HDTracks: What's Going On 24/96

[firedog]

Sounds good to me. Don't have another version to compare it to (how did that happen?).

 

Frequency Spectrum and dr look good too.

goin on foo_dr.txt

[AlainGr]

It seems good, but I have a question: is it normal that the frequency response is that steady up to 40khz or could it be that someone has kind of "equalized" what is beyond 20khz to show that this is hi-res ?

 

It's the shape of it that makes me wonder...

[Julf]

It seems good, but I have a question: is it normal that the frequency response is that steady up to 40khz or could it be that someone has kind of "equalized" what is beyond 20khz to show that this is hi-res?

 

It's the shape of it that makes me wonder...

 

Well, there is good reason to be suspicious, with the noise shaping and "spectrum restoration" tools available in modern sound processing packages. But in this case it does just look like a somewhat high noise floor that has then been filtered for the 48 kHz nyquist frequency. The next question is where that noise is coming from - analog tape/mixer, digital circuits prior to ADC, or filtered DSD noise - lots of possibilities.

[AlainGr]

Thanks for the info This kind of "regularity" over 20khz makes me suspicious. I see other "curves" that seem to fall more naturally, as I don't imagine such energy being conveyed that far...

 

Were the microphones and other devices that sensitive in the 70's ?

[Julf]

This kind of "regularity" over 20khz makes me suspicious.

 

For a good reason!

 

Were the microphones and other devices that sensitive in the 70's ?

 

Mostly no, and the tape wasn't up to anything that high either. I am assuming anything above 25 kHz is noise from the transfer. So the source was (in the best case) a tape that was played through a system with a noise floor around 85 dB. Unfortunately it could equally well have been a 48/16 digital recording played back through an analog step - no way to tell.

[AlainGr]

I have the vinyl version of this album... I don't listen to my LPs anymore, because there is always an audience eating popcorn while playing it But from what I recall, there was not much HF...

 

As you say, there are so many "tricks" to make a recording look a lot better than reality... I have been wondering for some times now at what point it makes sense to get hi-res music of that era... Well, apart from some exceptions, but which ones...

[Julf]

As you say, there are so many "tricks" to make a recording look a lot better than reality... I have been wondering for some times now at what point it makes sense to get hi-res music of that era... Well, apart from some exceptions, but which ones...

 

I have the same dilemma - have been getting much more conservative about buying hi-res downloads. These days I tend to go for them only if it is one of my top favourite albums, or if I know the provenance of the recording.

[AlainGr]

Indeed, that's how I perceive it now. I have bought "again" so many times and I feel... Well, I feel a lot more unsure than a few years ago when I started getting hi-res out of curiosity. I am now driven a lot more with disbelief than the contrary. At least for old recordings.

 

The only part that I can find interesting, as it "could be" less "artificial" maybe is the bit depth... Provided that it is faithful. I would like to say "It is possible to check it the supplemental 8 bits are not only padded with zeros" but I guess that there are specialists that can also "mascarade" this...

[Julf]

The only part that I can find interesting, as it "could be" less "artificial" maybe is the bit depth... Provided that it is faithful. I would like to say "It is possible to check it the supplemental 8 bits are not only padded with zeros" but I guess that there are specialists that can also "mascarade" this...

 

Unfortunately the bit depth is even easier to masquerade than the sample rate. Any digital-analog-digital step will mask bit depth, but even when staying purely in the digital domain, just adding a bit of dither and/or white noise will do it.

 

While tools like audacity can help to detect obvious up/resamples, they can't prove a recording *isn't* an up/resample.

[AlainGr]

Oops... I am showing my roots with the way I wrote "masquerade" ("mascarade").

 

It seems that apart from the graphs that we use to try to determine the veracity of a product, all is left is the reputation of a production house and a good old pair of ears...

 

No, not easy to separate the real thing from the "genetically modified organism" version... Or music with "steroids" or "growth hormones"...

[Julf]

It seems that apart from the graphs that we use to try to determine the veracity of a product, all is left is the reputation of a production house and a good old pair of ears...

 

And in the case of HDtracks, that reputation isn't too much to write home about...

 

No, not easy to separate the real thing from the "genetically modified organism" version... Or music with "steroids" or "growth hormones.

 

And in many cases, just "injected with water" (or empty bits)

[AlainGr]

I have found this: Upsampled FLAC 24Bit 96kHz | Hifitrack.com

 

Seems like some have decided at least to show their colors...

[Julf]

I have found this: Upsampled FLAC 24Bit 96kHz | Hifitrack.com

 

Wow! That's refreshing honesty!

