KeithL
Administrator 
Posts: 10,255
|
Post by KeithL on Mar 8, 2017 15:55:15 GMT -5
The problem, as I see it, is that a lossless digital CODEC is easy to verify - it's either lossless or it's not. And they were sort of trying to play MQA off as that in the beginning. However, now that we've all figured out or admitted that it isn't lossless, we're back to square one... They have a repair process for fixing up existing masters which they say makes them sound better (and many people seem to agree)... They also have a compression CODEC for sending content that they've processed to end users... And they're also claiming that, in order to get the best results, you really need to buy one of their special DACs.... Now I see a LOT of different claims, most of which aren't all that easy to verify... The only real problem is that you sort of need to buy into the whole thing as a package... although that depends on how you want to use it. For example, if you subscribe to Tidal, and that's what you want it for, then being asked to buy a $200 special DAC to make your Tidal subscription sound better doesn't seem unreasonable. (As long as it makes MOST of what you listen to on Tidal sound better rather than just a few albums.) However, since they're NOT claiming that it will make any of the CDs I already own sound better, and I don't happen to use Tidal (and Spotify is coming out with a lossless service).... And there are still only a few albums that have been processed with it that I might buy - and that MIGHT sound better that way - they still haven't convinced me. And, to be honest, when we were talking about a new and more accurate way of getting the content from point A to point B it was sort of easy.... HOWEVER, now that we're talking about something that's supposed to "make the music better", it's much more complicated. And, again to be honest, no amount of explanations about why it SHOULD sound better is going to take the place of proving that it actually does.... (And, no matter how that works out, we're still going to be talking about a SUBJECTIVE difference.... ) And, if they really want to start claiming that their process does a "more accurate end-to-end analog-to-analog transmission of the data", then they should be able to provide proof. We could start with some direct comparisons of the difference between the input and output with 24/192k PCM and with MQA. (Oddly, I haven't seen any such proof offered.) And, if such proof isn't possible, or won't be especially compelling, then they're going to have to settle for: "It sounds better, If you agree, then buy it." (And, if they're going to do it that way, then we really don't need to know how or why it sounds better, right?) |
|
|
|
Post by jefft51 on Mar 8, 2017 20:33:28 GMT -5
It seems to me that MQA 'could' make a master recording better if the master was not done well, otherwise to be Better it would have to be adding something not there in the first place and that is subjective. If my singing were Autotuned, it would make my singing better, but would not be accurate. You CAN test accuracy especially if you have access to some nice, expensive oscilloscopes, but it's much more difficult to test what sounds better. Jeff  |
|
|
|
Post by yves on Mar 9, 2017 5:36:46 GMT -5
You mean like reading 20,000 pages to understand the Affordable Health Care Act. I think what MQA expects is the same as what we were told regarding the Health Care Act - don't bother reading it, just vote for it and pass it. Think about how many people bought a DAC because they like to play and listen to digital music files, but they don't understand even the basic principles behind digital audio aka the Sampling Theory. Next, a guy named Bob Stuart comes along and he makes it even more complicated by combining the latest advancements in auditory neuroscience with it all among other serious stuff like coding so... yeah, of course it's going to be gruesome... kind of like trying to build your own set of coils for your rebuildable e-cig without wanting to learn Ohm's law (I always liked turtles, but nowadays I like vape clouds better as the latter are more acoustically transparent). |
|
KeithL
Administrator
Posts: 10,255
|
Post by KeithL on Mar 9, 2017 10:10:54 GMT -5
Absolutely..... The problem, as I see it, is that most people really don't have accurate and specific reference points. Even worse, in audio, sometimes the reference points themselves are just plain not worthwhile. How do you actually KNOW whether that recording sounds like the original performance in the studio or not? Were you actually there when it was recorded? And, before you try really hard to make your basement sound like a theater or concert venue, think long and hard about whether you really want it to. Personally, I go to lots of live concerts, and they're fun, but often they don't sound nearly as good as my home stereo system. My local iMax theater gets louder than my living room - but my living room sounds better. My point here is that very few of us will be in a position to KNOW whether that MQA recording really sounds closer to the original. In fact, the very idea of "the original" may be less certain than we think. With a modern