Meridian MQA at CES 2015
I wanted to make sure and have time to visit the Meridian room in the Venetian Tower and chat with Robert Stuart, the main main behind the MQA (Master Quality Authenticated) technology that dramatically lowers the bandwidth (and storage size) of digital audio files. And by dramatic, it seems a 192 kHz24-bit stereo PCM file (or even 384 kHz) can be delivered as a data stream within about 1 Mbps…that’s less than a CD! Already there are some big announcements associated with this technology.
TIDAL, the Scandinavian streaming service that has been online for a few months with CD-spec streaming is among the first to state they have a deal with Meridian to include “high-resolution audio” to their streaming service. It will be interesting to see how TIDAL, who has been billing themselves as a “high-resolution music streaming service” will modify their marketing message and pricing when they start to offer streams sourced from higher sample rates and longer word lengths (the kind of “so-called” high-resolution files that are widely available from HDtracks, PonoMusic, SuperHiRez, and other sites…and which total about 10,000-15,000 albums or 100,000 tracks.). If moving from lossy streams to CD spec audio bumped the price by a factor of 2 (from $10 to $20 per month) was appropriate, then how much of a premium will the move to high-resolution analog remasters cost us? I’m keeping an eye out for their next move. You can watch an interesting interview with Pal Bratelund, Tidal’s Strategic Partnership Manager at What HiFi?
Robert Stuart was a panelist on a session in the HRA Marketplace on Wednesday and talked about the MQA technology. I’ll be posting the video of that session asap. I chatted with him briefly as he exited the room and asked when I could get a chunk of his time. He indicated Friday morning things in the Meridian room would be calming down, so I made a point of visiting in the late morning. They closed the door and an hour later I exited. It got a thorough explanation, demonstration, and commitment to have some of my files converted to MQA files. Robert also told me that they would send one of their portable players to me so that I could check it out in my own studio. I look forward to that opportunity.
In the meantime, Robert walked me through the thinking that went into the MQA development. He laid out the production path from recording or mastering studio to home listening as an “analog to analog” signal path with a challenging digital chunk in between. Their goal is to preserve the exact sound of the that final analog experience (even if it is being sourced from a digital audio workstation like Pro Tools) including the individual characteristics of the Analog to Digital and Digital to Analog converters. If the engineer, producer, and artist are listening in a studio equipped with dCS or Apogee converters, then the metadata about those devices will be “encapsulated” into the MQA sound file so that the playback mirrors exactly the experience in the studio…and at approximately one tenth the bandwidth.
I asked if this was basically a technique for getting 10 pounds of audio into a 1 pound container? He replied yes. But I really wanted to get to the bottom of the alleged improvement in the “fidelity” of the sound. If Pal Brateland of Tidal called it a “codec”, then I wanted to know how an encoding scheme could change or enhance the sound of the music being decoded? That’s when Robert opened the distance between his hands and told me, “we’re expanding beyond just the digital to digital steps in the production chain to include the analog front and backends.” I get it. There isn’t really any sonic improvement involved. Meridian’s MQA technology isn’t really anything more than a great codec that links information from the original session to the delivery DAC via metadata. The ADC and DAC are talking to each other. That’s pretty clever.
Then came the demonstrations…lots of them. I’ll give you my impressions tomorrow.
I am what’s called a VIP subscriber to Tidal. My first month was free and I am now paying $13/month by paying for 3 months at once. I don’t know how this will continue, but for now I find it an excellent alternative to buying 1/2 of a high-res album every month. I use a Wadia DAC that has always sounded great with CDs and high-res tracks.
The economics for end users work out really well. I actually wonder if the TIDAL folks have the same arrangement with the artists and ultimately the artists. If there’s no upside to making the albums available as streams, the situation might be fluid.
Meridian’s MQA technology sounds brilliant. I am anxious to hear your impressions after actually listening, Mark.
‘That’s pretty clever’. And Tidal (over here it is WiMP) wants to join in.
I’ll keep that in mind for when the real HiRes files are available 😉
I fear it might take a looong time.
The fact is there just aren’t that many high-resolution files around…if you adhere to my definition. Certainly, not enough to support PonoMusic or HDtracks.
So (almost ) noone will benefit from the MQA for a long time.
I fear – in the hands of the ‘wrong’ people – this will just be another marketing gimmick.
The MQA technology might have some advantages, but the responsible people in the industry should focus much more on the quality of recording/mastering.
It’s a big deal for getting the archived analog tapes up to their maximum presentation…and for streaming services like TIDAL.
The question I have is whether you need to upgrade your DAC or whether the decoding can be done via software such as JRiver or Audirvana before the computer sends the stream to the DAC.
You will need upgraded hardware and software to handle the new MQA streams…or without upgrading you will get normal FLAC quality. I should ask Bob if the bandwidth of the FLAC files will be the limiting factor.
Hello Mark,
you need update hardware (DAC firmware) OR software (audio application) to handle the new MQA streams…
Hello Mark,
Is there any kind of DRM or copy protection feature in MQA for the benefit of the streaming and download service providers? Thanks.
