Skip to main content

Topic: Can you fake a hi-res track? (Read 3047 times) previous topic - next topic

0 Members and 1 Guest are viewing this topic.
Can you fake a hi-res track?
Reply #25
I seem to recall that there were a couple of sites that were pulled up for selling supposedly hi-res material that was then discovered to have no HF content. This was a minor embarrassment in some quarters, particularly where the tracks had already received favourable reviews.

It just occurred to me that an unscrupulous owner might find their website business so jeopardised by such a discovery, that rather than replace their inventory, they might seek to 'rectify' the situation by some form of batch processing.

Not that I'd seek to spread FUD amongst hi-res enthusiasts that they're listening to undetectable frauds.



The fact is that about half of the commercial SACD and DVD-A released between 2000 and 2010 were found to have no useful HF content.  I can't say no content at all because SACD inherently adda noise above 20 KHz, and if the upsampling is followed by processing in the analog domain, that will add some noise > 20 KHz.

  • bennetng
  • [*][*][*][*][*]
Re: Can you fake a hi-res track?
Reply #26
Not that I'd seek to spread FUD amongst hi-res enthusiasts that they're listening to undetectable frauds.
Proof of (your) concept to make a fake.
This old 1986 anime song doesn't have hi-res version, right?
No MQA'ish ugly gap and mirroring in spectrogram :))
https://youtu.be/Mmbc8E6rAmM
  • Last Edit: 27 November, 2017, 10:22:16 AM by bennetng

  • Wombat
  • [*][*][*][*][*]
Re: Can you fake a hi-res track?
Reply #27
Very well done indeed. If you know about it you may wonder at second one when the noisefloor gets a slight step but with synthetic music this can be caused by everything.
Is troll-adiposity coming from feederism?
With 24bit music you can listen to silence much louder!

  • bennetng
  • [*][*][*][*][*]
Re: Can you fake a hi-res track?
Reply #28
Yeah, that noise floor simulation is a lousy one, in Spek it even shows some comb filtering color bands in its mono mixed spectrogram. Run the signal through a D/A/D chain will make a better fake and make null test much harder due to clock drift.

There are a lot of ways to make fake hi-res and I am not surprised if some mastering graphics engineers are doing this already.


  • Porcus
  • [*][*][*][*][*]
Re: Can you fake a hi-res track?
Reply #29
Had it only been audible, then adding "fake" HF could very well have been for the better:
http://wiki.hydrogenaud.io/index.php?title=Perceptual_Noise_Substitution

I am sure it is possible to implement a clever level-adjusted ultrasonic noise injection as an on-the-fly DSP.  Eagerly anticipating foo_hirezifier to add noise according to the PRE_EMPHASISHI_REZ tag.

  • bennetng
  • [*][*][*][*][*]
Re: Can you fake a hi-res track?
Reply #30
Another fake 96k demo with fixed noise shaping "bug" and bonus faint 36kHz idle tone.

Original file at 44k:
http://media.soundsonline.com/wav/2432_Allegro-Agitato.wav

They are all MIDIs. Click the "Audio Demos" and find the others.
http://www.soundsonline.com/hollywood-strings

Fake orchestra and choir sound best with fake hi-res ;D

  • Porcus
  • [*][*][*][*][*]
Re: Can you fake a hi-res track?
Reply #31
and bonus faint 36kHz idle tone.

Is that a known vintage tape bias frequency?  You should definitely compare (sighted, of course) to a tone at 44.1 or 48, which is guaranteed to sound more harsh, cold and digital.

Re: Can you fake a hi-res track?
Reply #32
To answer OP's question.  Very easily, with this http://www.voxengo.com/product/warmifier/

  • bennetng
  • [*][*][*][*][*]
Re: Can you fake a hi-res track?
Reply #33
Is that a known vintage tape bias frequency?
What a professional assumption!! Sounds like a comment from a vintage gear guru from some random vintage audio forums. Maybe they can tell the tape and tape machine model as well ;D

Last year a moderator from one of the largest local (Hong Kong) PC enthusiast forum posted a picture of the Nvidia GTX1060, a forum member forwarded the picture to reddit and a lot of "experts" claimed that picture was photoshopped, and analyzed the "naturalness" of the picture's reflection and lighting and so on.

https://www.reddit.com/r/nvidia/comments/4q808n/gtx_1060_appears_in_hong_kong/

Just an example of how a real thing can be "detected" as fake, and vice versa.