MQA disappointing

Not everyone uses Roon. MQA is a fix at source… for everyone to enjoy and low cost MQA DAC’s are more and more available. Portable users benefit especially.

And were are they getting this AD/DA info from at best they have around 16 pre defined filters and they pick one, there is no way they could have fingerprinted such accurate data in all hardware in the last two decades, plus recordings can be made up from multiple files bit depths and frequencies, a lot of archived releases have already been stored using very high quality hardware, it’s hard to believe that MQA is making this sound better than the original. What studios archive in MQA none probably, it’s just a proprietary consumer file to revive a flagging revenue stream for the record companies and the consumer will pay for it in the end by taking choice away

4 Likes

According to Bob Stuart in the “white glove” section of Bob Talks, they’ve been collecting A/D, D/A and recorder specs from all over the world for years. Do take the time to read the White Glove section, it’s clear what a labor of love trying to restore some of these old recordings is, and you’ll appreciate better what has gone into the converter corrections.
The D/A corrections are inside the D/A for licensed MQA DACs and not part of the upsampling filters. The A/D corrections similarly are separate from the downsampling filters.
I don’t think this is a flagging revenue stream at all!! Bob Stuart comments that streaming of old CDs is around 50% of all streaming, though newer releases are in high res. Second, there’s been a lot of attention to streaming lately with the UK doing a Parliamentary Study Review of streaming right now; the labels say that revenues are beginning to return to what they were in 2001 (before piracy hit) and Sony UK made a point that streaming is 80% of their total revenue. So not flagging.
MQA is not directly an archive format AFAIK.

Do you realize that’s what all converters do, whether MQA or conventional PCM? All converters have to upsample to get to the input sample rate of the sigma delta modulator. There is further upsampling inside the modulator. Most of the PCM upsampling is done on chip, though it can be in an FPGA. MQA merely substitutes its own upsampling filter designs for the PCM upsampling filters that exist in your DAC.

I agree the white glove treatment is probably a good project for the dozen or so releases but the majority does not get converted this way take the latest Warner’s 16bit conversion, batch processed no MQA studio validation just a re work of a 16bit file not even going back to the archive because it probably cost too much money. Some 7000 titles were converted to 24bit a while ago but have not been retrieved for conversion, even Apple request 24/48 for MFIT

2 Likes

Yes I realize that, but a normal dac playing a 24bit 192kHz FLAC is not going to need 8x upsampling.

The label conversions were done over 4 years. I’m not sure the words “batch conversion”, which people use derisively, is accurate. MQA spent a lot of time learning how to do encoding and they might have refined the process to the point that it can be done reasonably well in an automated way, while perhaps flagging the situations that need operator intervention. The labels also had engineering teams involved I believe.

'cmon man, you’re usually better with your rationalisations than that…

Let’s take Tidal/MQA Ltd’s claims of “millions” of MQA files for granted, and let’s be generous, and say that’s two million files.

Now, you say it’s “over four years”. Ok, so that’s 500k files a year. 500k files a year, let’s make it easy on ourselves and call that 2000 files per workday on average, 8 hours a day, 250 files an hour, 4 files a minute. On. Average. Over 4 years. And that’s being generous.

I’d say that at that rythm, the “teams of engineers” were needed to cart the “masters” from the vaults, if anything.

[edit: my original math was twice too nice to MQA Ltd)

3 Likes

Seems not logical, if they did them over a period of 4 years, they could have released those while doing so. (They do so now). If you make 10000 cars and you want to sell them, do you wait for all of them to be produced before selling the 1st? hmmmm

1 Like

At least divide by 12-15 to get the number of albums. Files means tracks. I don’t know whether every track needs recalibrating when they’re all from a single performance, same mikes, same noise floor.

I have absolutely no clue what the business model was here.

You may find that depends on the DAC and the filter mode selected. If a DAC works best when the final stage pipeline is operating at 352-384, or even 704-768, then that is what the device designer may choose for everything and so over-sample everything to make use of the best filter. Often they provide a choice of filters and some of them may indeed by non oversampling.

Many modern DACs chips are no longer the simple R2R and early DS devices they one were and these days rely very heavily upon DSP to work at their best, or even at all.

