MQA is Vaporware

[Ryan Berry]

On 4/14/2021 at 5:53 PM, GoldenOne said:

Just as an update, MQA did respond to my email (AFTER the posts about censorship were made here and elsewhere).
Most of the content in their response was either marketing talk, or easily counter-able (or outright easily disproven) claims. 

My video is done, including my response to their response. Its up on my patreon atm as I put things on there early but it'll be up on here as well as a written version very soon.

 

I'm late to the show here, but I wanted to thank GoldenOne for putting the video together.  As a manufacturer, we we really off put by how difficult it was to really get to compare the technologies before having to sign on a line and had to arrange for our own listening tests using some masters we had access to.  Suffice it to say for that and several other reasons, along with Charley's already pretty outspoken issues with the technology, we never had much desire to come back to the table.

I can't fault manufacturers that decided to support the technology for whatever reason.  There's some places in a listening test that makes you say, "Oh, I get why people like that," but we found that overall, the bad things it seemed to do to audio tracks outweighed the few good, at least to our ears.  We also know that a GOOD Minimum Phase filter to us is preferable, so I can see understand people hearing an improvement designs that may not have it as an option.  However, more than anything, I understand the pressure one can get from the market.  There was a time we couldn't go a week without someone asking if we had or were going to support MQA, and there was a few conversations about entertaining the idea of throwing it on just to say we support it.  Ultimately, we decided to hold out and those phone calls eventually faded away to maybe one or two a year.  The idea of losing sales can always make it tough to stick to a decision.

I know Charley would have loved watching your video.  Again, thanks for your work on it.  Maybe they'll decide to take a look at what you've done and find a way to improve what their process (if it can be).  You never know.

[Ryan Berry]

On 4/17/2021 at 9:55 AM, MikeyFresh said:

For every ADC and combination of different ADCs ever produced and used on any given recording? I do not believe that one bit.

 

And as long as one can distill every DAC and ADC down to simply the chip being used and ignore the idea that analog circuit designs, power supply designs, etc. all have a factor in what a piece sounds like, of course.  

[Ryan Berry]

On 8/25/2021 at 4:29 PM, John_Atkinson said:

Before Charley passed he sent me an experimental complementary reconstruction filter for the Ayre D/A converters. This allowed perfect time-domain behavior throughout the recording-reproduction chain, just as is claimed for MQA. That was ironic indeed, given Charley's hatred of MQA. For reasons unknown Ayre never released this reconstruction filter.

I'm not really certain I'd call it ironic, John.  Charley's objections to MQA were never based around what filter they were using.  In fact, that bit of information wasn't really even available early on, at least as presented to us.  His issue was the DRM nature of the format, the obviously lossy compression despite what they wanted to claim it at the time, and the inability to deliver on the promises it was being set up to fulfill for high end audio, which, at least to everyone's ears at Ayre, never was close to living up to.  I'm rather limited on what I can talk about, but suffice it to say, Ayre's and MQA's core principles on audio have far more differences than similarities. 

[Ryan Berry]

On 8/27/2021 at 2:50 AM, firedog said:

But Charley also didn't like the sound. Note the early praise by several reviewers of the MQA version of  The Doors' "Riders on the Storm (M. Lavorgna and others). Charlie specifically pointed out that what they were describing as sounding superior was actually a version that hid some of the detail available in non-MQA listening. That came across to them as somehow better. 

 

You're exactly right.  In fact, I was able to get the first listen in on these before bringing Charley in, ironically with that very song.  I think the best way I could describe it was hitting the "Bass Boost" button on the old Sony Walkman's...it was fun for it a bit, but it certainly wasn't right in more ways than it was fun.  To get that fun, certain elements of the track were nothing less than completely missing that we were used to hearing in the song, and left us wanting to hear them again...not a tradeoff we considered worth making.

 

On 8/27/2021 at 3:00 AM, John Dyson said:

When doing such comparisons, the MQA source might have/have not started with the same master.   It is pretty easy (if the distributors/studios agree) to get a MUCH BETTER or WORSE copy of a recording.   Right now, we get an 'in-between' quality.  The MQA advocates just need to get whatever quality (not normally distributed) that might be internally available.

 

Agreed there.  I can't go into many details, but suffice it to say that we were in a fortunate position to compare apples to apples in our listening tests.

[Ryan Berry]

On 9/25/2021 at 2:14 PM, bogi said:

If MQA would widely succeed then it could become very tempting for other coons to try the same business model: forcing people to pay for nonsense, because manufacturers included it in their products.

