I do not know the hourly rate a white glove sound engineer charges. But for this amount I think this hardly covers clicking “add to queue for processing”.
4 Likes
On the other hand people complain MQA are making money off artists 
1 Like
I don’t hear this complain often.
But this fact shows what to think about the claims about the efforts was made to identify all the A/D converters used in the studio and correct all their faults.
That work has probably been done already and programmed into the system
1 Like
They are able to autodetect the A/D converters in the batch process? Now that is a great engineering achievement!
1 Like
Detail :
… minimum-phase digital filter proposal with a slower roll-off that reduced the overall amount of ringing. The resulting filter has no pre-echoes, and only about one cycle of post-ringing. …
The result is simply the most musically natural digital playback available today.
Filter 2 in my DAC
Indeed, we don’t need MQA at all.
Just the original PCM, select a minimum phase slow roll off filter and done.
And it even sounds better as nothing was lossy compressed.
5 Likes
Thank you. So MQA encoding =
1 Like
The thing is, you have no idea what they can do… The results sound sweet though
The idea of leveraging may have passed you buy but it is used in almost everything you purchase today.
Unfortunately, it is not that simple. The effort to programme a really good sounding minimum phase filter and to match it to the rest of the circuitry is very high.In most cases, the suppliers of even expensive DACs use the filters integrated in the DAC chip or the filters specified by MQA, which are rather suboptimal.
Watch the video and test an Ayre DAC, even if it’s just the “entry-level” CODEX or the PonoPlayer.
1 Like
And most ESS chipset has this selectable filter in one of seven available to the user so PCM + filter plus a bit of EQ warmth or DSP and a slight DB increase in volume and you have got it 
1 Like
Yes you’ve got one single filter for use in the DAC upsampling. Almost no relationship to the whole chain processing done in MQA.
2 Likes
The encoding program is probably both refined and mostly automated by this time, including A/D converter corrections. This is probably not so difficult as you assume given the amount of experience they’ve had.
But would you prefer that indie artists are charged much more? Would you then say it’s more valid, or just a rip off? Is there any price you’d see as valid?
1 Like
What chain as far as I’m aware there are 16 pre defined set filters available with maybe availability of 32 at best, based on the 1000’s of variations in building music in digital I find it very hard to comprehend they can find an exact match, most of this work has already been done at a professional level when archiving to PCM or DSD incl deblurring to achieve the optimum master copy.
4 Likes
Can I suggest you read Bob Stuart’s blogs on the MQA website. Correction of A/D and D/A converters, plus the sequence of complementary downsampling and upsampling filters, plus proprietary noise shaping, fractional bit dither, and subtractive dither throughout; those are the chain. The choice of filters is based on an analysis of the dynamic range and peak SNR by frequency of the file which is done before encoding. This is all covered in the blog but also in their JAES papers, which are free downloads.
1 Like
Maybe MQA should do some “deblurring” on their own website first.
“unfold”
“lossless compressed estimates” (what?)
Also all the techniques you just mentioned that sound so wonderfull are just there
to cover up the errors they created themselves by reducing bit depth and downsampling.
We don’t need these lossy corrections with a regular (non-mqa) pcm.
Noise shaping is a technique typically used in digital audio, image, and video processing, usually in combination with dithering, as part of the process of quantization or bit-depth reduction of a digital signal. Its purpose is to increase the apparent signal-to- noise ratio of the resultant signal.
4 Likes
As stated in the blog, the MQA approach always encodes the full dynamic range of the musical information with the highest possible precision. No, they do not create errors by reducing bit depth and downsampling. MQA is informationally lossless, although it throws away excess noise. This is fundamental to understanding the technique.
2 Likes
MQA cannot recover anything beyond actual 48khz even though certain microphones and instruments transients can achieve 60khz this is lost during the lossy compression
3 Likes
The problem I have with MQA: everything is blurred. Mostly the technical information. If I read other manufacturer sites, I get white papers, technical explanations, how tos and so on.
On the MQA site I only get information in the form of a management summary. Bob’s Talk is mainly joining technical terms, which sound good. In the end, however, he does not reveal any details
Can you please specify a few links, which explain in detail, what happens in the encoding process?
I have a problem with the expression “informationally lossless”.
When I’m not looking at the famous “Mona Lisa” with a magnifying glass under a spotlight, I could change dark brown to black. I can also look at a fake of “Mona Lisa”. It is informationally lossless. I do not know how the original really looks like. But it is not the same.
Of course is a recording never exactly the same as the live event. But after studying the details my conclusion is: MQA is only an approximation of the recording. It is informationally lossless.
4 Likes
They do encapsulate the noise floor beyond 48kHz (section “C”, discussed in the blog under Origami, the last mile). The problem with the music data beyond 48 kHz which some instruments produce is that it is very low in amplitude and lies at frequencies far above the audible range; it’s a question of whether your ears would ever respond to those frequencies or whether their presence would in any way affect music. I believe Bob Stuart and Peter Craven, along with mastering engineers they’ve worked with, would have answered that in the negative (although it’s not published that I’m aware). If they thought that region were audible, they’d have included it in the encapsulation.
2 Likes