Finally, music we can buy in 768 khz sampling rates.

[Sound Liaison]

what's that thing (singularity) at 150 kHz?

see answer above

 

[MakeMineVinyl]

I have the impression that it's a bit rounder. It seems to be a bit closer to the sound straight from the Studer.

Speaking of the Studer, what tape type do you use and at what fluxivity?

 

[Sound Liaison]

So how that version derived from the 768 kHz recording?

It was created from the analog master. Not from the 768 file.

 

[voodooless]

It was created from the analog master. Not from the 768 file.

Well, then it really not a fair comparison, is it? To many variables changed.

 

[BDWoody]

Why stop there? Make it a 20/48 can call it a day… hell, I proposed to ABX with 16/44.1

I guess we could make the appropriate samples from the source file and offer them for ABX. It doesn’t need to be the complete file, 20 to 30 seconds should be plenty. We probably would not need permission from the label/artist, since for research this would fall under fair use. But alway better to ask first anyway . Also the mods might have something to say about that as well, so @BDWoody?

No problem from my side.

 

[Sound Liaison]

Well, then it really not a fair comparison, is it? To many variables changed.

Hi Voodooless
all the different sample rates are created from the same master. However for the 768 we used the RME ADI and for the other we used Merging.

 

[Sound Liaison]

Speaking of the Studer, what tape type do you use and at what fluxivity?

Hi MMV
Recording Masters SM 900. I have to ask our tape engineer about the exact number.

 

[MakeMineVinyl]

Hi MMV
Recording Masters SM 900. I have to ask our tape engineer about the exact number.

Thanks. I would assume since that is a +9db tape, you'd be using it at that level to get the lowest noise.

 

[Sound Liaison]

It would be nice if these fancy recording companies got serious about their craft and spent more time on mikes, and mike placement, and setting up the musicians for sucess rather than using a silly mix of obsolete tape machines and overkill digital sampling.

Hi Levimax
we are not very fancy. We are a small engineer/musician cooperative. Please read here; Sound Liaison about us
Have a listen to our recordings, I'm sure you will apreciate that microphone placement and setting up the musicians is done very well.
Here are a couple of videos. One recorded with only one microphone and one with a more traditional but minimalist multi mic;rophone set up.

 

[levimax]

Hi Levimax
we are not very fancy. We are a small engineer/musician cooperative. Please read here; Sound Liaison about us
Have a listen to our recordings, I'm sure you will apreciate that microphone placement and setting up the musicians is done very well.
Here are a couple of videos. One recorded with only one microphone and one with a more traditional but minimalist multi mic;rophone set up.

Thank you for your reply. I have purchased and enjoyed some of your music over the years and appreciate your dedication to the craft. I do struggle however with the idea of mixing a 1970's tape machine with 768 Khz digital sampling which strikes me as being inconsistent and catering to current audiophile fads. I would be very interested in hearing the straight digital capture but really struggle with the idea of the tape machine in the signal path which acts as a high and low pass filter as well as adding in relatively huge amounts of noise and distortion. I understand some people may find the tapes filtering and distortion and noise "euphonic" (I would argue they find the "idea" of it euphonic rather than the reality). In any case I know many people on these boards would love to be able to compare the straight digital capture with the "tape machine" version and of course best of luck and a happy and prosperous new year.

 

[Blumlein 88]

Well, then it really not a fair comparison, is it? To many variables changed.

I'd say that is the best way. No resampling involved. Fewer variables in one sense.

 

[Sound Liaison]

double post, sorry.

 

[Sound Liaison]

Thank you for your reply. I have purchased and enjoyed some of your music over the years and appreciate your dedication to the craft. I do struggle however with the idea of mixing a 1970's tape machine with 768 Khz digital sampling which strikes me as being inconsistent and catering to current audiophile fads. I would be very interested in hearing the straight digital capture but really struggle with the idea of the tape machine in the signal path which acts as a high and low pass filter as well as adding in relatively huge amounts of noise and distortion. I understand some people may find the tapes filtering and distortion and noise "euphonic" (I would argue they find the "idea" of it euphonic rather than the reality). In any case I know many people on these boards would love to be able to compare the straight digital capture with the "tape machine" version and of course best of luck and a happy and prosperous new year.

Thanks for the compliments Levimax
Most of our albums are recorded in DXD and the files we sell are one to one copies of the master file.
So if you are hesitant about this album there are enough purely in the box recorded and mixed and mastered albums on our site.
The earliest albums are recorded in 24/96 or in 24/88
For the Ray! album we wanted to go "out of the box" and experiment with an analogue signal chain.
It was a very refreshing experience although it is a slower process. No automation.


Here is a bit from the liner notes;

"The 2 most important engineers in Ray Charles' oeuvre were Tom Dowd and his protégé Al Schmitt. The story goes that before Ray Charles, Tom Dowd and Al Schmitt started working together, tape recording levels were kept safely way below the limit. But Charles, Dowd and Schmitt realized that by carefully getting close to the limit, a beautiful light colorization of the upper harmonics took place. In tribute to the 2 engineers I decided to use an analog signal chain........

continue reading here; Liner notes Ray , scroll down

 

[radix]

Here's the average power spectrum of the sample track:
View attachment 176222

150 kHz is squarely in longwave HAM (140-300 kHz or so), which might still be used in Europe? It's an AM modulation. Smart power meter go out to 150 kHz. Power converts, like for solar, can also radiate up to 150 kHz (go to scholar.google.com and search for 150khz). Broadband over electric can also have 150 kHz noise. 153 kHz (and every 9 kHz thereafter) are licensed AM bands.

