Purity at the source
Join Our Community Subscribe to Paul's PostsIn yesterday’s post, I spoke of how CD quality streaming on Qobuz bested by a country mile the same on Tidal. But it makes one wonder. How could that be true? How could the identical file sent by one service sound different than by another? Technically, they should be the same. But, they are not.
One question that has long plagued me and others (thanks Kent!) is how streaming services get their original files. Are they simply handed over by the labels? I suspect not in many cases. Spotify founder Daniel Ek speaks volumes of how difficult his company’s beginnings were when not one label would give him the time of day. This suggests the entire Spotify library, at least at the time, was ripped independently from commercially available media.
Working with mastering engineer Gus Skinas for so many years I have come to understand how rare good sounding digital masters are. Just the act of transferring the original files—analog or digital—to a CD master has an enormous impact on how they sound. I have had this demonstrated to me too many times to even question it for a moment.
So, we understand bits aren’t bits. And we can easily demonstrate the transfer of digital audio to a commercially viable medium varies sound quality. Is it any wonder one streaming service sounds remarkably different than another?
It would be interesting to know how each of the two services got the bulk of their libraries.
It seems the insight that bits aren’t bits (in contrary to what was seen as a fact for the last 30-40 years) opens the doors for essential improvements in digital sound (not only related to playback equipment but mainly to processes from recording to media production)
Hi Paul
“Well, we understand bits are not bits.” Just as a “1” is a “1” and a “0” is a “0”, bits are also bits.
When do you finally understand that the bits are units of information rather than the electrical signal generated from them?
No one would think that a “Z” is not a “Z”, but when it comes to processing musical information in digital form, the phrase “bits are not bits” is brought out again and again.
Has anyone who claims that ever done a real analysis down to the level of the bits?
Assuming that there are actually audible differences, this is certainly not because the data contents are different, but the data is processed differently.
What would you say if you were to write a bill for a customer over $ 1,000 and the customer would claim that he received a $ 100 bill?
Regards
Bernd, I think you’re right with what you mean (thinking more in a digital way) while what I understood from Paul and what I meant myself is related to what probably goes beyond what we knew so far and probably influences the 0 and 1 reality.
Thanks, Bernd but I suppose it’s my way of speaking in shorthand. Of course bits are bits when we think of information. And we can agree that a 1 is the same if it is an on of light, a hole in a punch card, or an electrical impulse. It is, obviously, only a representation of a bit of information. I truly understand that.
But what I mean by these short words is the bigger concept I tried to refer to in the post like mastering, like how the files are transferred and downsampled, etc.
I think what you’re trying to say is the Bits which result from the process of mastering a CD are not necessarily the same as the Bits which are delivered by TIDAL, Qobuz, Spotify, etc. And, to the core of your original point, the Bits provided by TIDAL are not necessarily the same as the Bits provided by Qobuz.
In principle, if you had the time and resources to put into it, you could compare the Bit Streams of nominally identical tracks played on TIDAL and Qobuz and establish whether they are in fact different, but having done so, it would be nearly impossible to determine, from that analysis, the mastering processes used to generate them.
If I was of a mind to do so, I could download an album in 128kbps MP3 format and convert those files to DSD64. Even though they were now in the DSD64 format, they would still sound just like 128kpbs MP3 files. I might then attempt to sell those off on the open market. Apart from listening to the files and declaring “Hey! Those sound like crap” it would actually be remarkably hard to forensically establish the provenance of those files. Sure, they would be in the DSD64 format, but the Bits would not be the same as the Bits on a proper Studio Release on SACD.
Very well put. The whole thing is an inextricable PITA.
Providing they are accumulated in an adequate buffer, bits are just bits, and the source for the pattern is irrelevant – as long as they remain as bits. Since we then convert them to an analogue signal the conversion process is the important one and that can potentially be subtly affected by the system as a whole, which includes the various devices used to feed data into the buffer since they are normally galvanically connected and can act as sources of noise However I thought that Paul was referring more to the various mastering processes used to generate the original bit patterns, and the variations which could result from different processes. That seems entirely possible.
I’ve been told by hi-res download sites that they get their files from the distributors, not from the labels. I’d assume the same is true for Qobuz and Spotify. I think in general each service gets the same files (barring some kind of difference with legal rights/distribution agreements for certain geographic areas).
