The challenges of symphonic recordings

Started by miker, May 30, 2007, 03:50:33 PM

Previous topic - Next topic

0 Members and 1 Guest are viewing this topic.

miker

I enjoy big symphonies, but I am often disappointed by the recording quality. I also listen to Jazz. These recordings are usually super crisp and clear while many of the classical recordings are somewhat muddy; lacking clarity.

Is this the challenge of recording a large group?

Are there Symphonic recordings out there that make you feel like you are on the stage?

Thanks

hornteacher

There are recordings out there that are more crisp.  I actually look for those because I enjoy them more than the "distant reverb" sound.  Usually (but not always) the crisper rcordings are pure digital (DDD) and more recent.  Mackerras I've found tends to record symphony cycles (Mozart, Beethoven, Brahms, etc) with little reverb and, in many cases, a reduced sized orchestra which makes the sound very "clean", "crisp", "close", (not sure exactly what the word is but you get the idea).  Recent Abbado recordings have this as well as does most of the Telarc label.

miker


head-case

#3
I find modern recordings far inferior to the older ones because economics overrides technical and asthetic quality in modern recordings.  Now orchestras are recorded with extremely large numbers of microphones so that balances can be adjusted from the control room and millions of little takes can be spliced together to make the finished product.  If the trumpet screws up, they find a place where the trumpet was playing the same note and cut and paste it in.   These recordings do not have any artistic vitality and don't give a proper image of the stage, in my experience.  In the old days a small number of microphones was used and the hall was used to "mix" the sound and the orchestra had to play the movement through, with ocasional splices here and there (which were done with scotch tape).  If you want to hear good sound listen to any Mercury Living Presence recording, or Decca recordings from the 50's or early 60's.   Telarc purports to use the Mercury technique, but my experience is that they have such an obsession with bass response that listening to anything they release is likely to blow out your speakers and/or loosen your teeth.

Don

Quote from: hornteacher on May 30, 2007, 05:56:21 PM
There are recordings out there that are more crisp.  I actually look for those because I enjoy them more than the "distant reverb" sound. 

Yes, that reverb business can be a real pain, as it damages the clarity and definition of music as well as screwing up the musical dialogue among voices.  This "wet" sound can be found in many modern recordings from solo to large orchestra.  I guess the record companies think its public likes the reverb sound, but I'm not a happy customer.

MishaK

Quote from: hornteacher on May 30, 2007, 05:56:21 PM
There are recordings out there that are more crisp.  I actually look for those because I enjoy them more than the "distant reverb" sound.  Usually (but not always) the crisper rcordings are pure digital (DDD) and more recent.  Mackerras I've found tends to record symphony cycles (Mozart, Beethoven, Brahms, etc) with little reverb and, in many cases, a reduced sized orchestra which makes the sound very "clean", "crisp", "close", (not sure exactly what the word is but you get the idea).  Recent Abbado recordings have this as well as does most of the Telarc label.

This is a little too general. There are tons of really crappy DDD releases, particularly from the early 80s. It is really hit and miss as it depends on the acoustics of the hall, the record company and the recording engineer. A few recent recordings in outstanding sound:

Bruckner 4, Salonen/LAPO/Sony

Stravinsky Firebird, Boulez/CSO

Mahler 5, Chailly/Concertgebouw/Decca

Beethoven Symphonies, Barenboim/Staatskapelle Berlin/Teldec

Quote from: head-case on May 31, 2007, 08:14:59 AM
I find modern recordings far inferior to the older ones because economics overrides technical and asthetic quality in modern recordings.  Now orchestras are recorded with extremely large numbers of microphones so that balances can be adjusted from the control room and millions of little takes can be spliced together to make the finished product.  If the trumpet screws up, they find a place where the trumpet was playing the same note and cut and paste it in.   These recordings do not have any artistic vitality and don't give a proper image of the stage, in my experience.  In the old days a small number of microphones was used and the hall was used to "mix" the sound and the orchestra had to play the movement through, with ocasional splices here and there (which were done with scotch tape).  If you want to hear good sound listen to any Mercury Living Presence recording, or Decca recordings from the 50's or early 60's.   Telarc purports to use the Mercury technique, but my experience is that they have such an obsession with bass response that listening to anything they release is likely to blow out your speakers and/or loosen your teeth.

This is actually completely inaccurate on a number of levels.

