Giving Audirvana a try

Hello,
I have been a happy user of JRiver Media Center on Windows for more than 10 years, using mainly XMOS-based USB DACs (EmmLabs, Meitner, …).

Recently, I subscribed to Qobuz, which is fantastic! However, JRiver does not let you browse (or even connect to) Qobuz directly. My workaround has been to use BubbleUPnP server on the Windows computer and the BubbleUPnP remote app on Android to browse Qobuz and use JRiver as the renderer. This is not an optimal setup (app not always stable) and I need to switch remote apps when switching between Qobuz and my own library.

I happen to own many ripped SACDs in iso format, which play perfectly fine in JRiver. I thought I’d give Audirvana a try, since it claims to support SACD iso (Roon does not), as well as Qobuz.

Unfortunately, Audirvana takes literally minutes to start playing an SACD iso, stops in the middle of a track, or simply crashes.

I therefore extracted all stereo tracks from my iso images to dsf format (note that Roon now becomes a candidate, since playing SACD iso is not a requirement anymore :-). Audirvana can play the dsf tracks perfectly and even gapeless when using Wasapi (Windows 64-bit), but not when using ASIO. There is a half second interruption between each track when using ASIO.

I guess I could live with Wasapi (some claim ASIO sounds better), but the sample rate is limited to 192Khz and DSD64 (DSD over PCM), whereas my DAC definitely supports 384Khz and DSD128. JRiver has no problem playing these higher rates via Wasapi without downsampling.

I have tried several XMOS drivers, including a rather recent 4.13 version, without any difference. Note that Audirvana sees the correct higher sample rates when using the ASIO driver.

So here are my questions:

  1. Why is Audirvana not supporting SACD iso, while its published specs claim it does?

  2. Why are the sample rates artificially limited when playing via Wasapi? Is this a Microsoft Windows limitation? If so, how is JRiver able to bypass these limitations? Is there a way to convince Audirvana that the higher sample rates are actually supported?

  3. Why is the ASIO driver not playing gapless? Is this an issue with the driver (XMOS) or with Audirvana? I hear a faint click on JRiver, but not a large gap as on Audirvana.

Do not get me wrong, I love Audirvana so far. I think it actually sounds better than JRiver. But I wish the issues described above did not exist.

Thank you!

Here is my debug info, in case this helps (with Wasapi selected):

Audirvana 3.5.42

SIGNAL PROCESSING:

Polarity Inversion:
	Globally: OFF
	Per track: ON
Effects plugins NOT ACTIVE

UPSAMPLING:
SoX not in use
SoX filter parameters
Bandwidth at 0dB = 99.5
Filter max length = 30000
Anti-aliasing = 100
Phase = 66

AUDIO VOLUME:
Max allowed volume: 100
Replay Gain: None
SW volume control: OFF

LIBRARY SETTINGS:
Sync list: 4 folders
AUTO: \Obelix\Music\CDs
AUTO: \Obelix\Music\HRx
AUTO: \Obelix\Music\HiRes
AUTO: \Obelix\Music\SACDs_dsf
Library database path: C:\Users\xxx\AppData\Local\Audirvana\AudirvanaPlus\AudirvanaPlusDatabaseV2.sqlite

ACTIVE STREAMING SERVICES
Qobuz: Connected as Qobuz Studio

=================== AUDIO DEVICE ========================

Max. memory for audio buffers: 2432MB

Local Audio Engine:WASAPI

Preferred device:
DAC2X (XMOS XS1-U8 MFA (ST))
Model UID:DAC2X (XMOS XS1-U8 MFA (ST))
UID:\?\SWD#MMDEVAPI#{0.0.0.00000000}.{efd91921-07e7-4fdb-978a-0ede7ce3a613}#{e6327cad-dcec-4949-ae8a-991e976a79d2}

Active Sample Rate: 44.1kHz

Bridge settings:
Sample rate limitation: 384kHz
Sample rate switching latency: none
Limit bitdepth to 24bit: OFF
Mute during sample rate change: OFF