This topic merely scratches the surface.
I purchased several 96/24 downloads of albums where all they are is software upsampled CD level music, not remastered.
I quit downloading anything 44.1, because my own rip of the CD is always better.
I did a TIDAL trial, and it never sounded as good as my CD rip. Yet, the playback software engine varied the quality from poor to listenable enough to know whether or not to buy the CD.
The list of this type of stuff goes too long to post. Its a lot to put up with for a continually lessening percentage of decently mastered digital recordings. I hardly know if my vinyl is truely analog anymore. Many of those are transferred from digital masters.
The potential of digital medium is tremendous. The application of it is catastrophically polluted.
“The potential of digital medium is tremendous. The application of it is catastrophically polluted.”
Brilliantly put Sir.
Blimey Paul, if you can’t find out where streaming services get their material that worries me! The sources must differ so the ‘bits’ are not the same therefore different sound?
Qobuz explain all here:
https://www.qobuz.com/gb-en/audio-quality
Thanks but they don’t actually say anything. They just talk of their goals and that they use great care. There’s really nothing more to learn which is unfortunate.
“We do the preliminary work with recording studios, in partnership with record labels and film studios, to obtain 24-Bit Hi-Res files with an identical sound to that of the engineer’s studio master”
“We take special care with regards to the conformity of the audio files sent to us by our suppliers: sampling frequency, resolution…we seek to oust any cases of upsampling and other such adverse practices.”
You can always ask them at RMAF.
You’re talking about hundreds and hundreds of different labels and distributors here, varying from those who have audiophile sensibilities and a desire to ensure that the customer receives the highest quality product, to those who measure every transaction solely in terms of profit and won’t spend a single brass cent if they think they can get away with it.
There is another aspect. Particularly with the bigger labels, you will have your strategic discussions with one department whose job is to decide whether the streaming service you are offering is compatible with the business objectives of the label, and these will include concerns such as quality and provenance. Once the strategic objectives are agreed, the business deal will often be handled by different people entirely, who won’t care a rat’s ass about quality and provenance because their job performance is evaluated solely in terms of dollars.
High powered tech people are so in love with their machines and computer programs. But in their rush to get the most titles available streaming companies have done what they obviously had to do, get material from distributors. In other words copy tapes. Original masters are harder to find and access. Ever wonder why Bear Family CDs sound so great (and cost so much)? They only use original master tapes. If you want to hear good sounding recorded music, the best bet is always going to be a first pressing LP.
So if you want quality not just quantity (and convenience), then the source isn’t a streaming service. It’s a thrift store.
Get the feeling here that many people don’t want streaming to be as good as CD, and in fact CD never quite did it for them as vinyl was better, and vinyl was really only the poor relation to shellac, which was undeniably inferior to wax cylinders and their unequalled timbre, but nothing will ever replace the good old days when we lived in a cave and bashed away on a drum made from a mammoth’s scrotum because the sound was glorious and it kept us warm at the same time. Those were the days …
Now that is a drum I want to see, although it would have been nice to let the mammoth keep his drums intact.
I wonder if JRiver will offer an interface with Qobuz? I did some reading, and apparently after JRiver put a bunch of work into integrating Tidal, they turned them down.
Last night I downloaded the trial of Audirvana for Windows. I can fill a queue with a combination of my music files and Tidal. Haven’t come to any conclusions on sound quality, but I now know that all the burps, and weird behavior is definitely being caused by the Tidal desktop app.
Compared to JRiver and Tidal their user interface is clumsy.
I will probably try the Qobuz player, if they have the obscure stuff, it will be good bye Tidal. It would be nice if Qobuz would offer a Sublime light, the hi-rez without the discounted downloads, with a monthly fee.
Steven – good on you for tarring this community with the Luddite brush. From the guy with a 2nd mono arm on his TT ; )
Again, don’t knock it if you haven’t tried it.
I have a Fred Flintstone turntable (those of you in other countries may not be familiar)…..it has a solid stone plinth and platter, and features a live woodpecker that, when you tilt him over into the groove, serves as the stylus. Very organic-sounding. None of today’s exotic materials come close.
