 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Where we disagree is that they absolutely can check the files. It's an electronic process, something programmatic that can be set up to analyze any track coming in, validating that they match specs, and notifying Qobuz when any don't, then they employ manpower - and they should - before they advertise the product they are selling as having a higher quality than it does. It's something they very much could do, with little expense and effort.
The problem is there isn't a good way of checking that doesn't involve checking /every single file/ by hand. There's no program out there (at least last time I checked) than can do it properly automatically and actually get accurate results enough times for it to be trustworthy. Only so-called solutions I know can give BOTH false positives and false negatives.
Yeah, there is no way they can automatically confirm the accuracy of the files. Qobuz is just the vendor. Obviously, if they sell you a file as "high-res" and it is not, it is your perfect right to go for a refund and deal with Qobuz, but it is not something they could easily check and verify. They are actually a small company (like most high-end streaming providers). There are also many different types of genuine high-res files. Actual high-res recordings, remastered analog tapes, or even upsampled early digital recordings (like part of the Karajan Bruckner set; only the part of the analog recordings (which is the bigger and more important part) is actual high-res, the other "high-res" files are actually files carefully up-sampled by DG (and they are upfront about it). No way any company could automatically verify and accurately assess all that.
However, I would definitely try to avoid labels that provide "phony" high-res files.
|
|
|
|
|
|
|
|
|
|
|
|
 |
Posted: |
Jun 9, 2021 - 5:57 AM
|
|
|
|
By: |
slint
(Member)
|
Where we disagree is that they absolutely can check the files. It's an electronic process, something programmatic that can be set up to analyze any track coming in, validating that they match specs, and notifying Qobuz when any don't, then they employ manpower - and they should - before they advertise the product they are selling as having a higher quality than it does. It's something they very much could do, with little expense and effort.
The problem is there isn't a good way of checking that doesn't involve checking /every single file/ by hand. There's no program out there (at least last time I checked) than can do it properly automatically and actually get accurate results enough times for it to be trustworthy. Only so-called solutions I know can give BOTH false positives and false negatives.
I have a code that can do this, so surely music stores could do it as well. They can hire me if they want! For High-Res vs CD quality this is actually very easy to differentiate between pure noise and audio content at higher frequencies. For MP3 vs CD quality this is fairly easy as well. Re-encoded MP2, AAC and AC3 are a bit more tricky, but there is no harm in using a bit of statistics. If a particular source of data (label) triggers 50% fake files, then there is certainly something going on. If a label triggers 0.1% fake files, maybe there's an explanation, such as using a DVD source to replace a track missing from a master tape, etc. Also of course, some musicians may have decided to create music from MP3s, but that shouldn't happen for Golden Age films!
|
|
|
|
|
|
|
|
|
|
|
|
|
|
Posted: |
Jun 9, 2021 - 6:12 AM
|
|
|
|
By: |
thx99
(Member)
|
I have a code that can do this, so surely music stores could do it as well. They can hire me if they want! For High-Res vs CD quality this is actually very easy to differentiate between pure noise and audio content at higher frequencies. For MP3 vs CD quality this is fairly easy as well. Re-encoded MP2, AAC and AC3 are a bit more tricky, but there is no harm in using a bit of statistics. If a particular source of data (label) triggers 50% fake files, then there is certainly something going on. If a label triggers 0.1% fake files, maybe there's an explanation, such as using a DVD source to replace a track missing from a master tape, etc. Also of course, some musicians may have decided to create music from MP3s, but that shouldn't happen for Golden Age films!
But you're talking only about testing one of the two primary features of hi-res audio: the sampling rate. The other feature being quantization (i.e., is this true 24-bit or up-quantized 16-bit?). Not a difficult thing to do, but I'm not aware of a freeware tool out there that does it. As mentioned above, I've mocked up a tool internally to do it but if it's already out there, I'd like to know.
