|
|
|
Consonant with what Seth and Dan have said, I think my best sacds have been recorded at a lower level than usual. These have the most outstanding bass and dynamic range. Mahler is a composer who's music demands this approach to avoid overload. Further, apart from a trifle - and I really mean a trifling bit of trifle here - a trifle of dryness, there was nothing I could fault with this recording heard in multi-channel. The sound stage was W - I - D - E as wide could be and regarding depth, that is an artifact of stereo recording. A listener only a few rows back from the stage would not hear in the concert hall what we have come to expect from stereo recordings. He would hear a different kind of depth - that of the hall!
In mch, that is exactly what we very handsomely are presented in this recording. That and tremendous slam and about the sharpest transients I can remember hearing. The dryness noted has a slightly deleterious effect on the sweetness of the string sound yes, but in my opinion, this is a magnificent recording. I think it is my favorite Mahler 6th. Oh yes, I listen to it pretty darned loud!
|
|
|
Post by Dan Popp March 22, 2007 (42 of 52)
|
|
seth wrote:
The low level is also a result of the audio not being compressed.
Dynamic compression can raise the average level, but lower levels do not necessarily mean uncompressed, and higher levels may not be the result of compression. Any label employing dynamic compression on classical music for SACD release is "incompetent," to use George's word. From what I can tell this kind of incompetence is rare.
All the best.
|
|
|
|
|
|
Dan Popp said:
However, recording levels are generally a *subjective*, not an objective standard. If the engineer recorded it so that it clipped (too hot -> distortion), that would be incompetence, in my book. But it would have to be very low indeed to say that incompetence was at play.
Whether the digital representation of the signal is PCM or DSD, it will have its least distortion when the recording level is closest to maximum, which I shall refer to as 0dB. Ideally, one would record so that the loudest music gets as close to 0dB as it can be without exceeding it.
I don't think the recording level is subjective at all: most people measure. You go into the recording session and get the orchestra to play the few bars surrounding the beat the hammer in the last movement, and set you levels so that loudest sound gets right up next to 0dB.
Given the large dynamic range of SACD, you can certainly record at levels that never in the loudest moments exceed -6dB. You leave yourself plenty of headroom, the noise floor is still acceptably low, and no one is ever going to say you clipped. What's the point? Safety for a recording engineer who isn't sure of the maximum level?
If you don't care for the word "incompetence," it is still a mistake -- the recording engineers sacrificed the "best," highest resolution, lowest distortion 6dB of the dynamic range.
On the low recording level disc, my SACD player no longer puts out 2V max; it puts out only 1V. I turn up the volume 6dB on the pre-amp to match the levels. The background noise is raised by 6dB right along with the music.
Think of it this way: In a 16-bit linear PCM system (RBCD audio, for example) when you are up close to the maximum signal level you have a resolution of one part in 65536. Drop the level by 6dB, and you have a resolution of one part in 32768. Etc. You are essentially recording with a 15 bit system instead of a 16 bit system.
It is just a mistake. If you liked the sound of the disc, you probably would have liked it better if the full dynamic range of the recording medium had been used, and there had been less distortion in the digital representation as well as a better s/n ratio in the analog circuits.
|
|
|
|
|
|
Hello All,
I'm the one to blame for this SACD!
I can assure you that all the bits on this disk are full, both on the PCM and DSD layers. In fact, if I hadn't "pulled down" the hammer blows significantly the apparent level for the rest of the disk would be even lower.
Getting the full dynamic range of a piece like Mahler 6 onto a recording isn't so much a question of what the technology can do (with 120 dB of dynamic range, noise isn't a problem), but more a question of whether the playback systems can reproduce this dynamic range, and if people want to playback loud enough to reproduce the dynamic range of a full symphony in their listening room.
Many if not most CD's and lots of SACD's, even of classical music, have their dynamic range significantly compressed, either with a compressor or by manual gain riding. At Polyhymnia, we generally do little or no gain riding, and no compression, preserving the dynamic range of the original performance as much as possible. That why some of our SACD's may sound softer than some others. Rest assured that we are competent, and that every master we send out is recorded at full level.
Everett
|
|
|
|
|
|
PolyhymniaEverett said:
Getting the full dynamic range of a piece like Mahler 6 onto a recording isn't so much a question of what the technology can do (with 120 dB of dynamic range, noise isn't a problem), but more a question of whether the playback systems can reproduce this dynamic range, and if people want to playback loud enough to reproduce the dynamic range of a full symphony in their listening room.
Everett
My loudspeakers are pretty good and I think I am a strong advocate of "lifelike" recording. But am I really? I like to sit in the front right corner of the second balcony of Symphony Hall in Boston. When the Hammer falls, its pretty darned loud back there (in that hall at least) but my wife, who likes to watch as well as listen sits much closer to the orchestra actually overlooking the first violins. She assures me that its much louder there than where I like to sit!
While I think I would like to have the full dynamic range of the orchestra in my listening room, there are a couple questions apart from would I really like it that loud (in the perspective of up close that is typical of most recordings).
What I really wonder (worry) about is whether my speakers can do it without damage. I expect not and so I appreciate the care you take to give us the quality product you do. As I indicated in a post earlier on this thread, most of my best recordings require that I turn up the volume a bit and then there is with respect to the scale of things, a very satisfying likeness of an orchestra in a concert hall heard in mch.