Great question, Paul. I think many people (and labels) are confused by this question. I’m putting together a series of articles on the subject because I think it’s important to understand… not all music retailers offer the same sonic experience. Also because I want to know who’s messing with the music. it’s not always what we send out.
All music retailers who offer streaming build their own app. These apps can be coded to have automatic level adjustments so that the listener doesn’t have to reach to the volume knob or digital fader for each song. Unlike the days of buying an album where you might get 10 songs relatively at the same volume, single songs from various albums don’t flow easily volume wise. It’s one of the hardest things to do in mastering and many people differ on how to approach it.
So, if the app is control the volume, it has built in limiters/compression algorithms which means lower dynamic songs are going to sound more compressed than the producers intended. They may start with a decent 44.1 FLAC file but the app could change the sound. We haven’t done the formal tests yet, but I recognize my recordings sound different through different services.
In between the label and the music retailer is an “aggregator”. This is a 3rd party service that takes one file (generally 44.1 16 or 24 bit) from the label and manages/preps the delivery of the file to the retailer with metadata. The aggregator also receives the money and distributes back to the label from the retailer. This is another place the music can change.
You may not have heard these aggregator names… Tunecore, Dashgo, The Orchard, CDBaby, etc. It’s B2B. They deliver to Tidal, Apple, Spotify and hundreds of others. Interesting to note… Qobuz is new and generally not listed with these aggregators. For Qobuz, at this time the labels work directly with them (which can be good and challenging when a legal department is taxed with getting agreements signed). Similarly, MQA services do not go through aggregators. My understanding is that the label does the deal directly with MQA.
For instance, Blue Coast Records has an aggregator (several we’re testing, actually) for streaming, etc. Our aggregators ask for the 44.1 16 bit file. They send to Spotify, Tidal, Apple, etc. We are negotiating a deal with Qobuz and plan to send them our files directly in a much higher resolution than what our aggregator accepts. We put our own files up on Youtube (and frankly, when listening in the higher fidelities on youtube, for a streaming service it sound most realistic with less compression — again, when we do the uploading).
Funny, you’re probably right about the early days of Spotify and ripping from the CDs. That’s what we did at Liquid Audio in the late 90’s. Some of my research there involved comparing ripping applications, devices, etc. Believe me.. there are thousands of variables and none of these ripping tools sounded the same… just as conversion tools don’t sound the same. I tested hundreds of combinations and finally said, “enough already”. Settled on the best and that was that.
I’m sure you’ve done this, but… check to make sure you’re comparing apples to apples. Are you comparing MQA audio with Qobuz audio? Were the sampling rates the same on the masters provided? I’m looking forward to testing our Blue Coast Records on these services. MQA is not the same master even if some of you may enjoy it more… and that’s okay… as long as you like it, we’re happy… and no, Blue Coast Records is not on MQA at this time.
When comparing there are other issues like ‘caching’ that could affect the sound which may be dependent on the speed of the internet at that given time.
Here’s the first article in the series where I go into more depth of these concepts.
https://dsd-guide.com/why-does-song-not-sound-same-when-played-from-different-streaming-companies#.W69tsHtKjbg
Thanks Paul! Keep up the good work.
Cookie Marenco
Blue Coast Music
For your readers… 20% coupon code good through Oct 15, 2018, one time use
QUARTET20
https://bluecoastmusic.com/
We’ve just release a Laurence Juber recording of Beatle songs. I didn’t record it, but Al Schmitt did. Sounds darn good. 🙂
Thanks for that great explanation, Cookie.
My own experience is that the MQA masters on Tidal sound very good — pretty close to the equivalent high-res download (although if I were a manufacturer doing a demo, I’d use the download for ultimate sound quality). The regular Tidal catalogue is wildly variable in sound quality.
I do not go through the Tidal app’s volume control. I use the “Streamthrough” feature of Pure Music. PM uses a proprietary dithered volume control and gives me access to its 64-bit digital crossover for my subwoofer. Tidal sounds noticeably worse when using its volume control.
Great information. Would be interesting to find out if Paul was using Tidal’s volume control.
