pink fish media

Go Back   pink fish media > discussion > audio

Notices

Reply
 
Thread Tools
  #16  
Old 17-02-17, 11:59 AM
HarryB HarryB is offline
pfm Member
 
Join Date: Nov 2012
Posts: 1,054
Excellent. More opportunities to sell the same stuff all over again to the gullible, the golden eared and those with too much money and time.
Reply With Quote
  #17  
Old 17-02-17, 12:08 PM
radamel radamel is offline
pfm Member
 
Join Date: Nov 2012
Posts: 3,161
This should be especially directed at streaming services and not downloads.

I haven't paid a penny more for having access to Tidal's MQA music.
__________________
I refuse to join any club that would have me as a member.-
Groucho Marx
Reply With Quote
  #18  
Old 17-02-17, 12:35 PM
Dowser Dowser is offline
Learning to bodge again..
 
Join Date: Jul 2006
Posts: 4,479
Quote:
Originally Posted by jirij View Post
Right, you could easily bypass it by cutting away the "folded" encoded track, which kind of defeats the point for mqa for the end user (assuming the encoded track contains some kind of magic that makes the music better), so it's fairly safe to assume most people will just leave the files alone.
I have no technical knowledge on how MQA does its thing (and to be fair I am pretty anti digital full stop - vinyl sounds better than Cd - CD sounds better than file based digital ), I come at this from an overall IT perspective and how to protect data within public infra (secure the payload, not the file) - so is this enfolded track stored as meta data with MQA then? What happens if it is removed? A good way to protect data is to use something you know, with something variable/uncontrollable, and merging the first with the second such that the first is undetectable. Digital media is a pretty good variable source to use with something you know, but if MQA stores the bit people know as metadata, I give the format a very limited shelf life before someone breaches it.

In general digital data protection today, it is more about how you protect the "bit that you know" - this is generally a key of some sort, and that is the bit that must be protected...of course, any access to the data means that before payload access you must be provided to the key somehow...this is where it gets interesting.

I think either MS or Google will release the first scalable solution (MS Azure tokenised digital rights first release comes in March or April I think) - it won't work properly for the first few releases, but once it does then media digital content providers will have a sure fire way of controlling who can access their data.

However, anyone who can access that data, can then quite easily clone it and re-release it without any protection. So how do you protect your payload to ensure that if it is cloned, the end result payload is lower quality/inferior somehow? Legally, a digital signature of course - but practically, that does not not work with digital audio/video payload - there are enough of us out there that just don't care about the legal aspect. But if you knew that the best quality was only available via a paid source, wow - you'd be set for life I never really followed MQA until now, I will now go and try to inform myself what it is and how it works.

Digital rights/protection is an incredibly important and interesting subject currently IMO - the future has us storing all data files in a big public cloud (let's call it the Internet), and then restricting who can access the payload within those files. This is a sea change to previous approach of protecting access to the data files in the first place, which while the talk of protecting access to the data rather than the perimeter has been corporate security buzz wording for several years now, there is no end-end scalable solution that enables this ideal.

When it does it will have unexpected impact methinks
Reply With Quote
  #19  
Old 17-02-17, 10:35 PM
HarryB HarryB is offline
pfm Member
 
Join Date: Nov 2012
Posts: 1,054
Quote:
Originally Posted by radamel View Post
This should be especially directed at streaming services and not downloads.

I haven't paid a penny more for having access to Tidal's MQA music.

YOU haven't. Yet. But you will.

But how many new people have been sucked in already?
Reply With Quote
  #20  
Old 18-02-17, 12:04 AM
peperoni peperoni is offline
Officially now an Old Git
 
Join Date: Jul 2003
Posts: 308
No one is sucked in, we do have brains and can think it through ourselves. We really don't need a great overseer to protect us.
Reply With Quote
  #21  
Old 18-02-17, 12:14 AM
radamel radamel is offline
pfm Member
 
Join Date: Nov 2012
Posts: 3,161
Quote:
Originally Posted by HarryB View Post
YOU haven't. Yet. But you will.
I think you are speculating. Or are you saying that you KNOW for a fact that Tidal has decided to increase their price and that I'll be willing to accept that price increase?
__________________
I refuse to join any club that would have me as a member.-
Groucho Marx
Reply With Quote
  #22  
Old 18-02-17, 12:16 AM
johnhunt johnhunt is offline
pfm Member
 
Join Date: Jul 2003
Posts: 2,320
Quote:
Originally Posted by HarryB View Post
YOU haven't. Yet. But you will.

