Love Supreme at 24/96 ?? really ??

[Mike in MD]

HD Tracks says that they have Coltrane's Love Supreme at 24/96.

 

Has anyone been able to check if this has real high frequency content and if it has avoided the dynamic limits that come with 16 bits? Or is this just a conversion of a standard CD from 16/44 to 24/96 ??

 

Thanks,

Mike

 

 

[Audio_ELF]

While I've not seen any detailed review of the Love Supreme recording, I do believe that HD Tracks policy is that all high res recordings are high res transfers from Analogue Master tapes, or from digital masters of at least the quality claimed (maybe downsampled but never upsampled).

 

Eloise

 

[barrows]

HDtracks. Do you have any reason to believe that it would not be 24/96? In the description HDTracks states that this (A Love Supreme) is remastered from recently discovered first genereration analog mastertapes, by original engineer Rudy Van Gelder.

There have been a couple of problems with previous HDTracks 24/96 releases, due to non-transparency of the source of the high res files from the record company to HDTracks. When HDTracks discovered these problems they promptly fixed them (got true high res versions). At this point I believe that HDTracks is quite vigilant in being sure that the source of their high res download files is true high res, and not upsampled.

David Chesky is a great guy, and a true music lover, and we owe him a debt of gratitude for developing what is quickly becoming the best source for high res music downloads.

 

[carelessethiopian]

I have no evidence one way or the other but my subjective opinion is that the HD tracks download sounds significantly better than the rip from the CD I have.

 

mac mini-> hiface-> Bryston BDA-1-> Mcintosh M6500->Klipsch RF 83s

 

[rtrautner]

Having heard Love Supreme over many years from CD, cassette and vinyl sources, I can attest that this recording is superb! If you are interested, I wouldn't hesitate. I echo the sentiments expressed earlier about HDTracks-they seem to be truly committed to high resolution and using the best analog source material. While it would be difficult to prove that the material has not been upsampled, HD Tracks is a source I trust and if they say it's native 24/96, I believe them. To my ears it sounds like the real deal.

 

[firedog]

HD Tracks has had a big issue recently of discovering that a significant part of their hi-res catalogue was in fact upsampled and the offending files were removed from the catalogue. Some were replaced by true hi-res files.

 

As far as I know, the vast majority of the files in question were 24/88 downloads derived from SACD.

 

Bruce at Puget Sounds, who does the transfer for HD Tracks from SACD, has said that they have recently realized that many SACDs are derived from 24/44.1 masters that are simply upsampled to 88.2 for SACD production. Apparently this is very common in the SACD catalogue.

 

HD Tracks is supposedly now taking measures to ensure that such upsampled files aren't added to their catalogue. Their policy is that they never knowingly use an upsampled source for hi-res.

 

BTW, if you own SACDs, this means that you've probably been misled, at least on some of your SACDs - some of them probably aren't "native" hi-res.

 

As a rule, if HD Tracks says something was made from the analogue master, that means they did a hi-res transfer to digital. I have a few of these releases - they sound good and do have actual hi frequency material in them, acc'd to spectrum analysis of the files.

 

[ted_b]

it's from the analog master, not SACD (there is a Love Supreme SACD which is why I asked, but typically those are 24/88 as per firedog's comments above). BTW the day I ping'd Bruce he said he had failure rate on 50 of the 100 DSD (SACD's)files he'd evaluated that day!! Meaning 50% of them were faux hirez, upconverted redbook or similar.

 

[Synfreak]

All you need to check for the "real" resolution of any given file / download is a (good enough) FFT programm / tool.

Many audio-editor progs have one integrated (Steiberg Wavelab i.e.).

 

Most (surely not all) of the time it is quite easy to spot an upsampled track, because of the steep filtering above ~22khz (at 44,,1khz samplingrate) or ~24khz (on 48khz).

 

And as you are on HDTRacks, I've got some downloads from Linn, which are also just recorded at lower samplingrates (44,1khz).

