INTERNET BLIND TEST: MQA Core Decoding vs. Standard Hi-Res (24/88 or 24/96)

Testing a hypothesis sounds like a great idea to me.

 

Basic principle is to establish a null hypothesis. So in your case:

There is no difference between a MQA and a non MQA file of the same mastering (the test assumes all variables are controlled as constant except for the what is being tested ie MQA v non MQA).

 

I made a statement for effect but didn't spell out my hypothesis which was that they sound the same, as you can see from my other bullet points. My goal is to answer the question does MQA sound better. The route I've chosen has the ability to follow a logical progression and hopefully discount any other interpretation of the study along the way. I could have proposed MQA sounds superior it would actually make no difference to the route I would take, I should have made it clear rather than joke about it but I thought the joke was obvious, given the other bullet points.

My provisional assessment at the moment is that the chance of success is low given all the inherent problem but it's a nice little logical exercise, this is too similar to my day job not to have a bit of fun with it.

 

The question is: How would you define "better" in the context of audio comparison?

 

You've spotted what is probably Meridian's Achilles heel, my initial thought is that we definalty need more than binary data but I suspect we're stuffed.

Just had a quick look at the outcome and one interesting question why did 50% choose the worst sounding clip for NORTH COUNTRY II

 

I am curious if anyone on this forum purchases MQA downloads?

 

Fair enough, and jokes aside, the first hurdle is to demonstrate that the null hypothesis does not hold. This still doesn't prove that there is no difference between the two (in the same way we cannot prove unicorns do not exist) but at at least it provides a rational reason to stop there and not worry about whether MQA sounds better.

However if you do manage to demonstrate that the null hypotheses does not hold, ie there is a difference, then further time consuming experiments can be conducted to determine which one sounds better to the majority of listeners.

 

I've purchased a few (and I do have a MQA enabled DAC/streamer).

Apart from curiosity, I purchased one or two as they were made from better masters. It is the same reason why I still buy CDs, LPs and Hi Res formats. For any album I go with the format that was made with the best (to my ears) master.

 

Thanks for leaving this up. I finally got around to trying it. I thought the first track was obvious. I'd like to think the vocals make it easier for me to hear the differences. The second two were guesses for me. Guessed right on one and wrong on the other. Officially 2 out of 3 but the differences in the second two were very subtle if I heard anything at all, while I had a very high confidence on the first track.

 

I downloaded the file and listened to the 3 pairs of clips.
So, how do you see which clips were MQA/standard Hi-Rez?

 

Check the results.

 

Ah, thanks. I was confident I had all 3 sets correctly identified, but, was wrong on one of the sets.

 

Yes indeed, but it has to be the right hypothesis. The MQA folk have proposed a way to stream Hires content which, at the time at least, was not possible due to bandwith restrictions. So the correct hypotheses would be, does MQA sound better than CD (which can be streamed)? So the anti-thesis, which needs to be disproved, is that CD sounds at least as good. Unfortunately for the MQA folk, my conclusion after listening to quite a few MQA albums, is that CD does indeed sound at least as good. As to whether Hires sounds better then either of those two, well, that's another question ...

 

The idea that streaming high-res content was not possible until MQA came along is at odds with the fact that people have been easily streaming 1080p video (that includes full bandwidth, multichannel audio) for years (cable TV, Apple TV, etc.) and 4K video more recently, both of which require far more bandwidth and stream at far higher data rates than high-res audio alone. It has been well over a decade since any bandwidth limitation might have restricted high-res audio streaming. More important, neither TIDAL nor any of the other MQA sign-ups (2L, etc.) have even once mentioned bandwidth costs as a reason to pay for an MQA license. Consumers have simply stayed away from high-res audio in droves. Basically, the vast majority of consumers can't hear enough of a difference between CD, TIDAL HiFi streaming and high-res downloads. That audiophiles want high-res to be a pervasive standard is not in contention here, rather only the idea that its audible benefits can be only subtle at best for many, many people. MQA, as you've noted, improves nothing.

 

Are you actually for real?

 

I believe he is being honest. The problem with scientific papers is a known issue. It's worse in soft science disciplines like medicine and nutrition and better in hard science disciplines like physics, but it's something which is real. A good scientific BS detector to have is being able to spot which papers reports meta-analysis, observational studies which confuse correlation with causation, studies that provides "proof" by means of teleoanalysis and falls into logical fallacies or papers that confuse relative with absolute risk and the list goes on.

I was gratified to see that @Archimago experience led to results that matched my expectations and he published the results in a manner that is far more in line with the scientific method than the poetry waxed by some HiFi magazine writers and editors. But it's not a controlled study and there are many problems with basing research on answers from questionnaire . The other side of the coin is that at least he tried. He described the method, published the data, it's conclusions and therefore it's an experiment that can be replicated, confirmed or disproved and used to propose new theory to explain the phenomena. That's basically the exact opposite of what MQA did.

