What audio system do you have, or plan on getting?

[StudioGuy]

So approximately what you're saying is that even the most "live"-sounding recording involved a lot of llusion in its creation in the mixing & mastering process.  To that I say no doubt but anyway the best contemporary recordings are far ahead in terms of the "live" illusion that what was produced a few decades ago.

Stereo is itself an illusion (the stereophonic effect), so by definition even the most "live-sounding [stereo] recording" must involve illusion. Whether there was "a lot of illusion" depends on genre (classical music uses a lot less than popular genres), what you consider "a lot" to be and exactly how it's recorded, particularly in the case of classical music, and how it's mixed. Mastering has relatively little effect, it's mainly down to the recording, editing and mixing. Mastering is effectively just "tweaking" the final mix, in order to make the final mix sound as good as possible on the target audiences' playback equipment rather than only on the mix studio's playback chain. Modern recordings sounding more "live" is a consequence of several factors but is largely due to more options and "trickery" rather than less.

I have a few CD versions of the Mercury Living Presence records created by Bob & Wilma Cozart Fine.  The CD versions were remastered for the media by Wilma herself, (not so the SACD versions as I've heard).  These recordings seem good in comparison to other early recordings -- much better in terms of "liveness" than, e.g., Deutsche Grammophon recordings made in that era and quite awhile thereafter.

The fifties was a very interesting time in recording history. Technology invented during the war was further developed for commercial use and the release of stereo introduced a whole slew of recording considerations, the combination of which resulted in a wild flurry of recording experimentation, development and competition. Arguably leading the way was Decca but with the possible exception of EMI, each had some talented engineers/producers producing some excellent recordings although some were slightly ahead of others. ...

(My comment once was they they sounded like they were made in a high school gymnasium...OK that's a bit hyperbolic.)
Apparently most of those Mercury recordings employed a minimal microphone technique, i.e. only three mics.  But I'm told that that sort of minimal microphoning requires an inherently excellent recording venue and careful placement of the musicians as wall as the mics.

OK, this is where we get to the heart of the matter (particularly with classical music recordings) and unfortunately it's also very complex due to the technical/acoustic issues, the subjective/psychoacoustic issues and the relationship between them. On the technical side we've got acoustic issues such as the "inverse square law", Sabine's formulas, Schroeder frequencies, critical distance, microphone polar patterns and freq response, et al. Far too much to go into here but some salient simplifications; the closer the mic/s is to the sound source the more of the direct sound from that source is captured relative to everything else, specifically extraneous noise and room reflections/reverb and of course vice versa when the mic is placed further away. Using a few mics when recording a relatively large ensemble therefore raises the issue of recording a balanced sound, while at the same time accounting for the "critical distance" (the distance at which the reflected sound is equal in level to the direct sound). This requires a considerable amount of time/experimentation and often, the optimal distance for a balanced sound is not the optimal distance relative to the critical distance, so the engineers/producer had to compromise. This is almost certainly the cause of the "gymnasium" effect you described.

In addition to the above are the psychoacoustic effects and specifically the "cocktail party effect" ("selective hearing"); at a live event, what we see and hear informs/allows our brains to "focus" on specific sounds by reducing everything else. In a reverberant and/or noisy environment, we will perceive more of the salient sound due our brain reducing the amount of perceived reverb and noise. In other words, if for example we placed microphones at a particular listening position in the auditorium, it will record more noise and reverb than a human in that same location would perceive/experience. So, we would have to employ some "trickery" to alter the sonic reality to more closely match the "experience".

The solution to many of these issues is to use more (multiple) mics, although that brings other (more minor) issues, particularly in the 1950's when only 2 or 3 channel tape recorders were available. Decca largely pioneered this approach, developing the 3 mic Decca Tree in the early '50s, adding "outrigger mics" in the mid '50s and then 4 or more "spot mics" in the later '50s, despite the minor issue of requiring more "trickery". By the late '60s their orchestral setup was around 30 mics and all the other labels followed the same multi-mic course, with the only exceptions being one or two minor audiophile labels who managed to convince their gullible target audience that using just a couple of mics was somehow purer/more authentic and therefore "better", when the reality was that it was inferior.

