MQA: Fidelity That’s Better Than Sex?
Let’s assume that MQA delivers on its claims. According to Jason Victor Serinus’ article on the recent Stereophile.com site, the process has been used on the entire catalog of Morten Lynberg’s 2L catalog. If you’ve never heard one of his amazing productions, you owe it to yourself to download a track or a sample and listen. They are absolutely terrific recordings…with or without MQA. 2L has been nominated for multiple Grammy awards over the years…the label’s work is that good. So why would recordings this good need to be “scrubbed and clean and born anew with MQA”?
I asked myself the same question. I have a catalog of almost 100 albums that might undergo the same MQA process. For me it might mean getting some terrific recordings onto TIDAL or other “high-res” music streaming sites. But until proven wrong, I’m hoping that my tracks will sound exactly like they do when played in my studio. I honestly don’t want the fidelity to be changed…in any way.
Yesterday, I hosted a close friend and one of his important musician friends. I don’t want to mention his name but trust me this guy is the real deal. He was the music director for one of the world’s biggest music celebrities for over a decade (he’s played the SuperBowl). He’s been spending a lot of time in China working on a variety of projects in the high-end audio and streaming distribution.
They arrived at about 12:30. I gave them the royal tour of the new studio construction, the Astound studio in the middle of the building (front cover of MIX magazine a couple of years ago), and the smaller project studios located in the rear of the building. Finally, we arrived at my main room. The size of the room and the array of 5 B&W 801 Matrix III speakers made an impact. He is endorsed by PMC. He and Greg Morgenstein, the engineer for Astound, got into a lengthy conversation about how amazing the PMC line is.
I sat the guest of honor in the center position and proceeded to unleash a few of my favorite recordings. I played John Gorka (“Let Them In”, The Latin Jazz Trio (“Mujaka”), Jennifer Warnes (“So Sad”), and a few others. Within the first few minutes of hearing real high-resolution, surround music tracks made without any artificial processing or mastering, he sat back in the chair and uttered, “These recordings are the best I’ve ever heard, this is better then sex!” I’ve never had anyone make that comparison but it’s pretty high praise, indeed.
He liked the sound of the B&W speakers too.
I heard a sample of a 2L track for orchestra and choir in the MQA suite at the show. It sounded amazing, really! But Morten’s unprocessed tracks sound amazing as well. Is the world of real high-resolution music the target for the MQA process? Most of what I’ve heard from them has been archival stuff from the 50s and 60s. Those sources originated on analog tape…which might be more fertile ground for the process. Again, with the goal of preserving every bit of fidelity that existed on the masters…but not “enhancing them”.
But remember that the MQA process results in a new file that has the ultrasonic stuff and metadata about the ADCs cleverly encoded into a standard bandwidth container. That means that the catalogs of the major labels will have to undergo MQA processing. Think about it for second. PonoMusic and HDtracks sell the “high-resolution” output from the major labels. At last count, there are about 7500 albums that have undergone the process of “remastering” from analog tape to “hi-res” files. This has happened over the past 8 years! And now to move to MQA versions, all of those albums will have to be sent back into the studio. And what happens when MQA 2.0 comes along?
Does anyone else see a problem here?
Dear Mark,
It would be nice if you would give – from time to time – an indication of the progress you make on your book.
Kind regards,
Walter
Walter, I’m trying to post an update every month through the KS page. It’s coming along. I’m trying to complete a chapter per week…and am almost current with that goal.
Hi Mark, What disc is Jennifer Warnes’ ‘So Sad’ on?
This is the project that can’t be shared. It’s the finest thing that Jennifer has ever done, but sadly she won’t approve of its release. It’s a track that I play at my studio when I want to demonstrate the very best I can do with an amazing artist.
It is said that Frank Sinatra was just as obsessed with the sound quality of his recordings as he was about the songs themselves. It would be interesting to hear if the process could do anything with what many consider state of the art at that time. So little is known about how he went about ensuring that quality. It would make an interesting story in itself I’m sure.
MQA is being written up as a huge improvement in sound quality. I’ve heard MQA encoded files through speakers and headphones and can’t say that I was wowed. The promise of the technology is that we will be able to experience the fidelity of the original source with less degradation than ever…and have that via streaming.
I may be missing something but I thought the real plus of mqa was that you got the high resolution file compressed enough for streaming?
You’re not missing anything. That’s what MQA is all about…taking so-called “hi-res” music and allowing TIDAL to stream it.
I don’t know if I see a problem, but what I do see is a revenue stream. A big one.
Wow ,you seem to be the problem here. New technology that sounds better then old technology is a problem because it’s inconvenient?
I’m not sure I follow your comment. The new technology doesn’t sound better than the old technology. If you were to listen to the analog master of the Bob Dylan tracks they’ve used in their demo AND then A|B switch to the MQA version, they will sound the same. The MQA is not an enhancement as some writers and proponents have claimed.