1. Splicing, cutting and studio recording generally has actually decreased due to the costs of recording orchestral music. Whereas in the heyday of large-scale industrial classical recording in the 60s you would have studio sessions for days on end where never anything longer than perhaps a six to seven minute stretch of music was recorded in one single take, nobody can afford such productions today. Today, for cost reasons, a large number of recordings (by now I'd venture to guess, the majority) are made as "live" recordings. Now, these often aren't single takes either. Excepting the occasional one-night-only on-tour performance captured on disc, most concerts are subscription concerts performed on three or four evenings. The recording engineers will take the best of the three (plus possibly a patch session where a few select passages are recorded to eliminate bloopers, cellphones, coughs or premature applause) and make one single "live" performance out of it. Most of today's classical recordings are far closer to the live experience than anything put out by Karajan, Solti, Szell or Ormandy, who recorded almost everything in a highly controlled studio environment in countless takes completely removed from the live experience.

2. Number of microphones: I won't generalize here, but most of the major concert venues have fixed microphone setups (e.g. Philharmonie Berlin, Orchestra Hall Chicago) that are used for live broadcasts as well as recordings (they might add a mike for a piano or a harp or something like that). I haven't counted the mikes at the CSO, but they don't strike me as excessive. While you might prefer the sound of 50s and 60s recordings, compared to what is possible on good audio equipment today, they pale to what is possible with a proper multi-microphone setup. See point 3 below.

3. What you describe sounds like nostalgia combined with inadequate stereo equipment. Certain older recordings (e.g. anything by Solti from the 70s) sound gripping on a mediocre stereo but very shrill and tinny on a very good one. Most older analog recordings have limitations that don't apply to more modern digital ones, but that many listeners find preferable. The limitations are primarily in dynamic range. In a live symphonic concert, the difference between a solo ppp and a tutti fff can be over a hundred decibels. Analog recordings have a much more limited range. A good sound engineer will adjust the performance such that the ppps are audible but that the fffs don't blow you away. But the result on analog recordings is often that the ppps are unnaturally prominent while climaxes seem to plateau and thickly orchestrated tuttis sound congested. The performance sounds constricted and unnatural. You might not hear this on a mediocre stereo, but get a good set of speakers or headphones and it becomes apparent immediately. Now, from a technical standpoint of course we want to try to capture the live experience as closely as possible. Indeed, some newer recordings have an enormous dynamic range. But then we get people like you who complain that they have to turn up the stereo to even hear the ppps and then have to turn it back down to prevent the fffs from blowing away the neighbors two houses down the road.

head-case

Quote from: O Mensch on May 31, 2007, 08:49:05 AM

1. Splicing, cutting and studio recording generally has actually decreased due to the costs of recording orchestral music. Whereas in the heyday of large-scale industrial classical recording in the 60s you would have studio sessions for days on end where never anything longer than perhaps a six to seven minute stretch of music was recorded in one single take, nobody can afford such productions today. Today, for cost reasons, a large number of recordings (by now I'd venture to guess, the majority) are made as "live" recordings. Now, these often aren't single takes either. Excepting the occasional one-night-only on-tour performance captured on disc, most concerts are subscription concerts performed on three or four evenings. The recording engineers will take the best of the three (plus possibly a patch session where a few select passages are recorded to eliminate bloopers, cellphones, coughs or premature applause) and make one single "live" performance out of it. Most of today's classical recordings are far closer to the live experience than anything put out by Karajan, Solti, Szell or Ormandy, who recorded almost everything in a highly controlled studio environment in countless takes completely removed from the live experience.

2. Number of microphones: I won't generalize here, but most of the major concert venues have fixed microphone setups (e.g. Philharmonie Berlin, Orchestra Hall Chicago) that are used for live broadcasts as well as recordings (they might add a mike for a piano or a harp or something like that). I haven't counted the mikes at the CSO, but they don't strike me as excessive. While you might prefer the sound of 50s and 60s recordings, compared to what is possible on good audio equipment today, they pale to what is possible with a proper multi-microphone setup. See point 3 below.