In general, in the studio and my home we tend not to use the volume controls for many devices and rely on one master out from an expensive controller. I paid thousands for this device and it colors the sound (which I complained about after purchase) but it was a solution I needed to compare multiple devices. At least all devices have the same added ‘stuff’. Hope that makes sense. 🙂
Cookie
Thanks, Cookie. As always, a wealth of information. I do not use their volume controls and make a point of keeping them at 100 so there’s the best chance of being bit perfect.
The aggregators were an unknown to me but they make sense. Thanks for that info. I am hopeful you can make a deal with Qobuz to get Blue Coast uploaded and available – that’s great but I wonder if it will be a wise financial move. Doesn’t the idea of your library freely available mean I won’t have a reason to buy?
Thanks Paul. Reply coming in 2 parts.. Part 1
I figured you’d not use their volume controls, Paul. When I do a more comprehensive test, I’ll let you know what I think. I’m going to talk to the aggregators, too. I suspect aggregators make copies (without much concern to converters or sonics) and probably Tidal makes a few copies on their end. Then the sound gets rammed through their app.
Have you compared Tidal’s sound to the same file played through Roon? Could be interesting to hear your results. Are you using the hifi service for Tidal?
Whatever file you heard (and assuming it is the same file as sent to Qobuz, which is 50/50 that it’s the same audio master) is that Tidal was limited to the CD WAV file and Qobuz made efforts to get a higher resolution file, then did the conversions themselves. Since Qobuz makes a point of looking for dynamics, it’s highly possible they got a file pre-CD mastering without the compression.
Knowing the source of the material is the only way we can really judge. I plan to do that.
Cookie
Blue Coast Music
Part Two… Reply to Paul…
How does a label make money when streaming is free?
The answer is you don’t. We look at the relationship as good exposure to possible new ears who may become customers later. Qobuz doesn’t have the mechanism for DSD (I’ve been told) and we will probably only give them up to 9624 for downloads.
I can’t disclose how the money is divided in the contract but I can say this…. they’ve figured out a way to compensate for the increase in bandwidth charges that they will incur and it will come from the labels. Good for them…. sad for the labels.. I support them because it’s a real cost. We’re in the same boat as a retailer everytime a DSD256 file is downloaded. Imagine that cost everytime the album is played… it’s substantial.
We look at the relationship with Qobuz as promotional. They’ve offered us opportunities to participate which profiles us and we’re happy to help.
They understand their profitability rests in downloads, not streaming (no one survives in streaming except the end user who gets free music until the companies go belly up or sells to Apple/device company)
We could streaming DSD256 right now (and we have), but the bandwidth costs are prohibitive. Customers would be better to buy the DSD download at this period in time.
Qobuz knows I’m a competitor. Since we generate our revenue from DSD sales, I’m hopeful it could work. I noticed Channel Classics signed with them. Sure they’re thinking the same thing for the PCM audience. I have the agreements and plan to sign.
Recently, we’ve have a boost in our Master Quality Disc CD sales (hand made, 24k gold, exceptional sound). I believe CDs will come back like vinyl.
Speaking of vinyl, we’re are in negotiations to release the Blue Coast Collection 1 & 2 in vinyl next year.
It’s a very tough time for the industry as a whole. Eventually, the consumer will have to pay for the music… especially great quality sound. Either it will be in a new technology development or through the service provider or higher rates for streaming or watching ads. Quality of sound is all but lost except for a few holding the ground. I’m not sure I like the idea of Spotify being a record label, but that’s what’s happening.
Sorry for the gloom n doom day. There’s still time to plug up the holes in the bottom of this boat called the music business.
Thanks Paul!
Cookie
Cookie, as ever your input is invaluable and comprehensive. It would seem anybody basing their music buying on backing streaming is killing the very thing they supposedly enjoy. Why do we think we can simply have music at next to no cost when everything else in life has to be paid for? – if you see a painting, ceramic pot, chair, piece of jewellery you would like to own do you ‘stream’ that?
Thank you, John. I think many of us see that forest beyond the tall trees. It’s scary. Fantasy Studios just closed two weeks ago. We’re receiving more and more unacceptable quality audio for release recorded at home or in a basement studio mixed on headphones. Income from recorded music is not a reality for most musicians… $35 for 12,000 plays a month isn’t going to cut it.
