misterdog
Not the canine kind
So we need to invest in an MQA cartridge for our turntables ?
MQA
- Thread starter Rosewind
- Start date
- Status
So we need to invest in an MQA cartridge for our turntables ?
Amber Audio
This is the Day
Maybe
I was reading about other stuff and saw this article, I get regular emails about new vinyl releases and MQA mastered ones had never flagged up so was curious about this guy and his Doors release. Not done much in depth searching but it doesn't appear to be a popular option yet.@DimitryZ, the post really was an attempt at humor, and was not aimed at any one participant in this thread. Winning the MQA thread was meant to be interpreted as both the pro or anti position. As with cable threads I am surprised by the number of hours spent, and passion involved in trying to change the other side's mind, knowing that it probably won't happen.
I don’t actually have much of an opinion about the subject of this thread, and I haven't read much of it because I'm not that interested. And I can't promise I won't make another image the next time a cable thread gets over 500 posts!
Robert
Tapehead
My real issue with MQA is that it's proprietary closed source and requires licensing.
it alarms me that streaming platforms would encode their libraries with a proprietary codec, and that the consumer cost of streaming devices increases because licence fees are paid to MQA.
We just don't need this - it's bad.
Does it bring anything I want to the party in terms of sonic benefits or required net bandwidth? - no.
it alarms me that streaming platforms would encode their libraries with a proprietary codec, and that the consumer cost of streaming devices increases because licence fees are paid to MQA.
We just don't need this - it's bad.
Does it bring anything I want to the party in terms of sonic benefits or required net bandwidth? - no.
Tony L
Administrator
This is my fear turning into reality! Total marketing BS hype turning a lossy format into a PR benefit. Dominic Cummings/Leave.EU-grade audio.
PS Thankfully I already have all the Doors albums I want on Steve Hoffman-mastered DCC vinyl or original Elektra, so no digital stuff of any description anywhere in the signal path.
Jim Audiomisc
pfm Member
Yes. The snag then being that - to be of use to end-users when a given implimentation is in use those who wish to use it need a 'compatable' process that deals with details that aren't in the patent. The bun and the sixpence. For some decades now this has become quite common in patents I've read.
The drawback for the patent holder is that it means a 'clean room' devising of a compatable method isn't protected by their patents. The challenge for 3rd parties, though, is to find one.
However the use of 'imperfect' reconstruction filters is hardly novel, so this could be done openly and used with DACs that are not themselves MQA types.I'm less sure about the other side of MQA because details are so vague. And TBH I'm not even sure they use it given that HDCD claimed things it never did.
Jim Audiomisc
pfm Member
Erm it could be functionally identical so long as it used 'clean room' methods given some limits. One is that any methods openly used before the patent could be used as they can't be claimed by the patent. e.g. using 'imprefect' filters to generate aliasing. The other is that a method may differ from the *novel* ones in the patents.
Using a an old fashioned NOR gate in a patented arrangement doesn't stop other people from using NOR gates.
Hence it seems to me to be quite possible that we could find someone devises an open source - at least partial - MQA decoder. Even if it can't use MQA as a brand name. it might perhaps be called a FAQE system, though with E = Enhancer.
Interesting to speculate if the bitflagging pattern that indicates MQA is protected. i.e. No-one could use that bit pattern. But the use of such patterns again isn't new. So an open FAQE creator could probably also be employed if someone thought a neat ideal.
That said, I guess a FAQE enhancer would be more useful and who knows, perhas each DAC producer could offer their own variety of reconstruction filters for the user to experiment with. A wider choice for users. Should be welcome for those interested in this sort of thing...
Jim Audiomisc
pfm Member
You can't make and sell something claiming it is the MS Windows OS or GUI, etc. But you can make desktops that look and behave in similar ways. Linux does this all the time, and MS contribute to Linux code. That said, none of my Linux boxes look much like `doze as I much prefer the RO WIMP and desktop. Much more user friendly.
Timcat
pfm Member
I agree, but isn't the patent simply doing its job of protecting the proprietor from unlicensed competition, including by end users? However, a competently drafted patent should still do so even if the specification included full details of the commercial product.
The drawback for the patent holder is that it means a 'clean room' devising of a compatable method isn't protected by their patents. The challenge for 3rd parties, though, is to find one.
Your first point is not strictly correct. A "clean room" devised method that worked in the manner claimed in a patent would still infringe the patent despite the fact that it was not copied from the original. Patents, unlike copyright, provide an absolute monopoly, there is no need for there to have been any copying for there to be infringement.
As I mentioned upthread, at least one third party, Auralic, has devised their own method of reconstruction or decoding (call it what you will!) MQA that they claim does not infringe MQA's patents. I don't think that they have been sued, so this may indeed be possible!
