American sound quality - what went wrong?

[mathias broucek]

Last weekend I listened to the wonderful Reiner Heldenleben.

It comes from 1954 and yet still sounds amazing. His Concerto for Orchestra also sounds fantastic.

So, why do many US recordings from the 1960s and 1970s sound so terrible?

[Stunsworth]

My experience is that many of them are very close miked.

I would guess because it was cheaper - balancing errors could be fixed later on in post production, or because that was the 'immediate' sound that the consumers liked and sounded initially impressive.

Thinking about it, although threy may have been mixing desks to cope with a lot of mics, the recorders with a similar number of tracks weren't available at the time, so rather than the first suggestion I'll settle on the second.

[mathias broucek]

.... or because that was the 'immediate' sound that the consumers liked and sounded initially impressive.

IIRC the Beach Boys aimed for a sound that was good on car radios and other cheap equipment. (Of course their creative genius only had one working ear, in any case....)

[Gordon]

I'd agree with Stunsworth on this one. Most early 1950s to say 1960 mastering was done with simple 2 or 3 channel tape machines with simple mixers. Mercury and RCA in the US did this using Ampex 1/2 inch tape or film recorders. The US companies tended to favour spaced omnis approach whereas EMI still tended to Blumlein and Decca used a compromise in their famous 3 mic Tree.

The trend everywhere from around 1964 was for multimicing but with not too many of them. Phase 4 was Decca's 20 channel system and EMI had their Studio Two series. As time went on tape machines with 24 or even more tracks were available but were these fully used for classical recordings? Look at the photos for the Decca Ring recordings for example and it is easy to see lots of mics about but the tape machines look as if they were mixed in session to 2 track stereo. They definitely has separate mixer sections for orchestra and singers.

If everyone was using the same technology why were American recordings worse? It what way are they bad? Well, US pressings of LPs were not thought to be as good as European or Japanese but I don't think that's what MB means. And then there was the dreaded RCA Dynagroove of course. I think it has to do with constraining ambience and confusing acoustic background when multiple close mics are combined.

The point about making music sound good on radios is a good one. We might complain in the UK about FM processing in the US it is even worse. Optimod was developed for the US where FM stations have to compete on loudness and coverage in a very congested waveband and so fine tuning of the modulation including compressing the sound dynamic range were all done. A system that CBS came up with called FMX was tried as well as Dolby noise reduction.

FM is a very complicated process and needs very careful management to keep clean. Broadly speaking it means keeping the volume down but that doesn't give as good coverage.

[Ferretfancy]

Gordon

Decca certainly mixed directly to two track stereo, certainly until the 1970s. James Lock gave us an interesting talk at the BBC's Wood Norton training school about this philosophy. There was some use of extra tape machines for off stage effects etc. but these were always cued in to the live mix at the sessions, and not added at post production.

Dynagroove was not part of RCA's studio recording technique,it was actually a device for use in the disc cutting process. The geometry of the pickup arm and the stylus configuration in use inevitably produce distortion on playback, especially at the end of the LP side. To combat this tracing distortion, Dynagroove added a signal to the disc cutter head which was designed to be out of phase with the distortion components, thus cancelling them out when the disc was played. The problem was that this made assumptions about the dimensions of the stylus used for playback at a time when there were numerous types of elliptical stylus, all of which could give different results in real life.

[Gordon]

There are several ways to degrade audio quality in tape masters, assuming the machines are maintained and set up correctly for the tape stock in use:

early tape was not too good - low coercivity, low remanence, large grains so giving lumpy noise, distortion,
use low linear tape speeds [15 instead of 30 ips] for economy but lose dynamic range and noise performance, having said that 15ips was standard from the 50s, but the point is still valid,
overdrive tape and get distortion - useful to give a brightness to the sound, Decca did this judiciously - and print through
underdrive tape and get too much noise. Enter Dolby.
use multi-track machine with standard tape width - ie 1/4 tape with >4 tracks - and get noise from tracks that are too narrow, not to mention crosstalk. With 16-24 tracks you need 1" or 2" tape. Tricky to edit with a razor blade!! Anyone remember Edivue?? [yes I know that was more to do with video tape - Ah! those quadruplex leviathans].