[AlainGr]

I think it would be a nice idea to start a thread about "hi-res: the morning after"

[les pako]

And in the case of HDtracks, that reputation isn't too much to write home about...

 

 

 

And in many cases, just "injected with water" (or empty bits)

 

 

"Mastering was completed by Kevin Reeves at Sterling Sound NYC, using the original masters from the Motown Records vault. The masters were played on a modified Studer ?A820 with Wolke Butterfly heads and converted to digital at 192khz/24bit resolution using the DCS 904 converter and Sterling’s proprietary mastering systems. As always, the most direct signal path was maintained throughout the mastering process"

 

So are you saying HD Tracks is printing lies?

[goldsdad]

Here's a matching image from the Red Book layer of the MFSL hybrid disc.

 

A spectrogram (red and blue graph) of the HDTracks 24/96 would be far better than the 8 seconds averaged spectrum (purple graph) for identifying the signal distinct from noise.

 

[Julf]

Here's a matching image from the Red Book layer of the MFSL hybrid disc.

 

Thanks!

 

A spectrogram (red and blue graph) of the HDTracks 24/96 would be far better than the 8 seconds averaged spectrum (purple graph) for identifying the signal distinct from noise.

 

Yes, the spectrogram shows the signal vs the noise. Unfortunately it doesn't necessarily show "spectrum corrections", as those can be done in an amplitude-dependent way (basically generating artificial harmonics and intermodulation noise to fill up the upper part of the spectrum only in peaks).

[Julf]

"Mastering was completed by Kevin Reeves at Sterling Sound NYC, using the original masters from the Motown Records vault. The masters were played on a modified Studer ?A820 with Wolke Butterfly heads and converted to digital at 192khz/24bit resolution using the DCS 904 converter and Sterling’s proprietary mastering systems. As always, the most direct signal path was maintained throughout the mastering process

 

Thanks! That is useful information. The Studer is specified as 30 - 20,000 Hz at 15 ips and 40 - 22,000 Hz at 30 ips, so anything above that is noise. But at least it seems the source is a proper analog master tape.

 

So are you saying HD Tracks is printing lies?

 

I don't think they have any reason to lie in this case. But when I was referring to their reputation, I was thinking of the multiple cases where we have seen that they have sold material upsampled from 44.1 kHz as "hi res 96/24" (as documented on this forum).

[goldsdad]

[…] it doesn't necessarily show "spectrum corrections", as those can be done in an amplitude-dependent way (basically generating artificial harmonics and intermodulation noise to fill up the upper part of the spectrum only in peaks).

 

A thread on "spectrum corrections" and their signatures in graphs would be interesting. Hint hint...

[Julf]

A thread on "spectrum corrections" and their signatures in graphs would be interesting. Hint hint...

 

Hint taken...

 

Let's see if I can produce a couple of examples. Might take a day or two.

[Julf]

Let's see if I can produce a couple of examples. Might take a day or two.

 

Right. Only took 30 minutes (OK, had I really polished out the edges it might have taken 10 minutes more):

 

http://www.computeraudiophile.com/f14-music-analysis-objective-and-subjective/spectrograms-how-cheat-12232/#post158787

[firedog]

version 2

 

But as I said, I think it's a good sounding digital remaster. In the end, all I really care about is getting the best digital transcription of the original that I can, not whether it has material in the 30k-40k range. Seems like they did a good job with this.

[Julf]

But as I said, I think it's a good sounding digital remaster. In the end, all I really care about is getting the best digital transcription of the original that I can, not whether it has material in the 30k-40k range.

 

Right. But maybe the conclusion is that energy beyond 22 kHz, and resolution beyond 16 bits doesn't matter?

[les pako]

I don't think they have any reason to lie in this case. But when I was referring to their reputation, I was thinking of the multiple cases where we have seen that they have sold material upsampled from 44.1 kHz as "hi res 96/24" (as documented on this forum).

 

 

Yes, I've been reading the forum for many months now and appreciate the warnings regarding HD Tracks material. But in the last month or so they've been including statements of provenance for at least some of their new releases. I'm consequently feeling more confident about purchasing said releases until someone proves the statements to be falsehoods.

[Julf]

Yes, I've been reading the forum for many months now and appreciate the warnings regarding HD Tracks material. But in the last month or so they've been including statements of provenance for at least some of their new releases. I'm consequently feeling more confident about purchasing said releases until someone proves the statements to be falsehoods.

 

True. Let's hope they have heard the feedback and are being more forthcoming about the provenance of the material.