As I'm given to understand, many DGG recordings were remastered for CD using their "Original Image Bit Processing" method which involved remixing the numerous original tape tracks with subtle time delays to simulate a more concert hall ambiance.

No, DGG may have added reverb or "subtle time delays" to some of their recorded tracks and indeed that is generally required when recording a large ensemble with a main array and spot mics but that was incidental to their "Original Image Bit Processing" method. In the 1990's studio digital recorders evolved to 20bit recorders (and then to 24bit) but the distribution was still CD and therefore it was necessary to reduce the bit depth to 16bit for distribution, which required the use of dither. After the publication of the Noise-Shaped Dither Theorem, companies started to produce products that employed that process, initially Sony with their "Super Bit Mapping", Apogee's "UV22" and others followed. The labels then started using their own marketing and marketing slogans for this, this is what DG's "Original Image Bit Processing" is, along with HDCD (high definition cd) and various others but all of them are just noise-shaped dither, which is inaudible at any reasonable listening level and way below the noise floor of old analogue recordings.

[Fëanor]

As I'm given to understand, many DGG recordings were remastered for CD using their "Original Image Bit Processing" method which involved remixing the numerous original tape tracks with subtle time delays to simulate a more concert hall ambiance.

No, DGG may have added reverb or "subtle time delays" to some of their recorded tracks and indeed that is generally required when recording a large ensemble with a main array and spot mics but that was incidental to their "Original Image Bit Processing" method. In the 1990's studio digital recorders evolved to 20bit recorders (and then to 24bit) but the distribution was still CD and therefore it was necessary to reduce the bit depth to 16bit for distribution, which required the use of dither. After the publication of the Noise-Shaped Dither Theorem, companies started to produce products that employed that process, initially Sony with their "Super Bit Mapping", Apogee's "U22" and others followed. The labels then started using their own marketing and marketing slogans for this, this is what DG's "Original Image Bit Processing" is, along with HDCD (high definition cd) and various others but all of them are just noise-shaped dither, which is inaudible at any reasonable listening level and way below the noise floor of old analogue recordings.

Thank you for this interesting correction & clarification.

I've heard audiophiles insist the DGG's 'Original Image Bit Process' CDs are "a big improvement over media without the process.  I have no personal opinion (or impression) since I have few 'Before' examples and those LPs.

FWIW, I have an original 1975 Carlos Kleiber Beethoven 5th LP, (to which I haven't listened in years).  I believe this was recorded on analog tape.   CD versions today claim to be Original Image Bit Processed;  presumably the tape recordings are converted at a higher bit rate and down converted to 16/44.1 with the advanced dithering process.

I also have the Klieber performance on SACD;  I generally listen to the ripped stereo CD layer of the SACD.  If I can find the disc I might have a listen to the SACD layer on my Sony Blu-Ray with outputs 5.1 channel in PCM format.

As for the Kleiber performance, I seems almost OVERLY dramatic, something that would probably have surprised Beethoven himself ... but I'm not a musician or any sort of musical expert, so what do I know? 

[71 dB]

No, DGG may have added reverb or "subtle time delays" to some of their recorded tracks and indeed that is generally required when recording a large ensemble with a main array and spot mics but that was incidental to their "Original Image Bit Processing" method. In the 1990's studio digital recorders evolved to 20bit recorders (and then to 24bit) but the distribution was still CD and therefore it was necessary to reduce the bit depth to 16bit for distribution, which required the use of dither. After the publication of the Noise-Shaped Dither Theorem, companies started to produce products that employed that process, initially Sony with their "Super Bit Mapping", Apogee's "U22" and others followed. The labels then started using their own marketing and marketing slogans for this, this is what DG's "Original Image Bit Processing" is, along with HDCD (high definition cd) and various others but all of them are just noise-shaped dither, which is inaudible at any reasonable listening level and way below the noise floor of old analogue recordings.

That's what I thought "Original Image Bit Processing" was, basically shaped dithering, but I wasn't 100 % sure. The way these companies describe these methods to us customers is marketing nonsense, not exact technical description. 

