the case of electronics, a pseudo-impulse is generated as a wav (or similar) file, and that signal is fed to the device so that its impulse response can be measured. From the impulse response proper, a wealth of information about what the device does to the signal can be understood.

 

So again, the dual for this can be seen in architectural acoustics, and in the case of the simulation software (ray tracking) these impulse responses are modeled via ray-tracing so that the response at the receiver's postion can be synthesized. If you root around the pages for CATT and ODEON, you'll see that they start with a 'dry' signal (recorded in an anechoic chamber) and then once the model is done, they feed that recording to the model - pretty cool stuff.

 

Yet, and to your point (and what I touched on before), it simply amazes me that many of the best sounding halls every built were done without the benefit of computational modeling techniques. I can't say why some forms (architecturally speaking) are what they are these days, but at the same time, if properly modeled (and faithfully executed during construction), it's entirely possible to build a non-traditional-shape hall, and still have it deliver favorable or even great acoustics. The truth is, it takes some very clever and attentive people a great deal of time, thought, and effort to realize that potential...and yet...look at what the Greeks and Romans did years ago.

 

This stuff is just plain out fascinating to me...

 

Oh, and for those who want to know more about Angelo Farina and his work with impulse response measurement and convolution reverb, peruse the web. I have, somewhere, a link to some of his presentations and or papers on the subject of convolution and impulse response, and I stronly urge anyone who has an interest to seek these out. IF I can manage to find that link, I'll update this thread.

 

Another 'gem' that's out there is "The Scientist and Engineer's Guide to DSP" written by Steven W. Smith - and get this - you can download any or all chapters for...free. That's right, free. Have a look: http://www.dspguide.com/

 

PS: Here's a shortcut to the bit on convolution and impulse response: http://www.dspguide.com/ch6/1.htm


Edited by immersifi - 6/19/14 at 10:08am
post #22 of 24
Quote:
Originally Posted by immersifi View Post

Quote:
Originally Posted by eyeresist View Post

Thanks, very interesting, Mark. Odd that, despite all the praise for the sound of the Musikverein over the years, modern concert hall designers prefer more avant garde shapes rather than replicating what sounds best.
You're most welcome.

It is interesting in a way. I mean, to think that some of the great halls and amphitheaters were designed hundreds if not millenia ago, and none with the help of modern technology. It's been said that some of the amphitheaters in Greece (and, I believe, Sicily and other parts of Italy) have acoustics that allow a person seated well back from stage to hear the sound of a match being struck.

Still, for those not familiar with the modern approach (most rely upon a technique known as ray-tracing (pretty easy to visualize - think of a laser pointer pointed towards a mirror as opposed to a wall - one reflects much more light than the other - ray tracing sort of works this way)), you should look into the CATT and ODEON pieces of code ( http://www.odeon.dk/archaeological-acoustics-open-theatre# ) as well as their auralisation demos. This is some pretty powerful stuff that allows, often with a very high degree of precision and realism, to simulate halls before they are built. The thing is, like any tool, the simulation (and resulting auralisation) really comes down to just how well the space has been modeled. Gettting all of the material properties right is absolutely paramount. After all, computers only do what we tell them to.

Here's an example of an auralisation demo that some may find interesting:

http://www.odeon.dk/sound-musikverein

Apart from the field of modeling, people like Angelo Farina (University of Parma) have spent years collecting impulse responses of various well-known concert halls. Why do this? Well, those who work in audio recording can make use of these impulse responses using a technique known as convolution (reverb). Convolution reverb differs from traditional simulations in that you are using the actual impulse response of a venue to create a more mathematically-correct reverb. Again though, just how you actually get the impulse response is itself a pretty heady process. In essence though, the idea is to use some sort of stimulus in a space with microphones located around the venue. Sometimes these are stereo pairs, sometimes mono, and sometimes, binaural mannequin head microphones. Many different forms of stimuli have been used (and there are many papers written on the subject), but the most common ones re things like a starter's pistol, a firecracker, a balloon burst, or a dodecahedron speaker, into which is sent either Gaussian noise, or a chirp (a rapid sweep).

Why is the impulse so important?

Mathematically speaking, an impulse is (more or less) defined as a signal that is 'infinitely high' and 'extremely short-lived' (a Dirac delta function, or in related sense, the Kroneker delta) - you can think of a single hand-clap (or a single finger-snap) as a kind-of-sort-of-mostly-in-a-way-but-not-really an impulse response. Again, a true impulse in the time domain is basically "nothing" then the impulse, then "nothing" once more, so that hand-clap and finger-snap aren't really true impulses, but...you get the idea.

Now, if you take the transform of the impulse response from the time domain into the frequency domain, you will end up with a perfectly flat spectrum; all frequencies appear at equal power. However, this is often difficult to realize in practice - making something that fits such a definition is all but impossible. However, getting close is often enough to properly model the space. So by hitting a device's input with an impulse, you are imparting all frequencies at the same magnitude - you can see how this could be useful. Incidentally, this is the very same reason why during structural testing (buildings, planes, brdiges, what have you), the stimulus is often a blow from a special type of hammer that happens to be equipped with a load cell (which measures force). Why do this? Because if the impact is 'clean' you will impart all frequencies to the device. As such, if resonance frequencies exist, they will be excited by the blow from the hammer (and of course, things like the damping of the structure (how easily vibrations decay) can be measured.