multi-tracked album, we may hope to hear the same thing the mixing engineer heard when he mastered it, but that's probably about as close as we can get. Most modern recordings are what computer folks would refer to as "virtual" - the mix is put together from separate tracks - each of which is heavily tweaked and customized. The final mix isn't INTENDED to sound like the original tracks that were recorded. It's distinctly possible that the original tracks weren't even recorded in the same room, or on the same day. So, forget "original" or "accurate to reality" - your best goal is "accurate to what the engineer wanted you to hear". The reality is that we try to tune our sytems to sound as close as possible to what we IMAGINE the original sounded like (or we just give up, accept that we're chasing will 'o' the wisps, and make it sound the way we like it). Back to the point, though..... Since you have no way whatsoever of knowing if that DAC sounds like it should, you're going to have to deduce whether it does or not.... So, does it have a bunch of impressive numbers that show that it SHOULD sound right? (And WHICH numbers should you be paying attention to?) And, does your favorite pop star or recording engineer endorse it as "the best sounding damn DAC he's ever heard"? And do "4 out of 5 recording engineers swear by it"? Or do "most of the people we've surveyed rate it a 9 out of 10 in sound quality"? (With audiophile equipment, a long technical explanation, with lots of numbers, is likely to head the list.) The problem in this particular case is that, because it involves things like "neuroacoustics", there are no numbers that are even likely to describe what it sounds like. (By the numbers, or by an oscilloscope trace, MP3 files are appallingly inaccurate, but in real life they don't sound all that bad.) My point (yeah, finally) is that we keep seeing longer and more detailed explanations of why MQA files SHOULD sound better. And, yes, I personally understand enough of the tech jargon to say that their arguments make some sense. (But, from what I read, "cold fusion" made sense - it just didn't work.) Also (unfortunately) the MQA guys (or their marketing department) keep changing the details of story. (And, no, that does NOT mean that it isn't true, or that their product isn't good.) As far as I'm concerned, their marketing guys are trying very hard to set the EXPECTATION that MQA files sound better. They're working very hard to make sure that, if you hear a difference, you'll assume that it's the MQA version that's right - or "better". And they're doing this by EXPLAINING to you how the MQA version SHOULD sound closer to the original. And they're building on the momentum that the recent interest in "high-res" files has started (the idea that new file formats can actually be an improvement). And that's good marketing (I'm honestly impressed that MQA hasn't gone the way of HDCD already). So, personally, NO, I haven't heard any MQA files yet (I'll get around to it when they have MQA versions of some of my favorite albums). And, if they sound better because of the post-processing (on a normal DAC), then that will probably convince me to see if it sounds even better on an MQA DAC. BUT............ If I hear a few MQA versions of songs on Tidal that sound really great, will that PROVE that EVERYTHING they're saying is true? Errrrr.... no..... Because, as well as being MQA, they are also remasters, and I've heard lots of remasters that sound better than the original. (And I've got a couple of recordings on plain old CDs that sound scary good.) So, if they come up with some new remasters that sound better, I'll be happy. And, if they come up with a whole bunch of remasters that sound better, then I'll start believing that MQA has something significant to do with the improvement. And, if I ever get a chance to listen to an analog master, and compare it to the MQA reproduction, and compare it to a 24/192k PCM copy, then I'll get to find out if that claim is true. But, I'm rambling again..... The real point I wanted to make was that no amount of awesome explanation about how it SHOULD be great is going to convince me..... (And, to be quite blunt, I don't see anything in their explanation or claims that anyone is going to "debunk" either, so trying to puzzle out the details is sort of a waste of time.) You're NOT going to figure out whether it really works (sounds good) or not by reading about it.... You mean like reading 20,000 pages to understand the Affordable Health Care Act. I think what MQA expects is the same as what we were told regarding the Health Care Act - don't bother reading it, just vote for it and pass it. Think about how many people bought a DAC because they like to play and listen to digital music files, but they don't understand even the basic principles behind digital audio aka the Sampling Theory. Next, a guy named Bob Stuart comes along and he makes it even more complicated by combining the latest advancements in auditory neuroscience with it all among other serious stuff like coding so... yeah, of course it's going to be gruesome... kind of like trying to build your own set of coils for your rebuildable e-cig without wanting to learn Ohm's law (I always liked turtles, but nowadays I like vape clouds better as the latter are more acoustically transparent). |