If that's not enough, mechanical factors such as wow etc from eccentric capstans or poor motor torque control and flutter from friction, guide roller bearings etc and non-constant speed reel end to reel end. It's miracle they worked at all!

[Eine Alpensinfonie]

Some of the CBS/Sony Bernstein recordings left a great deal to be desired with close miking. The Bernstein "Der Rosenkavalier" was a late John Culshaw recording, but it sounds as though there was some kind of post-recording tampering, for it lacks the depth of sound that made the same producer's Decca recordings so special.

[Ferretfancy]

Decca lent their team to CBS for the Rosenkavalier, and also for the Falstaff with Fischer-Dieskau. I don't think any of them were happy with the results. I had an acquaintance who had been a disc cutting engineer for Decca at the time when they had a close tie up with RCA. Apparently discs for the American market on the RCA label were recut with an upper frequency boost to please American tastes in loudspeaker response.
I find reissues which were originally on CBS easier to listen to on CD, with more depth of sound than their LP equivalents.

[cloughie]

Decca lent their team to CBS for the Rosenkavalier, and also for the Falstaff with Fischer-Dieskau. I don't think any of them were happy with the results. I had an acquaintance who had been a disc cutting engineer for Decca at the time when they had a close tie up with RCA. Apparently discs for the American market on the RCA label were recut with an upper frequency boost to please American tastes in loudspeaker response.
I find reissues which were originally on CBS easier to listen to on CD, with more depth of sound than their LP equivalents.

Ferret you beat me to it. To my ears CBS and RCA recordings on LP have been considerably improved in the remasters for CD - Bernstein, Walter, Szell, Leinsdorf, Ormandy all sound less 'spiky'. Also many of the Mercury issues which came out in inferior pressings on bargain Philips labels - some were swingers, others deteriorated in sound towards the end of sides probably because of too long sides. Many eg DetSO Paray ans MinSO Dorati have emerged on superb CDs.

[richardfinegold]

Some of the CBS/Sony Bernstein recordings left a great deal to be desired with close miking. The Bernstein "Der Rosenkavalier" was a late John Culshaw recording, but it sounds as though there was some kind of post-recording tampering, for it lacks the depth of sound that made the same producer's Decca recordings so special.

Many of the Bernstein recordings from that erea were indeed dreadful. I have an lp of Mahler's 7th that is unlistenable. Comparing it to the recent Sony remastering of the Bernstein/Mahler/NY Phil box is a revelation. The remastering sounds like a real orchestra playing in a real space, not a freaking telephone booth (as the lp suggests).
I read recently that Columbia mastered many of their 1960s recordings as they would for pop music. They assummed that people would be listening on AM radios in their cars and compressed the sound accordingly.

[mathias broucek]

Ferret you beat me to it. To my ears CBS and RCA recordings on LP have been considerably improved in the remasters for CD - Bernstein, Walter, Szell, Leinsdorf, Ormandy all sound less 'spiky'. Also many of the Mercury issues which came out in inferior pressings on bargain Philips labels - some were swingers, others deteriorated in sound towards the end of sides probably because of too long sides. Many eg DetSO Paray ans MinSO Dorati have emerged on superb CDsat'.

That's very true!

[LeMartinPecheur]

Back in the early 70s when I first got interested in music and hi-fi, I heard somewhere that the characteristic CBS LP sound (scrawny, brittle, very bass-light) was partly explained by American hi-fi manufacturers tending to produce speakers with a dull treble and too much bass, so that more musical classical recording engineers made bright, bass-light LPs to compensate. Anyone know if this might be true?

I've seen plenty of postings from US classical lovers which suggest that nowadays, ideas of the proper sound of high-end hi-fi reproduction both sides of the bond are equally dedicated to neutrality. In which case we might expect CD transfers of CBS master tapes from that period to be a lot more to UK tastes, after making all possible allowances for close-miking and other such ineradicables. My impression is that this is true but I still tend to keep clear of most ex-CBS orchestral recordings, but I'd be grateful for others' views. Ought I to retune my prejudices??

EDIT Sorry, I'd missed ferretfancy's post #8 on Rosenkavalier which seems to confirm at least the treble end of my story.