 

I think it would be difficult to see it succeed twice.  The studios are desperate for another get rich scheme.  MQA provides a means to set up an "exclusive streaming service", akin to what we see with Disney+ and the like for video, and lets them effectively sell their catalog to us all over again in a constant monthly revenue stream.  I don't expect that they'll want to adopt 2, 3, 4 different formats.  They don't need to.  The bigger concern for me is the belief that if they're able to lock down a DRM format like MQA, we'll never see music again in a format that doesn't, "replace noise with other very similar noise," while pretending to be a magic format that's better than the original.

[Ryan Berry]

I was really hopeful when I saw the tier news from Tidal and that MQA was "HiFi Plus".  I thought I could just get the lower tier and MQA would stop infiltrating my listening tests (I do everything I can to shut it off), but if we're just getting lower rate MQA files even in the plain "HiFi" tier, it looks like I still need to be really cautious.

 

As a story: just the other day I was doing some pretty serious listening with one of our QX-5's that the customer said just didn't sound very good to him.  He purchased it and was really excited to get into digital audio and we were really scratching our heads at what may have been wrong with the unit.  It measured perfectly, it sounded great, I had multiple people (including visitors) listening to it and never heard a problem.  We just couldn't figure it out.  Finally decided to roll through some other music sources and did some testing with the Ethernet using our Tidal account.  The first song sounded TERRIBLE.  We had the same song on our network, so I switched from the Tidal library to the network library with BubbleUPnP and it was night and day different.  The song stopped sounding like it had distorted bass and that it was being sampled from a MP3 file.

So back I went to Tidal and found another song.  The next song was great!  So now I was really scratching my head.  Maybe it was a bad recording?  I went back to the original song as I knew there was about 10 different hits pop up when I searched for it and that's when I saw the dreaded red "M" next to the song.  I looked back at the other song in Tidal and sure enough, no red M.  I thought I had changed my settings so that it wouldn't play MQA files, but I believe I did that running Tidal through MConnect, not BubbleUPnP, which uses a different app to interface with Tidal.  At least it was a good reminder how bad the technology sounds.  It's a shame it's so difficult to keep Tidal from forcing it into my system.  For the end of the story, found out the customer was using Tidal as well, so he plans on trying it with another service to see what he thinks.  I suspect he'll have good results.

[Ryan Berry]

55 minutes ago, Axiom05 said:

Simple solution Ryan, dump Tidal. 😁 Talk about an unreliable source for evaluating equipment.

Yeah, unfortunately I need it for testing purposes as a manufacturer.  Not my go-to right now, for certain.  It's a pity, it had potential.

[Ryan Berry]

4 hours ago, Abtr said:

OK, you heard what you heard whatever that was with some systems. I don't think you are being dishonest, but a controlled scientific study indicates that there is no audible difference and so does my own experience. I think there's nothing in MQA that permits a vast audible difference from one playback system to the next as you suggest. Maybe you should do the work and try a controlled (blind) test on a system that you think makes MQA sound like garbage relative to e.g. redbook.

 

I agree with your point that it makes no sense to go to MQA from a consumer's point of view because it doesn't improve the music. But it also doesn't vastly degrade music (on any system), which is my point.

 

There's a lot of moving parts at play here that make comparative studies extraordinarily difficult compared to comparing other formats, not the least of which has been the obvious alteration of tracks to sound more impressive for reviewers when MQA was pitching their technology.  There was some uncomfortable shuffling when we let them know we have a copy of the original master files from our connections that we were going to use for A-B listening tests.  I felt this was even more to test appropriate at Ayre as we were already using a Minimum Phase filter in our products for years at that point, so there was less chance of being "wow'd" by introducing the filter into a product that may otherwise not have one when switching over to MQA mode.  

I can say that I have yet to hear a single track offered in MQA format that sounds anything like the the equivalent high-rez version from the same album.  I don't know if I've been unfortunate and just picked tracks at random that happen to have had a heavy remastering hand or had something go horribly wrong in the conversion process, but I can spot them instantly as they all suffer from sounding overly bass-inflated and lacking the same level of detail to my ears.  At best, it was "better MP3" to us.  I can also say that I've ran into accidentally listening to a MQA track enough times at this point where I only had to look at the file playing after the first 10-15 seconds to confirm it was one that snuck into a playlist on me in Tidal to feel confident that it's not a case of listener's bias at play.  Listening to the non-MQA version easily confirms this.  Similarly, artists who don't have the "Master" tag, such as Shawn Colvin, never seems to suffer from the same issue.  Why the McGill study find differently, I'm interested in looking into.  The differences were so stark, frankly, that I really haven't followed up much on what others have found since as I expected it would be obvious to any listener as tracks became more available and more companies continue to adopt a similar Minimum Phase filter as an option.