Audiolab 9000N

Do you plan to integrate Audirvana with Audiolab 9000N. I am already an Audirvana user . I bought Audiolab 9000N and I do not want to spend for Roon …Thank you in advance…

Hello @Leventgergin, from the specs page I see that this device implements the UPnP protocol, it should work with Audirvana afaiu. Is it connected to the same network as the device which runs audirvana?

Hi @GioF71 …I use Audirvana over wifi but at the screen of Audiolab 9000N format is displayed as WAV.I think program is converting it. When I (free) tried Roon. The original format is kept.(as Flac or MQA.) I assume that those programs are running in different ways…

Hello, I am just making hyphotesis now: maybe the device does not decode flac, but this looks quite strange in 2024, I would not bet on that. Or for some reasons Audirvana decided to transcode to wav. If the sampling rate and bit depth are as expected, this should not be an issue.
About Roon, according to this old thread, it looks like Roon streams pcm (so uncompressed streams). The fact that the display shows flac might merely depend on metadata.

Just my 2c, but of course maybe other audiolab owners and/or people from Audirvana might be able to clarify.

1 Like

Not an expert on this but my understanding is as follows:

Think this issue is one of the 9000 not reading the metadata sent to the device correctly. Audirvana takes a FLAC or ALAC file or whatever. Extracts the pcm file within the FLAC/ALAC container, decompresses it and sends it as a WAV file to the streamer over UPnP, it also sends the metadata extracted from the FLAC/ALAC container for the endpoint to display the original file data. Roon does something similar, extracts and uncompresses the file and sends it as an “uncompressed PCM file” to the streamer via its own proprietary RAAT transfer protocol. The exact details are the companies respective IP so they are never going to lay bare all the details of the process.

Both systems do this to reduce the processing done by the endpoint and therefore “improve SQ”

Have always believed somewhere in the different software manipulation of this process is the reason I hear a difference in SQ between AS and Roon. But again I am coming from a layman’s perspective.

Obviously DSD files are handled differently.

1 Like

In other words, it works?

Maybe according to the manufacturer but most display the original source file and metadata not what is actually received, but yes you could say it works.

1 Like

Just to visually show this. A copy of the debug info whilst playing a 24/44.1 FLAC file to a pair of KEF speakers. As you can see the file transmitted is an uncompressed wav file. Below this, the file meta data.

1 Like

Thank you for your response…Today I tried something different and changed setting : Forced upsampling type to DSD128 (Audirvana) and I surprised how the SQ changed…I will keep listening with this new setting… (at the display of 9000N is DSD128) I also tried force to max upsampling which 768/32 was good enough…

Been through that experiment myself. Whilst subtly different I had no preference one way or the other so decided to leave upsampling off to reduce the size of the files sent to the KEF’s as I feel that is possibly a route to limiting the risk of transmission glitches. The less data sent over my part wired/part wireless network the better I think… I do like the sound from the R8 Brain upsampling algorithm however.

1 Like

I see…now it perfectly make sense why wav is displayed at 9000n

Yes, it does. When using a Network player (UPNP) Audirvana converts the format to wav 32bit (for pcm only of course)

The 9000N is PCM-centric… It uses the ESS9038Pro DAC chipset which does not have a pure 1-bit PDM (DSD) path to the output circuit… It converts 1-bit PDM to multi-bit PCM (DSD-Wide) for output… I suggest using a “Custom” upsampling scheme to convert LPCM 16/24/44.1kHz through 24/192kHz,with a maximum sample-rate setting of 705.6kHz and 768kHz respectfully, and play PCM 352.8kHz, 384kHz, 705.6kHz and 768kHz in the native sample-rate… Audirvana will play DSD files at the native sample-rate, however these are converted to multi-bit PCM in the ESS chipset. You will never get an unfettered 1-bit DSD output signal…

A custom upsampling scheme example:
44.14kHz → 705.6kHz
48kHz → 768kHz
88.2kHz → 705.6kHz
96kHz → 768kHz
176.4kHz → 705.6kHz
192kHz → 768kHz
352.8kHz → 352.8kHz
384kHz → 384kHz

:notes: :eye: :headphones: :eye: :notes:

1 Like