I’ve built a few USB based D/A converters in the past which have been fun projects, lately though I've been reading up on the Asynchronous USB controllers but found the makers of these controllers to be very reluctant to provide any details with datasheets to be non existent. 
I know the marketing guys for the big companies seem to love to get their teeth into new technology & hype everything up in an effort to make us all believe that Async USB is the one & only way to get good sound from a USB socket, but I have learnt to take this with a large pinch of salt because a lot of the hype for Async USB comes from companies who have stuff to sell, so of course they are going to be biased (until the next big thing comes along & then be all over that like a rash).
What is strange to me though & I wonder if anyone else has noticed this? is the mystery or almost cloak & dagger approach that is taking place with the makers of these Async USB controllers, which all seem to be made in China or somewhere similar, companies are said to have non disclosure agreement contracts or NDA’s in place to stop anyone from disclosing any details including specifications, example circuit & so forth of their products & how they work, WHY? It’s not as if we are about to copy such a complex device & make our own.
There have also been a few rumours floating around the internet that some of these Async USB controllers which are protected by these NDA’s are said to take the 16bit 44.1khz of a standard USB & merely raise it internally up to 96khz or 192khz, rather than having pure native support for those frequencies, how much of this is rumour or whether its based on anything I cant say, but nobody who has used these chips are going to be able to disclose anything because of the NDA.
It is also very odd why none of the makers of these Async USB controllers have their products available with any of the usual channels like the main distributors (RS, Farnell, Mouser etc), every other product is usually stocked at either of these places with full datasheets available, when I asked one of the Chinese makers of these Async USB controllers they just replied “no we do not have any distributors, everything is by agreement of NDA policy”. They did not want to answer any questions & seemed very sheepish. Its not because they are a small company or anything like that either, because the amount of cheap Chinese stuff on sale that makes use of Async USB controllers are quite prolific, so they could easily supply distributors but instead prefer to keep in the shadows if you like with all this NDA.
Is it because these larger distributors always have datasheets available for products & this would require the companies producing these Async USB controllers to provide detailed datasheets on how they work, perhaps which could expose something?
To add to the mystery is the fact that Texas Instruments which lets face is a much larger concern has discontinued their TAS1020B Async USB controller, with a TI employee confirming that TI have no plans to introduce any other Async USB controllers, so despite all the buzz around Async it seems like they are getting out of this game?
I know that when it comes to audio its never ever all down to just a single component, in this case Async or not, as anyone knows its much more to do with the entire design, & while there seems to be a kind of universal marketing phrase doing the rounds, something like “if it hasn’t got Async USB then it will be like listening though a keyhole” or perhaps “if it doesn’t have Async, you DON’T WANT IT” & I have seen many other silly remarks, when in fact I've heard commercial USB products that do not use Async controllers at all & found many to sound better than ones using Async.
I just wondered what others thought on here about this or am I just rambling?
I know the marketing guys for the big companies seem to love to get their teeth into new technology & hype everything up in an effort to make us all believe that Async USB is the one & only way to get good sound from a USB socket, but I have learnt to take this with a large pinch of salt because a lot of the hype for Async USB comes from companies who have stuff to sell, so of course they are going to be biased (until the next big thing comes along & then be all over that like a rash).
What is strange to me though & I wonder if anyone else has noticed this? is the mystery or almost cloak & dagger approach that is taking place with the makers of these Async USB controllers, which all seem to be made in China or somewhere similar, companies are said to have non disclosure agreement contracts or NDA’s in place to stop anyone from disclosing any details including specifications, example circuit & so forth of their products & how they work, WHY? It’s not as if we are about to copy such a complex device & make our own.
There have also been a few rumours floating around the internet that some of these Async USB controllers which are protected by these NDA’s are said to take the 16bit 44.1khz of a standard USB & merely raise it internally up to 96khz or 192khz, rather than having pure native support for those frequencies, how much of this is rumour or whether its based on anything I cant say, but nobody who has used these chips are going to be able to disclose anything because of the NDA.