|
|
|
Post by Boomzilla on Mar 9, 2017 10:39:04 GMT -5
Well... It seems like we're all in the dark about how MQA sounds, none of us having heard it. I've politely requested that the Secrets website send me a MQA DAC for review. When I get it, I'll do some blind (not double blind) testing using Mr. garbulky, another audio amigo, and, hopefully, Mrs. garbulky. All of these audiophiles have strong experience with live acoustic music (all music teachers - one cello, and the other two multiple instruments). I propose to switch between non-MQA WAV files using the DAC without the MQA switched on and MQA encoded files WITH the MQA switched on. The listener will have to mark "A" or "B" on their score sheet (defining which files are being played) and whether the sound is better or worse with "A" or "B," (whichever they chose). This isn't scientific, and the sample size will be small. BUT the sample quality (being an entire group familiar with live music), should leaven the loaf. Boom |
|
|
|
Post by monkumonku on Mar 9, 2017 11:00:19 GMT -5
Well... It seems like we're all in the dark about how MQA sounds, none of us having heard it. I've politely requested that the Secrets website send me a MQA DAC for review. When I get it, I'll do some blind (not double blind) testing using Mr. garbulky, another audio amigo, and, hopefully, Mrs. garbulky. All of these audiophiles have strong experience with live acoustic music (all music teachers - one cello, and the other two multiple instruments). I propose to switch between non-MQA WAV files using the DAC without the MQA switched on and MQA encoded files WITH the MQA switched on. The listener will have to mark "A" or "B" on their score sheet (defining which files are being played) and whether the sound is better or worse with "A" or "B," (whichever they chose). This isn't scientific, and the sample size will be small. BUT the sample quality (being an entire group familiar with live music), should leaven the loaf. Boom Question - How do you know the non-MQA file is using the same master/mix as the MQA-encoded file? I thought MQA is supposed to "improve" even files that were not MQA encoded. If my understanding is correct, then shouldn't you also be comparing non-MQA files between having MQA switched on versus off? And also MQA-encoded files between MQA switched on versus off? In other words, we all welcome comparisons but they need to be apples to apples. |
|
|
|
Post by Casey Leedom on Mar 9, 2017 11:14:06 GMT -5
Yeah, I was going to ask the same thing. To be honest, I'd be happier if they gave you their software to review along with a good microphone and a coulee of DACs. Then you could try running their "repair" algorithm on known digital audio streams and listen to the original/repaired. You could try recording the same performance in two different modes and then compare them back, etc.
Casey
|
|
hemster
Global Moderator
Particle Manufacturer
...still listening... still watching
Posts: 51,950
|
Post by hemster on Mar 9, 2017 11:14:31 GMT -5
Well... It seems like we're all in the dark about how MQA sounds, none of us having heard it. I've politely requested that the Secrets website send me a MQA DAC for review. When I get it, I'll do some blind (not double blind) testing using Mr. garbulky , another audio amigo, and, hopefully, Mrs. garbulky . All of these audiophiles have strong experience with live acoustic music (all music teachers - one cello, and the other two multiple instruments). I propose to switch between non-MQA WAV files using the DAC without the MQA switched on and MQA encoded files WITH the MQA switched on. The listener will have to mark "A" or "B" on their score sheet (defining which files are being played) and whether the sound is better or worse with "A" or "B," (whichever they chose). This isn't scientific, and the sample size will be small. BUT the sample quality (being an entire group familiar with live music), should leaven the loaf. Boom Question - How do you know the non-MQA file is using the same master/mix as the MQA-encoded file? I thought MQA is supposed to "improve" even files that were not MQA encoded. If my understanding is correct, then shouldn't you also be comparing non-MQA files between having MQA switched on versus off? And also MQA-encoded files between MQA switched on versus off? In other words, we all welcome comparisons but they need to be apples to apples. The monk brings up a valid point. Also, if your audition audience consists of music teachers, then should the DAC's target market not be music teachers, and not laymen? |
|
|
|
Post by Boomzilla on Mar 9, 2017 12:03:58 GMT -5