I assume Mansr did not see that spike with no audio playing, so the spike is definitely in the recording? Anyway, perhaps the studio needs better shielding or grounding? I know we used to pick up some broadcast stations when we ran long outdoor cables until we upgraded to better shielded cables.

There's all sorts of other radio interference you'll run into at super-sonic frequencies. Beginning at 190 kHz you also have navigational beacons. Time signals are usually in 40 - 70 kHz range and transmitted at high power. France has a very high power one at 162 kHz.

 

[KSTR]

At such a high sample rate editing is not possible , at least I am not aware of any software that can do so, so it was not possible to "fix'' the 150kHz peak generated somewhere in the analog chain , probably the Studer. It's a straight one to one copy from the tape.

Well, for example, Adobe Audition can handle that easily (I'm on version 3.0).
Simple example: Lowpass, squared Bessel 6th order, linear-phase, @80kHz-6dB, frequency and impulse response:

More than 20dB reduction at 150kHz and all of the useless HF (see below) reduced. And the IR is almost apodizing type (sliding window averaging), almost no ringing and no "time smear" (if that is what you are after with that high sample-rate) so it will fully preserve the characteristics of the tape machine.

At 768kHz the noise floor at HF will be dominated by the agressive noise shaping used in the ADC (AK5574). The RME's input is lowpass-filtered and is -3dB down at 180kHz and even then the RMS values and spectrum look like this:

I would think (as that is my experience) there is little difference to be found, if any, when comparing your original recording with a properly lowpass-filtered one. The key is that we have full control over that lowpass. Effectively moving a reasonable reconstruction filter of choice right into the distribution format so to say. Wasteful on resources but otherwise fine in an approach to make the rendering more independant of DAC chip's built-in digital filters. Then again, is there any improvement going beyond 192kHz sample rates? I don't think so....

 

[MakeMineVinyl]

There is still a residual bias signal on recorded tapes when played back, and this signal has been put to uses such as the Planget Process. Residual bias was also used by the Watergage investigators to suss out which tape recorders recorded what Nixon tapes. Sometimes if you rock the tape slowly past the heads by hand you can hear it. 150kHz is not a multiple of the bias or erase on the Studer though.

 

[Sound Liaison]

150 kHz is squarely in longwave HAM (140-300 kHz or so), which might still be used in Europe? It's an AM modulation. Smart power meter go out to 150 kHz. Power converts, like for solar, can also radiate up to 150 kHz (go to scholar.google.com and search for 150khz). Broadband over electric can also have 150 kHz noise. 153 kHz (and every 9 kHz thereafter) are licensed AM bands.

I assume Mansr did not see that spike with no audio playing, so the spike is definitely in the recording? Anyway, perhaps the studio needs better shielding or grounding? I know we used to pick up some broadcast stations when we ran long outdoor cables until we upgraded to better shielded cables.

There's all sorts of other radio interference you'll run into at super-sonic frequencies. Beginning at 190 kHz you also have navigational beacons. Time signals are usually in 40 - 70 kHz range and transmitted at high power. France has a very high power one at 162 kHz.

Hi radix
we will look into it you might have a point.
we have had problems in the past with the fire detector and various alarm systems. We and the Metrople orchestra engineers, who also records here in the MCO studios, are constantly busy trying to reduce any outside disturbence. The studio was build in 1929 and together with Abbey Road is the oldest studio in Europe it's part of the charm but unfortunately maybe also a course for problems.
Here is the metropole orchestra in studio 5;

 

[Ron Texas]

Other than take up a lot of space on your hard drive, does this really do anything 96/24 can't?

 

[radix]

Hi radix
we will look into it you might have a point.
we have had problems in the past with the fire detector and various alarm systems. We and the Metrople orchestra engineers, who also records here in the MCO studios, are constantly busy trying to reduce any outside disturbence. The studio was build in 1929 and together with Abbey Road is the oldest studio in Europe it's part of the charm but unfortunately maybe also a course for problems.
Here is the metropole orchestra in studio 5;

EDIT: I looked at the wrong graph. Ignore this 153.6 kHz stuff.

I see from @KSTR post that it's actually 153.6 kHz. That really looks like a radio signal.

Also, the nRF903 ISM band devices (433 or 868 or 915 MHz) have a 153.6 kHz data rate, so maybe you have a wireless device near a cable or piece of equipment?

I've also seen 153.6 kHz RC oscillators used in some devices as their low-power clock frequency. Also UART 9600 x 16 = 153.6KHz, so some devices might use this over a serial connection. It's used in some RS-422 connections.

 

[mansr]

I assume Mansr did not see that spike with no audio playing, so the spike is definitely in the recording? Anyway, perhaps the studio needs better shielding or grounding? I know we used to pick up some broadcast stations when we ran long outdoor cables until we upgraded to better shielded cables.

It's a digital analysis of the file, no playback involved. I doubt Matlab is picking up stray radio signals.