Have you ever noticed, perhaps while cleaning out an almost empty garage or a shed, just how loud a broomstick can sound when it meets the floor? You can try this for yourself by placing a foot on the broom and pulling it back at the top (like a spring) and then releasing it. When it hits the concrete, something similar to an impulse is created. Because there is very broad-band energy produced, your ears hear (almost all at once) this very broad-ranging sound, and consequently, it seems quite loud to your ears (if you think about the hearing mechanism and such, and read up on things like temporal masking, you can start to get an appreciation for just how complex is our hearing system).

Anyway, the impulse response technique isn't all that new, but it is extremely powerful. In fact, this very same approach is used in audio gear to essentially measure the properties of that system - only in the case of electronics, a pseudo-impulse is generated as a wav (or similar) file, and that signal is fed to the device so that its impulse response can be measured. From the impulse response proper, a wealth of information about what the device does to the signal can be understood.

So again, the dual for this can be seen in architectural acoustics, and in the case of the simulation software (ray tracking) these impulse responses are modeled via ray-tracing so that the response at the receiver's postion can be synthesized. If you root around the pages for CATT and ODEON, you'll see that they start with a 'dry' signal (recorded in an anechoic chamber) and then once the model is done, they feed that recording to the model - pretty cool stuff.

Yet, and to your point (and what I touched on before), it simply amazes me that many of the best sounding halls every built were done without the benefit of computational modeling techniques. I can't say why some forms (architecturally speaking) are what they are these days, but at the same time, if properly modeled (and faithfully executed during construction), it's entirely possible to build a non-traditional-shape hall, and still have it deliver favorable or even great acoustics. The truth is, it takes some very clever and attentive people a great deal of time, thought, and effort to realize that potential...and yet...look at what the Greeks and Romans did years ago.

This stuff is just plain out fascinating to me...

Oh, and for those who want to know more about Angelo Farina and his work with impulse response measurement and convolution reverb, peruse the web. I have, somewhere, a link to some of his presentations and or papers on the subject of convolution and impulse response, and I stronly urge anyone who has an interest to seek these out. IF I can manage to find that link, I'll update this thread.

Another 'gem' that's out there is "The Scientist and Engineer's Guide to DSP" written by Steven W. Smith - and get this - you can download any or all chapters for...free. That's right, free. Have a look: http://www.dspguide.com/

PS: Here's a shortcut to the bit on convolution and impulse response: http://www.dspguide.com/ch6/1.htm

...thanks for the great write-up. I myself experienced the acoustics of several antique "theatres" in greece and in fact, some of them have amazing acoustics. Especially if you consider that there are just ruins left... So how might have this sounded in it's original condition...?
Edited by musikaladin - 6/18/14 at 9:25pm
post #23 of 24
Quote:
Originally Posted by musikaladin View Post


...thanks for the great write-up. I myself experienced the acoustics of several antique "theatres" in greece and in fact, some of them have amazing acoustics. Especially if you consider that there are just ruins left... So how might have this sounded in it's original condition...?

Well, there is one way that this could be answered...

 

In the Test and Measurement world, there are often correlation measurements that take place to validate a model. For example, in structural testing, a finite element analysis is performed (FEA) but then a modal test will be done to correlate. Why do this? Because the FEA model predicts what the mode shapes will be in the structure, whereas the modal analysis test tells you what they actually are. If they line up (the mode shapes) then this is a pretty favorable indication of the accuracy of the FEA model.

 

Likewise, using ray-tracing they could work the problem 'backwards'. That is, if one were to go to the venue and get the requisite measurements in various locations of the theater, they could tweak the model until they got the results that correlate with the actual data as measured in-situ. So, once the model (material properties, geometry etc) was validated, I would think it would be possible to extrapolate that and re-model the system with all of the seats in-place, with more angular shapes (to the steps etc - the things that have been eroded through the forces of water, wind, and time). I'm not a 'ray-tracing guy', but I think this is more or less how it would work.

 

It's a fascinating premise, and again, as long as the model is really accurate, so too should be the results. After all, the sound pressure levels being dealt with would certainly be outside the realm of non-linear values, and I think it's fair to say that the systems are LTI (linear and time-invariant), so my guess is that they could pretty accurately synthesize the sound of those venues when they were 'new'. Again, if they could determine the impulse responses, then convolving a 'dry' sound with those impulse responses should render a pretty representative time domain signal.

 

Mark

post #24 of 24

...and since this thread is about live classical concerts...check out this post. It covers some live A Cappella material, as well as Symphony Band, so I figured it was a pretty good fit as the latter was recorded in a hall (devoid of an audience), and these tracks have some pretty good dynamic range. I recorded the Symphony Band material in February of 2014, whereas the A Cappella material was done in a different theater:

 

http://www.head-fi.org/t/223165/legally-download-able-binaural-recordings-links/225#post_10633564

 

Mark

New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Music
Head-Fi.org › Forums › Misc.-Category Forums › Music › Live classical concerts?