Has anyone compared an auralic ‘MQA’ track with the same official MQA track?
Russ
Yep saw all that and agreed
Their approach sounds not far off pass through…
Having now read the Auralic statement, I would infer that they are managing to decode the MQA file with their own in-house methodology. If they are not doing that, they should have said something along the lines of “Here’s our alternative to MQA…”. I suppose further clarification will come in time.
Not sure how you can infer that, when Keys are required to Decode MQA material…and Auralic have clearly said that they are not licensing those keys from MQA
For your inference to be correct, then Auralic would need to admit to either hacking or reverse engineering the Keys required to Decode MQA…and I really don’t think that a public company will want to admit to doing that, even it were true
And at no point in their statement, do Auralic say that they are “Decoding”…merely carefully using the words “MQA Decoding support”
Just by way of clarification, this is AURALiC’s CEO, Xuanqian Wang, quoted in the DAR article:
“We do not use any MQA technology, this is not MQA certificated or MQA licensed. We are up-sampling the file using our algorithm, applying our own in-house developed filter, to optimize sound quality, not just for a particular DAC but all devices.”
To me, this sounds very similar to what you can do in HQPlayer with its “poly-sinc-mqa” filters.
I certainly agree with @joel and others that AURALiC is walking a fine legal line here. IMO, based on recent history, Wang tends to get a little too far out in front of the MQA stuff. I’m guessing he and Bob Stuart don’t have a really close personal relationship at this point.
I infer it from those words included in the Beta Firmware Release Note.
In one of these MQA related threads I think someone gave a link to an article indicating that reverse engineering may not necessary be against the law or copyright infringement. It depends on the circumstances. Maybe Auralic believe themselves to be within the law.
If those words do not mean they are decoding MQA, do you believe that they mean that passing through the stream unmolested to an MQA DAC is the support they refer to. If that is the case I think they are being disingenuous.
No, I don’t believe that Auralic are just allowing the stream to just “pass thru”
Let’s look at this another way
Right now, you can take an MQA Track [or indeed any FLAC Track] from Tidal / Roon…and ask Roon to Upsample it to 8x 353 / 384k…and apply one of a range of Filters to it
Do you think it would be correct if Roon were to describe the combination of those two processes above as “MQA Decoding”??
Because that is what Auralic themselves are claiming to do
Or to put it in another way…
-
MQA Decoding / Unfolding is about “re-assembling” the original stream as it was before the Encoding process…[i.e. no upsampling or data interpolation]…whereas
-
Auralic’s method seems to be about CREATING a new 192k [4x] or 384k [8x] stream from a 48k [1x] stream…by using their Upsampling and Filtering
Like I said earlier, if they are not MQA decoding then they are being disingenuous. From the little that I understand about this I would not say that upsampling and filtering adequately covers MQA decoding. They may believe that they have a process which adequately mimics what MQA decoding is. How anyone comparing true MQA decoding against their method will fare will no doubt open up even more debate.
Agreed. If they are not decoding/unfolding then it is wrong to call it MQA support.
If they are truely decoding/unfolding MQA then you have to wonder how they managed to do this without using technology/information that MQA made available when they were working together.
Is Aurialic elevating each MQA track automatically to 352.8/384 kHz? Or is it unfolding, upsampling, or doing whatever to increase each MQA track only to its respective full sampling rate? If the latter, then Auralic is decoding MQA in band signaling to extract configuration information.
AJ
Auralic is careful and stated ‘proprietary resampling and de-blurring method’ the word is ‘resampling’ in audio term seems to be different from upsampling. I want to share this article below:
In fact Auralic implementation got good review here:
There’s no doubt they probably have code on hand when they demonstrated at CES 2016, a fully functional MQA on Aries Mini. Probably to avoid licensing infringement, they do it differently, again this my speculation.
There’s more evidences begin to surface that MQA does not actually ‘decode’ but rather ‘resample’ or ‘render’ with selectable ‘hidden’ impulse filters, to be exact there are 30 of them! Check out:
There’s more…
Hmmm - I wonder if Auralic uncovered some inner workings of MQA as claimed in this article.
In order to get it to work in the way that so upset MQA Ltd, exposing unfolded MQA to SPDIF, there had to be some degree of reverse engineering.
This may explain why I can’t hear the difference between Auralic ‘in house’ developed MQA vs a fully decoded Mytek Brooklyn DAC!