It is also very odd why none of the makers of these Async USB controllers have their products available with any of the usual channels like the main distributors (RS, Farnell, Mouser etc), every other product is usually stocked at either of these places with full datasheets available, when I asked one of the Chinese makers of these Async USB controllers they just replied “no we do not have any distributors, everything is by agreement of NDA policy”. They did not want to answer any questions & seemed very sheepish. Its not because they are a small company or anything like that either, because the amount of cheap Chinese stuff on sale that makes use of Async USB controllers are quite prolific, so they could easily supply distributors but instead prefer to keep in the shadows if you like with all this NDA.
Is it because these larger distributors always have datasheets available for products & this would require the companies producing these Async USB controllers to provide detailed datasheets on how they work, perhaps which could expose something?
To add to the mystery is the fact that Texas Instruments which lets face is a much larger concern has discontinued their TAS1020B Async USB controller, with a TI employee confirming that TI have no plans to introduce any other Async USB controllers, so despite all the buzz around Async it seems like they are getting out of this game?
I know that when it comes to audio its never ever all down to just a single component, in this case Async or not, as anyone knows its much more to do with the entire design, & while there seems to be a kind of universal marketing phrase doing the rounds, something like “if it hasn’t got Async USB then it will be like listening though a keyhole” or perhaps “if it doesn’t have Async, you DON’T WANT IT” & I have seen many other silly remarks, when in fact I've heard commercial USB products that do not use Async controllers at all & found many to sound better than ones using Async.
I just wondered what others thought on here about this or am I just rambling?
This is something I've poked into quite a bit, precisely because I'd like to build my own.
It's true that the documentation is about as impenetrable as the human mind can conceive. This is something I've encountered in connection with other communications protocols, notably CanBus.
I think that these two technologies are similar in that there are considerable financial vested interests.
The whole question of USB audio is difficult to understand. Why was such a problematic implementation chosen for the initial commercial offerings? RS232, the previous serial standard in wide use on computers had a simple mechanism to allow unsynchronised clocks to be employed at RX and TX. USB audio, by way of contrast, initially employed a scheme where embedded timing information is recovered from the transmitted bitstream, commonly by a PLL. Of course in this it is doing no more than mimic the professional and domestic standards for digital audio embodied in the S/PDIF and AES/EBU protocols.
Perhaps this is a partial explanation of why this course was chosen, since it was probably not anticipated that even these standards would come under attack by audiophile paranoia.
I think that the powers that be actually do regard the move to employ asynch USB for audio as being prompted by irrational desires on the part of a small section of consumers, and that it is a minority interest which will die out if not encouraged, and that even if this does not prove to be the case, keeping the subject obscure preserves the commercial interests of a section of manufacturing industry, in much the same way that automotive manufacturers require that you join the club (SAE) if you want access to the inside dope on CanBus.
Definitely not in the spirit of open source.
The basics of Can Bus are out there for the physical implementation, as they are for USB.
Where you do have to get a bit more involved and often sign NDA's is J1939 with can bus. A lot of the standard is available but as we live in a competitive world companies like to keep some information close to there chest, this is the same with a lot of interfaces, you can easily create the physical interconnections etc, but often have to have firmware/software to implement things and as large amounts of money are spent on developing such things you can realise why it is not all out there in the public domain.
Where you do have to get a bit more involved and often sign NDA's is J1939 with can bus. A lot of the standard is available but as we live in a competitive world companies like to keep some information close to there chest, this is the same with a lot of interfaces, you can easily create the physical interconnections etc, but often have to have firmware/software to implement things and as large amounts of money are spent on developing such things you can realise why it is not all out there in the public domain.
The Audio-Widget project is an open-source async USB audio interface. You can read more at www.qnktc.com or in the Audio Widget thread here.
Linux's USB audio driver supports async operation natively and the source code is out there for anyone to look at.
So, some async stuff is cloak and dagger but not all. You can have an open-source system from end to end if you want.
Linux's USB audio driver supports async operation natively and the source code is out there for anyone to look at.
So, some async stuff is cloak and dagger but not all.
You can have an open-source system from end to end if you want.- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.