Seriously though I agree I would not and have not spent a penny on MQA yet. Just stream it from Tidal and do the first unfold in Roon no need for an MQA DAC and no need to spend anything extra on MQA.
1 Like
This review was done in 2016 during its early years of infancy but I guessed everyone has moved on… The bad marketing, confusion and controversies still haunt everyone today.
Even Roon offer MQA decoding, it’s still lack the rendering part and the only way to get all the ‘benefits’ you still have to get a MQA DAC. Without doubt, I will stick with PCM and DSD downloads as I can get full benefits of lossless and master quality with my DAC.
1 Like
Nobody is haunted here. Maybe haunted by Trump or Brexit but not by a music format!
That said I thought Darko became less and less enamoured with MQA subsequent to that review, not necessarily because of its sound but because of the doubts many express here. Some for good reason but some unfounded too. Personally I go on how it sounds, and it doesn’t sound universally bad to me. I don’t deny it might do to others but the worst properly authenticated MQA I have heard was down to the wonky master used which I think had stretched.
Perhaps worthy of discussion is the value of the second stage of MQA processing. My take is it is an adjustment for your DAC used. My take is not helpful to MQA hardware vendors. And that is what matters is DAC quality, and if you have a real issue with the filters used with an MQA DAC then the first unfold (compensating for the ADC) followed by your own choice of DAC and any processing you choose to use in Roon is the very best of all worlds.
I have a few MQA albums and CDs just for curiosity’s sake.
MQA is a lossy format. In my example it compresses the HiRes file to a lower samplerate (i.e. 352.8kHz to 88.2kHz or even 44.1kHz). It then “first unfolds” to 88.2, applying its magic sauce. Then in the second unfold, the resulting file gets upsampled to 352.8kHz.
The 44.1kHz file actually has a bitdepth of around 13bit with the other bits being used to somehow store the > 20kHz high frequency components that are missing in the 44.1 signal. How the original 24bit/352kHz signal can be reproduced accurately (as MQA claims) from the 16bit/44.1kHz file is beyond me. You are definitely losing information re. the original 24/352.8 file (just look at the differences in file size). MQA claims you are only losing things you can’t hear anyway and the MQA process is improving certain aspects in the time domain. Whether the resulting gain outweighs the things you lose is highly debatable
For sure, MQA is a good technology if the task is to transport a highres file through a limited bandwidth channel, such as when streaming with TIDAL. If your goal is to have the best possible music track quality, get the original highres file. I don’t know of a way to improve on the original (unless you remaster). You wouldn’t buy an aac or mp3 file instead of a CD would you, even if its bitrate is beyond the transparency threshold.
BTW, any differences I detected between an MQA track and the original were down to the different masterings used. 2L have different versions on the website. Original master with / without MQA, new remastering with / without MQA. You can compare yourself.
Any chance you can comment on the sound rather than the technology which has been batted to and fro to death.
You say it’s Lossy, implying MP3 type losses and MQA state No Musical information is lost. It just goes on. How does it sound is what interests me?
Particularly MQA CDs. I don’t have acces to any but would be interested in sound impressions… Decoded, undecode first unfold…
I don’t how to say this @Chrislayeruk, but I would say it is ‘musically lossless’ without the use of psychoacoustic compression like MP3/AAC. This is the area where MQA is not clear on this, so I guessed it is bad marketing?
Unfortunately many people are not convinced and the fact is, it is still a lossy codecs (technically lossy). My point is, this is not going away and will pop up here and there in any forums discussion.
I find none of the MQA albums / CDs I have sound better than the original. That’s enough for me to buy the (lossless) original and be done with watered down MQA. If the time domain magic MQA do did something worth writing home about, things might be different. But it doesn’t (at least to my ears listening to dCS Rossini).
Sorry if I am not more precise, but to me it is simply not worth doing any lengthy comparison.
MQA was developed by Bob Stewart, of Meridian fame. Hmmm, I wonder why MQA sounds so good on a Meridian?
It sounds damn good on my Bluesound Pulse 2 also…
sounds darn good on my Cary…
trust YOUR ears!
1 Like
Though I kind of wonder what you’re hearing, after reading this test.
But I think I have this figured out. There’s no point to buying MQA versions of recordings – it’s a secret format with no obvious advantages, and several obvious disadvantages, like lower resolution and how to decode it being a secret. Maybe once ffmpeg has codecs for it, I’ll reconsider. Meanwhile, I can rent it via Tidal or Qobuz and see if I can tell the difference with good headphones.
2 Likes
Patented so presumably this won’t happen?
No, they could release it for free use, but still retain the patent.
They could, but that’s not my read of what they plan to do based on what I’ve read. I mean, once you can get it from FFmpeg for free, how do they charge DAC manufacturers for royalties?
I don’t know anything about their company, but my understanding is that their format is only part of what they do. They remaster fixing issues with the previous masters, and they get everyone in the performance together to “authenticate” the particular recording. So there’s more to their value proposition than just the format, which doesn’t seem to really bring much to the party, compared to straight 24/88.2 PCM. So perhaps they concentrate on recognizable value, and perhaps a new format, like digitally signed FLAC.
Dunno about royalties. Perhaps the manufacturers are just as happy to have a new tech thingie to embed in their systems, so they can sell new systems to people whose old systems really work just fine.
Is this only true for MQA Studio? Or do you get a blue light with a nod from the copyright holder (who is neither artist, mastering engineer or producer)?
This must be the case. I cannot imagine that every MQA title available included a substantive approval from the artist or engineer. I am pretty confident that many of these were assembly line jobs. That doesn’t make them bad, but there’s no way that artists and engineers have sat around and accomplished approval of the thousands of titles available.
Well, then, what’s the point? It’s the creatives that matter.
[Perhaps I’m letting my personal bias intrude here.]
Well I have always thought the main point was the deblurring from the time domain damage caused by early ADCs. Of course the engineer already accounted for that in how they mastered the recording, but we can play along with MQA’s idea that there is damage to be undone.
It may be that someone like the artist/producer/engineer listened to and approved the MQA versions. I just bet that was a fairly shallow process and highly doubt they were in the studio working on it with MQA or providing substantive input of any real value. The Blue Light is supposed to authenticate no further changes from the approved version…I think somehow we are attributing value to that approval, and that may be where there is a disconnect. Why is the blue light so exciting if it just means it wasn’t changed since a fairly meaningless event?
1 Like
Nothing sounds good on Meridian…