Obviously, I need to clarify... Source A = Just a plain old WAV file (44.1/16) played by the DAC with MQA totally OFF Source B = The same (44.1/16) file but MQA encoded & played by the same DAC with MQA enabled What I'm trying to determine, simply, is 1. "Do MQA encoded/decoded files sound audibly different from plain WAV?" and 2. "Of the two, which (if either) sounds more like real acoustic music in real space?" I want the music teachers because, on a daily basis, they ARE exposed to real acoustic music in real space AND they hear live music in a variety of venues. They should know (regardless of the music on the recording, regardless of the recording's original venue, etc.) which, to their ears, sounds more realistic. Now if I had the equipment and the gumption, I'd first have the musicians all bring their instruments and record a live performance of them in my living room (two simultaneous recorders - one with MQA, one without - coming from the SAME microphone feeds). Then play back the two recordings - one "straight," the other with MQA decoding. Then, the musicians would know how the original performance sounded. But alas, I don't have any of the necessary equipment, and although garbulky does, he doesn't have an MQA-active recorder. So we aren't going to get any "prove it by the numbers" results from this experiment - only subjective assessment from a team of professional musicians. Boom |
|
|
|
Post by pedrocols on Mar 9, 2017 12:13:13 GMT -5
Is there a reason you want to make you guys so miserable? Can you just listen and enjoy music?
|
|
|
|
Post by monkumonku on Mar 9, 2017 12:18:06 GMT -5
Obviously, I need to clarify... Source A = Just a plain old WAV file (44.1/16) played by the DAC with MQA totally OFF Source B = The same (44.1/16) file but MQA encoded & played by the same DAC with MQA enabled What I'm trying to determine, simply, is 1. "Do MQA encoded/decoded files sound audibly different from plain WAV?" and 2. "Of the two, which (if either) sounds more like real acoustic music in real space?" I want the music teachers because, on a daily basis, they ARE exposed to real acoustic music in real space AND they hear live music in a variety of venues. They should know (regardless of the music on the recording, regardless of the recording's original venue, etc.) which, to their ears, sounds more realistic. Now if I had the equipment and the gumption, I'd first have the musicians all bring their instruments and record a live performance of them in my living room (two simultaneous recorders - one with MQA, one without - coming from the SAME microphone feeds). Then play back the two recordings - one "straight," the other with MQA decoding. Then, the musicians would know how the original performance sounded. But alas, I don't have any of the necessary equipment, and although garbulky does, he doesn't have an MQA-active recorder. So we aren't going to get any "prove it by the numbers" results from this experiment - only subjective assessment from a team of professional musicians. Boom Notwithstanding Pedrocols comment, which I agree with because the proof is just in do you like how it sounds or not, but regarding your Source B above - when does the encoding take place? Do you encode it yourself or is this something that you receive that has already been encoded? The reason I ask is, how do you know that the Source A and Source B files are actually identical prior to one of them being MQA encoded? If you encode it yourself then that's one thing, but if it arrives in an encoded state how can you verify it is the same master as the Source A file? |
|
|
|
Post by garbulky on Mar 9, 2017 12:27:45 GMT -5
Obviously, I need to clarify... Source A = Just a plain old WAV file (44.1/16) played by the DAC with MQA totally OFF Source B = The same (44.1/16) file but MQA encoded & played by the same DAC with MQA enabled What I'm trying to determine, simply, is 1. "Do MQA encoded/decoded files sound audibly different from plain WAV?" and 2. "Of the two, which (if either) sounds more like real acoustic music in real space?" I want the music teachers because, on a daily basis, they ARE exposed to real acoustic music in real space AND they hear live music in a variety of venues. They should know (regardless of the music on the recording, regardless of the recording's original venue, etc.) which, to their ears, sounds more realistic. Now if I had the equipment and the gumption, I'd first have the musicians all bring their instruments and record a live performance of them in my living room (two simultaneous recorders - one with MQA, one without - coming from the SAME microphone feeds). Then play back the two recordings - one "straight," the other with MQA decoding. Then, the musicians would know how the original performance sounded. But alas, I don't have any of the necessary equipment, and although garbulky does, he doesn't have an MQA-active recorder. So we aren't going to get any "prove it by the numbers" results from this experiment - only subjective assessment from a team of professional musicians. Boom It would be hard to get the plain old wav file. Because... The wav file has the MQA data already on it. Even with MQA off, it is not the same file as having no MQA. The wav file is now lossy whether mqa is turned on or off versus the same wav file that is lossless without mqa in it. The other problem is let's say you have a CD of Dolly Parton's "Coat of many colors". Then you can get the wav file off there. But the MQA file is not necessarily that same file - but with MQA added. It could be a comletely different mix from the Master. I.e. it's whatever the record company says is the "definitive" version. It's not necessarily the same one that's on your CD. However both types of comparisons would be valid - just in different ways. We could figure out if it sounds better. But I have to mention....the setup matters. So if there was no difference in your setup, that doesn't mean there isn't a benefit. It just means there's no difference in your current setup. However if there was a difference...then we can argue for something positive. But first.....you have to have an MQA source like Tidal. Do you have this? Because if you do, then you can use your Oppo to hear at least the first fold decoded of MQA. |
