Jump to content


  • Content Count

  • Joined

  • Last visited

  • Country


Everything posted by FredericV

  1. Regarding the deliberate crippling by dithering / adding noise / band limiting the signal during the MQA encode step to 17/96 at best, I found an actual ground breaking article how such damage could potentially be undone: https://petapixel.com/2021/08/30/googles-new-ai-photo-upscaling-tech-is-jaw-dropping/ Why is this tech interesting? Their upscale tech is based on a trained model which has learned from the deliberate crippling of the signal by adding noise until only pure noise remains, and knows how to do the inverse: MQA encoding adds noise to the signal, and in t
  2. Once again, our well known influencer keeps spamming old "unscientific" HB videos, in an attempt to educate his audience: He probably has a list of canned articles he keeps spamming. In his failed attempt of debunking GoldenSound, HB has admitted he can only guess, and this video example is a simulated example of time smear and more proof of the same. HB can only guess how mQa tries to correct this, as mQa has never provided before- and after files or has explained the process: Still waiting for scientific proof ...
  3. Denying access to certain content because of contractual requirements, even though the device is capable of playing the content, is a form of DRM. Here the R in DRM obviously stands for both "restrictions" and "rights". It reduces your rights by denying you to listen to regular PCM and restricting you to the mQa versions. mQa is managing your rights. And then on the HB site, we found this quote: Furthermore, mQa contains a CAS system, as it requires a decoder to access the pseudo high-res part of the data, and without a licensed decoder you get a crippled version which
  4. What a funny paradox by HB, but I can at least confirm those among the club winning best of show in Munich, do not use MQA at all to reach this award. We personally asked several reviewers. Seen in the secret mQa group: I remember the same paradox with SACD vs CD all the way back in early 2000. PS: not convinced about the shift in cross-over point: https://www.aes.org/e-lib/browse.cfm?elib=19396
  5. Propably the isolator is not fully compliant with USB standards (or the combination of the isolator with his cables) and is messing with bits which will confuse the mQa decoder. I once had something very similar with an iFI USB power plant and a dual USB cable from another brand: with regular PCM there would be not so subtle dropouts, but with their own dual cable it would not occur. From this perspective mQa works as designed. I actually keep a Mytek in stock for this purpose. It's an easy tool to visually confirm bitperfect output of music servers.
  6. Now we need a blue light on record players, when an MQA record is being played .... Lossy++
  7. The fact that this question is being raised ...
  8. The problem with the GPL is that the data processed by GPL tools, is not protected by the GPL. It only protects the software, not the data. Adding MQA to existing open source players is clearly wrong: https://www.musicpd.org/commercial.html
  9. Amir has a HDCD backstory: https://www.audio “science” review/forum/index.php?threads/mqa-bad-for-music.1378/page-2#post-36197 Amir's team got the HDCD light working, but it was actually fake, not doing any decoding: Even more interesting is the fact that foobar has an actual open source HDCD decoder: https://git.lopez-snowhill.net/chris/foo_hdcd which was also used for ffmpeg: https://hydrogenaud.io/index.php?topic=112225.0 https://opensource.com/article/20/7/hdcd
  10. Is Mansr no longer active here? Looking around on ASR, this post is interesting: https://www.audio “science” review/forum/index.php?threads/tidal’s-new-lossless-tier-says-goodbye-to-mqa-but-does-it-really.22968/page-9#post-769891 https://distrokid.zendesk.com/hc/en-us/articles/360059827614-How-I-Can-Send-My-Music-in-TIDAL-Master-Quality-MQA- The sausage machine for $17.99 per album ...
  11. 1. why did Apple, Amazon HD, Qobuz and many others chose NOT to implement mQa? 2. a new world which is based on the old world? How is that possible 5. for 1X rate sources such as 24/48 and 24/44.1, the encrypted part of the file does not compress well with flac, so it's actually not saving Tidal any bandwidth, whereas Qobuz would be saving bandwidth serving the non-encrypted 24/48 or 24/44.1 flac 6. MQA CD and those batch encoded Tidal redbooks into MQA CD versions, are using upsampling on several decoders .... To dig a little bit deeper why MQA cannot defeat info
  12. They also tried to remove the GoldenSound video from the wiki page ...
  13. Just found evidence on Wikipedia, MQA is trying to have "lossy" removed and thus hide the truth, and rewrite the narrative: https://en.wikipedia.org/wiki/Talk:Master_Quality_Authenticated#Disputed and then their proposed changes were countered by the wikipedia editors: They are also trying to remove the @GoldenOne GoldenSound video from the wiki page:
  14. Something with the time domain went wrong ...
