Audio frequency range of LP sv. CD (YouTube)

the mix tape forum
Post Reply
13hm13
junior member
junior member
United States of America
Posts: 16
Joined: 17 Oct 2016 18:13
Location: Los Angeles

Audio frequency range of LP sv. CD (YouTube)

Post by 13hm13 » 24 Oct 2018 02:03

This video was posted on YT about 2 yrs ago.
Give it watch ....
https://www.youtube.com/watch?v=9MxJtMcRidw

I'm going to assume the above 22khz energy content is (are) harmonics.
Correct?
Please share your opinion.

Thx!


==============
Description of video posted by vlogger on YT:
This is my response to the video at https://youtu.be/4eC6L3_k_48 .
So...what happens if you take a vinyl record that was mastered on all digital equipment in 1985 with a known master recorder frequency response of 20 to 20,000kHz (16bit@44.1 or 48kHz) and look at the frequency spectrum? You'll see it here. Remember, it is impossible for there to be frequencies above 20kHz on this due to the session and master recordings. Impossible, that is unless there is something being introduced to the playback.
While true that it is possible to have frequencies above 20kHz on a vinyl LP, most of the time, that's wasted noise that you can't hear. Vinyl - like the CD - was engineered for 20 to 20,000Hz and the RIAA curve ends at 20,000Hz. (There were exceptions for CD-4 quad LPs using a 45,000Hz carrier, but those were cut at 1/2 speed (or below) to make them work and rolled off the low end at 40Hz. Plus, they were recorded at a much lower level and were not durable.)
==============

13hm13
junior member
junior member
United States of America
Posts: 16
Joined: 17 Oct 2016 18:13
Location: Los Angeles

Re: Audio frequency range of LP sv. CD (YouTube)

Post by 13hm13 » 24 Oct 2018 16:37

Sorry: that title should read Audio frequency range of LP vs. CD (YouTube)

not: Audio frequency range of LP sv. CD (YouTube)

analogaudio
long player
long player
United States of America
Posts: 4686
Joined: 06 Feb 2007 23:58
Location: Monroe NY USA

Re: Audio frequency range of LP sv. CD (YouTube)

Post by analogaudio » 26 Oct 2018 01:16

There is some musical energy above 20kHz, trumpets produce some, so do percussive sound sources. While CD can in theory reproduce a full level 20kHz signal and LP cannot, LP can reproduce sounds a little above 20kHz but with loss of intensity and CD cuts off hard at about 19-20kHz. Human hearing sensitivity falls rapidly above 10kHz even in the young and 20kHz is inaudible to all.

Here is the behavior of human hearing shown in a chart called Equal Loudness Contours ISO226:


32134

Hanuman
long player
long player
Posts: 1419
Joined: 12 Jun 2007 07:59
Location: Bangkok

Re: Audio frequency range of LP sv. CD (YouTube)

Post by Hanuman » 01 Jan 2019 04:58

A little bit old hat, this subject, but here's a recap (from 2011):
16195
This is from a digitally recorded album (at 50KHz sampling frequency) from 1980. The shelf at the brick wall is clear and anything above it can only be noise and distortion.


Another:
16193
This comes from the c. 1987 reissue of Rubber Soul, which is known to have come from a 44.1KHz master.

You need to know a lot about the provenance of an LP before drawing conclusions about its genuine frequency content.

13hm13
junior member
junior member
United States of America
Posts: 16
Joined: 17 Oct 2016 18:13
Location: Los Angeles

Re: Audio frequency range of LP sv. CD (YouTube)

Post by 13hm13 » 02 Jan 2019 20:21

Hanuman wrote:
01 Jan 2019 04:58
A little bit old hat, this subject, but here's a recap (from 2011):
16195

You need to know a lot about the provenance of an LP before drawing conclusions about its genuine frequency content.
What's the diff. between the YELLOW and GREEN trace?

Hanuman
long player
long player
Posts: 1419
Joined: 12 Jun 2007 07:59
Location: Bangkok

Re: Audio frequency range of LP sv. CD (YouTube)

Post by Hanuman » 03 Jan 2019 01:06

What's the diff. between the YELLOW and GREEN trace?
Yellow: Peak hold
Green: Average

13hm13
junior member
junior member
United States of America
Posts: 16
Joined: 17 Oct 2016 18:13
Location: Los Angeles

Re: Audio frequency range of LP sv. CD (YouTube)

Post by 13hm13 » 03 Jan 2019 02:26

Hanuman wrote:
01 Jan 2019 04:58
A little bit old hat, this subject, but here's a recap (from 2011):
...

You need to know a lot about the provenance of an LP before drawing conclusions about its genuine frequency content.
Did you capture these spectrums yourself?
If so, what TT/cart. and sound card were used?