[Ferretfancy]

LeMartinPecheur

I'm sure you are right about the American penchant for loudspeakers with dull treble. As i mentioned in an earlier post, Decca cut their LPs destined for the American market with a boost at about 7k to compensate. A good example on LP was the wonderful Pierre Monteux recording of the Enigma Variations issued by RCA in the States. The domestic pressings here were warmer in sound
with more sense of the recording acoustic, complete with a distant whistle during the penultimate variation - a train perhaps ?

[Gordon]

Having listened to a few CBS originated CDs this AM of Szell [his Mozart, Haydn and Beethoven] and Bernstein [is Mahler with NYPO] I can hear no serious problems with the source material. They were all recorded before the main age of multitrack [beyond 4] but probably did use multimoics [sorry, I live in Hampshire!].

Of course the remastering will have done something to the original tapes [if that is what was used and, for this era, I think this is so] to reduce tape noise [using CEDAR eg] and to perhaps adjust dynamics. Who knows. However all these CDs are clean and far from being scrawny. I deduce from that that the master source recording was done well, for the time at least, and the complaint about poor sound relates primarily to the vinyl.

I used to have some of the above on LP and so can no longer check and memory is a deceptive thing. Some of those LPs would have been almost certainly pressed first time around [ie not re-issues on budget labels, often done from original metals] in the UK by Philips and/or EMI on their own labels around the pre-1960 era at least [I remember Szell's from the US Epic label on EMI Columbia] until CBS [and RCA too] started to arrange for the pressing of its own. All EMI's Columbia issues from the CBS catalogue were given EMI catalogue numbers [SAXxxxx stereo] and the pressing matrices had EMI labels [YAXxxxx for stereo] masters cut at Abbey Road. So when we assert that US sound is poor is the UK LP as bad? I don't remember any such complaints.

Now the question is: where did they get the master pressing tapes from? Once a master is approved by artists and production staff a copy is used to send to the pressing plant for cutting and that may or may not have been doctored to get the sound required on playback on domestic equipment of the time and may be subject to market tastes as has been decribed in previous posts. On the other hand the cutting engineer who produced the pressing master could also put in his pen'orth. In the UK/Europe the record companies had the attitude that what was on the tape should end up on the disc. It is possible that master metals from the US were used in which case the doctoring could also be audible over here. Were US masters cut differently for domestic market than export? Sounds as if they were.

If the agument about market taste and also HiFi manufacturers' predilection for a certain sound are true why did they not end up doctoring CD for that same taste in sound/reproduction equipment? Or did they give that up by the time CD came along? Is there any evidence that US CDs are also lacking to UK ears? I have quite a few CDs obtained from the US and labelled as being made there and I hear no difference. Vinyl, being an essentially analogue medium, is more vulnerable than CD to processing adjustments and is/was actually subject to deliberate adjustment to deal with some of the problems of reproduction [eg watching the depth of cut by limiting the L-R difference signal]. In the 50s and 60s Abbey Road had a magic box covered in knobs that was used in the cutting room to prepare the master tape transfer to disc! If the tape sound was to get to disc unchanged what were those knobs for?

Here's a photo of the front panel - there was a stereo version too. Note the labels - you could do a lot of damage with this device indeed it was banned from the studios and confined to cutting rooms until the Beatles demanded it and got what they wanted as they always did.



Here is a cutting room at Abbey Road with the device and a limiter underneath and note the BTR1 tape machine.

[Gordon]

.... complete with a distant whistle during the penultimate variation - a train perhaps ?

Well it was recorded in Kingsway Hall [Kenneth Wilkinson] with the ever present tube beneath. I doubt however that the faint bleat of LT tube trains would get into the studio. More likely to come from the street outside, it was June. No surface steam trains for miles - Charing Cross some distance away with trains on high bridges over river. 1958 was well in the steam era though. Must listen to that recording again, never noticed the whistle.

PS: Done so and there's a bird tweeting, starting at about 28 seconds!! The summer probably had doors/windows open and let the creature into the Hall!! Not as bad a drunks from the Mission next door. However at about 1.03 into the variation there is a single short toot audible too for all the world like a steam engine or a penny whistle out in the lobby!! I wonder how many other recordings have these extra-musical contributions. Tell you what though doesn't that tape hiss.