Selected device:
Local audio device
DAC2X (XMOS XS1-U8 MFA (ST)) Manufacturer:
Model UID:DAC2X (XMOS XS1-U8 MFA (ST)) UID:\?\SWD#MMDEVAPI#{0.0.0.00000000}.{efd91921-07e7-4fdb-978a-0ede7ce3a613}#{e6327cad-dcec-4949-ae8a-991e976a79d2}

6 available sample rates up to 192000Hz
44100
48000
88200
96000
176400
192000
Volume Control
Physical: Yes
Virtual: No
MQA capability
Auto-detect MQA devices: Yes
Not a MQA device, user set to not MQA
DSD capability: DSD via PCM 1.1
Device audio channels
Preferred stereo channels L:0 R:1
Channel bitmap: Ox3, layout:
Channel 0 mapped to 0
Channel 1 mapped to 1

Audio channels in use
Number of channels: 2
Use as stereo device only: No
Simple stereo device: No

1 output streams:
Number of active channels: 2, in 1 stream(s)
Channel #0 :Stream 0 channel 0
Channel #1 :Stream 0 channel 1
2 ch Integer PCM 16bit little endian 44.1kHz
2 ch Integer PCM 24bit little endian aligned high in 32bit chunk 44.1kHz
2 ch Integer PCM 16bit little endian 48kHz
2 ch Integer PCM 24bit little endian aligned high in 32bit chunk 48kHz
2 ch Integer PCM 16bit little endian 88.2kHz
2 ch Integer PCM 24bit little endian aligned high in 32bit chunk 88.2kHz
2 ch Integer PCM 16bit little endian 96kHz
2 ch Integer PCM 24bit little endian aligned high in 32bit chunk 96kHz
2 ch Integer PCM 16bit little endian 176.4kHz
2 ch Integer PCM 24bit little endian aligned high in 32bit chunk 176.4kHz
2 ch Integer PCM 16bit little endian 192kHz
2 ch Integer PCM 24bit little endian aligned high in 32bit chunk 192kHz

Local devices found : 2
Device #0: Digital Audio (S/PDIF) (High Definition Audio Device) Manufacturer: Model UID: Digital Audio (S/PDIF) (High Definition Audio Device) UID: \?\SWD#MMDEVAPI#{0.0.0.00000000}.{a61bf125-a053-4204-b627-25a5d3ca5540}#{e6327cad-dcec-4949-ae8a-991e976a79d2}
Device #1: DAC2X (XMOS XS1-U8 MFA (ST)) Manufacturer: Model UID: DAC2X (XMOS XS1-U8 MFA (ST)) UID: \?\SWD#MMDEVAPI#{0.0.0.00000000}.{efd91921-07e7-4fdb-978a-0ede7ce3a613}#{e6327cad-dcec-4949-ae8a-991e976a79d2}

UPnP devices found : 1
Device #0[AV] Samsung Soundbar MS750 UID: uuid:c21e5fdf-a431-4025-a4a5-85ac26c6f79d Location: http://x.x.x.x:9197/dmr

Audirvana do support iso files, what matters is how the file has been created. Since you have crash when you try to play it, can you send it to us at support@audirvana.com (if the file is more than 20mb, please let me know by mail) so we can take a look at it?

The DSD over PCM standard has been developed to send raw DSD in a fake PCM stream.
To identify it as such, there are 8 bit markers in it.
And the presence of these markers doubles the need of bandwidth, thus explaining why on a mac only half of the max DSD rate is available.

It should be gapless if the driver has been properly created, the fact that you hear a click in JRiver and a gap in Audirvana indicate that it could be a driver issue.

Damien,
Thank you for your prompt reply. I messaged you privately about providing an SACD iso to debug.

The DSD over PCM standard has been developed to send raw DSD in a fake PCM stream.
To identify it as such, there are 8 bit markers in it.
And the presence of these markers doubles the need of bandwidth, thus explaining why on a mac only half of the max DSD rate is available.