3. What you describe sounds like nostalgia combined with inadequate stereo equipment. Certain older recordings (e.g. anything by Solti from the 70s) sound gripping on a mediocre stereo but very shrill and tinny on a very good one. Most older analog recordings have limitations that don't apply to more modern digital ones, but that many listeners find preferable. The limitations are primarily in dynamic range. In a live symphonic concert, the difference between a solo ppp and a tutti fff can be over a hundred decibels. Analog recordings have a much more limited range. A good sound engineer will adjust the performance such that the ppps are audible but that the fffs don't blow you away. But the result on analog recordings is often that the ppps are unnaturally prominent while climaxes seem to plateau and thickly orchestrated tuttis sound congested. The performance sounds constricted and unnatural. You might not hear this on a mediocre stereo, but get a good set of speakers or headphones and it becomes apparent immediately. Now, from a technical standpoint of course we want to try to capture the live experience as closely as possible. Indeed, some newer recordings have an enormous dynamic range. But then we get people like you who complain that they have to turn up the stereo to even hear the ppps and then have to turn it back down to prevent the fffs from blowing away the neighbors two houses down the road.

I don't know about the CSO, but when I attended a SFO concert which was being recorded for CD release their normal microphone setup was gone and at least 50 microphones were on the stage.   I remember reading about a recording made by EMI of the New Years Concert in Vienna which was done with a 128 channel mixing console, essentially one microphone for each instrument.  With old technology splicing was not practical in most cases because you couldn't get the reverb from the two takes to match up, and there would be a noticable glitch at the splice point.  If there was a splice, you heard it.  Now you can do a digital fade on the reverb mikes while punching in the tracks for individual instruments at a splice point to blend two takes together, or splice in an alternate take for an individual instrument.  If you listen to a typical old recording with headphones you will hear maybe 2 or 3 splices.  I have a DVD of a recording session and it shows the music being recorded literally 3 or 4 bars at a time, with hundreds of fragments being spiced together.

I don't know who you consider a "good audio engineer" but the publically described procedure used by Mercury Living Presence employed no spotlight microphones and no "gain-riding" whatsoever.  The resulting dynamic range taxed the limits of the origional LP releases, but sounds splendid on CD.
 

MishaK

#7
Quote from: head-case on May 31, 2007, 09:42:37 AM
I don't know about the CSO, but when I attended a SFO concert which was being recorded for CD release their normal microphone setup was gone and at least 50 microphones were on the stage.   I remember reading about a recording made by EMI of the New Years Concert in Vienna which was done with a 128 channel mixing console, essentially one microphone for each instrument. 

Fair enough. I have never been to the SFSO's home venue. But re: the New Year's Concert, what splicing is done there? I mean, the same thing is issued on DVD and broadcast live. It's pretty much a single take.

Quote from: head-case on May 31, 2007, 09:42:37 AM
With old technology splicing was not practical in most cases because you couldn't get the reverb from the two takes to match up, and there would be a noticable glitch at the splice point.  If there was a splice, you heard it.

But you could of course record in a hall with little reverb and get rid of the problem that way.

Quote from: head-case on May 31, 2007, 09:42:37 AM
Now you can do a digital fade on the reverb mikes while punching in the tracks for individual instruments at a splice point to blend two takes together, or splice in an alternate take for an individual instrument.  If you listen to a typical old recording with headphones you will hear maybe 2 or 3 splices.

Or more. Also, you can always splice during a long rest without any reverb issues.

Quote from: head-case on May 31, 2007, 09:42:37 AM
I have a DVD of a recording session and it shows the music being recorded literally 3 or 4 bars at a time, with hundreds of fragments being spiced together.

What performance? Watch the Making of Solti's Ring DVD, whatever it's called, you'll see that their takes were very short as well.

Quote from: head-case on May 31, 2007, 09:42:37 AM
I don't know who you consider a "good audio engineer" but the publically described procedure used by Mercury Living Presence employed no spotlight microphones and no "gain-riding" whatsoever.  The resulting dynamic range taxed the limits of the origional LP releases, but sounds splendid on CD.

I beg to differ on the "splendid". Splendid for its time, yes. Compared to more recent technology, no. Again, there is just no comparison in dynamic range or spaciousness of the sound.

