MQA - CD

[GUTB]

Here's one of the tracks that authenticates:

04 Track04.wav

[mansr]

2 hours ago, GUTB said:

Here's one of the tracks that authenticates:

04 Track04.wav

Could you share a track that doesn't authenticate as well?

[GUTB]

This one doesn't authenticate.

 

01 Track01.wav

[wklie]

32 minutes ago, GUTB said:

This one doesn't authenticate.

 

01 Track01.wav

 

I tested this track on Lumin and its MQA indication works as expected.  Other users also tested several ripped MQA CD they bought from Japan successfully.

 

For Roon specific issue, please submit a problem report in Roon forum by flagging @support there.

[MetalNuts]

I may be over worrying about DRM, but will it be the DRM that causes the ripping problem of MQA-CD?

[mansr]

8 hours ago, GUTB said:

This one doesn't authenticate.

 

01 Track01.wav

Feeding those files to my MQA scanner.

01_Track01.wav
0000c210: MQA signature at bit 8
0000c210: [5] datasync
          magic                     36: 0x11319207d
          stream_pos_flag            1: 1
          pad                        1: 1
          orig_rate                  5: 0x03 [352.8 kHz]
          src_rate                   5: 0x00 [44.1 kHz]
          render_filter              5: 8
          unknown_1                  5: 0x04
          unknown_2                  1: 0
          pad                        2: 0
          mqa_level                  8: 0x90
          items                      7: 2
          size                       8: 0xc2
          size                       8: 0x20
          type                       8: 0x00
          type                       8: 0x01
          stream_position           32: 0x00021a8b
          [type 0]
          unknown_3                  2: 2
          unknown_4                  4: 0
          unknown_5                  7: 15
          unknown_6                  7: 127
          start_offset              27: 0x00010eb
          pad                        1: 0
          data                     146: [skipped]
          [type 1]
          unknown_7                 32: 0x06473a63
          checksum                   4: 0xd

 

04_Track04.wav
0000288f: MQA signature at bit 8
0000288f: [5] datasync
          magic                     36: 0x11319207d
          stream_pos_flag            1: 1
          pad                        1: 1
          orig_rate                  5: 0x03 [352.8 kHz]
          src_rate                   5: 0x00 [44.1 kHz]
          render_filter              5: 8
          unknown_1                  5: 0x04
          unknown_2                  1: 0
          pad                        2: 0
          mqa_level                  8: 0x90
          items                      7: 2
          size                       8: 0xc2
          size                       8: 0x20
          type                       8: 0x00
          type                       8: 0x01
          stream_position           32: 0x03481d5a
          [type 0]
          unknown_3                  2: 2
          unknown_4                  4: 0
          unknown_5                  7: 15
          unknown_6                  7: 127
          start_offset              27: 0x01a4100
          pad                        1: 0
          data                     146: [skipped]
          [type 1]
          unknown_7                 32: 0x080ffa63
          checksum                   4: 0xf

 

The first MQA signature is much further into the file for Track01 than Track04. Perhaps Roon simply doesn't search far enough. The 'stream_position' field indicates the absolute position of the MQA sync word ('magic') counted in samples from the start of the stream. For Track01 the difference between this and the offset it was actually found at is very close to 2 seconds, the standard length of the CD lead-in. The absolute position of Track04 is roughly 20 minutes. This suggests that the CD is coded as a single MQA stream in contrast with Tidal and downloads that restart, and often change MQA parameters, for each track in an album.

 

We also see that the MQA control stream is in the LSB of the CD (the scanner treats everything as 24-bit). This means there is no high-frequency information encoded anywhere.

[mansr]

On 7/13/2018 at 8:51 PM, mansr said:

http://hqcd.jp/uhqcd_eng/

 

Apparently they use a low-viscosity polymer that more accurately replicates the pits in the stamper, and the reflective layer uses something (undisclosed what) superior to the usual aluminium. These modifications result in a signal from the optical pickup with higher contrast and less jitter than with a regular CD. That's all quite reasonable, though I'd like to measure the improvement for myself rather than trust the pictures they provide. The reason this is still snake oil is that every CD player is designed to work with the signal parameters specified by the "red book" standard, IEC 60908. A correctly working CD player can retrieve the data from an undamaged standard disc with 100% accuracy. Further improving the signal quality will not result in better than correct readout of the data, and it certainly won't have any of the claimed miraculous effects on the sound.

I did some testing with a UHQCD:

 

[GUTB]

Well I have picked up a Mytek Liberty along with this Chinese LPS. A $1k DAC shouldn't come with a SMPS but I do appreciate the DC line in at least. I chose the Mytek because it was the only affordable DAC which supports MQA-CD, or MQA over SPDIF.

 

I've been burning it in for over 200 hours now so it should be about ready for critical listening. I have some MQA-CDs coming in from CD Japan but they're taking their sweet time shipping them. I have some MQA-CD samplers and Tidal Masters of course.

[Ralf11]

Looking fwd to your listening impressions, or - better - blind testing...