Hanuman
long player
long player
Posts: 1419
Joined: 12 Jun 2007 07:59
Location: Bangkok

Re: Audio frequency range of LP sv. CD (YouTube)

Post by Hanuman » 04 Jan 2019 03:11

Did you capture these spectrums yourself?
If so, what TT/cart. and sound card were used?
Yes:

Dynavector Karat 17D3
Fidelity Research FR64s
SP10
EAR324
Digidesign 002

13hm13
junior member
junior member
United States of America
Posts: 16
Joined: 17 Oct 2016 18:13
Location: Los Angeles

Re: Audio frequency range of LP sv. CD (YouTube)

Post by 13hm13 » 04 Jan 2019 03:44

Hanuman wrote:
04 Jan 2019 03:11
Did you capture these spectrums yourself?
If so, what TT/cart. and sound card were used?
Yes:

Dynavector Karat 17D3
Fidelity Research FR64s
SP10
EAR324
Digidesign 002
Nice gear! So if there is noise above 22k for digitally-mastered vinyl records, your high-end setup should capture it!!

Have you tried a spectrum analysis on an open-reel tape deck? If not, try this: record any CD onto a good open-reel tape deck (or a very good cassette deck, like Nakamichi). Then re-run the same test as you did with vinyl. Post those results.

SUGGESTION: Your graphs only extend to 32khz. But in other experiments (see first post), the spectrum was extended to at least 96khz. If you re-run your tests, try to graph out to 96k (there's a lot of interesting information out to that freq).
See:
http://www.channld.com/vinylanalysis1.html

http://www.channld.com/dreamer.jpg
http://www.channld.com/dream192v.jpg

Hanuman
long player
long player
Posts: 1419
Joined: 12 Jun 2007 07:59
Location: Bangkok

Re: Audio frequency range of LP sv. CD (YouTube)

Post by Hanuman » 05 Jan 2019 03:45

SUGGESTION: Your graphs only extend to 32khz.
48K (half sampling rate), 32K is just the highest marked.
If you re-run your tests, try to graph out to 96k (there's a lot of interesting information out to that freq).
My Digi 002 (and Mytek ADC) max out at 96K sampling rate so pointless to measure higher than 48,000 hertz.

Channel D is misleading us with that Supertramp analysis. The title "Analysis of Vinyl LP High-Frequency Content" is false because what is shown is actually the frequency content of the entire process including playback. My tests (and others) show quite obviously that the process of playback (and conceivably cutting although cutting engineers seem adamant that it's the playback that supplies almost all unwanted artefacts) adds a non-trivial amount of upper frequency content to the signal. This means that Channel D's logic is bad. You can't deduce the frequency profile of a vinyl record (and certainly not a master tape) by observing the profile that emerges from the vinyl replay system.

To this I would add that the performance parameters of the studio equipment most likely in use for the "Crime of the Century" sessions are pretty well known and in no way was that chain ever going to supply actual recorded output out to 96 kilohertz. That's an outlandish claim. Contemporary photos of Trident Studios, for example, seem to show a mix of Studer A80 multitrack and mastering recorders. The 2-track version of this model has a high-frequency response to 18 KHz at 15ips. There would've been Dolby A on all channels and it, too, was not particularly wide bandwidth, only specified to 20KHz.


This is interesting though, from 2002:

https://positive-feedback.com/Issue2/mastering.htm

Stan Ricker provides pretty decent evidence that he cut 122KHz onto a lacquer during the development of the later MOFI mastering chain. Spoiler: it wasn't music.

13hm13
junior member
junior member
United States of America
Posts: 16
Joined: 17 Oct 2016 18:13
Location: Los Angeles

Re: Audio frequency range of LP vs. CD (YouTube)

Post by 13hm13 » 05 Jan 2019 23:19

Hanuman wrote:
05 Jan 2019 03:45
To this I would add that the performance parameters of the studio equipment most likely in use for the "Crime of the Century" sessions are pretty well known and in no way was that chain ever going to supply actual recorded output out to 96 kilohertz. That's an outlandish claim.
The cutting lathe can be set to cut out to 50khz (but that wears it down fast, so most mastering engineers filter).
Also there are harmonics that ring in the "electrical domain" -- and these are orders of magnitude (multiples) of the highest "recorded" frequency. I think that's what Channel D is showing with the bell sound recorded on the "Dreamer" track.

