Looks like MQA finally got its act right…
http://www.mqa.co.uk/customer/how-it-works
‘The first unfold recover all direct music related information. Output is 88.2kHz or 96KHz’
So what happen if the input sampling is greater than 88.2kHz or 96kHz? Somehow when one add a renderer will ‘magically’ shows the original sampling of music recorded, say 176.4/192kHz or 352.8/384kHz. Now, isn’t it ‘cheating’ when there’s no information can be recovered beyond the limitation of MQA Core decoding?
If you read the MQA technical paper (I don’t have a link but it has been posted multiple times before) or other MQA technical introduction, MQA claims that only musical information inside a specific triangle (in the graph in the paper) is audible, and their process fully captures the audible information. MQA Core restores that triangle. If you watch MQA YouTube video, you’ll find that they claim running the DAC at a higher rate is good for sound quality with ultrasonics.
That is fulfilled in the first unfold, I’ve no dispute on that.
Running at the higher rate is good? No, this corresponds to the original sampling rate if it is higher than MQA core. I don’t have issue when the original master is recorded on 88.2/96kHz but if the original master is higher say 192kHz or 384kHz, it doesn’t reflect the actual information in these sampling rates, or I say these are actually ‘up-sampled’ (inside the renderer) in order to correspond to the original recorded sample rate. And this is misleading the consumers!
For example for sampling rate higher than 96kHz:
24/192kHz FLAC file is equal losslessly conveys to a DAC during conversion. All information pertaining at this sampling rate is recovered!
MQA 24/44.1/48kHz -> first unfold to 88.2/96kHz -> renderer (upsampled) 192kHz to DAC.
Only up to 88.2/96kHz information is recovered but consumers see 192kHz and thought all information pertaining 192kHz is recovered!
There’s nothing new here?
All this was even explained in Bob’s original infamous origami video, no?
My argument is it doesn’t reflect the actual information of the original recording when the sampling rate is higher than MQA core output.
Yep understood but that’s not a new thing, no?
As an example, “Bob Marley - Live!” is available in Tidal MQA at 192k.
But when played via Audirvana to my non-MQA DAC (Hugo2), A+ outputs at 96k obviously. But all the ‘music content’ is still there, as per Bob Stuart’s own words in the origami video but obviously the other >96k content is missing.
But this has been known for a long time, no?
Unless I am mis-understanding your point (apologies in advance if I am).
My argument is simply why not display 88.2/96kHz which is MQA Core output instead of ‘misleading’ consumer when it is higher than that, when there is no information?
This scenario is exactly the same when you purchase a up-sampled files and sell it as Hi-Res? Cheating?
MQA certified devices always display the sample rate of the master, not how the DAC renders it, which is not necessarily playing at the same sample rate as the master.
Ok I get you. Bob says there is no ‘music content’ above >96kHz, but the record labels are the ones producing tracks at >96kHz and selling them.
Some claim that while there’s no music content above >96kHz, there’s other stuff happening >96kHz that affects the listening experience. It’s technical and over my head.
And then there’s those that says the un-folding >96kHz is just DAC specific up-sampling by the renderer.
I’ll stop since we have Peter here - he’s much smarter than me on this stuff lol
the first unfold only restore the FS to 88.1kHz and 96kHz, the deblurring must be perform on Hardware level so only happening in MQA Renderer and Full Decode Hardware
If there’s no information above 88.2/96kHz then it should done in the first unfold which MQA is doing right now in Tidal app. I’m happy with that. Anything above the sampling rate, please don’t show it to the consumers, even the original recording is recorded higher than this.
As claimed by them, no information above 88.2/96kHz, so stop there.
But MQA Ltd need to charge more to hardware partners for the 2nd and 3rd unfolding. It’s about making more money, no?
I’m with you - just give me the ‘music content’ up to 96kHz (@danny !) and I’ll let Roon and/or my DAC up-sample to higher rates, however I want to.
De-blurring is also done after the ADC, during mastering. In the DAC side, another de-blurring is done plus a ‘up-sampled’ (inside the renderer) to match the original recorded sampled if it is higher than MQA Core, i.e >88.2/96kHz.
It makes things more confusing when someone decode using Tidal app is limited to 88.2/96kHz and another person get a MQA DAC see it display 176.4/192k and above.
It’s not about the musical information that higher resolutions can capture its about the ability to use less harmful filters That is the benefit of higher sampling rates not more musical information.
http://www.indiana.edu/~emusic/etext/digital_audio/chapter5_rate.shtml
Great, finally I get my message across!
Yes but I asked this same thing quite a while ago too 
As soon as I heard Bob say there’s no music content >96kHz, that’s all I wanted!
Right. For MQA, some additional reasons are discussed here:
The idea of DSD has been around since SACD is launched in 1999, it use ‘less harmful filter’ then PCM. In fact DSD has almost a perfect impulse response to an analogue ‘click’.
So we have it NOW, enjoy the ‘analog’ sound…
No we don’t, I have access to no DSD content at all.
Sure. The only problem is that if you record using DSD, you can’t mix with it without resorting to PCM (DXD). The greatest weakness of DSD is that it is 1-bit. Two bits would literally have been infinitely better, because you can use dither to achieve infinite resolution, which is critical if you want to manipulate it.
So: DSD nice, but essentially useless in a manner of speaking. Let me also say that many SACD releases sound wonderful, but is that inherently DSD, or the care with which they’ve been (re-)mastered?