1 Topic by ozark2001 (edited by ozark2001 2023-12-15 10:34:41)

  • Registered: 2022-03-06
  • Posts: 10

Topic: AVB to other digital formats lossy?

Hi,

I was thinking if there is any loss/difference when a signal is sent via AVB from e.g. RME Digiface AVB to an RME M1610 or AVB Tool and then "converted" to any other format like Madi, ADAT or AES to send the signal digitally further, of course staying at same bitrate and Samplerate of e.g. 96khz/24bit throughout?

Like should a signal nulltest if just passed through a Digiface AVB to AVB Tool via AVB, then from AVB Tool to M1610 via Madi, then from M1610 back to Digiface via AVB and back to computer?

Or are those digital format conversions like in the above example always somewhat lossy?

  • Max
  • Max
  • Moderator
  • Offline
  • Registered: 2007-05-09
  • Posts: 316

Re: AVB to other digital formats lossy?

Hi Ozark,
if clocking is configured correctly, those signals pass fully transparent with full 24bit resolution.
Max

3 Reply by ozark2001 (edited by ozark2001 2023-12-15 11:19:20)

  • Registered: 2022-03-06
  • Posts: 10

Re: AVB to other digital formats lossy?

Good to know!

Is this true because digital format conversions in general are fully transparent  so only clocking or bad cables can cause signal issues?

Is that also true for any conversions between the commonly known digital formats like Adat, AES3, Spdif, Madi and AVB/Dante or just true for above example formats because of how RME specifically handles these conversions in their own line up/ with the mentioned devices?

Just curious how digital format conversions are generally done if implemented/offered within certain devices.

  • From: Germany
  • Registered: 2007-04-10
  • Posts: 9,784

Re: AVB to other digital formats lossy?

There's no real conversion as such. The audio format remains the same, e.g. 24 bit 48kHz LPCM. MADI and ADAT are not actually different audio formats, they are just the carrier/container, as are AVB and other network protocols. The audio content as such is no different.

Regards
Daniel Fuchs
RME

RME Support's Website

  • Registered: 2022-03-06
  • Posts: 10

Re: AVB to other digital formats lossy?

Thanks Daniel, I did not know that!

So the only potential source of altering the digital signal would be jitter, caused by bad clocking or cables?


With regards to the Clocking:
For a null test, to find out if cables and clock are fine throughout a digital chain, is creating any direct "loop shortcut" by passing a digital out connection directly back to its digital in on the same device okay? Or would that create an immediate clock issue, since the device would be "synced to itself"?

As examples, would looping a signal through the AES output of e.g. an UFX+ directly back to its AES input via a short 110ohm digital AES cable be expected fully transparent?
Same idea for connecting a stream of ADAT out directly to the ADAT in on the same device?

  • From: Germany
  • Registered: 2007-04-10
  • Posts: 9,784

Re: AVB to other digital formats lossy?

Jitter affects AD and DA conversion, but not digital signal (transfer).

Incorrect clocking will usually produce audible clicks.
You can loop an AES signal, but the interface or card would have to be in master mode and should not try to sync to itself.

Regards
Daniel Fuchs
RME

RME Support's Website

  • Registered: 2022-03-06
  • Posts: 10

Re: AVB to other digital formats lossy?

Thanks for clearing that up with the jitter!

RME Support wrote:

You can loop an AES signal, but the interface or card would have to be in master mode and should not try to sync to itself.

Would an external word clock also be fine for the device with the shortcut loop?

My idea is to use this chain for a nulltest:

PC -> (USB) -> Digiface AVB -> (AVB) -> AVB Tool -> (Madi Coax) -> Ufx+ -> (Adat1 out) directly connected back to (Adat1 in) -> Ufx+ -> (Madi Coax) -> AVB Tool -> (AVB) -> Digiface -> (USB) -> PC