[Todd]

By the late '60s their orchestral setup was around 30 mics and all the other labels followed the same multi-mic course, with the only exceptions being one or two minor audiophile labels who managed to convince their gullible target audience that using just a couple of mics was somehow purer/more authentic and therefore "better", when the reality was that it was inferior.

Like Water Lily Acoustics.  Back in the 90s, a local iconoclastic stereo dealer didn't buy the marketing nonsense and did a demo of one of their recordings and a Decca recording to prove the point.  His advice was to ignore audiophile claims, which seems counterintuitive.  That dealer is still around, unlike most others from the time.

[StudioGuy]

Thank you for this interesting correction & clarification.

You're welcome!

I've heard audiophiles insist the DGG's 'Original Image Bit Process' CDs are "a big improvement over media without the process.  I have no personal opinion (or impression) since I have few 'Before' examples and those LPs.

FWIW, I have an original 1975 Carlos Kleiber Beethoven 5th LP, (to which I haven't listened in years).  I believe this was recorded on analog tape.  CD versions today claim to be Original Image Bit Processed;  presumably the tape recordings are converted at a higher bit rate and down converted to 16/44.1 with the advanced dithering process.

It's entirely possible, in fact expected that the "Original image bit process" CDs are a big improvement but not because of that process, just because it was remastered and then distributed digitally. Media without that noise-shaped dither process (EG. Just with standard TDPF dither) would be audibly indistinguishable. Consider that the noise floor of studio tape recorders in the '50s was at around -50 to -55dB, by the mid '70s that was down to around -68dB and at their peak in the 1980's, down to around -70dB to -74dB. CD using standard dither has a noise floor around -92dB and with noise-shaped dither around -120dB (in the critical hearing band). So that's dither noise that's about 16 times and 316 times respectively below the noise floor of the tape they're recording from and therefore in both cases completely inaudible.

As for the Kleiber performance, I seems almost OVERLY dramatic, something that would probably have surprised Beethoven himself ... but I'm not a musician or any sort of musical expert, so what do I know? 

Interesting question. There's no way to know of course and I would guess similarly to you but even if he were surprised, I think he would have been supportive, especially given the context. Beethoven obviously knew his 5th would be surprising, if not shocking and reports suggest he may have deliberately emphasised that fact. Holding the baton raised for several minutes until the audience was silent, to increase the impact of the opening phrase, baring in mind that prior symphonies (by say Haydn or Mozart) tended to start quiet and build to a climax. So Kleiber being overly dramatic is arguably more true to Beethoven's intention and the experience of concert goers of the time, given today's context where it's such a standard of the repertoire that it's no longer surprising/shocking.

Like Water Lily Acoustics.  Back in the 90s, a local iconoclastic stereo dealer didn't buy the marketing nonsense and did a demo of one of their recordings and a Decca recording to prove the point.  His advice was to ignore audiophile claims, which seems counterintuitive.  That dealer is still around, unlike most others from the time.

Actually I don't recall hearing of Water Lily Acoustics, I was thinking more of Sheffield Labs but I know there were a few others over the years and reading up on them, yes, exactly like Water Lily Acoustics. And, I'm glad that dealer is still around, good advice!

[aukhawk]

Even the best possible recordings sound different than live performances.  I have heard live recordings of performances I attended.  They sounded different.  The only way to have a live experience is to attend a live performance.

As a colleague of mine once said - "I don't go to concerts any more - not enough top"   
Like me he was a sound engineer working for a large broadcasting company.  At the time he was a bit senior to me, a music specialist, and a musician himself (playing in an amateur string quartet).  So it was an 'interesting' remark (and much-repeated by other colleagues down the years). 
Really of course it was a comment on the studio monitoring conditions that we worked in day in day out - speakers which were by design a bit merciless at the HF end.

[Kalevala]

As a colleague of mine once said - "I don't go to concerts any more - not enough top" 
Like me he was a sound engineer working for a large broadcasting company.  At the time he was a bit senior to me, a music specialist, and a musician himself (playing in an amateur string quartet).  So it was an 'interesting' remark (and much-repeated by other colleagues down the years). 
Really of course it was a comment on the studio monitoring conditions that we worked in day in day out - speakers which were by design a bit merciless at the HF end.