We peons will never get to hear the “analog master” so we have to deal with what we can get. The question will be “is MQA better than PCM?”. And, of course, everyone who buys it will say “yes”, just like they do with 192/24 of anything…
MQA is PCM with some clever coding tricks…
“Does anyone else see a problem here?”
Maybe, but how hard is it to do the MQA encoding. If it’s as simple as trancoding a file or up-down sampling, it only takes a couple minutes?
But to me the truth is the groundwork is already laid. The JVS and JA articles at Stereophile and all the rest of the “gurus” from the mover/shakers of High End audio have already been pushing this process for a year now. I don’t know the whole reasoning but like the Regens, Jitterbugs, DSD, etc; it’s a hustle that’s looks to be just about a done deal. The same guys that buy the $3k power cables are already drooling and have their wallets out waiting for it’s release. I don’t see the path but I’m not in the industry, as they say “follow the money”.
I could quote something from Aczel again but I won’t do that to you.
My brain hurts. LOL
I haven’t see the MQA encoding tools but the process can’t be any less complex than doing a new “high-resolution” transfer, which the labels are already doing. And that process has only resulted in 7500 albums over the course of 7-8 years.
I was glad you played “Let Them In” by John Gorka, I purchased the blu ray The Gypsy Life and I was really only interested in transferring the audio files to flac. But didn’t realize how great the visual and audio just playing on my cheap 5.1 Sony system and Visio big screen, I could only imagine what it would be like watching it in your studio. Thanks for making these artists come alive.
John is an amazing artist and his project remains one of my favorites. His songs, his voice, and the heart that he puts into each tune are absolutely first rate.
The emperor’s new clothes?
No, I think there’s merit to the process. Bob Stuart is very smart and I believe that MQA used on new real “Hi-Res” recordings would be a good thing. It’s the over the top descriptions by the usual reviewers that MQA somehow “delivers better-than-hi-res sound quality” that amuse me. The inventor told me that MQA doesn’t enhance the fidelity of the source file, it makes sure that nothing is lost from source to destination.
The claim is that with the current process, the fidelity is *not* preserved from the original performance to the release. By profiling the original ADC and removing what is purported to be distortions inherent in the current releases, the MQA process is said to produce final output truer to the original performance / master than is currently available. So, by definition, it would be “better” than what is now available to consumers.
But not better than the original analog master! I guess it depends at which point you consider the true source. As an engineer, I would expect the sound of my final master in my studio to be accurately reflected through whatever technology to the end user. That would represent no loss of fidelity or change. Nothing about that process would be seen as “better” than my master. The same would be true for the analog master heard in the mastering rooms of the major labels. If they played a Bob Dylan master, I don’t believe that MQA (or any other technology) would be an improvement on that fidelity. As I understood Bob Stuart, by knowing the “imperfections” of the transfer/distribution chain from source to final consumer delivery…it is possible to match the sonic fidelity of that original master. That’s not an improvement to me. That’s maintaining the fidelity that the original engineers achieved.
When I read a senior audiophile editor state that MQA “delivers better-than-hi-res sound quality”, it elevates MQA to something that I don’t believe it is. It is not a process that improves the sound of the master recording. It makes it possible to deliver that original fidelity with less loss.
Hello Mark
Under the headline: You Know Your Old When….. To answer your question, yes I see a problem here, and it’s the never ending re-packaging of products you already own. I’m saving my old vinyl just in case.
Thanks Robert. I have my boxes of vinyl but I don’t have a player.
You make mention of a recording, you have of Jennifer Warnes “So Sad”. I have purchased ” Nitty Gritty Surround”, mainly to get Jennifer Warnes. Even though this is now older I think her voice & your recording methods are ” as good as sex”. Does AIX have more Jennifer Warnes?
Barry, yes I do. I have an entire album of Jennifer Warnes. It was recorded about 8 years ago and is among the best things I have ever produced and engineered. However, she decided to prevent me from releasing the project for reasons that I can’t fathom. It’s probably the best thing she has ever done. When she and her lawyer first heard the rough mixes everyone knew that we had something really special. Too bad that she’s unreasonably refused to let me share with the world.
I have heard a couple of the JW tracks. tHey are stellar. One would think that JW would be interested in selling some albums. I guess not.
Mark, just a remark about your mention of PMC speakers. I listen to PMC speakers in my home, specifically IB2 SE. Most of the world is not aware of just how good PMC speakers are. If you want to truly hear a recording just as the artist intended then PMC should be your pick. Many of the better recording studios in Europe use PMC speakers exclusively.
You’re right they are terrific speakers. Not my favorites…but right up there.
