MQA - One Filter to Rule them All?


Hi Everyone,

Just thought I'd start another flame....er, discussion. I've been reading some about MQA.  It has several components, but I want to focus on one in particular. The digital filter compensation.

The other two parts are compression and authentication.

We don't have a lot of DAC's to listen to with MQA right now, but here's my understanding.

By measuring time or amplitude errors in ADC's AND DAC's MQA seeks to correct the behavior, making the entire A/D --> D/A chain closer to ideal. It's pretty ambitious. What I'm wondering is, assuming this is real and not snake oil, does this mean all MQA DAC's will begin to sound alike? Will otherwise mediocre DAC's step up, and great DAC's not have that much to contribute anymore?

If so, maybe this will usher in another great era of tone controls being built into our preamps or DAC's instead of having to make tonal changes via cables and tweaks.

What say you? Assuming MQA is not snake oil, (could be, haven't heard it) doesn't it mean all DAC's will sound the same?

Best,


Erik
erik_squires

Showing 8 responses by erik_squires

@ahender you are not alone, however I think there's a bit of a mistake there in terms of the generic process.

It's a 2 part process.  During encoding they attempt to remove any effects from the ADC but after this it's expected that the MQA decoding process takes into account the DAC and attempts to compensate for it.  It's most certainly NOT generic. The DAC must be measured, corrected and submitted to Meridian for certification. 

Meridian _do_ have a generic ADC compensator however, for encoding previously recorded high resolution sources when the correct filters are not known. I think that's maybe what you were thinking of. 

Best,


Erik 
@kijanki Great, thanks! :)

Interesting reading. 99% of that writing has nothing to do with sound quality but marketting and branding concerns, which are very very real. I do worry it’s just a way of taking money out of the market, like GMO corn seeds.

If I read it correctly for Schiit it boils down to being some one new to pay money to, in addition to questions about how well they would be treated as a minor (in volume) manufacturer.

I think Schiit is playing it safe. MQA is brand new and by 2017 could be useless. If Tidal goes down, or worse, gets bought by Apple, it may be all over.  One other point is differentiation.  MQA could very well be the opposite of this, if all DAC's will be forced to sound much more similar than before. The value proposition is going to have to be high, because honestly I don't think most people will fully understand what's going on with it besides compression. 

There’s also the possibility that the algorithms will be analyzed and, using that knowledge, better, open source solutions will be able to leap frog over them. It’s a very dynamic situation.

Best,


Erik
@cvalle as far as I know you are mostly correct, however, I have a personal connection ( cough cough cough, bs) with Berkeley Audio Design and they have stated they hope to release MQA compatability this year (not bs).

If Tidal support is real, AND the benefits are real, it will be like selling a good pie. You don’t have to sell it, the wafting smell will sell itself. Otherwise, it will be like Windows 10, which requires Microsoft to take over your computer, possibly illegally, and push an update you didn’t want.

You are right, of course, all of the current models of the Mytek DAC’s have MQA support, including the Brooklyn which I’m listening to right now. The older, Stereo 192 DSD does not, but it’s also insanely cheap via the website right now, $1,000 as a discontinued product.

Best,


Erik

Hi @bombaywalla

Haha, I don’t have that kind of cash flow, and I’m not the biggest BAD fanboy. I like them a lot, but overall I think I prefer warmer sound.

According to what I read online, yes, Mytek did in fact submit the DAC’s for approval, or measurement or whatever. I’m not sure how it works, or what exactly they must measure. It sounds like a pain in the butt honestly. I doubt every applicant has the same or appropriate measurement devices, so this must mean that Mytek has to submit the DAC somehow, get Meridian to measure it, and provide refreshed code or coefficients, have those implemented, and re-measure. What a mess.

As I understand it, they measure the behavior of the entire DAC, not just the chip. It would be really convenient if it was just per chip so there’s just one set of adjustments for a BB 20459 (or whatever). This is one reason why I read Schiit is not on the bandwagon (among others). Perhaps in the future as they get more experience this is how things will go, so it will make it a lot easier for small-volume makers to get pre-approved decoding.

It may even be possible MQA decoding will be build into streamers, so you can configure your steamer and say "My DAC uses an AKxxxx chip" and voila. Then you could be all Schiit happy and have MQA. On the other hand, once the type of benefit is known (if any), maybe an open-source DAC adapter will be written which can provide at least the proper de-coding adjustment.  That is, if Meridian is measuring something like a pre-ringing filter, adjusting for that in the stream could be done by others without stealing their code, just the idea.

After the Alpha with MQA comes out maybe i’ll find a fellow San Francisco Bay audiophile to bring it and compare though. Right now I am just hoping for construction to finish so I can finish breaking in the Brooklyn. :)

Best,


Erik
@tomcy6 I wonder. ;)  Im far too jaded to believe in price being completely tied to performance.

Good to know about a bargain player. 

thanks!!
@mike_in_nc

Good points.  I'm still waiting to see if there will be an open-source alternative. To my knowledge, you cannot patent or copyright decoding, or a format. You can patent your specific code, but you cannot, for instance, prevent some one from writing a Zip, Jpeg,  MPEG or MP3 decoder. I may be wrong on this, please don't take this as sound legal advice.

It's quite possible that if the benefits of MQA are real, we'll see just such things. If nothing else, the promise of compressing my current music collection to half it's size and being able to take it along on my portable is a very nice thing.



@mike_in_nc

No heresies.  I remember HDCD, there was also a Windows decoder, which I once thought about reverse engineering and converting to Java, or C for Linux but lost interest. :) It actually decompiled fairly nicely.  I believe Microsoft ended up with the HDCD license (lots of good it did them) and that's how the decoder came about.

I don't remember over-the-top reviews. Where there any?

I'm not sure what the lesson is except that you can only tell a fad in retrospect. :) The same could be said for SACD and yet the format remains.

MQA does have the undeniably useful feature of compression, even if we ignore all the sonic claims. I think that one way or another, their view on how to compress audio will remain relevant for years to come. Whether it lives on in a proprietary format or in open source formats remains to be seen.  Technology, specifically bandwidth and storage, may have moved onto the point that it has become irrelevant.  20 years ago, MQA compression would have been a godsend and could have saved high end music from the MP3 era. 

Best,


Erik

OK, I just gave up on MQA. Stereophile's latest article starts with such shameless hyperbole that it's gone off into the land of cold-fusion, dark energy reactors, and healing cancer with magnets.

It's not like I didn't know Stereophile has to make a buck to stay in business, but this level of bs is higher than my waders. I can't be polite and smile anymore.

Now I'm going to go try to do some DSD streaming and see if I find a compelling reason to eat up all my hard disk space with DSD downloads.

Best,

Erik