Which speakers (I suspect various ones over the years) did your employers use?

K

[Spotted Horses]

As a colleague of mine once said - "I don't go to concerts any more - not enough top"   
Like me he was a sound engineer working for a large broadcasting company.  At the time he was a bit senior to me, a music specialist, and a musician himself (playing in an amateur string quartet).  So it was an 'interesting' remark (and much-repeated by other colleagues down the years). 
Really of course it was a comment on the studio monitoring conditions that we worked in day in day out - speakers which were by design a bit merciless at the HF end.

Reminds me of a remark by Mark Anstendig, an idiosyncratic poster to one of the classical forums that preceded the current lot, to the effect that he had to wear earmuffs when attending SFO concerts because the hall had been designed to sound like CDs (which he loathed).

My own impression is that typical recordings on typical equipment have too much top, and I have an EQ setting on all of my playback software with a gentle rolloff (maybe half a dB per octave) across the entire audio spectrum.

[Fëanor]

Reminds me of a remark by Mark Anstendig, an idiosyncratic poster to one of the classical forums that preceded the current lot, to the effect that he had to wear earmuffs when attending SFO concerts because the hall had been designed to sound like CDs (which he loathed).

My own impression is that typical recordings on typical equipment have too much top, and I have an EQ setting on all of my playback software with a gentle rolloff (maybe half a dB per octave) across the entire audio spectrum.

If you EQ then gentle roll-off is the way to go.

I use MathAudio's Room EQ in my Foobar2000 player.  It has two setting, "Bright" which is flat, and "Neutral" which has a gentle roll-off.  I always use the latter.

Most speakers, used without EQ, are designed to have gradual roll-off of highs in a typical listening room with typical reflections take into account. ...e.g.

[aukhawk]

Which speakers (I suspect various ones over the years) did your employers use?

Searching for an image of the then-ubiquitous LS 5/1 - I had to raid the British Science Museum 
The less said about these awful studio monitor loudspeakers the better.  You wouldn't want speakers sounding like this at home.
Towards the late '70s the design evolved (5/5, 5/6 and so on) but it was not until the 5/8 in the mid-'80s that the BBC gave us a monitor that I could really enjoy listening to - by then of course we had our hands full with the transition to digital - 'interesting times'.


Science Museum Group. BBC LS5/1 Studio Loudspeaker with AM8/1 Amplifier. 2012-5118/308 Science Museum Group Collection Online.
https://collection.sciencemuseumgroup.org.uk/objects/co8355784/bbc-ls5-1-studio-loudspeaker-with-am8-1-amplifier

[Valentino]

My own impression is that typical recordings on typical equipment have too much top, and I have an EQ setting on all of my playback software with a gentle rolloff (maybe half a dB per octave) across the entire audio spectrum.

Measured in the listening position my speakers drop about 7 dB from 100 Hz to 10 kHz. In addition I have pulled the presence area back one dB or two.

[StudioGuy]

As a colleague of mine once said - "I don't go to concerts any more - not enough top" 
Like me he was a sound engineer working for a large broadcasting company.  At the time he was a bit senior to me, a music specialist, and a musician himself (playing in an amateur string quartet).  So it was an 'interesting' remark (and much-repeated by other colleagues down the years). 
Really of course it was a comment on the studio monitoring conditions that we worked in day in day out - speakers which were by design a bit merciless at the HF end.

It maybe that the "studio monitoring conditions" were part of the problem but only a part, there are various other factors. In addition to the loss of level proportional to distance (roughly according to inverse square law) we also have the loss of high freqs due to "air damping", which is proportional to both distance and frequency. We will lose 8kHz content (assuming 20°, 45% humidity) at the rate of 3.5dB per hundred feet, which increases to about a 12dB for 16kHz content. In addition to that, in a concert hall the audience is also going to hear a relatively high level of the concert hall's acoustics (reflections/reverb), which due the longer distance (more air damping) and absorption of the walls will have very little content above about 8kHz and pretty much none at all above about 12kHz. A musician on the other hand will be hearing their instrument from just a few inches away, there will be virtually no air damping effect and they'll also hear a far lower level of reverb (relative to the direct sound). So a musician will hear (and expect to hear) a far greater amount of HF than will the audience who are many more feet distant.