|
|
|
Post by novisnick on Mar 9, 2017 12:38:04 GMT -5
Obviously, I need to clarify... Source A = Just a plain old WAV file (44.1/16) played by the DAC with MQA totally OFF Source B = The same (44.1/16) file but MQA encoded & played by the same DAC with MQA enabled What I'm trying to determine, simply, is 1. "Do MQA encoded/decoded files sound audibly different from plain WAV?" and 2. "Of the two, which (if either) sounds more like real acoustic music in real space?" I want the music teachers because, on a daily basis, they ARE exposed to real acoustic music in real space AND they hear live music in a variety of venues. They should know (regardless of the music on the recording, regardless of the recording's original venue, etc.) which, to their ears, sounds more realistic. Now if I had the equipment and the gumption, I'd first have the musicians all bring their instruments and record a live performance of them in my living room (two simultaneous recorders - one with MQA, one without - coming from the SAME microphone feeds). Then play back the two recordings - one "straight," the other with MQA decoding. Then, the musicians would know how the original performance sounded. But alas, I don't have any of the necessary equipment, and although garbulky does, he doesn't have an MQA-active recorder. So we aren't going to get any "prove it by the numbers" results from this experiment - only subjective assessment from a team of professional musicians. Boom It would be hard to get the plain old wav file. Because... The wav file has the MQA data already on it. Even with MQA off, it is not the same file as having no MQA. The wav file is now lossy whether mqa is turned on or off versus the same wav file that is lossless without mqa in it. The other problem is let's say you have a CD of Dolly Parton's "Coat of many colors". Then you can get the wav file off there. But the MQA file is not necessarily that same file - but with MQA added. It could be a comletely different mix from the Master. I.e. it's whatever the record company says is the "definitive" version. It's not necessarily the same one that's on your CD. However both types of comparisons would be valid - just in different ways. We could figure out if it sounds better. But I have to mention....the setup matters. So if there was no difference in your setup, that doesn't mean there isn't a benefit. It just means there's no difference in your current setup. However if there was a difference...then we can argue for something positive. But first.....you have to have an MQA source like Tidal. Do you have this? Because if you do, then you can use your Oppo to hear at least the first fold decoded of MQA. I think your last statement may be incorrect my friend. In order to hear MQA from Tidal you must use a PC or Mac computer to play Master recordings (MQA). But I may be mistaken . |
|
|
|
Post by monkumonku on Mar 9, 2017 12:40:33 GMT -5
Okay, I am convinced. MQA is the real deal for me. And for all of you, too.  |
|
|
|
Post by pedrocols on Mar 9, 2017 12:56:37 GMT -5
It would be hard to get the plain old wav file. Because... The wav file has the MQA data already on it. Even with MQA off, it is not the same file as having no MQA. The wav file is now lossy whether mqa is turned on or off versus the same wav file that is lossless without mqa in it. The other problem is let's say you have a CD of Dolly Parton's "Coat of many colors". Then you can get the wav file off there. But the MQA file is not necessarily that same file - but with MQA added. It could be a comletely different mix from the Master. I.e. it's whatever the record company says is the "definitive" version. It's not necessarily the same one that's on your CD. However both types of comparisons would be valid - just in different ways. We could figure out if it sounds better. But I have to mention....the setup matters. So if there was no difference in your setup, that doesn't mean there isn't a benefit. It just means there's no difference in your current setup. However if there was a difference...then we can argue for something positive. But first.....you have to have an MQA source like Tidal. Do you have this? Because if you do, then you can use your Oppo to hear at least the first fold decoded of MQA. I think your last statement may be incorrect my friend. In order to hear MQA from Tidal you must use a PC or Mac computer to play Master recordings (MQA). But I may be mistaken . You are correct. I can stream MQA from my pc but not from my iPad. |
|
|
|
Post by garbulky on Mar 9, 2017 13:55:10 GMT -5
Y'all are right. Sorry!