  15. 88.2 MQA decoding includes an upsampling step of the baseband 1X signal from a 24/44.1 distribution file to 24/88.2 and then add the missing ultrasonics to the upsampled 24/88.2 version, recovered using lossy origami. which means that the impulse response should look like test05b in this post, which looks exactly like the plot from Craven's filter In our case we did x2 from 96K to 192K, but doing 44.1 -> 88.2 or 96 -> 192 is basically doubling the samplerate, and the impulse is no longer one sample, but spans several samples. If we just see one pole at 88.2, that can'
  16. When argumentum ad verecundiam has become a religion ... Many responded the HB video did not debunk the claims of @GoldenOne / GoldenSound, instead he admitted he is guessing. He did not even question mQa's answer, or try to peer review it ... just make the BT reply sticky.
  17. There are other ways for "bastardized PCM". http://jpinsoft.net/DeepSound/ Just watch Mr. Robot season one, where he demonstrates the tool "deepsound" which he uses to hide photos inside the noise floor of PCM files. You can imagine the flac encoder having a very hard time encoding non-audio data bits from a bastardized pcm source file to which this garbage (from the viewpoint of the encoder) was added, as the flac prediction algo can't do a good job on predicting the next sample. It will contain more noise, so it will need to store a bigger residual, thus larger files. This is
  18. If MQA is claimed to be end-to-end in the analog domain, please show us the MQA DAC output of an MQA encoded square wave on an analog scope, and originally sampled from an analog function generator. So the MQA version of Monty's video: Please prove that input and ouput square waves in the analog domain are end-to-end. Bet mQa can't do this ...
  19. He did not even link to the original video in his response. Why would he not mention the original content he is trying to debunk? This one is amusing: as he goes on in appendix 4: and then he shows some plots of .... obviously computer generated signals which do not exist in real music, except maybe in computer generated music: These impulses are clearly computer generated. So how does MQA test it's claims when users are NOT allowed to encode computer generated signals, while MQA tries to prove MQA can encode them? Why the dual standards? Why is
  20. How can MQA be 9 mbit? As all MQA is decimated to their infamous 17/96 coding triangle for 24/96 and any other source mulitples of 48 kHz such as 24/192 and 24/384, the bitrate of their 24/48 distribution files is always very similar no matter to which fake resolution they unfold. True 24/192 without such decimation is usually around 5 mbit, so how does MQA fit in the 9+ mbit advertising?
  21. When Tidal MQA users realize they are on the Titanic .... Even if they lower the non-MQA tier to the same price point as Apple, for that amount competitors are offering real hi-res without the origami. The shelf life of proprietary origami has expired a long time ago, as Edholm's law predicts that the bandwidth and data rates double every 18 months, which has proven to be true since the 1970s. Furthermore, since 5G, we are starting to see unlimited data plans, e.g. in Belgium we now have unlimited 5G plans from Proximus, for which the fair use is defined as 100GB. At 5 mbit for
  22. They look like actual tools but used as props to add authority to the video. Because those tools in real use, would not result in such a clean background settings: When Hans was still writing for Vi-Fi, he had one of the few publications which actually showed measurements. So he probably has some skills, but he never bothered to look into the technical arguments of the independent MQA researchers - which is a bit contradictory.
  23. Hans did not debunk anything. From watching the video we learned: he does not understand the science, he just assumes MQA works in a way he cannot prove, such as the usage of splines he is guessing he is using the argument from authority (argumentum ab auctoritate) he only reads official publications from MQA, AES, the patents - he does not investigate the research of independent researchers he looks a bit angry His video can be summarized by this still: He admits not going into any talking points from the GoldeSound video. Bas
  24. 10 ouf of 12 comments are invisible from this suggested FB post, so I took a similar action. I also did cancel Tidal as I could not accept existing redbook files being replaced by crippled batch encoded MQA CD versions. I won't miss much, these are the 2 comments which were allowed:
  25. Just found this example in their secret group: "I compare many tracks from Qobuz hi-res and tidal mqa and does unfold to the same sample rate as these fake hi-res file they sell these day. " Fore sure MQA adds something which is not in the 24/192 version on Qobuz. Does he actually believe Qobuz is the fake version, and MQA is the real version, while it shows the aliasing from the leaky upsampler in the MQA version which was not in the Qobuz version?
  • Create New...