• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Asynchronous USB is actually synchronous (Isochronous)

yummy

Active Member
Joined
Oct 21, 2017
Messages
142
Likes
188
I finished an article last year, took me two years to finish it, and it's gone (actually I lost the whole drive, just stopped working one day)
plus four or five interviews with more experienced buddies (interviews were not important), since I remember (trying) main part, I write it again

here make it short

title being a bit misleading, should be isochronous
USB has four modes, bulk, control, interrupt, isochronous
audio data doesn't allow "asynchronous", Asynchronous is for USB fingers and portable drives

what asynchronous is just the clock singal, not the data, data is always synchronous (should be isochronous)
Asynchronous means
acting as own clock source, that's pretty pretty old stuff, but marketing makes it wow we're new king, no

pretty pretty early usb input (guess 2010, 2011? actully USB input appeared a lot earlier but I don't remeber which year, I used firewire)
is synchronous

IEEE1394, we call it Firewire from here and Thunderbolt, are asynchronous since the beginning, and they never ever make those marketings (the Firewire guy who made and revised the standard for those year was forced to resign, he now becomes some kind of a soulless guy, you know, FW didn't survive in the wave of USB, audio isn't important while facing survival)
I would say Thunderbolt is totally dead for audio, new mobos are all thunderbolt 3, only few ADDA included Thunderbolt (but only one of them is famous, Lynx), and all are TB gen 2. (adapter? adapter is taboo in hifi world)

since they are all asynchronous
now they are at the same starting point (firewire's dead, thunderbolt's too high-pitched, you can't find computers with tb2 so it's dead too, and you gotta find tb cable, pretty expensive)

the obvious mediocrity of USB is that you can't get a sound with completed highs/mids/lows, some are ok highs, worsen lows and some (actually xmos) the highs always doesn't sound well, mid and lows pretty good. with newer driver did improved a bit, but highs still lacks a lag.
one of explanations is that the layout of USB in the mobo are just too long, like 10 meters in a single piece of small mobo (data going a long road from the drive to the usb port), audio frequency gets kinda "trump" at the high frequencies while running at 480 Mbit/s, and usb isn't the top priority in the OS system.
TB is basically PCI-E, FW needs to install a PCI-E card (you could install PCI-E cards for USB as well, I have one..), the "road" of FW might be shorter, should be one of the reasons.

after a long audition of USB, and switch to FW, the highs are back, trebles are back, can't go back to USB anymore, although USB is the fastest, great-easiest-connecty-est method right now. it can't sound too good.


(dante is different thing, more asynchronous than any asynchronous, network doesn't require clock signal at all
but you need a route/switch to connect it or you are gonna need a mobo with dual net ports, and it doesn't sound comparably good like the FW/TB)
 

andreasmaaan

Master Contributor
Forum Donor
Joined
Jun 19, 2018
Messages
6,652
Likes
9,399
I don't know enough about these various interfaces to understand the claims, but would like to understand how you suggest this would show up in measurements of USB audio devices?
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,522
Likes
37,050
Yes, there is so much wrong information in the OP it's asking a lot for someone to correct it.

Let us just say the OP is all kinds of wrong-headed. Would you like to ask some more pointed questions @yummy ?

Is this article from 2012 the source of confusion:
https://www.psaudio.com/pauls-posts/myth-asynchronous-dac/

If so the source of your confusion is Paul McGowan is saying you shouldn't call a DAC an Asynchronous DAC if it is asynchronous only on the USB input. Because it relies on external clocks for other inputs like SPDIF even if USB is Asynchronous. There are methods to buffer any of the inputs and have the clock be local only and not reliant on external clocks the way SPDIF normally is even on an SPDIF, Toslink or AES/EBU input. And those were and still are rare.

The middle part I'll skip as too messed up to address.

The part about Thunderbolt.......it looks like Thunderbolt will become a sort of common input on recording gear used by studios. It is expensive, but that is hardly an impediment to people who blow hundreds of dollars on a cable. There are interfaces that allow you to use Firewire over Thunderbolt with an inexpensive adapter cable. Thunderbolt is starting to show up on a number of Windows computers this year. Thunderbolt's real advantage is being able to handle more channels of sound (important to recording) and having nearly zero latency (again important to recording). Whether it makes it into DACs I don't know. My guess would be not. My guess is USB C if it finally becomes ubiquitous will be the new input most gear will be equipped with sometime in the near future.

The rest of your description about USB is simply and neatly refuted by measurements of USB driven gear at the analog outputs.
 

garbulky

Major Contributor
Joined
Feb 14, 2018
Messages
1,510
Likes
827
@yummy, you are not an English speaker right? I can read Cantonese, Chinese and a bit of Japanese, if your articles are in one of these languages can you post some links of them?
Not trying to be rude, but the impression I got was he was some sort of European.
 
Top Bottom