Timcat
pfm Member
Erm it could be functionally identical so long as it used 'clean room' methods given some limits. One is that any methods openly used before the patent could be used as they can't be claimed by the patent. e.g. using 'imprefect' filters to generate aliasing. The other is that a method may differ from the *novel* ones in the patents.
Using a an old fashioned NOR gate in a patented arrangement doesn't stop other people from using NOR gates.
Hence it seems to me to be quite possible that we could find someone devises an open source - at least partial - MQA decoder. Even if it can't use MQA as a brand name. it might perhaps be called a FAQE system, though with E = Enhancer.
Interesting to speculate if the bitflagging pattern that indicates MQA is protected. i.e. No-one could use that bit pattern. But the use of such patterns again isn't new. So an open FAQE creator could probably also be employed if someone thought a neat ideal.
That said, I guess a FAQE enhancer would be more useful and who knows, perhas each DAC producer could offer their own variety of reconstruction filters for the user to experiment with. A wider choice for users. Should be welcome for those interested in this sort of thing...
Working around a patent without risking infringement is not a simple thing. However, it can be done in many cases, but it requires engineers to work with patent attorneys (who will always be technically as well as legally qualified), as the solutions required are not always that obvious or simple. Fun though
.
gustav_errata
pfm Member
Hence it seems to me to be quite possible that we could find someone devises an open source - at least partial - MQA decoder. Even if it can't use MQA as a brand name. it might perhaps be called a FAQE system, though with E = Enhancer.
I think a similar situation was in place for MP3 for a long time: the decoding/encoding methods were under patent, but FOSS implementations existed. GNU/Linux distros were reluctant to include the codecs by default because of the threat of patent litigation, although they were easily installed anyway.
Jim Audiomisc
pfm Member
I can understand that as it was similar here until Open Reach finally insalled FTTC here. Before that even streaming the 320k aac streams from BBC radio was unreliable at times.
My solution was to download things before 9am in the morning when speeds were higher[1], or simply to buy CDs, etc. I prefer to pay to 'own' anyway rather than 'rent access'. I still download large items before 9am to avoid them getting added to my 'monthly cap'. By buying I suspect I also put more money into the pockets of the musicians than they get from streaming.
[1] Automated the process so they get fetched as I make breakfast, etc.
Jim Audiomisc
pfm Member
No idea if it is true. Seems mad. But raises the curious question of how someone will 'de-blur' the 'tip resonances' of the cutter and their stylus. 8-] However...
Oh dear! An LP of MQA?! At least HDCD on an Audio DVD was prefectly decodable back into LPCM! But perhaps just another sign of the behaviour of some in the music biz. If it sells another 'version', great, count the money!
Jim Audiomisc
pfm Member
I don't. And one of my worries is that - yet again - 'new' issues of Joni Mitchel material will be farted about. Is here recent set of early recordings plain CD or something else? Given experience from earlier years we can't always trust what it may say on the tin!
Jim Audiomisc
pfm Member
However - as your last point shows - a method may be different to the original.
Another point this brings out is: If they took you to court saying you'd duplicated their method, they'd then be telling everyone "This open method is exactly what we use" and thus it would no longer be confidential.
Their patent *can't* cover methods like using a non-Nyquist reconstruction filter because many people have used those for many years in many forms.
So 'manner' in this case really would mean a duplicate. And if you did that without knowing their hidden details you can't be said to have stolen details they kept hidden. Nor be blamed for choosing them lacking the knowledge of which specific filter values, etc, they use.
Have you read the HDCD patents? They make an interesting comparison with MQAs in this area.
Jim Audiomisc
pfm Member
Working around a patent without risking infringement is not a simple thing. However, it can be done in many cases, but it requires engineers to work with patent attorneys (who will always be technically as well as legally qualified), as the solutions required are not always that obvious or simple. Fun though
Agreed. You need to know what you are doing. In this case it would also help to have some background in signals areas like steganography and using methods like autocorrelation to determine the significant characteristics. Mathematically it is quite interesting as an exercise But as you say, can be facinating and enjoyable.
However the aim of FAQE wouldn't be to 'decode MQA' but to produce an alteration to the HF content that people 'like'. i.e. similar in terms of the subjective reactions. Which does not need an exact replication of the MQA process.
Mr Underhill
pfm Member
I thought this was an interesting review:
M
M
Tony L
Administrator
I haven’t bought the early demos etc box set so have no idea. I do know that I prefer the earlier/first CD issues of her core catalogue to the remasters, though I can’t be arsed farting around with HDCD so will inevitably be hearing some in the box set with obvious error. I’m lucky in that I have the albums I care the most about on original issue vinyl and it is better than any CD mastering I’ve heard with the possible exception of the couple of audiophile cuts Hoffman did for DCC. I’m still looking for a mint 1st press of Hijira though, its one of my favourites and the remastered CD is all kinds of wrong, just bright, thin and gutless to my ears.
- Status