EDIT: I wrote about verifying quantization in a previous post on page 11 of the TWISTED NERVE-(Bernard Herrmann)- STYLOTONE thread here (scroll down until you see some quantization level graphs):
https://www.filmscoremonthly.com/board/posts.cfm?forumID=1&pageID=11&threadID=114145&archive=0
|
|
|
|
|
|
|
|
|
|
Posted: |
Jun 9, 2021 - 6:28 AM
|
|
|
|
By: |
slint
(Member)
|
I have a code that can do this, so surely music stores could do it as well. They can hire me if they want! For High-Res vs CD quality this is actually very easy to differentiate between pure noise and audio content at higher frequencies. For MP3 vs CD quality this is fairly easy as well. Re-encoded MP2, AAC and AC3 are a bit more tricky, but there is no harm in using a bit of statistics. If a particular source of data (label) triggers 50% fake files, then there is certainly something going on. If a label triggers 0.1% fake files, maybe there's an explanation, such as using a DVD source to replace a track missing from a master tape, etc. Also of course, some musicians may have decided to create music from MP3s, but that shouldn't happen for Golden Age films!
But you're talking only about testing one of the two primary features of hi-res audio: the sampling rate. The other feature being quantization (i.e., is this true 24-bit or up-quantized 16-bit?). Not a difficult thing to do, but I'm not aware of a freeware tool out there that does it. As mentioned above, I've mocked up a tool internally to do it but if it's already out there, I'd like to know.
Oh, right. The code I've made is not doing that, but then so far I've only been interested in detecting lossy sources, because 16-bit/44.1khz is good enough for me. That was mostly a fun way to test machine learning using only the frequency spectrum. My collection is not large enough to require something 100% automatic, but it seems to work.
|
|
|
|
|
|
|
Interestingly iTunes has all of these as Apple Digital Master. For Back the future all I could find was part three in Apple digital master format.
Thanks jkannry for the heads-up. I would have preferred to have a physical copy but haven't had much luck finding one. But for $4.99 on iTunes, I can't complain. Too bad they don't have the 2-disc version for the first movie.
|
|
|
|
|
|
Kudos to DG for being upfront about their "carefully up-sampled" files, but honestly, why bother upsampling them? They're likely doing it just so that all of the files in the set are at the same sampling rate, but there is absolutely zero benefit to upsampling them. Leave them at their original sampling rate. Upsampling them just makes them larger files without adding any true frequency content to them.
Well, yeah, the obvious reason why they upsampled them is so they have the same sampling rate as the rest of the set. It's all on one Bluray set, so the file size is not an issue on that set. They are all remastered, so the original 16 bit/44.1 kHz of symphonies 1-3 were upsampled to fit the rest of the set. One benefit is that it is (at least from legitimate sources) easy to distinguish this new mastering from the previous mastering, which are also available in their original resolution.
But as I said, it's not "deceiving", as DG clearly states that that is what they have done (and the high-res masters of symphonies 4, 7, and 8 sound indeed much better than their earlier CD incarnations... have not listened to the others yet).
https://www.deutschegrammophon.com/de/katalog/produkte/bruckner-the-symphonies-karajan-9555
|
|
|
|
|
|
|
I don't know if it's that easy to tell the real from the fake. Excessive noise reduction will sound similar to damaged tape or compression noise.
I'm also annoyed about tricking people into buying upsampled audio files. I'm not sure, however, if it is the vendors' job to check that. To come back to the earlier example: there are plenty of Blu-Rays that are sourced from tape instead of new scans from negatives/release prints. Should amazon stop selling them?
I think the best bet is to buy only from reputable labels, but even then you may have surprises.
The bottom line is that digital music distribution is really a missed chance due to the lack of content and quality control.
|
|
|
|
|
|
The bottom line is that digital music distribution is really a missed chance due to the lack of content and quality control.