I think this Mahler recording is mostly excellent to outstanding! Thanks again.
|
|
|
Post by fafnir March 27, 2007 (46 of 52)
|
|
PolyhymniaEverett said:
Hello All,
I'm the one to blame for this SACD!
I can assure you that all the bits on this disk are full, both on the PCM and DSD layers. In fact, if I hadn't "pulled down" the hammer blows significantly the apparent level for the rest of the disk would be even lower.
Getting the full dynamic range of a piece like Mahler 6 onto a recording isn't so much a question of what the technology can do (with 120 dB of dynamic range, noise isn't a problem), but more a question of whether the playback systems can reproduce this dynamic range, and if people want to playback loud enough to reproduce the dynamic range of a full symphony in their listening room.
To put 120 dB of dynamic range in perspective in terms of the capability of home equipment:
120 dB is equivalent to a voltage ratio delivered to the loudspeakers of a million to one and a power delivery ratio of 10 to the twelfth to one. If the loudspeakers were plugged into the wall outlet instead of the amplifier, approximately 115 volts RMS would be delivered @ 60 Hz to the loudspeaker - not that anyone in their right mind would do this - and would certainly destroy the speakers. The actual power delivered would depend on the loudspeaker impedance - a frequency-dependent complex number, but would very likely exceed 1000 watts.
The voltage delivered at 120 dB under this speaker-popping level would be only 115 microvolts and in the vicinity of 0.001 microwatt. I suspect that the sound of my breathing would mask it.
Bravo to Polyhymnia for ensuring that the dynamic range is suitable given the constraints of the home listening equipment and environment. IMHO the Mahler 6 is a fantastic recording.
|
|
|
|
|
|
PolyhymniaEverett said:
I'm the one to blame for this SACD!
Getting the full dynamic range of a piece like Mahler 6 onto a recording isn't so much a question of what the technology can do (with 120 dB of dynamic range, noise isn't a problem), but more a question of whether the playback systems can reproduce this dynamic range, and if people want to playback loud enough to reproduce the dynamic range of a full symphony in their listening room.
I apologize for calling your effort incompetence, particularly since you have stated that it was a conscious choice. I have learned something about both your recording and my own preferences.
I think I qualify as a person who perhaps does not want to "reproduce the dynamic range of a full symphony" in my listening room. For me, this is not the best way to enjoy music. Since I go to concerts regularly, I am aware that there /is/ a difference in playback level, and since I get to hear a grand piano in the house on an almost daily basis, I can experience the difference anytime I need to refresh my memory.
There are not too many pieces of music that have one loud event that pushes down the other 80 minutes of sound to accommodate the quick peak level, and there are certainly few in my collection. I want no more. I don't know if Mahler intended the hammer blows to be /that/ loud -- I recall reading somewhere that it was supposed to be a big thud, a dead blow.
Some fool will put a Howitzer in there next to get still more impact. Where I went to school, one of the composition students put a separate line of music at the bottom of his score, and it remained unidentified until the final page. It was revealed to be a nuclear weapon, and it was just marked at the end with a
_
|
|
X
.
Marking it staccato and an 1/8 note was a nice touch. (It looks better in courier typeface.)
|
|
|
Post by seth March 27, 2007 (48 of 52)
|
|
georgeflanagin said:
There are not too many pieces of music that have one loud event that pushes down the other 80 minutes of sound to accommodate the quick peak level, and there are certainly few in my collection. I want no more. I don't know if Mahler intended the hammer blows to be /that/ loud -- I recall reading somewhere that it was supposed to be a big thud, a dead blow.
It's not loud -- there's a lot of impact.
The recording is on a SACD, done by audiophile engineers, so not to be rude, but they're purposely appealing to a niche market in terms of sound quality/dynamic range, though I think anyone with any audio system can appreciate the recorded quality.
|
|
|
|
|
|
Hi,
thanks for all the responses! It's always good to hear that people are enjoying listening to what we work hard to (re)create!
We could go on for ever discussing the "right" dynamic range for recordings. There's no correct answer -- it depends completely on the playback situation. Classical radio stations compress everything drastically (sometimes adding extra compression during rush hour), because their main audience is in the car, where the noise mandates a tiny dynamic range. Many people at home listen to classical music primarily as background music (though probably not a Mahler Symphony), and don't want their conversations obliterated by loud music. Even for those of us who like to listen to the full dynamic range some of the time (me included) there are times when it's not practical -- for example late at night if you have neighbors (or family). The best option would be to be able to control the dynamic range on the PLAYBACK side -- but there's very little consumer equipment that can do this, with the interesting exception of dolby digital equipped home theater equipment.
The problem with compression on the playback side is that most compressors have to carefully adjusted to deal well with different sorts of music, and tend to have the most problems with classical music.
The best way to solve this would be to record the disc at full dynamic range, but to include a gain riding curve along with it, so that the dynamic range can be sensibly reduced on the playback side as desired. Unfortunately, the only current system that includes anything like that is Dolby Digital (in a limited form), leaving us having to do it on the record side.
This means that recording and mastering engineers have to determine the best dynamic range for a particular recording, also taking into account the competition (you don't really want your discs to sound softer than someone else's, especially for popular music). I tend to think the the SACD public wants the most realistic experience -- thus the largest dynamic range. Glad to hear that most of you seem to agree!
Everett
|
|
|
Post by Phil W March 28, 2007 (50 of 52)
|
|
Everett: Could you tell us whether the Ondine/Philadelphia discs are DSD or PCM recordings? Thanks.
|
|