And the days of ‘infinite play’ will be gone to protect the companies from labels rigging the system. Remember infinite play? When you could listen to the same song 100 times in a row? I did that… yeah, crazy, but sometimes, I’d listen to the same song for days.
We’re all holding out for something to happen. Maybe we’re back to the days of patrons. Here’s our donate button… 🙂
https://bluecoastmusic.com/donate#.W6-yAntKjbg
When Spotify gets bought out by Google or Samsung or a Chinese company that attempts to cut costs, that will the danary in the coal mine. The rest will follow. The companies and their employers will do fine. The artists, labels and craftsperson… well, see you behind the counter of your local coffee shop.
To all… find your favorite artist or music store and buy something today. Download, CD, even mp3. One mp3 sale can make a difference.
Thanks all,
Cookie
Blue Coast Music
https://bluecoastmusic.com/
Napster was the starting shot and then came “Don’t be evil”. This corrupted half a generation with illegal downloads.
Most comsumers pay big money for their I-somethings but expect music for free. I admire the ethos of your small company in this cruel situation.
Funny enough, I was working at the “legal” counterpart to Napster back then… Liquid Audio. We launched 2 years before Napster…. funny enough ended up with the same VCs….. conflicts? oh my… I should write a book. Nails in the parking lot, screaming matches… the works.
You’re right.. it started a generation of “free” expectations.. and commoditization. Daniel Ek’s (founder, CEO Spotify) early investor was Sean Parket, one of the early investors for Napster and Facebook. Any wonder why this is happening can ask Sean… he’s worth billions now….
These guys are souless when it comes to compassion for the creators.
oh, don’t get me started… well, I guess you already did. 🙂
cookie
Even though this is a crucial topic for someone who claims to love music, it is difficult to discuss here. I have tried to in vain.
I’ll open a expensive bottle of wine and play Revolution Number 9 if it happens.
Because we grew up listening to the radio.
To my knowledge the broadcasters were not corrupt, at least in Europe (exept the few pirate radios in NL and UK in the 60ties). Till today they pay a fair share to musicians and composers, a lot more than Spotify et al. Of course not everybody bought the promoted records, especially when you had a tape machine. These private copies though were legal and there was no possibility to spread them worldwide.
That’s not my point. My point is that growing up listening to music on the radio establishes the fair market value of popular music (defined as the music they might play on the radio) as being free.
Great post, Cookie, lots of good insight into how the manufacturing chain works. I’m curious, what did you settle on for ripping CDs in terms of software (and hardware if you discovered it matters). If you were ripping a large CD collection today, what would you recommend?
Karl, I’m hopeful better software came along and honestly, I don’t remember what I ended up with. One thing I do remember is that none of the CD rips sounded as good as the CD itself.
If you were ripping today, I would suggest never throwing away your originals and use the rips for non quality listening experience. Keep the discs in a box, if you have to. That way if someone actually creates a software program that does a fantastic rip, you won’t regret it. I’ve regretted moving my masters and rips too many times… now I keep the originals — especially the 2″ tapes.
Cookie
Blue Coast Music
What an eye opener today`s topic has been for me, especially your posts. Luckily, I never bought into ripping, downloading or streaming. I have all my discs and that’s all I play-the good old-fashioned way. Maybe in my next life.
Thank you. I do what I can to help with what I know. I don’t think I’ve met one person in the industry who truly understands the complete supply chain… I don’t. I was an artist, worked for a label and ran a commercial recording facility. Once I called 2 entertainment attorneys and a publishing exec to review a contract and got three very different financial results. Sometimes, all you can do is chuckle and drive to the beach. 🙂
Cookie
Thanks much for your valuable input once more Cookie!
Reading about your experience with rips I wonder if it might be the opposite (files for streaming better than the CD) with files sold for download coming from the source used before making the CD? I assumed so so far.
Some labels like ECM call them “pro studio masters”.
Jazznut, thank you for your comment.
Files for streaming can be very different than the CD sold if the aggregator were provided files before CD mastering. Excluding our work with Blue Coast Records… as a studio we routinely provide the artist with a more compressed CD master for replication — at their request. Many often provide the aggregator with the files “pre-cd” master so they are less loud and more dynamic.