|
|
|
|
Post by flamingeye on Mar 14, 2017 21:36:57 GMT -5
Errrrr. yeah...... The confusion is due to the fact that MQA is a whole bunch of different things which all sort of go together. It is actually a collection of technologies and an overall philosophy for a workflow. Each piece can, at least in theory, give you an improvement in sound quality, and most of the pieces can be used independently of the rest of the pieces. Here are a few of the more prominent claims and conclusions. As you will see, many are independent, and may or may not live up to the individual claims made for them. 1) As a digital CODEC, MQA is LOSSY. If you're starting from a digital original, some of the information is discarded, and you CANNOT get it back. That is the definition of a lossy CODEC. (MQA readily admits this.) 2) Starting from an analog original, MQA claims to preserve MORE of the IMPORTANT information than other formats with equivalent bandwidth requirements. This would assume that your analog original was encoded on "MQA certified" equipment and played back on "MQA certified" equipment. They therefore claim that, starting from an analog master, ALL digital encoding is lossy anyway, and that MQA is LESS audibly lossy than uncompressed PCM. (Their claim that something is lost whenever you encode to a different format is true; their claim that their process yields a result that is audibly closer to the analog original is yet to be proven.) 3) Starting from a digital "master", MQA claims to be able to reverse engineer the digital recording and remove some of the flaws caused by the original encoder. They claim that you'll hear a net improvement in quality because the problems they fix are more important than the information they discard. They also claim that this process can be done using an "automatic encoder" or a higher-quality encoder that is manually operated. 4) As with most lossy encoders, the producer may choose to trade off quality against bandwidth. With MQA, they claim that you can produce a stream that is audibly slightly above CD quality, but takes up a lot less space. They also claim that you can produce a stream or files that are the same size as CD quality, or even 24/96k PCM files, but are of audibly higher quality. 5) Any MQA encoded file or stream can be played on "ordinary equipment", but you won't get the benefits of the MQA encoding. Any MQA encoded file or stream can be played using a software MQA decoder, and you'll get SOME of the benefit of the MQA encoding. To get ALL of the benefits of MQA encoding, you'll need to play your MQA content on an MQA certified DAC. 6) There will be a process whereby the studio or performer MAY "certify" that a given MQA version of a track has been "approved". The folks at MQA have made it quite plain that they will "take the word of the studio" that the version they're encoding is the best quality one available. The automated MQA encoder MAY also be used to produce "MQA encoded and improved copies" of various tracks without human intervention or any approvals/certifications. (I've heard descriptions of various LEDS which might light to indicate a "certified" track....) 7) Finally, because the MQA certification performed on hardware includes design factors intended to improve transient accuracy, MQA certified DACs MAY sound better when playing ordinary content. (Note that this certification involves paying MQA for a license which may entail design review or modification of the product itself.) 8) If everything lives up to their claims, then there would be an excellent reason to want a home encoder. Because the encoder is claimed to be able to detect and correct certain error that occurred during the original digitizing process, a home encoder could theoretically be used to process and "fix/improve" digital content you already own, and so make it sound better. However, I have NOT heard any claims that this would ever be practical, or that they plan to make one available. My understanding is that it is a lossy compression scheme that allows the provider, Tidal for example, to squeeze higher resolution audio down an increasingly busy internet connection. If you have an audio system that can resolve high resolution sources then it is a good thing as you should hear the difference when compared to MP3. If you are streaming to a mobile phone, I don't see any point. As with all compression schemes, the decoder is in the receiver and will require a licence. Even if the hardware existed at a domestic level, I don't see any benefit from being able to encode at home. Disc space is cheap and if you'd want to encode to put music on your phone, you might as well use FLAC or MP3. this just all sounds awful why not just have lossless and be don |