 

No, no, no. You misunderstand. There is no inherent issue with scientific papers. The problem is ones flawed interpretation of data.
I will admit I havnt read the whole thing. But to discredit this work by saying science is faulty without giving any reasonable argument besides stating conspiracy theory level possibilities like that people are just voting without listening is the lowest of blows.
There are no proofs outside math and we dont need any.
Im also sure there are more tests like this one which can be used as additional data to support whatever conclusion they point to.

 

I'm not sure we are on the same page.

 

Without wishing to stray too far off the topic

When scientific papers are redacted, how does it impede the progress of science?

 

If I somehow misunderstood whats discussed here then I apologize, but the only argumemts I see directed toward the study are strawmen.

If you think the study is flawed then you need a good reason for why.

 

Well, I can only speak for myself, of course. Until I upgraded to Verizon FiOS (fibre-optic) earlier this year I was on DSL for about 15 years. I could only watch Standard Def video and Tidal MQA would stutter badly. Now I can watch 1080p streaming but not 4K and certainly not 8K. But I think the 1080p is aggressively compressed, nowhere near the image clarity of a Bluray disc.

Regardless, I don’t think MQA is setting out to please aging audiophiles such as myself, simply because that’s not where the money is. Rather, it is all about getting the youngsters weaned off mp3 and onto a higher-quality sound, and incidentally, that means signing up paid-for services. This requires compression as the kids don’t care about cable or fibre-optics – they stream on handheld devices off WiFi stations all over town. You may have heard that ESS Technologies is bringing an MQA enabled chip for use on smartphones and that, my friend, is no coincidence.

 

I was one of the participants in @Archimago experiment, so I'm hardly an objective and non-biased party here. But why don't you PM him directly, or post on his site. I'm sure he will be able to explain himself what are the limits and weaknesses of the protocol he used.

 

Your TIDAL experience is unfortunate, but it has little to do with datarates or bandwidth required by TIDAL and everything to do with the awful service being provided by Verizon on a local network that appears to have been resegmented so often that it had turned into a cranky useless mess for anything other than email and web surfing. The fact remains that even compressed 1080p requires more bandwidth than high-res streaming. If your new Verizon FiOS connection is not allowing you to watch 4K, then the implementation and connection is deeply at fault. Verizon should be spanked. Seriously.

TIDAL, Qobuz, and indeed all the high quality pay-for-play services that require hardware of a certain cost and monthly content fees of a certain cost are absolutely depending on disposible inome of Boomers and their 30-40-something children for viability.

The MQA gang doesn’t care about weaning MP3 users off that low quality format. There’s no money in that - Apple owns the iTunes business, MP3 is free, most end users have a hard time telling the differrence between 128Kbps and 720Kbps through ear buds, and adding the initialism MQA to a portable music player or smartphone or tablet doesn’t mean anything to anyone because it’s just another group of three letters amongst a sea of acronyms and initialism that have nothing other than a vague marketing influence on consumers. A widely marketed Super Duper Pooper Scooper is still just a cheap tool to clean up after your dog.

Anyway, weaning MP3 users has never once been mentioned in any significant way by anyone from the MQA gang. It has only been mentioned in passing as a marketing bullet point. The MQA gang doesn’t care about high quality sound because they’re certainly not providing it. The MQA gang is interested in licensing fees. Stuart and company have devised a clever method of and vehicle for generating licensing fees from audio companies and music companies. Mazel tov to them - every new schtick deserves at least a single round of applause - but none of it has anything to do with better sounding music or improving the lot of MP3 listeners who are happily enjoying their music and who aren’t going to pay (for example) $20 for something they already have right now for $10 or less, or more likely ad-supported for free. Problem? What problem?

Boomers and others with disposible income spend $20 a month for TIDAL HiFi and Qobuz. The audiophiles who are unable to afford TIDAL and Qobuz for now, will either be able to in the not to distant future or will find satisfying alternatives if their budgets are still too tight for the expense.

High-res streaming is a luxury. MQA does nothing whatsoever to change that, and may in the future actually make it unnecessarily more costly than it is today in addition to inevitable price increases generated or demanded by increasing industry costs in general.

 

"It is difficult to make predictions, especially about the future" as Mark Twain (or was it Yogi Berra, or someone else entirely) famously said . But it seems to me that Dodo Bird is about to come around to HiFi land again. He already has his collection of SACD, CD & HDCD and now his sights are trained on Bluray & UHD Bluray. Another plate of Betamax, anyone? Those guys at SONY sure know how not to treat a customer...

I saw a quote by the marketing guy at MQA stating that the goal is to convert all available music to this format, and I think he means it. Sort of like how Wikipedia means to digitize and store all printed knowledge (adding their own 2 cents in the process, just as MQA does). When and if that has been accomplished, they will collect royalties regardless if content is distributed free, on a subscription basis, or pay-as-you-go. But then again, it may not play (pun intended) this way.

Personally, I don't care as I will always have my physical media (but what if that good old Oppo konks out)!.

 

That's certainly entirely logical, question becomes - how would you know before buying?