At Blue Coast Music we post the files for download as converted from the highest quality file. That includes the 44.1 16 bit that we sell. If someone made a CD from that file, it would be very different from the CD.
Our MQD series (Master Quality Disc… sorry, MQA, we had the name long before you) of CDs are made from the non compressed files. It sounds like ECM does the same. I don’t know for sure.
As a label, (Blue Coast Records) we are beginning to consider life beyond the ‘audiophile’ community and may have to address the growing trend for smart speakers. I suspect we will be creating a different master (more compressed) in the future to satisfy lossy compression (Spotify, applemusic, google play) and the dynamic versions for audiophile listening (which we prefer).
Reality is 90% of the listening audience (sadly) doesn’t care about the quality. If audiophile go to streaming, it leave little reason and add complexity to have multiple masters to listen to. We’re working on a designation that can identify more dynamic recordings that will require a playback boost.
hope that helps, though it probably does more to confuse. apologies.
Cookie
Thanks much Cookie!
The state of things before a common disc is made (and what online streaming services receive as sources) seems even worse than I guessed, not only due to mastering aspects of the final product, but also due to processes before.
So I hope I don’t draw false conclusions, but after all I read here and there so far, they are the following:
Pre CD mastering files available to redbook or hires file selling vendors can sound much better than the physical media.
Online streaming services (with as it seems currently one exception) usually have even clearly inferior sources available than both music file vendors and physical media producers. On the long run, due to less and less care for sound, online streaming services will probably rather go further down in quality and quicker than other alternatives.
The reason some say physical media sounds better than streaming files and some the opposite probably is because they compare physical media to streaming services, ripped files and files sold for download with varying results.
Aside of this and regarding vinyl:
The reason why I personally often hear clearly better results from vinyl masterings compared to digital releases can have 3 reasons from what I remember:
– many of the best (re)mastering engineers mainly work for vinyl releases and do less digital versions
– those engineers usually use non-compromised sources for producing the final product (in opposite to the usual sources for CD or streaming service albums or similarly produced LP‘s)
– even for less audiophile produced releases, analog/vinyl mastering by nature allows less options of harmful mastering techniques than digital to be applied by sound engineers who don’t care much for sound.
On the other hand digital releases can sound better than vinyl if both releases are produced from the same source in a comparably audiophile way and if the dynamics of the music make limiting vinyl mastering steps necessary or lead to mechanical impacts.
Meanwhile I also have an assumption why even many professional or very experienced listeners who do know about the above mentioned things are convinced of the general superiority of digital. I assume it’s because (aside of convenience aspects) when comparing, they refer to really perfectly produced digital releases of very demanding dynamics AND they usually don’t make any such comparisons of the majority of music material.
Those are pretty good assessments.
I wouldn’t go so far to say one format is better than the other. We all have different tastes in music and systems we listen from. I’m not sure extreme dynamics are good all the time.
I love my chamber acoustic folk music recorded at DSD256 with lots of space and full dynamics. At the same time, I’ve mixed reggae dub dance music to be loud and obnoxious. I love that too. One night, I mixed a bunch of scratchy noise on the left channel only just to have a laugh. And I’ve mixed the bass in the center to abide vinyl rules and then do an out of phase disco mix for CD.
What is true in our current time, is there is less quality control now that artists and small labels can upload directly to download and streaming services. I’ve heard about hundreds of remixes of one song being available for sale.
Where before the mastering engineer delivered files to replication houses, it’s now someone ‘at the label’ or the artist uploading files to the aggregators. There are automated mastering systems at aggregators that will “master your song for $6”.
Nearly 40% of our current music listened to comes from independent labels going to digital aggregators – and anything goes. There is no rule book for “right” with digital. Vinyl had a rule book or the disc didn’t play.
I’ve never heard an artist say their vinyl record needs to be louder. But I do hear it about CDs and mp3’s. That said, I have my doubts vinyl could absorb dynamics that we are able to sell in DSD256. That doesn’t mean it sounds better.. it means I can make dynamic music if I choose and create the best master for each format.
Next year, you might find us making even more different sounding masters to optimize sales for new platforms … Our job at the label is to sell music to all customers who want to purchase…. smart speakers, here we come! (Don’t worry, we’ll always have DSD256 available).