I certainly don't think it's a missed chance, it is doing quite well really, and I have purchased quite a few high-res files from various labels. It's like CDs or LPs: obviously, there are bootlegs, bad pressings, badly mastered recordings, etc... those have always been there. That is unfortunate, but those things have always been part of music distribution. I have not had problems so far.
I've had far, far more trouble with poor vinyl pressings than I ever had with wrongly declared download files.
|
|
|
|
|
|
|
|
|
|
I bought the Quartet re-recording of THE BRIDE WORE BLACK on CD, but it is also available as lossless/high-res download at Qobuz.
Quite a few Quartet releases are available there, such as BASIC INSTINCT or TOTAL RECALL.
That's why I didn't get the CD of their re-recording of ENDLESS NIGHT so far, waiting to get it as high-res download instead. However, that one has not shown up there yet. So I have not yet bought any recording of it at all, even though I actually want it.
Sometimes, it is the counter-intuitive release strategy that works against the labels, it seems. I mean, I definitely want this recording, and I will eventually get it, but have as of yet not purchased it.
|
|
|
|
|
|
This is where these digital sites fail me, they lack any infomation about what they're actually releasing. Picking up off my previous threads in search of the highest quality materials of Star Wars, take THE EMPIRE STRIKES BACK:
There's this:
https://www.hdtracks.com/#/album/5dd65ced848038447729b9bf
SONY CLASSICAL release of the LP presentations in 96kHz · 24bit
And this:
https://www.hdtracks.com/#/album/5df1fc900bee25c09bc18efa
DISNEY RECORDS remastered release of the LP presentations in 192kHz · 24bit
Listening to samples, they are completely different mixes, so what are their sources?
The Sony version is way sharper and clearer in the highs with lead instruments, with little ambience and missing some of the instruments.
The Disney version is more ambient, with less defined lead instruments but more audible supporting orchestration.
And then what is "Horch House" claiming in 2016 to have struck a deal with Sony to have the original master tapes converted to a digital format:
https://daviddenyerpr.files.wordpress.com/2016/01/press-release-horch-house-new-releases-2016-01.pdf
Which master tapes? The 35mm Mag Film reels or 16-track tapes for A NEW HOPE? The 24-track or 8-track tapes for EMPIRE and JEDI? The remixed multi-track tapes by John Neal for the LP releases?
It's frustrating because it wouldn't be difficult for these companies to document this and be accountable for including this information on their sites.
|
|
|
|
|
|
|
This is where these digital sites fail me, they lack any infomation about what they're actually releasing.
Agreed, that is by and large one of the most annoying issues I have with many digital file sites/releases.
|
|
|
|
|
|
|
|
|
|
The one labelled from Sony Music is the original 1980 album mix released on RSO/Polydor.
The Disney Music one is the 2018 album remaster which (I believe) uses the 1997 Special Edition mix as a source, edited to match the 1980 album master, though there are differences based on the fact that the 1980 album mix and the 1997 mix came from different elements.
That's what my ears tell me. If anyone has more accurate information please correct me.
Thanks for looking into this. What I've found through more research is this:
SONY CLASSICAL (2015/2016) was a remaster of the original LP masters. It is identified by the original album art for the LPs. This Sony remaster was released confusingly as two products: digital, called the "Ultimate Digital Collection", and on vinyl called the "Ultimate Vinyl Collection". (Confusingly a third product that did NOT use these masters was called the "Ultimate Soundtrack Collection" and simply featured a re-release of the 1997/2004 2-disc recordings of the original trilogy).
This interview with the engineering team describes the LP masters as the source:
https://www.youtube.com/watch?v=YuH7uznb2bU&ab_channel=TalksatGoogle
DISNEY RECORDS (2018) was apparently a remaster of the original FILM masters. It is identified by the cover art of Star Wars logo, film title on solid color, the label "remastered", all placed over a black background.