|
|
|
Post by jmilton on Mar 15, 2017 7:30:09 GMT -5
|
|
|
|
Post by monkumonku on Mar 15, 2017 9:05:20 GMT -5
My two cents, I found that article adds nothing to the discussion except that the title is misleading: MQA (MASTER QUALITY AUTHENTICATED). WHAT IT IS, AND WHY YOU WILL WANT IT! The author presents no real arguments as to what makes MQA any better than CD quality. He makes it clear that there is a huge difference between mp3 versus CD or MQA, but he also concludes that there is (or he thinks there is because he does not sound adamant about it) a very subtle difference between CD versus MQA, with the latter just having a bit "more" of everything, whatever that means. Does he mean it is more organic? More chocolatey? In the end, what he is adamant about is that one should definitely subscribe to Tidal with the MQA option because that quality is so much better than mp3. I won't argue there, but is that really any different a conclusion than would have been before Tidal partnered with MQA? If CD and MQA are that close and it seems the author really can't distinguish one from the other in a blind test, then the recommendation would still stand to subscribe to a pre-MQA Tidal using CD streaming quality. Also, the explanation of how MQA works is similar to what I have already read - it describes the packing and unpacking but none of what really makes it "better" than CD quality. He says it packs the original file to an even higher resolution, but how can that be? That's no different than upsampling, and your true resolution is not going to ever be better than the original. Then when it is unpacked, the resolution goes even higher. That again makes no sense, practically speaking. So the way I read the article, MQA is still voodoo. |
|
KeithL
Administrator
Posts: 10,255
|
Post by KeithL on Mar 15, 2017 9:21:10 GMT -5
I would also suggest reading this analysis by Archimago of the results (linked from the hometheaterhifi article): archimago.blogspot.com/2017/02/comparison-hardware-decoded-mqa-using.htmlHOWEVER, I would again remind you that MQA is promising TWO VERY DIFFERENT BENEFITS. 1) First, it is promising to deliver high quality audio, particularly in streaming situations, very EFFICIENTLY. By that I mean that it claims to deliver an ACCURATE rendition of the original content using less bandwidth or file size than other formats. This is important, for example, if you have limited storage space, or if you want to limit your data usage on a portable device. Archimago's testing seems to confirm that MQA is capable of delivering a copy of a digital original pretty accurately. 2) They claim that the MQA encoding process is REMOVING ERRORS CAUSED BY THE ORIGINAL A/D CONVERSION.
To be very plain here - they are claiming that the MQA version will sound BETTER than the digital original (and more like the analog original). They claim that their encoder actually "reverse engineers the original encoding process" and reduces or eliminates flaws caused by the original conversion process. Note that this is a very specific claim.... and also one that is virtually impossible to test unless you have access to THE ANALOG ORIGINAL. Without access to the analog original, you can say that you like their altered version better, or that you don't like it better, but you cannot confirm that it is closer to the original. (Even if you claim that "it sounds closer to real" or "closer to a live performance" you cannot know whether it is improving accuracy, or simply UNDOING a change that was done deliberately in the mastering process.) I would also note that the things that MQA claims to fix involve transient response. Therefore, when you compare two versions of the file, and look at the AVERAGE difference between them, it may not accurately reflect the audible difference. (Small alterations in the shape of transients may result in differences of significant amplitude, but very short duration, which may be audible, even though they have a low AVERAGE power level.) |
|
By the way, yves posted a note in a different thread with a references to a bunch of papers which may include part of what I've been looking for:
