Do remember that MQA is taken from Hi-Res PCM stereo mixed down. The recording studios don’t have access to MQA encoding. They need to send the copy (Hi-Res PCM stereo mixed down) to MQA Ltd for processing. What you are listening is a ‘processed’ and ‘altered’ sound from the original PCM masters.
All recording studios uses either multi-tracks PCM or DSD or both for their recording sessions. There’s no MQA being involved at this stage. If you can hear such a big deviation from PCM vs MQA that suggests MQA adds it own sonic signature and this does not reflect the original SQ of the recordings. We must not conclude that anything that sounds ‘good’ (subjectively) reflects the actual performance.
There’s a good explanation of minimum phase filter. Ironically, this group of filter is used for ‘listening’ mode not ‘monitor’ mode.
Moreover, since this type of filter introduces distortion at higher frequencies due to its long tapped delay, the distortion is similar to what I called ‘even’ or closely resembles to 2nd harmonics distortion. This type of distortion is generated by vacuum tube amp or device which have this type similar characteristics curve. Many perceived this as ‘sweet’, ‘warm’ and easy to the ears.
It is okay to like this type of sonic signature but don’t forget this does not necessary reflect the SQ of the original recording or performance.
Yes Archimago has done excellent work. Above I simply tried to translate these kind of technical observations into a description of what we hear and in the context of how we know that our hearing functions.
I am not sure minimum phase dustortion is quite like tubes - I don’t agree with that. Tubes have very good transient response. I believe it is purely the apodizing filter that MQA uses that will introduce tube like distortion (mild compression). This compression aspect of MQA may indeed sound more like really old analog. Certainly MQA seems a little rolled off in the HF - at least subjectively,
Tube distortion is essential for additional musical clarity. Again it is because our hearing is so sophisticated. If you have more harmonics and they are well audible then the ear/brain can BETTER detect the fundamental. We don’t just hear individual fundamental frequencies but we pick up or focus on the harmonics too. Possibly this is a requirement to understand speech.
I find tubes essential for listening to hi-hat. Often the hi-hat is buried on everything but jazz. Tubes allow you to hear the hi-hat clearly even on a busy rock mix. They also make vocals and guitar richer but that isn’t usually needed as they are often already very clear with solid state gear. Tubes are able to make soft sounds more audible due to the addition of harmonics.
Anyway, everyone in the audio industry knows that minimum phase must be avoided if you are to preserve good soundstage and imaging. Only Linear Phase is an acceptible choice for a DAC filter - and we knew this 40 years ago!
The reason that minimum phase exists is that studios may apply aggressive filters ( high Q) at frequencies WITHIN the audible band to an individual instrument on a single track of a multi-track recording and this definitely causes audible pre-ringing if Linear Phase is used. Pre-ringing from brick wall filters for D to A is NOT audible if done correctly with the filter operating entirely outside the audible range. MQA seems to have carefully constructed fear over a non-existent problem so that they can offer a solution.
Both will preserve the phase relationship of all frequencies.
The smooth filter may have some slight roll off and it may have some aliased frequencies (noise) in the high frequencies.
The precise filter will have stronger but inaudible ringing at the frequency of the filter and ONLY if there are data issues (like intersample overs which is sometimes present on pop rock loud masters). This is totally inaudible in a good design HOWEVER if the DAC has poor IMD (Intermodulation Distortion) performance in the HF then ringing might create side bands around audible frequencies. Stereophile usually do a test with a pure HF tone to look for IMD.
So what sounds better to you may depend on your preference for roll off and the IMD performance of your DAC.
This is an example of excellent IMD performance - a 19 Hz full scale pure tone mixed with a 20KHz full scale pure tone - the distortion sidebands are down by more than 110dB.
Fig.9 Benchmark DAC3 HGC, HF intermodulation spectrum, DC–30kHz, 19+20kHz at 0dBFS into 100k ohms, 44.1kHz data (left channel blue, right red; linear frequency scale).
Note how consistently the reviewer notices an “expansive soundstage width” on MQA processed files vs a narrower width with unadulterated audio. This is EXACTLY my point that I made a few posts above. MQA is blurring the soundstage by delaying high frequencies with respect to low frequencies!
“Expansive Soundstage width” is just a euphemism for a hole in the soundstage! It is a nice way to describe a major issue with MQA - Phase Distortion!
Hole in the soundstage, you are kidding me as I listen to Mavis Staples new live album in MQA. It’s amazing, and she is right out front and Center. Ears please, ears…
This is certainly not an accurate representation of the soundstage of most MQA material as I hear it on any of my DACs - those DACs being Linn Klimax DS/1, Mytek Brooklyn+ or Chord Hugo.
The “hole in the soundstage” is relative to the more accurate precise soundstage of non-MQA files played on a DAC with a traditional linear phase filter.
Most people should hear it. Certainly young people will. Your hearing needs to go well above 10 KHz to hear the MQA distortion as only high frequencies are delayed significantly. A phase accurate loudspeaker will also help make it more easily audible.
It is well known that phase distortion can produce an “expansive soundstage”. Guitarists use this all the time - a phaser or flanger - a simple pedal effect. I guess audiophiles might be fooled into thinking this is a more accurate rendition but it is just an effect from added distortion.
I totally agree with your rationale and share your view and am surprised with qualitative belief in some cases without evidence which is akin to “bullying” - your views are well put mate!
We get the usual hyperbole about MQA but no measurements to back it up. This could be a review of ethernet cables made from silver, magic rocks you affix to your cables or any other type of nonsense. It just happens to be one guy with a bias towards MQA telling the world he enjoys MQA.
It either is or it isn’t interesting. You can’t have both but clearly they have as much authority as anyone else and they like MQA. You don’t and that ok also.
Like so many others, I have nothing new to say on this subject…