Thread: Recording Technique: Harm but No Foul

Posts: 3

Post by Dan Popp November 14, 2003 (1 of 3)
I had some thoughts in response to Beardawg's review of the Beethoven/Karajan recordings that might be more fit for the forum.

BD writes:
I’ve read on a quite a few places that record companies tend to pump up the volume for the SACD layers and make them sound ‘better’.

I'm really skeptical about that. There is an absolute limit to digital recording levels - PCM or DSD. So the Peak level can't be increased. The Average level could be increased to get greater perceived volume (what I call "density") but the audience for classical music would revolt at any sign of compression, and rightly so. I don't believe there's any chicanery going on here. I think it's probable that you've got two different types of recording coming out of 2 different sets of D/A converters. All things being equal, the SACD layer should be both louder at the peaks and softer in the valleys. But all things are not equal - that's the point of SACD, right? 8-) My guess is that this is just a byproduct of trying to A/B two different sets of converters playing two inherently different signals.


Beardawg wrote:
Recording (re-mastering) is a much more complicated issue. I’m pretty sure there was far too many microphones involved and the stereo image was created artificially in the control room.

IMO, it's impossible to make a stereo recording of an orchestra that will sound right to everyone on all systems. Perhaps the "purest purist" (!) approach would be the "dummy head" technique: 2 mics imbedded in a wooden sphere or dummy head where human ears would be. But there are disadvantages even to that method. The best reproduction for a dummy head recording will be on headphones; speakers throw a bit of a wrench into the works. Mono compatibility won't be perfect. And most importantly, the sound you hear in the concert hall is interpreted by your brain differently than the sound the dummy head hears, as played back in your living room. The true ambience of the hall is compensated-for during the actual experience, but not (as well) during playback of the recording.

Any technique using more than 2 mics is going to involve some mixing and panning - i.e. judgement calls by the engineers. Your phrase, "created artificially," seems to have an aura of "cheating" about it. But I think that most orchestral recordings today use a combination of (near)coincident pair for the main signal, "spot" mics for solos, and ambience mics. It's just that most of them sound a lot better to you than these apparently did.

Post by beardawgs November 16, 2003 (2 of 3)
Dear Dan Popp,

I usually don’t compare SACD and CD layers of my discs, but concentrate on SACD. For this Beethoven cycle I was intrigued to check out the CD layer, as it is my first ever Beethoven symphonies cycle with Karajan in any format. My first ever SACD was Mahler’s symphony no. 6 with Tilson Thomas, before I got my first SACD player. So I was familiar with the performance and redbook transfer before I had a chance to hear it as a SACD. The sound levels are equal throughout, and that is the disc we usually play to our friends for demonstration of sound quality differences. First CD layer, then SACD layer, and the levels are always the same! Now, to be honest, out of some 100 SACDs we have, neither me or my partner really bothered to switch between the layers and compare sound levels, but when I did that with Beethoven I was quite surprised that there was a significant difference.
Now, during the years I spent in a sound studio as a trainee music producer my tutors, experienced in pop and classical music recordings were always very aware of the fact that louder they play the recording, it tends to ‘sound’ better. It’s a little trick I also learned very well, that I’m not surprised if someone else is using it for the same reasons. And the reason I mentioned the differences of sound levels in my review was just stressing the fact, not trying to say that is good or bad thing to do. From my correspondence with DG I’ve learned that DSD and CD transfers were made directly from stereo master tapes, so I’m pretty sure there was already some sort of compression used before the final master tapes were produced. And I didn’t hear any significant difference in relative dynamic range on CD and SACD layer here, so my conclusion is that the PCM and DSD signals were just transferred in different levels.

As for stereo image, you are perfectly right – most of modern recordings are mix of different types of microphones and positioning and all those signals have to be perfectly blended in the control room. After all, that mix is what makes the difference between good and not so good recordings – natural transparency and depth of the big orchestra being the most difficult thing to achieve. Call me purist, but (and I’m talking from my actual experience) the less microphones used – better the recording. One of my favourite recordings ever are Mahler’s symphonies on DENON with Eliahu Inbal, recorded from 1985-1992 with only one stereo pair of B&K microphones. No dummy head used, but the mics were positioned up and away from the orchestra. But making recordings like that is more complicated than is seems – first of all, they have to be made in acoustically superior halls and they require hours and hours of very precise positioning, i.e. orchestra playing many days just for the sake of the sound technician. Multi miking is much cheaper solution, just drop as many mikes as you have, record separate sections of the orchestra on multi-channel tape, dismiss the orchestra and mix it all later on. That approach is also more appealing for the sound technicians, as it gives them the opportunity to “do” something, not just plug in the cables and press the record button once the microphones are set.
Back to Karajan’s set, we’re talking here of early stereo days when the fascination with the new technology was as exciting as the music itself. Keeping in mind that Herby himself was very technical man, personally supervising microphone positioning, sound, colours and perspective. Also, he himself never hided the fact that he also overdubbed certain passages and certain orchestral groups (if I remember well, second violins were mostly overdubbed) in these recordings. In the sixties that was a perfectly acceptable approach, as the recorded orchestral sound was treated as something different from the concert orchestral sound.
I wasn’t clear in my original review when I said ‘too many microphones used’… what I meant was remiking, - changing microphone positions for different sections of the same piece. It is interesting to compare early Karajan’s stereo recordings with the later ones, especially digital – with the change of the orchestral sound he was producing, the recordings were more natural and concert-like.

Post by Dan Popp November 16, 2003 (3 of 3)
Dear Beardawgs,

Thanks for the info on the '60s overdubbing. That's interesting!

One of my very first CD purchases was Von Karajan's recording of Beethoven's 9th with the Berlin Philharmonic. I thought it sounded fine then, but it sounds like nails on a chalkboard to me now. It must have been one of the very first "DDD" recordings made. I wouldn't have been interested in the older (analog source) recordings if I hadn't read your review.

Yes, I see that it's possible that the engineers transferred an analog master at different levels to the two formats. I would consider that a bit sloppy because, theoretically, you do want to use that last bit... to get the best performance out of the amplifier's signal-to-noise ratio, if nothing else. But it's entirely possible.

I am very much a purist myself when it comes to the simplicity of miking. But there is something to what I said about how the brain focuses-through some of the ambience of the hall that makes me also appreciate judicious spot miking. I have not gotten to do as much of it as I would like, so I'm no expert. I think the hardest thing to deal with in stereo is the ambience. No matter what mic setup you use, at least half of the ambient information is "incorrect" - i.e. coming from the wrong direction!

Thanks for your polite and informative response.

Closed