This promo video with the engineering team describes the Film Masters as the source:
https://www.youtube.com/watch?v=WSOzvCtaHmg&ab_channel=DisneyMusicVEVO
While it's not clear what masters specifically were used, the video contains footage of a few key artifacts: Eric Tomlinson's engineering notes for the 16-track master from STAR WARS. A reel of 24-track tape from THE EMPIRE STRIKES BACK. What looks like a reel of master tape from EMI / Abbey Road for RETURN OF THE JEDI (image is too close to discern specifics).
The same people who worked on the SONY CLASSICAL release also worked on the DISNEY RECORDS release. My guess is that during the process for Sony's release a couple years prior, the discovery of elements led to the desire for a more accurate remaster and while the SONY CLASSICAL was rushed out to coincide with the release of THE FORCE AWAKENS, the subsequent DISNEY RECORDS release was just given a bit more time. Having seen the long interview for the first Sony release, clearly everyone involved is passionate about the project so I wouldn't be surprised if they did what they could to preserve the original masters.
I will have to compare the two, but what is fascinating is the idea that SONY's release used the LP masters while the DISNEY release used the film masters. To what extent I have no idea, but a comparison of the two would be helpful to determine the recordings closest to Eric Tomlinson's original live mix. It's just a bit confusing because the film mix would have been pre-LP, which would have to mean that for the DISNEY release the engineers would have digitized the film masters and then COPIED the work on John Neal for recreating the LP presentation again.
|
|
|
|
|
|
|
|
|
|
Posted: |
Aug 28, 2021 - 2:05 PM
|
|
|
|
By: |
tyuan
(Member)
|
Come on, can You REALLY hear such a big sound quality difference between 16 and 24 bit?
I wish you could randomly try to listen to both 16 and 24 bit tracks, and then challenge You to recognize them with a flawless result.
It's only marketing, gentlemen! Majors are trying to take Your money again, and again, and again, for the same stuff You have already purchased. No more.
If We were dogs (or bats), maybe that "24 bit mania" might make sense, but we are only poor human beings, and also almost deaf, due to our average old age.
24 bit is a good sample rate when You are mixing multi tracks or doing sound engineering, in order to preserve the best editing quality ever; but when it's time to listen to the music 16 bits are more than enough.
|
|
|
|
|
|
Come on, can You REALLY hear such a big sound quality difference between 16 and 24 bit?
I wish you could randomly try to listen to both 16 and 24 bit tracks, and then challenge You to recognize them with a flawless result.
It's only marketing in order to take Your money again, and again, and again, for stuff You have already purchased . No more.
That's partly true, but varies based on what equipment people are using to listen to the music, and what the general health of their hearing is in the first place. Shawn Murphy, the recording engineer for the Star Wars prequels does acknowledge the diminishing returns of higher and higher quality. In general, I think the big question is what masters are being used to create the digital files. In the case of Star Wars, which I am exploring, the bitrate is less of a concern in comparison to the MIX itself.
|
|
|
|
|
|
|
|
|
|
|
|
|
Posted: |
Aug 28, 2021 - 2:23 PM
|
|
|
|
By: |
Ny
(Member)
|
The same people who worked on the SONY CLASSICAL release also worked on the DISNEY RECORDS release. My guess is that during the process for Sony's release a couple years prior, the discovery of elements led to the desire for a more accurate remaster and while the SONY CLASSICAL was rushed out to coincide with the release of THE FORCE AWAKENS, the subsequent DISNEY RECORDS release was just given a bit more time. Having seen the long interview for the first Sony release, clearly everyone involved is passionate about the project so I wouldn't be surprised if they did what they could to preserve the original masters.
Weren't the Disney releases rushed out to coincide with their takeover of the Star Wars IP? I would've assumed, especially given the very poor reception the mixes on those releases received, that the decision to remaster was an arbitrary choice rather than an opportunity to use better sources.
Also, I'm curious, and maybe not understanding something obvious, but in comparing versions, how do we know what Eric Tomlinson's original live mix sounded like?
|
|
|
|
|
|
|
|
|
|
|
|
| |
|
|
|
|
|
|
|
|
|