I don’t believe so.
… so if metadata about the characteristics of the ADC and DAC used in the studio is embedded in the signal, my cheap DAC will sound like a Benchmark? I thnk not. There are ways that this could be done, for example the coefficients for the filters in the studio DAC could be included as metadata to be loaded into the filter in my DAC, assuming compatible architecure. But I’ve seen nothing in the patent or presentations to indicate that MQA is anything more than a sophisticated sompression scheme.
Don, Robert specifically mentioned several brand names of converters during his presentation to me. I was curious how hid new “codec” improves the fidelity…and that’s what he said. It does make sense but there are so many other factors (room acoustics, speakers, amplifiers etc) that would need modelling or measurement that I’m still waiting to see.
He appears to be saying that any compromises made by the ADC when digitising the input can be communicated to the DAC and compensated for when doing the D-to-A conversion. But to do this you have to quantify your ADC behaviour. And if you know this, you can do the compensation within the ADC, rather than at the DAC. It also doesn’t work when the input is already digitised, as in all the examples in his patent. If he has developed a system for linking the ADC to the DAC, I would have expected him to patent it because he intends to license the MQA technology. Hopefully all will become clear in time.
You’re right…I’ll be following up with him and try to keep you posted.
Hmmmm….
http://www.stereophile.com/content/ive-heard-future-streaming-meridians-mqa
Scroll down to “Correcting The Source” and read the next two paragraphs.
(Apologies if you’ve seen it already.)
It implies using DSP to correct for “temporal smearing” caused by the (necessary) filters in the ADC and DAC.
Temporal smearing is the current digital bogeyman, replacing jitter.
Thanks
How to correct for ADC behavior after the recording is already in the digital domain was covered in a comment added by John Atkinson to his Stereophile article on MQA. Basically, Meridian has modeled the behavior of a few of the most popular ADCs used in transfers and original recordings. Often , the specific equipment used is in the notes that go along with the digital master. If the notes indicate one of those modeled ADCs, then the correction can be applied. That’s the genesis of the claim that it improves the sound. Anyone interested in more details on MQA should give that article a read – it’s still the only one I’ve found by anybody with technical credentials. Meridian would do well to put more of that information on their own website.
Andrea, it might have application where the intent is to “flat transfer” an analogue master, but in almost every other situation it would be a modification of the originally approved work. The engineer/producer achieved and approved the sound as captured by the ADC and heard in the studio. Making it sound “better”will make it sound different from the approved version. (Assuming, of course, that the difference is actually audible.)
Actually, there’s just no need in making shifts in sound quality for still smaller file sizes: rumanian OptimFROG does it all lot better than anything similar & is very apropos for storing moderately upsampled audio data wherever! Further improvements can only be obtained with PAQ.
And since Meridian with their MLP and now MQA are unable to contend with OptimFROG then why deceive people ?
We don’t need and fidelity doesn’t benefit from upsampling…all you get is a bigger file. And OptimFROP give you about 25% reduction…MQA gives you much more (10 times better). Why bring up things that aren’t relevant?
192 kHz 24-bit stereo with 1 Mbps bit rate equals to HD MP3 + some amount of interpolation and probably noise shaping {but then again dithering must be applied to avoid sound deterioration} that ousts quantization noise {generated by the binary word} just up to your cherished higher frequencies thus making a PCM recording a DSD one {otherwise the noise will be very annoying since it is digital by nature, unlike nonlinearities}
So.. OptimFROG is a fully lossless compression technique whereas MQA is lossy but trying to get higher by means of some kind of upsampling {by the way, the Benchmark DAC resamples to 110 kHz}.
Seems like Meridian are trying to cheat with anti-alias filter steepness to squeeze better sounding . . .
I can’t say I’ve looked into HD MP3 but I’d be surprised if it could take a 384 kHz file and losslessly shrink it to 1 Mbps. MQA is not lossy.
Just look into a FLAC or an APE, both being considered lossless, and you’ll see typical bit rates around 800-900 kbps compressed CD format, implying Meridian to be utilizing their MLP which, as known, performs quite poor, hereof the ~ 1 Mbps.
In fact, Meridian Lossless Packing works very well…that why the DVD forum chose it for the DVD-Audio format.
This MQA idea has legs….I look forward to hearing more about it from you (you cut through all the hype).
Sorry I missed you at CES. Too much to see/hear…too little time. I did get to the Dirac room. This is the technology you mentioned that is built into the miniDSP processor you noted in another blog.This Swedish room correction technology is going to be built into some popular AV equipment as well as cars like Bentley and Audi. The before/after demo was very impressive.
Here’s an article you may have missed. Right conclusion….wrong reasoning.
http://gizmodo.com/dont-buy-what-neil-young-is-selling-1678446860
Keep fighting for better audio!
I just read the Gizmodo post…and I agree that they guy doesn’t know the reasons but he came to the right conclusion.
More info on how it works, in a Q& A by Bob Stuart:
http://www.meridianunplugged.com/ubbthreads/ubbthreads.php?ubb=showflat&Number=227413