Mark,
I apologize if this is a little off topic. I ripped my entire CD collection about 5 years ago into ISO files (virtual version of the disc). I was bored the other night and decided to play around with some of my music files. I use dbPoweramp for converting music files when needed, etc. I mounted the ISO file of Guns N’ Roses Appetite For Destruction and proceeded to rip into WAV & AIFF 192kHz/24bit. I do realize this will not provide additional audio fidelity over the standard 44.1kHz/16bit file.
However, when playing back Paradise City on both versions, right as Steven Adler is coming in with the drums, I noticed an audible difference in sound from the WAV file versus the AIFF file and how the drums sound. At first I didn’t believe this, but then played them both back multiple times.
The WAV version has a little more weight to it or feels slightly more boomy on the drums. The AIFF version sounds more natural and not as weighted. I use iTunes to play back my music. iTunes is set to output at 192kHz-24bit and my Schiit Dac does the D/A.
Any thoughts on how this can be. I understand that WAV and AIFF are both uncompressed file formats, but shouldn’t they sound exactly the same?
Would like to get your thoughts on this. BTW, really like your site and I am reading a lot of your posts from over the years, great stuff!
Thank you,
Mike
Mike, it’s impossible to say what you’re hearing. Under controlled conditions, an uncompressed WAV and AIFF file would contain exactly the same raw data. It’s only the header and the arrangement of the data stream that are different. This would result in files that would convert into identical fidelity.
Mark, thank you for the feedback.
Mike, if you have a Mac, you may want to use an audio player such as Audirvana , JRiver or Amarra and see if you still here the difference. I use Audirvana with a Gumby and just about all of my files are WAV or AIFF. I don’t hear any difference between them.
Joe,
I appreciate the the information on the other audio players. I have used JRiver before and unfortunately was not a fan of the layout/presentation of the player. I will give the other two a try.
Mike
Mark, care to share your preferred speaker?
I’ve been a fan of B&W for a very long time. But I think the new JBL M2 Studio Reference model is about the best I’ve heard. Would love to have a set in the studio.
” I don’t want to mention his name but trust me this guy is the real deal. He was the music director for one of the world’s biggest music celebrities for over a decade (he’s played the SuperBowl).”
Oh, well that proves it. BTW, when he played the Super Bowl, was he on the winning side? 🙂
“Does anyone else see a problem here?”
Yeah, me. It all flies in the face of everything I learned from four years of very hard work in engineering school. Not one word of any of it makes any sense to me. However, I will admit that the codec is clever but not infallible because it is based on the assumption that there are areas in the information channel that are not ever used for anything else.
Mark,
Please, clear up this doubt. Is Meridian MQA a process to upgrade the audio of original recordings, or it is simply a way to compress the audio stream without quality loss in order to fill in smaller buckets?
Ron, MQA could be considered an “upgrade” to what we have now…or a method to preserve the original fidelity of the source. And they do it in a much smaller footprint. I don’t regard MQA as an enhancement technology.
MQA is a lossy compression scheme that takes 24/192 audio down to 16/44 (a nearly 6x compression factor, impossible to do without losing information) and then packaged in a lossless 16/44 container like FLAC. Same idea as HDCD, really.
Of course the container has to be lossless, it would be impossible to decompress. say, MP3(MQA(original)). A 24/192 signal, even with lossy compression, will sound better than lossless 16/44, just like a 600kbps MP3 will sound better than 1980s Mac vintage 16-bit 22kHz audio. The idea is to take existing CD-quality streaming services like Tidal and retrofit them for HRA without having to change any of their infrastructure other than the player software.
What I object to is the weasel marketing that dances around the fact MQA compression is lossy, by misleadingly insisting on the lossless quality of the container. It’s deceptive, and the fact it is used to push a proprietary algorithm (with probably nice Dolby-style royalties for Meridian) makes it doubly shady. Disk space and bandwidth is inexpensive nowadays, there is no reason not to just use lossless 24/96.
Those were my thoughts exactly:
1) basic DSP theory tells us that MQA compression is lossy
2) these claims are being used to push a proprietary algorithm (unnecessarily)
3) the streaming bandwidth argument is hokey: CD quality is 1.4Mbps – to go to 24bits would be 2.1Mbps – to go to 96KHz would be 4.2 Mpbs – contrast this with Netflix needs 5Mbps for HD, and 25MBps for 4k video: seriously, why exactly does one need streaming to fit in a CD bandwidth container in this day and age???
4) if the impulse function of older ADC converters were not that good – and the first step of MQA is to fix this – well that is a linear operation – which guess what? – means we could fix this in an existing 96KHz 24bit stream just fine -no need to do subsequent compression
Overall there might be some good stuff going on (such as the latter) – but agree it feels a bit deceptive in the way things are being represented