 

This isn't very temting, and one might loose the interest in HighResolution too, but at the end of the day there is only one point that counts:

Good quality music - with tempting sound - even if it's "only" redbood CD ... ;-)

 

Cheers

Harald

 

[ted_b]

and that redbook can sound marvelous. However, if you are going to charge me more for upsampled redbook than the redbook counterpart also on your site I find it to be misleading or outright deceitful. That being said, HDTracks is doing a great job weeding out and/or uncovering these faux hirez files.

 

[Soundproof]

I asked this question here last year, and got an answer from Puget Sounds.

http://www.computeraudiophile.com/content/HDtracks-bumps-or-drawn-masters

 

It's a short thread, probably worth reading.

I appreciate the service HD-Tracks is providing, but would recommend that they have a specific provenance for their various albums. Just as we know where and how an LP was prepared and pressed, we should have access to information about how a High-Res version of a well known album came to be.

 

Cheers!

 

 

 

[audiozorro]

record label is upconverting Redbook to produce multi-channel SACDs. Do you have any examples or are you just referring to stereo SACDs that like many other so-called high rez formats may have a dubious past?

 

[Mike in MD]

Harald,

Does the stripped-down Wavelab Essential 6 have an FFT tool that can work on 24/96 and 24/88 WAV files, or do I have to get one of the (much) more expensive versions? I couldn't figure that out from the information on the website. I would really like to check my hiRez(?) files.

 

Thanks,

Mike

 

 

[Synfreak]

Hi Mike!

 

According to:

http://www.steinberg.net/de/products/wavelab/cubase5_featuredetails1001.html

 

The "Essential" version doesn't have the FFT tool.

 

I'll try to take a look on some other software packages and report back ...

 

Cheers

Harald

 

[Mike in MD]

I couldn't find the table of features that you located. The most expensive Wavelab has some really useful tools. In addition to FFT, there is a loudness distribution tool and an audio file "comparer." I expect that Wavelab can record the output from Foobar a check for bit alterations.

 

Mike

 

[Synfreak]

Mike (and others)!

 

I think I've found a method to get you to use an FFT:

 

http://www.voxengo.com/product/span/

 

This is a AudioUnits (Apple MacOS) or VST (Windows) PlugIn which should do exactly whats needed here .... for free.

 

All you need is a AudioUnit or VST capable "Host" software (Audio Editor).

 

i.e., audiacity could use VST (with an optional "VST-enabler" plugin).

 

I haven't tested it yet, but I will give it y try and report back how it performs.

 

Cheers

Harald

 

[tribeca_77]

Hello Dears,

 

I don't think that the FFT would be a good test. Usually these people are not stupid so if they have done something like that they injected some out-of-band noise. There is simple and effective way to see whether a signal has been upsampled.

 

You just have to look at the signal in the time domain on small windows (say 5-10 msecs), if the signal locally varies much more than its is red-book counterpart then it's highly probable that it has been upsampled. Unfortunately there is no objective measure that can say whether a a wave has been upsampled.

 

best wishes

Pietro

 

[firedog]

BTW, I have the standard res re-release "deluxe" edition (has bonus tracks from the cessions on 2nd CD) remaster, and the sound is excellent.

 

If the hi-res version sounds even better, then that is saying something.

 

[DiminishingReturnsOfficer]

Are you able to hear the difference between hi-res and redbook? 2 thoughts here: #1) Since HD Tracks charges over 40% more for the 88 or 96 kHz / 24 bit files, certainly they should be selling the real thing, but #2) Since you are paying over 40% more, you should be able to hear the difference between formats, if not, save your money. So far I've bought 2 albums from the "Hi-Res" store at HD Tracks; Vassar Clement's 'Livin' With the Blues' and an 88/24 version of Mahler's 6th. The former sounds spectacular; spacious, detailed, and very convincing. The latter, leaves something to be desired. It sounds closed in, there's some artifacts present during the First Movement, and I wasn't able to audibly discern the sledge hammer blows noted in the review. I have a recording of Mahler's First Symphony on CD that sounds much better. In fact, I recently streamed another performance of Mahler's 6th from Radio Symphony, at 256kbs, and it sounded roughly equal to this hi-res recording. I guess what this boils down to is; you can still have a poorly recorded Hi-Res file, just like you can have an exceptionally well recorded CD. There's no doubt in my mind as to which I'd choose, I'll take the well recorded 16/44 over a sloppy 24/96 dub any day.