Another point: I forget in which booklet this is, either the Decca Original Masters Schuricht set or the IMG/EMI Great Conductors Schuricht set. In it there are extended quotes from John Culshaw who produced Schuricht's Schubert 8th and how frustrated he was with Schuricht because he didn't want to do the music in short little bits that he would then splice together (the result, as audible on both sets, is an oddly ferklempt performance without a long line, rather uncharacteristic for Schuricht). Very much in contrast to Solti, whom Culshaw alwys hailed as the first great stereo recording conductor because he was acutely aware to the unique issues of performing specifically for a recording session. I think you greatly underestimate the amount of editing that went into most of the older recordings.

bhodges

I posted this on another thread, but it's appropriate here, too.  This is one of the best-recorded CDs I've ever heard, and coincidentally, here is a review on the Good-Music-Guide.  The label is Reference Recordings.



I also second the Stravinsky Firebird and Mahler Fifth suggestions from O Mensch -- spectacular sound.  ;) 

--Bruce

hornteacher

Quote from: O Mensch on May 31, 2007, 08:49:05 AM
This is a little too general.

I know, he caught me on the way to work.   :)


Greta

#10
QuoteThere are tons of really crappy DDD releases, particularly from the early 80s. It is really hit and miss as it depends on the acoustics of the hall, the record company and the recording engineer.

Yeah definitely, there are so many variables. I remember when I bought my 1st Carmina Burana (Previn, London SO), I wanted great sound so I bought the most expensive DGG (hey, I was 15) which boasted a live performance in "unprecedented DDDD sound" that had a whole page in the liner notes devoted to what 4D sound was and how it was now "actually possible to eliminate the listener's awareness of the technical medium".

Well, it's a nice recording but nowhere near all those claims, at times it sounds harsh in the upper timbres and the middle voices don't have as much clarity as I would have hoped for. DGG can be hit or miss. Two recordings I have of theirs that do sound fantastic are Gardiner's Planets with the Philharmonia and Sinopoli's Elgar 2nd, also with the Philharmonia.

QuoteA few recent recordings in outstanding sound:

Bruckner 4, Salonen/LAPO/Sony

Ack! I don't have this yet. His LA recordings with Sony do have wonderful sound I find, the Shostakovich PCs 1 & 2 with Bronfman, Lutoslawski 3rd, 4th and his M3 stick out in my mind, all fresh, clear, resonant, abundant in detail. 

Telarc is a label who has consistently great sound, and I would perhaps consider their symphonic recordings to be the feather in their cap. With the Atlanta Symphony, (and Cinncinati too) they made some real stunners. Very spacious, but powerful, lush, clear sound. Gorgeous. Usually flawless playing, and the interpretations are if not my top recommendations, quite solid.

miker

Is there a connection between Telarc and Teldec?

71 dB

BIS always records with a microphone pair. It's up to the label and their engineer how many mics are used. A mediocre modern recording is superior to an old 50's recording.
Spatial distortion is a serious problem deteriorating headphone listening.
Crossfeeders reduce spatial distortion and make the sound more natural
and less tiresome in headphone listening.

My Sound Cloud page <-- NEW Jan. 2024 "Harpeggiator"

karlhenning

Quote from: 71 dB on June 01, 2007, 06:07:05 AM
A mediocre modern recording is superior to an old 50's recording.

In some ways, sure.

head-case

Quote from: miker on June 01, 2007, 04:46:49 AM
Is there a connection between Telarc and Teldec?

no.  Teldec is a fusion of Telefunken and Decca.

karlhenning


MishaK

Quote from: head-case on June 01, 2007, 08:41:56 AM
no.  Teldec is a fusion of Telefunken and Decca.

I don't beleive that is true. Decca (known as London in the US) is part of Universal Classics (along with DG and Philips). Teldec (the old Telefunken) is part of Warner Classics and I don't think the label is even being used anymore. All newer recordings that would have been Teldec or Erato are now simply Warner.

head-case

#17
Quote from: O Mensch on June 01, 2007, 09:15:44 AM
I don't beleive that is true. Decca (known as London in the US) is part of Universal Classics (along with DG and Philips). Teldec (the old Telefunken) is part of Warner Classics and I don't think the label is even being used anymore. All newer recordings that would have been Teldec or Erato are now simply Warner.

It would be good if you could check your facts before posting.  The joint venture between Telefunken and Decca started in the 1950's.  If you look in old vinyl stores you will find numerous Decca recordings that were distributed in Germany under the Teldec name.  This venture continued up until the 80's when Teldec became independent and was subsequently bought by Warner.

http://rateyourmusic.com/label/teldec_schallplatten_gmbh