But how many new people have been sucked in already?
Do you mean people taking out tidal subscriptions when you say sucked in? if so I was sucked in months and months ago. By the way have you listened to Mqa yet?
Reply With Quote
  #23  
Old 18-02-17, 01:57 AM
Jim Audiomisc Jim Audiomisc is offline
pfm Member
 
Join Date: Mar 2016
Posts: 2,259
Quote:
Originally Posted by Dowser View Post
But if you knew that the best quality was only available via a paid source, wow - you'd be set for life I never really followed MQA until now, I will now go and try to inform myself what it is and how it works.
I did an outline of the public info on this a while ago. You may find these pages help.

http://www.audiomisc.co.uk/MQA/origa...reAndBack.html

http://www.audiomisc.co.uk/MQA/bits/Stacking.html

Basically, MQA employs three tricks. The above shows what their patents on the first two describe. The third is some kind of magic metadata to tell the playout system how to 'adjust' for the choice of ADC/processing at the start of the chain. There is, of course, also a magic bitpattern to tell the system the data is MQA. Someone on the CA Forum has spotted this xored into the LSB of the Left and Right channels.

The snag is that - surprise, surprise! - the patents and public statements omit many vital details. Thus the real details of how the system works are buried under a layer of secret sauce.
Reply With Quote
  #24  
Old 18-02-17, 02:24 AM
davidsrsb davidsrsb is offline
pfm Member
 
Join Date: Jan 2013
Posts: 6,121
Quote:
Originally Posted by Jim Audiomisc View Post
Basically, MQA employs three tricks. The above shows what their patents on the first two describe. The third is some kind of magic metadata to tell the playout system how to 'adjust' for the choice of ADC/processing at the start of the chain.
The tests carried out and published so far show now evidence of the part 3 filter adjustment being done at all.
This reminds me of HDCD where nobody ever used some of the features listed in the patent and marketing blurb
Reply With Quote
  #25  
Old 18-02-17, 02:51 AM
Jim Audiomisc Jim Audiomisc is offline
pfm Member
 
Join Date: Mar 2016
Posts: 2,259
Quote:
Originally Posted by davidsrsb View Post
The tests carried out and published so far show now evidence of the part 3 filter adjustment being done at all.
This reminds me of HDCD where nobody ever used some of the features listed in the patent and marketing blurb
Yes, the phrase "HDCD on stilts" has come to my mind once or twice. :-/ Both seem to try to pull off a magic trick of producing a result that is presented as 'compatable' with both LPCM replay and "magic decoder ring" playable. The problem being that the data is *not* now plain LPCM, so may have sacrificed *something* wrt LPCM.

For 24bit files/streams that may not matter as the magic may have replaced some of the excess noise bits. And *may* have no audible effect upon LPCM replay. But I am particularly wary of the part of MQA that presents it as possibly suitable for use on 16bit/44.1k audio CDs. That would be dream for some big music companies as they could decide sometime in the future 'single inventory' and flog high priced CDs on a "one size fits all" claim.

Given the music biz's messy track record with HDCD...

Personally, given the way they have often fouled up things in the past I confess I don't feel that I'd feel very 'assured' by them saying they have 'assured' me of the results.
Reply With Quote
  #26  
Old 18-02-17, 03:29 AM
Stunsworth Stunsworth is offline
pfm Member
 
Join Date: Apr 2012
Posts: 3,620
Quote:
Originally Posted by Jim Audiomisc View Post
The snag is that - surprise, surprise! - the patents and public statements omit many vital details. Thus the real details of how the system works are buried under a layer of secret sauce.
I don't know how the fuel management system in my car works, but I manage to get to work every day.
__________________
Steve
Reply With Quote
  #27  
Old 18-02-17, 03:41 AM
Jim Audiomisc Jim Audiomisc is offline
pfm Member
 
Join Date: Mar 2016
Posts: 2,259
Quote:
Originally Posted by Stunsworth View Post
I don't know how the fuel management system in my car works, but I manage to get to work every day.
Its always shakey to try making analogies which have also cherry-picked one statement out of a long set of points. All too often the attempted analogy isn't a reliable one. However since you've done that already, consider the following...

And you're still happy if it turns out that the car company sold you one that fiddled its test results and you've burned more fuel and had lower efficiency than they let you believe?

The real problem here is that you may not know what you have lost or missed out on. HDCD is a classic example of this. As has been the selling of cars based on engine management systems that gamed the tests.
Reply With Quote
  #28  
Old 18-02-17, 06:41 AM
radamel radamel is offline
pfm Member
 
Join Date: Nov 2012
Posts: 3,161
I'm eager to listen to artists like Sokolov, Argerich or Pires in MQA format.

IMHO piano pieces transients are very hard to reproduce so comparing the MQA versions to plain redbook should be interesting.
__________________
I refuse to join any club that would have me as a member.-
Groucho Marx

Last edited by radamel; 18-02-17 at 07:05 AM.
Reply With Quote
Reply

Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -7. The time now is 08:48 PM.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2017, vBulletin Solutions Inc.
pink fish media