 

[wgscott]

I bought Coltrane's "Lush Life (SACD)" a few weeks ago. Check this out:

 

 

I just noticed this is no longer featured on their website, and sent the following email:

 

 

Order Number: HD96640

 

On Sept 14, I purchased what was said to be a 96kHz sampled "Lush Life". I looked at the frequency spectrum with Audacity today and was a bit surprised to see that the signal is abruptly truncated at 21kHz (corresponding to a 44.1kHz standard CD sampling frequency).

 

I noticed it is no longer in your catalogue. Is this why?

 

I checked this with a free program called Audacity with one click of a button. Presumably you folks can do a similar quality control test on your offerings. If you do, you might find other anomalies. For example, the 88.2 kHz sampled Pizza Tapes has a significant noise band at 28kHz in both channels throughout.

 

 

[manisandher]

"For example, the 88.2 kHz sampled Pizza Tapes has a significant noise band at 28kHz in both channels throughout."

 

Yep, if you call -70dB FS 'significant'. I don't think it is - the noise falls off sharply after 30KHz and there is still significant signal above 30KHz (certainly no truncation @21KHz or the like).

 

Not sure if the noise is due to tape hiss or the ADC. I suspect it might be the latter - I notice a similar noise profile when I record on my Model Two, but only rising to a maximum of -105dB FS @28KHz.

 

In any event, these hirez files sound superb to me.

 

Mani.

 

[wgscott]

They do sound good, but it might be a wee bit stronger than you say:

 

 

But as for the Coltrane, they ain't "high res."

 

[Mike in MD]

wgscott,

 

thanks for the screen shot. Is there a program that costs less than $50 that can do this spectrum analysis??

 

I have tried audacity (or however it is spelled) and have not been able to analyze hiRez files. Or have not been able to figure out how to do it.

 

The screen shot above seems to have reasonable sound above 22K. Which music is it from?

 

Thanks,

Mike

 

 

 

[wgscott]

Sorry, this is just "Audacity", which is free software and available on all major platforms.

 

All I did was this:

 

1. Opened the file using Audacity

 

2. On the left, where you see that little inverted triangle next to the song name, click on that, and it reveals a menu. Then click on "spectrum" and it does an FFT, and you get the frequency spectrum in the main window.

 

The Mac OS X version works at least with flac and apple lossless (and mp3).

 

The second screenshot is the 88.2 kHz Garcia/Grisman "Pizza Tapes" from HD tracks. Obviously I can't hear the 28 kHz noise, and it doesn't seem to bother the dog, but he is pretty much deaf at this point.

 

My main point is what is going on with the allegedly 96kHz "Lush Life" Coltrane download. If it is sampled at 96kHz, it should have some signal going up to 48kHz, but it is abruptly truncated at 22 kHz, which means it was originally red-book CD sampled.

 

[barrows]

thread. HDtracks originally trusted the provider to actually provide true high res files for distribution on their site. When they became aware of some "high res" files that were actually appeared to be upsampled from red book sources, they removed those files from their site (and/or replaced them with true high res versions). Now they know not to trust the providers (record companies) and are testing all high res files before making them available for download.

WG, as you noted, it appears that Lush Life is no longer available in their 24/88.2 catalog-it is likely that they have discovered the problem.

 

[wgscott]

I assumed that is why they removed it from their website. I should have checked it two weeks ago when I bought it. However, it does raise the question of why they haven't contacted people who purchased the music. I understand that these things happen, but if the reason they removed it is they recognize it is problematic, then they should at least inform the customers.

 

Also, this thread started in July, so presumably the problem was known to exist for at least some of their stuff for at least 3 months, and this is a download I bought in mid-September. (I bought this because I thought the problem was a thing of the past.) They presumably are just as capable of clicking a button on free software for quality control as I am. It takes 30 seconds, max., and they probably only need to check one track per download to see if there is a problem. (The SACD ones seem like they would be a good place to start checking.)