 
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
|
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
173.52.249.160
In Reply to: RE: The 24/96 FLAC download 1st movement cadenza is at 445, so is the SACD's CD layer posted by David Smith on September 13, 2015 at 07:27:41
>Please provide timings of the A's you measured at 446.77 and explain what
>you mean by "appear to be close"? Please also provide the timing of the
>first movement from the beginning of the first note to the end of the last
>note, not including hall decay.
Will do so after you have answered my question posted earlier about how you
measure frequency on the recordings you mention.
John Atkinson
Editor, Stereophile
Follow Ups:
I used the Clear Tune app for frequency readings and a Korg chromatic tuner to check for agreement.
Dave
> I used the Clear Tune app for frequency readings and a Korg chromatic
> tuner to check for agreement.Thank you.
The A at the start of the cadenza on the 24/96 FLAC download and the CD
layer of the SACD is tuned to A=442.4Hz, whereas the orchestral A's
throughout the first movement are at 446.7Hz. (In each case the frequency
is that of the highest-level bin in the FFT.)The D at the first violin entry is at 152.3Hz rather than the correct 147Hz.
I checked this with the Tuner Lite app on my phone, which indicated that
the note was sharp of D at A=440 pitch.It isn't possible to download the file from YouTube that you posted, as it
is copy-protected, but when I have a moment, I will capture the audio
track to a recorder so I can perform FFT analysis on the first note of
the cadenza. However:The start of the final chord of the YouTube version starts at 22:24, ie,
22.20 after the orchestral entry at the start of the movement, for a total
file playing time of 22:36.The final chord of the FLAC version starts at 22:10, ie, 22:06 after the
orchestral entry, for a total file playing time of 22:23.This ratio between the two timings is very close to that required to
flatten A=445Hz (as marked on the master tape box) to A=440Hz.This suggests that the version posted to YouTube has been slowed-down
to correct the pitch.
John Atkinson
Editor, Stereophile
Edits: 09/13/15 09/13/15
SACDs marketed online don't include music samples (in contrast to CDs and MP3 files). So I'd be curious to hear what you and John have found in the SACD version, if you can upload a snippet from the recordings in question. Thanks much.
I did order the standard Redbook CD (the SACD is too expensive).... It would be a good addition to my collection, especially if the pitch/speed is OK. I will mention whether this version has the pitch/speed problem.
Hi John,
Thanks for answering and I appreciate your willingness to discuss.
A couple of things about your post above. If you read my OP, the A's I am referring to are specifically open-string A's on the solo violin because there is no left hand involved in playing them and thus they will be consistent at the tuned pitch. Because they are played in the cadenza it allows us to hear the pitch alone unobscured by other instruments.
The first note in the cadenza is not that A, it is an octave lower (~220 Hz) and played on the G string with the left hand raising the pitch to A. So the precise pitch is a function of player, not the tuning of the instrument. As such it is not a good indicator of pitch of a recording.
Other orchestral A's are in the same boat, the precise pitch is a function of the player and there is a degree of human error/choice. As such it is not definitive proof of anything in terms of the pitch of the recording. It's also worth noting that a good musician would not play all A's at 440 Hz, that would be out of tune in a D chord, for example (should be higher), and perfect equal temperment at A 440 is not desireable in an orchestra or on a piano.
On the other hand, the open-string A's at 17:43 of the first movement in the cadenza (using the YouTube timing) will be at a pitch that was set in the tuning of the instrument and are not subject to human error or variation when played.
I'm not sure what the basis of your claim that the YouTube clips have been slowed down (both of them) is, both were posted before John's article was published and could be easily referenced to a consumer CD if someone here could help out, though I have one on the way to me. What is the basis of that claim? I'll add that on my F4 side A the last chord begins right at 22:20 after the first note.
If the SACD is in fact somehow faster than previous versions that ought to be easy enough to verify, if you could confirm the 22:06 timing hopefully we can find someone else to corroborate. John's article references everything released after the pre-1973 LP's, not limited to the SACD. John also claimed (in print) that the timing discrepancy was more than a minute for the first movement - if you didn't see his letter in The Strad - which he later admitted was made without factual basis.
I'll get some clips and examples together to show what I'm talking about with the A's.
Dave
Aside from a possible tendency to play out of equal temperment, the human ear wants to hear notes in the higher treble range sharper than in tune, resulting in "octave stretching". I've seen analysis of recordings of famous soloists that demonstrates this. But that alone probably wouldn't explain a gap from a=440 to a=446.5.