When we record an orchestra (or other ensembles) the mics will typically be far closer to the orchestra than the audience, the main mic array just above the conductor for example, and in addition, various "spot mics" just a few feet away from individual instruments (or groups of instruments). Being far closer to the instruments the mics will pick-up far more HF content than most of the audience will hear and be closer to the sound expected by the musicians. So, the sentiments of "not enough top" are not uncommon among musicians (although not also being engineers, they may not express it exactly that way), while the opposite sentiments of "too much top" are not uncommon among some consumers.

The above, plus the inevitable timing discrepancies when using multiple mics, was the rationale behind the tiny handful of audiophile labels like Sheffield Labs (and Water Lily Acoustics) just using a stereo pair, nearer to the position of an ideal audience seating location. Unfortunately, that rationale fails because it almost completely ignores psychoacoustics, the fact that the sound entering our ears is significantly different to the sound we "hear" (perceive/experience), which the multiple mic approach accounts for.

In addition to the psychoacoustics effects such as Selective Hearing (the "cocktail party effect"), there are numerous other factors that affect our perception of a live gig vs a recording, for example the effects of expectation and anticipation of the audience when queuing, taking their seats, waiting for the performance to commence and then seeing a famous orchestra/musicians in person. We obviously can't record or reproduce any of these factors with an audio recording, all we can do is make the recording as subjectively good as we can and somewhat more "vibrant" than was actually the case, to hopefully provide an enhanced illusion that somewhat compensates for the psychoacoustic effects and other factors we can't record.

[Fëanor]

...
When we record an orchestra (or other ensembles) the mics will typically be far closer to the orchestra than the audience, the main mic array just above the conductor for example, and in addition, various "spot mics" just a few feet away from individual instruments (or groups of instruments). Being far closer to the instruments the mics will pick-up far more HF content than most of the audience will hear and be closer to the sound expected by the musicians. So, the sentiments of "not enough top" are not uncommon among musicians (although not also being engineers, they may not express it exactly that way), while the opposite sentiments of "too much top" are not uncommon among some consumers.
...

I've only ever been a "consumer"; I'm not a musician and never have been.  For decades I've found a too high proportion of (Classical) recordings to have "too much top", too bright, too hot or whichever term serves.

I can assure you recording and mixing experts that I have no interest being amongst the performing musicians.  I want that "6th row" perspective, no matter whether it orchestral, chamber, or choral.

[StudioGuy]

I've only ever been a "consumer"; I'm not a musician and never have been.  For decades I've found a too high proportion of (Classical) recordings to have "too much top", too bright, too hot or whichever term serves.

I can assure you recording and mixing experts that I have no interest being amongst the performing musicians.  I want that "6th row" perspective, no matter whether it orchestral, chamber, or choral.

No one mixes in a way that puts you (the listener) "amongst the musicians", we only record and mix in a way that has somewhat more top end and feels somewhat more "present". In addition to the "spot mics" and the main array we will have room/ambience mics that are much further away than the ideal listening position and we mix these all together according to subjective taste. It's somewhat of a balancing act, between what the sound would ideally be like at the ideal audience listening position and what it would be like much closer, with the clarity/detail and frequency content one would expect nearer to say the conductor's rostrum.

To a certain extent this produces a sound more similar to what one might expect to perceive, if one includes psychoacoustic effects such as "Selective Hearing". It's an illusion though and one that is applied entirely subjectively, so it will work better according to some consumer's perception and preferences than others. However, it was mainly Decca who pioneered this approach in the 1950's and then other labels either copied or were influenced by that approach or arrived at it independently and by the mid/late 1960's pretty much all the mainstream classical labels were doing this simply because the market dictated it. The sales using this approach were higher and the plaudits greater. In other words, those for whom this approach was better/preferable was a far larger group than those for whom it didn't work so well.