Their press release talked about millions of tracks. My arithmetic was already off by quite a bit, let’s not make it even more complicated for me here :wink:

Oh you can’t possibly be serious…

3 Likes

That’s true, but MQA makes applying alternative/bespoke correction nigh on impossible if you want a second unfold. My feelings is that MQA restricts my freedoms rather than expand possibilities.

5 Likes

Maybe not :slight_smile:

1 Like

It’s vague, but the exact details may change with 1.8

Bingo!

Digital Room Correction offers much greater sound-quality improvements than even the wildest MQA fans can promise. But that is fundamentally incompatible with MQA decoding in your DAC.

Roon circumvents this by splitting off the MQA data before applying DSP and then adding back the MQA data afterwards. The “authentication” is ruined, but unfolding can still take place.

If you’re not using Roon, however, …

2 Likes

The two most populare responses I got from people who claim MQA is better
when I say to them it’s actually sounding worse:

  1. “You probably listened to a different master in PCM that happened to be better”

Yeah right. I compared hundreds and ALL PCMs sounded better and the chances are very low they were from different masters as for all the track lengths in number of samples were exactly the same for PCM and MQA PCM. Not huge differences in sound but subtle. Different masters sound way different.
For most albums there even isn’t a second master available. So cut the crap please.

  1. “MQA sounded worse on headphones… You should listen with speakers”

Really? Why is that? So with MQA we cannot use headphones any more?
Another reason to ditch MQA then. A headphone rules out room reflections. It is the perfect tool to make comparisons. If it sounds worse using a headphone, then it certainly IS worse.

Ahh, I see, the room reflections must cover up the flaws, is that it?

The only ones that seem to keep pushing MQA are the ones that benefit from it :
Hardware selllers, studios, … and ignorant people who are already brainwashed by those.

The few that actually really like it are probably the ones that used the Dolby B function on old school cassette deck players. Not minding important details get lost. MQA reminded me of that today.

Of course you have to do an effort to compare, and Tidal left us no choice for doing so there.
If MQA was really that better, they would have given us a chance to compare.
I would suggest people to take a Qobuz account description for one month and do a comparison.
I did, and it took me 5 minutes to realize MQA is crap.

One of the funniest thing I saw was the release of MQA STUDIO tracks from Neil Young on Tidal
some time ago. Shining blue light … only … Neil Young was not happy with them at all.
So even that provenance light : :lying_face:

But of course the MQA likers have an answer to that : NY’s PONO player project. That must be it why he left the sinking ship. Or maybe he just doesn’t like the sound altering of MQA ? I don’t like it neither.

For me an MQA sounds like the original PCM is played through brand new speakers,
only they forgot to unwrap the speakers and their still in the wrapping material.
Important details went missing. MQA lovers call that “less fatiguing”… right.

I tried to believe MQA was better, but I compared, and MQA is worse for me
and apparently it’s worse for a lot of other people as well…

the AudioPhile Style thread “vaporware” is not that long because MQA is better.
this thread is also becoming long; and also not because MQA is better…
look at the title of the thread, the answer is there.

6 Likes

MQA is obfuscation at the source. There’s no “fix” because there is nothing to fix. The only possible advantage is that maybe there is some sort of pedigree that what I’m listening to is what the artist wanted. But so much of the music I love is recorded so awfully that it doesn’t really matter.

As someone who builds my own gear, what do I do with MQA? I can’t design around it, I can’t unpack it myself. It’s just more crap that I have to navigate around when looking for music.

Sheldon

6 Likes

There is plenty to fix with CD quality as is commonly known, pre ringing or blur at digital gateways is the issue. If you build your own DAC, at this time, you miss out on MQA unless you contact MQA LTD and ask their advice.
You do not lose anything over normal CD with MQA versions as they benefit from MQA at the front end and will sound better anyway.
MQA at far from being an exclusive source of music, so all the music you own so far is un affected and always will be and you can stream from other sources and continue to buy content.
I expect there is also a lot of negative expectation bias in the anti MQA world as people have made a decision they won’t like it. As much as there is positive expectation bias. Personally I am happy in a positive world…

1 Like