More to the point, it appears the CD, the SACD, and my f2/f2 copy of the LP all time at about 22:23 for the first movement, excluding all dead time at the beginning and end. (My turntable is accurate enough for these purposes.) So there shouldn't be any significant pitch discrepancies for these versions.
> Thanks for answering and I appreciate your willingness to discuss.You're welcome.
> The first note in the cadenza is not that A, it is an octave lower
> (~220Hz) and played on the G string with the left hand raising the pitch
> to A.Yes I know, I played violin in youth orchestras through high school and
occasionally played until I was in my mid-20s when I switched to
recorders and viola da gamba in order to perform in an early music
consort. But it is difficult to believe that a violinist as skilled as
Oistrakh would have played that stopped note out of tune.And I did try to distinguish when I was referring to pitches related
to A=440Hz and the actual note A. Obviously I did not do so with
sufficient clarity.> So the precise pitch is a function of player, not the tuning of the
> instrument. As such it is not a good indicator of pitch of a recording.I disagree. The difference between the overall pitch of the recording in
question and a true A=440Hz is greater than the margin of error in the
FFT analysis. There is also the fact that the master tape box itself is
marked "A=445Hz."> On the other hand, the open-string A's at 17:43 of the first movement in
> the cadenza (using the YouTube timing) will be at a pitch that was set
> in the tuning of the instrument and are not subject to human error or
> variation when played.Of course, and I suspect that Oistrakh retuned before recording the
cadenza. Even so, the pitch on the first note on the version John Marks
was referring to is sharp.
> I'm not sure what the basis of your claim that the YouTube clips have
> been slowed down (both of them) is,As I said, the YouTube version of the first movement is longer by almost
the exact amount to reduce the pitch from A=445Hz to A=440Hz. For me that
is the clincher.
John Atkinson
Editor, Stereophile
Edits: 09/14/15
"There is also the fact that the master tape box itself is marked "A=445Hz."Except that that doesn't explain the YouTube clips and presumably CD issues. If the machine that made that master was in fact slow it does not necessitate reissues being sharp.
"As I said, the YouTube version of the first movement is longer by almost the exact amount to reduce the pitch from A=445Hz to A=440Hz. For me that is the clincher."
That should be, really, all that is necessary. Why wasn't timing info included in the original article? Is the SACD timing different than CD versions? Having said that, your claim that it was "slowed down" implies that the YouTube version began at a faster speed, that is what I'm asking about in terms of your claim.
Dave
Edits: 09/14/15
> > There is also the fact that the master tape box itself is marked A=445Hz.
>
> Except that that doesn't explain the YouTube clips and presumably CD
> issues.No-one has analyzed the CD reissues.
> If the machine that made that master was in fact slow it does not
> necessitate reissues being sharp.Why not? All the reissues that have been analyzed _with the exception of
the YouTube clip_ are sharp of A=440Hz and faster. It looks to me as if
someone slowed-down the clip before posting to YouTube in order to bring
it back into correct pitch.> your claim that it was "slowed down" implies that the YouTube version
> began at a faster speed, that is what I'm asking about in terms of your
> claim.That is correct. The YouTube version is both slower and lower in pitch
than the reissued versions that John Marks wrote about.
John Atkinson
Editor, Stereophile
Edits: 09/14/15
the idea that someone deliberately altered a youtube clip in order to lower pitch to exactly A440 seems a tad odd to me. Ever run across something like that before?
> the idea that someone deliberately altered a youtube clip in order to
> lower pitch to exactly A440 seems a tad odd to me. Ever run across
> something like that before?
No. But what other explanation could there be for both the change in pitch
and the increase in the track's timing?
John Atkinson
Editor, Stereophile
nt
"No-one has analyzed the CD reissues."
Considering the breadth of the claims in the article that's a rather glaring omission.
Dave
> Considering the breadth of the claims in the article that's a rather
> glaring omission.
With all due respect, you're very quick to criticize and call others frauds
but slow to actually provide more evidence. I have analyzed 2 versions now,
John Marks several, but you just one, the slowed-down YouTube clip. I
suggest you purchase the CD and SACD reissues and get back to us with
what you find before you continue questioning others.