13hm13
junior member
junior member
United States of America
Posts: 16
Joined: 17 Oct 2016 18:13
Location: Los Angeles

Re: Audio frequency range of LP sv. CD (YouTube)

Post by 13hm13 » 05 Jan 2019 23:46

Hanuman wrote:
05 Jan 2019 03:45
[...]Studer A80 multitrack and mastering recorders. The 2-track version of this model has a high-frequency response to 18 KHz at 15ips.
I think the upper freq. cutoff was adjustable (to some extent).
Also, many home O-R decks can easily record to 30khz at 15ips.
Here's an Audio magazine (Jan. 1984) review of an ASC deck.
http://gammaelectronics.xyz/audio_01-1984_asc.html

When the FR was tested in the Audio labs, it was confirmed to what the manuf. specified.

Hanuman
long player
long player
Posts: 1419
Joined: 12 Jun 2007 07:59
Location: Bangkok

Re: Audio frequency range of LP sv. CD (YouTube)

Post by Hanuman » 06 Jan 2019 03:57

Also there are harmonics that ring in the "electrical domain" -- and these are orders of magnitude (multiples) of the highest "recorded" frequency.
That's harmonic distortion (at exact multiples of the original but not "orders of magnitude", a single order of magnitude being a factor of 10). We know it's there, both intuitively and empirically; my first trace in the thread shows distortion components outside the recorded and audible range very clearly. There are playback-induced harmonic distortion products both inside the audible range and outside and I accept that it's a valid archival approach to want to capture it all regardless of the generating source.
Also, many home O-R decks can easily record to 30khz at 15ips.
Here's an Audio magazine (Jan. 1984) review of an ASC deck.
http://gammaelectronics.xyz/audio_01-1984_asc.html
Not particularly relevant to the case. Anyway, domestic decks were always optimised differently to studio recorders. You'll notice that the curves show a progressive roll-off of HF response as the recording level is increased. Analogue studio decks and tapes were engineered more for headroom than outright HF extension and it was not unknown to see identical curves at all recording levels from -20 to as high as +10 (IIRC - can't cite any examples).

13hm13
junior member
junior member
United States of America
Posts: 16
Joined: 17 Oct 2016 18:13
Location: Los Angeles

Re: Audio frequency range of LP vs. CD (YouTube)

Post by 13hm13 » 06 Jan 2019 19:45

Analogue studio decks and tapes were engineered more for headroom than outright HF extension and it was not unknown to see identical curves at all recording levels from -20 to as high as +10 (IIRC - can't cite any examples).
Disagree .... some pro analog decks had extended spectral response, like the Studer A820
https://www.gearslutz.com/board/attachm ... i-freq.jpg

https://www.gearslutz.com/board/music-c ... onses.html

Also see:
http://www.endino.com/graphs/

And finally:
http://www.tangible-technology.com/media/media_2.html
For a perfect example, check out Figure One which is a frequency response chart that shows the performance variables of a Studer B67. This is a quarter-inch two track — with new heads — at its three speeds (from top to bottom): 30 IPS, 15 IPS and 7 1/2 IPS. ( This format is equivalent to a two-inch sixteen track.) Note that increased tape speeds extend high frequency response (well beyond that of 44.1 kHz digital) but also accelerate low-frequency roll-off. If phatness is key, the extended low end of 15 IPS is for thee.
http://www.tangible-technology.com/media/tapelo3g.JPG

===================

A thought experiment:

If I have a piano that has a key which plays 15khz (primary freq. of that key's piano string), what happens to the harmonics as they:

(1) go thru the analog tape deck (with 18khz top freq.)?
(2) go thru the cutting lathe (yes: the cutting lathe is downstream of the analog tape deck)?

zxcvb
junior member
junior member
Posts: 10
Joined: 07 Aug 2010 07:36
Location: Earth

Re: Audio frequency range of LP sv. CD (YouTube)

Post by zxcvb » 01 Feb 2019 01:05

Vinyl production master tapes are almost always 15 ips so they can fit the whole album on one 10½" reel. And of course they are copies of copies of copies, so take that into account.

For your first question, you might want to ask on the tapeheads.net forum. I would assume they will tell you that flutter and bias beats (intermodulation distortion) wreak havoc on the audible treble range. You will also be adding hiss / reducing SNR for every generation of tape. There is also (generally) a tradeoff between SNR, frequency response, and distortion. There's no way to minimize one without making the others suffer.

For the 2nd question, the real-time audio will be going to the cutting head control circuitry which will be making adjustments to protect the cutting head and keep from cutting an untraceable groove. The cutting head itself will be getting a half-a-rotation delayed copy of the audio from a delay unit, almost certainly a separate digital converter (1980s+), or via a very distant 2nd head on the tape deck (e.g. Studer A80, which could not go above 15 ips, by the way), or possibly by a separate tape-based delay unit, which of course would be yet another generation of tape in the mix.

As for how well the cutting head responds to the signal it receives, that might be one for the lathetrolls.com forum.

Post Reply