Some Little Truths about AES/EBU. - Hi-Fi Stereo & Home Cinema forums | ATC Forums

Announcement

Collapse
No announcement yet.

Some Little Truths about AES/EBU.

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Some Little Truths about AES/EBU.

    Hi all,


    Have you wondered why there are some many types of digital audio transfer benchmarks?

    Anyways, so it's said that AES/EBU(AES3) is the the most common alternative to the S/PDIF, and typically used to transmit PCM and Dolby Digital 5.1, the cable is formed by 110 ohm shielded twisted pair (STP) cable with XLR connectors up to a distance of 100 meters, and it's got longer life span .


    Fun fact, most AES3 interfaces are self-clocking, they carry an embedded clock from the source. that being said, a receiving device can be synchronized just to the embedded clock from the source device in a simple audio system.

    A DAC could automatically synchronize itself to the incoming signal; and so does a CD recorder when it's recording, some might use an external World Clock to as a master to slave other devices to achieve better clock synchronization, which effectively lowers the jitter effect and levels up the digital audio quality, per se.


    It serves 2-channels of 24bit/192k audio, but it also has the capability of carrying 4-channels of 24bit or 20bit/96k digital audio(I would't say optimally), and those may be related "stereo" channels or independent mono channels or 1-stereo plus 2 mono, however, when it comes to " Sound Quality ", it's somewhat personal & subjective haha..., so it's not really my place to conclude which digital audio format can be the best fit to an audio system, still all depends on the setup and the preference of one's ears.


    Hope my understandings are correct, please do share anything below if interested, I'd like to know more about the digital audio stuff for sure!


    Best,

  • #2
    Yes that's all OK AIUI.

    The AES/EBU (balanced) interface to a DAC is technically very similar to SPDIF (coaxial) and TOSLINK (optical) - they primarily just use different connectors. I think SPDIF was adapted for consumer use from professional AES/EBU and TOSLINK followed from SPDIF.

    In normal consumer DACs with these interfaces the DAC has to extract a clock from the signal to do the D/A conversion. This clock must have low jitter to avoid adding noise to the audio. This needs a rather good clock extractor to ensure the jitter is low enough to be audibly blameless. Lots of audio commentators were unhappy about this when DACs got separated (long ago) from CD players. However, good receiver integrated circuits have been available for about 20 years (but some DAC makers still seem to do it themselves and not always well).

    In audio studios with lots of DACs that must work together, a common clock gets distributed around the studio (a 10 MHz "word clock" I think) to make sure the audio outputs are all synchronized. For most home audio systems this expense is not needed.

    The I2S interface was invented as a means to transfer data into a DAC chip inside a product but seems to have become yet another external interface today. I have a copy of the specification but no experience of it. However, to act as an external interface, some common technical choices have to be made at independently designed transmitter and receiver ends, and I don't know how this is done to make the interface interoperable.

    The USB interface to a DAC can work in three modes: synchronous (generally not used in DACs); adaptive (used in early USB DACs); and asynchronous (used in modern USB DACs). The asynchronous version fetches data to convert in advance of needing it, stores it in a buffer, and has its own independent low jitter internal clock to do the conversion. To me that's the right way to get a clean DAC clock. ATC's CDA2 DAC has TOSLINK and SPDIF interfaces and also asynchronous USB using an Amanero interface board. Here's a photo of three of Amanero's USB boards:


    The two dark silver components on the left sides are the low noise crystal clocks for the 44.1xN kHz and 48xN kHz sample rates, and the interface delivers the data to the DAC chip over the I2S interface at the top (the double row of unconnected connections) - just as I2S was intended - internal to the product.

    Comment

    Related Topics

    Collapse

    Topics Statistics Last Post
    Started by Tobes, 09-11-2021, 08:20 AM
    6 responses
    0 views
    0 likes
    Last Post Tobes
    by Tobes
     
    Started by Garfield, 07-15-2020, 07:19 PM
    8 responses
    0 views
    1 like
    Last Post Garfield  
    Started by Papa Zita, 07-30-2020, 08:54 PM
    2 responses
    0 views
    0 likes
    Last Post fr.jazbec  
    Started by SKY1950, 01-08-2022, 06:44 PM
    2 responses
    0 views
    0 likes
    Last Post SKY1950
    by SKY1950
     
    Started by Aretos, 07-11-2020, 07:05 PM
    15 responses
    0 views
    3 likes
    Last Post Nopiano
    by Nopiano
     
    Working...
    X