John Atkinson
Editor, Stereophile
Actually I have analyzed the two different YouTube versions (though they are as close to identical as I can discern) and my early LP. I have explained why all of those are at A 440, and the YouTube versions are there for anyone to corroborate and you seem to have. Your repeated claims that they have been "slowed down" (identically and identically to my LP) remain without evidence. If the SACD is in fact fast there is no evidence that those versions were at one point fast as well. I have a CD on order, should be here this week. I have not found an SACD for sale. John Marks' comments were not limited to the SACD, were they I would not have commented as I don't own it. He spoke broadly about versions of the recording since 1973, which the YouTube clips (one of which was provided by Warner Music) are in contradiction to his claims, hence my comments on the subject. Furthermore he made claims without providing the most basic and definitive evidence, the timings of the music. The methods he used and you used are not definitive for the reasons I have explained, in any case it's really simple, if the SACD timing is different than others, they're different! If they are not, they are not. Your comments in the Stereophile column about the timing were hypothetical and did not refer to actual timings.
If the SACD is different I will retract my comments that apply to it and apologize to John to that end. My use of the word "fraud" was harsher than I intended, but John's refusal to address fundamental flaws in some of his claims and other incorrect statements certainly give the appearance of a lack of concern for the truth.
Dave
> Your repeated claims that they have been "slowed down" (identically and
> identically to my LP) remain without evidence.
Only if you ignore all the measurements that I posted and the fact - fact,
please note - that the master tape for the reissue was marked "A=445."
And when you say that
> The methods he used and you used are not definitive for the reasons I
> have explained...
I am afraid that all you are doing is revealing your own lack of
understanding of Fourier analysis.
John Atkinson
Editor, Stereophile
What measurements did you take of the YouTube clips before they were posted?
It is a simple task to adjust a tape speed, if the open-string A is at 445 on that master there is no reason that subsequent issues couldn't be at any pitch.
Dave
Atkinson did post a link to a snippet of the pitch elevated version and speed corrected version (link below). This is the only evidence I've come across of the pitch being elevated. (I think the lack of examples of the pitch-elevated version on the web was what made you jump to the "fraud" conclusion. Too bad SACDs don't include listening samples like CDs or MP3s.)
The snippet totally changed my perspective on this issue. Prior to that, I was on the fence, I thought the YouTube clips were at the original speed. Now I think they might be at the "fast" speed, pitch adjusted to A=440 in software. For the speed corrected part on the Stereophile snippet sounds "slower" to me, relative to the passage on the YouTube clips. (It also sounded a lot more realistic to me.) The non-speed corrected part sounds similar to me, speed wise (in spite of the pitch differences). But the playback time on the snippet is too short to be certain on this.
Whether the CD was done this way remains to be seen. As I stated elsewhere, I want to compare the CD to the original vinyl (both are at A=440), mainly whether the playback times are the same or different.
The bottom line, I think this issue might be more convoluted than a lot of us here initially suspected.
After reading the linked article, it appears that there may be pitch problems on some LPs as well (or a bad turntable on an LP rip).
All the more reason to identify each LP pressing, and count revolutions between two known points (start of first and last chords) of the first movement when these disks are actually being played.
It is not possible to adjust the timing separately from the pitch without loss of sound quality. This is even true with solo instruments, but definitely true when multiple instruments play together. It could be the YouTube MP3 artifacts mask these defects. Accordingly, the use of any MP3 or otherwise deliberately mangled (compressed) audio should be forbidden as the basis of intelligent discussion. MP3 is for deaf idiots who are walking on the sidewalk and about to step off the curb into the path of an oncoming truck.
Tony Lauck
"Diversity is the law of nature; no two entities in this universe are uniform." - P.R. Sarkar
As much as it pains me to say so. ;-)
Notice that the first movement timing on the EMI CD is also given as 22:36 (ed.: or 22:35, according to amazon.com). On my Angel LP, with stampers F2/F2, the first movement times at exactly 22:23. (I have a modest belt drive turntable, but it does spin at very close to 33 1/3 rpm, fortunately for me. Kids' college tuition comes first. Donations happily accepted. ;-)) Ed. no. 2: Having said that, the youtube audio linked by Dave gives a timing of 22:36, but there is between 3 and 4 seconds of dead time at the beginning and 8 seconds at the end. I excluded that dead time in my timing of the LP.
I don't have your sophisticated equipment, but my LP is noticeably sharp, I estimated at a=443 for at least a portion. I have my suspicions as to what happened here, but I'm not going to say and get into arguments. I want to hear the original source tape.
Edits: 09/13/15 09/13/15 09/13/15
Post a Followup:
| FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
||||||||||||||
This post is made possible by the generous support of people like you and our sponsors:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2025, Audio Asylum, all rights reserved ]