That was a lot of information today… more than is taught in most audio schools..
Thanks, Paul, for having a safe place to exchange. 🙂
Thanks everyone who asked intriguing questions that took me away from the Giants Dodgers game.
Cookie Marenco
Blue Coast Music
signing off for the day… 🙂
As I replied on your post yesterday, with Tidal, I’d read that Universal was sending watermarked files to Tidal for streaming, and I have heard this for myself firsthand with one of their titles. (It was one of those situations where I kept messing with my computer system trying to figure out why it sounded so bad…until I realized it was Tidal’s version of the album that was flawed.)
Any person who can hear a change of 2 parts in 10 billion in the same design can surely hear the difference between one design and another.
“you have better ears than I Gunga Din.”
I have studied MQA’s technology and I have just two things to say about it.
1. Based on my understanding of the audio origami process it violates the Shannon Nyquist criteria and so can’t possibly work as advertised.
2. Based on my own knowledge and experience but equally importantly on the reports of Mark Waldrup using direct experiments on himself and other audio engineers comparing true HD recordings and the same recordings down converted to RBCD standards there is no audible differences to humans between RBCD and true HD recordings. It should also be noted that most recordings advertised as HD aren’t because they are made from original masters that did not contain any usable signal above 20 khz. To be HD every element of the chain from the microphone to the file must be HD. Most buyers of HD recordings are getting screwed paying a lot of money for a larger bucket which contains a lot of extra zeroes. Having studied the Red Book standards briefly, I am awed by the redundancies and error detection and correction systems in it. For me RBCD is as good as it gets and any further improvement even if it is audible is not attributable to the extra dynamic range or frequency range of the recordings but to other reasons.
Here’s a primer on MQA.
https://www.youtube.com/watch?v=r_wxRGiBoJg&t=4s
https://www.youtube.com/watch?v=T5o6XHVK2HA&t=1071s
Regarding Mark Waldrep and the HD vs CD quality files test one of the most interesting aspects is the lack of people reporting their findings after downloading them – read into this what you will…
I have a technical question for Paul. If you don’t know the answer then you might ask Ted Smith. It’s something I’m curious about. You say that RBCD has no master clock yet when you insert a disc in any player it knows the time length of every section, where it is in a section, How much time has elapse, and how much time is left. If you have A->B repeat where you define a starting and stopping point it will always go back to the very same spot in time on the disc and it will end and repeat at the same time every time. If there is not clock to control the timing of the bits how does it do that?
SM, from memory, the CD contains subchannel data to control frame addressing and allow for movement between tracks.
This discussion these last two days has been enough to make one’s head start spinning at 33 1/3 rpm’s.
Opinions proffered:
— cd sounds better than streamed
—cd better than ripped
—streamed better than cd
—cached offline streamed better than online streamed
—cached offline streamed better than cd
—cached offline equal to download?
—Qobuz may be better than Tidal because it may not go through an aggregator….yet.
— hi quality streaming is hit or miss, dependent on a multitude of factors, and not economically sustainable for a dwindling audience.
I feel the same way. Let’s hope the industry doesn’t stop making disc transports or we’re in real trouble.
Different digital blu-prints to start with, different gear, different ears = different opinions on the sound differences…… simple really 🙁
Soundmind – saying that RBCD can essentially find its own ass with both hands is in no way the same as it having a high resolution master clock (as in Ted’s DACs) by today’s standards. It is 16/44.1. Sorta like saying, “Wow, my car can tell me how far it drove today with totally acceptable accuracy!”
What is becoming clear is that one of the things separating good from great digital these days is timing on a scale that is multiples (perhaps even orders of magnitude) of the resolution of what is required for RBCD.
Glad you bought up the timing issue, seems to me 96/24 will give you all (more than) the frequency capture you need and all the dynamic range but it falls short in the timing domain even cranking to 192 doesn’t quite do it but 384 does or DSD 256.
Great discussion, I need to reread and study all. Actually, I just signed up with the ROON trial service and don’t know what to make of the fact that ROON tried to get me to sign up with TIDAL. Well I am just starting to listen to the services; haven’t come to any conclusions; dam it takes time….