I am not familiar with MacOS and its support for USB audio. I use Audirvana on Windows, but the DoP (DSD over PCM) implementation should be identical on Windows, MacOS, or Linux. My understanding is that the 8-bit marker in DoP increases the bit depth from 16 bits to 24 bits, i.e. a bandwidth increase of 50%, but it does not require a higher sampling rate. Usually, 16-bit, 24-bit, and even 32-bit PCM words are usually supported via USB.

Stereo DSD64 can be transmitted with a 2-channel 24-bit PCM at 176.4Khz, and DSD128 with 24-bit PCM at 352.8Khz (also called DXD).

My DAC supports DXD, as you can see in the debug info below, when switching to the ASIO driver.
I just do not understand why Audirvana does not support the higher sampling rate in Wasapi (JRiver does). See my first posting in this thread for the debug info with Wasapi selected.

Regarding gapless playing when using ASIO, I agree it could be an issue with the driver. Hopefully, we can resolve the Wasapi issue so I can forget about ASIO.

Thanks again!

=================== AUDIO DEVICE ========================

Max. memory for audio buffers: 2456MB

Local Audio Engine:ASIO 2 driver version 1037

Preferred device:
XMOS USB Audio 2.0 ST 309B
Model UID:XMOS USB Audio 2.0 ST 309B
UID:XMOS USB Audio 2.0 ST 309B

Active Sample Rate: 44.1kHz

Bridge settings:
Sample rate limitation: 384kHz
Sample rate switching latency: none
Limit bitdepth to 24bit: OFF
Mute during sample rate change: OFF

Selected device:
Local audio device
XMOS USB Audio 2.0 ST 309B Manufacturer:
Model UID:XMOS USB Audio 2.0 ST 309B UID:XMOS USB Audio 2.0 ST 309B

8 available sample rates up to 384000Hz
44100
48000
88200
96000
176400
192000
352800
384000
Volume Control
Physical: No
Virtual: No
MQA capability
Auto-detect MQA devices: Yes
Not a MQA device, user set to not MQA
DSD capability: DSD via PCM 1.1
Device audio channels
Preferred stereo channels L:0 R:1
Channel bitmap: Ox3, layout:
Channel 0 mapped to 0
Channel 1 mapped to 1

Audio channels in use
Number of channels: 2
Use as stereo device only: No
Simple stereo device: Yes

1 output streams:
Number of active channels: 2, in 1 stream(s)
Channel #0 :Stream 0 channel 0
Channel #1 :Stream 0 channel 1
2 ch Integer PCM 32bit little endian 44.1kHz
2 ch Integer PCM 32bit little endian 48kHz
2 ch Integer PCM 32bit little endian 88.2kHz
2 ch Integer PCM 32bit little endian 96kHz
2 ch Integer PCM 32bit little endian 176.4kHz
2 ch Integer PCM 32bit little endian 192kHz
2 ch Integer PCM 32bit little endian 352.8kHz
2 ch Integer PCM 32bit little endian 384kHz

Local devices found : 2
Device #0: JRiver Media Center 27 Manufacturer: Model UID: JRiver Media Center 27 UID: JRiver Media Center 27
Device #1: XMOS USB Audio 2.0 ST 309B Manufacturer: Model UID: XMOS USB Audio 2.0 ST 309B UID: XMOS USB Audio 2.0 ST 309B

Actually if you want to play DSD files on MacOS you will need DoP. For Windows 10 it’s a bit different as ASIO drivers has been made to be able to decode DSD streaming natively. If you play a DSD track in WASAPI, you will need to enable DoP (if you don’t have a dedicated WASAPI driver). It doesn’t rely on the capability of your DAC, it rely on the operating system architecture.

DoP is enabled in Audirvana on Windows, using Wasapi. Once again, the problem is that Audirvana does not allow higher sample rates supported by my DAC, whereas JRiver has no issues. I’ll try Roon next.

I wished Audirvana was letting you try any sampling rates with Wasapi, rather than guessing the wrong maximum rates.

I have not tried Roon yet, but this Roon support page (https://help.roonlabs.com/portal/en/kb/articles/audio-on-windows) points to the exact problem I am encountering with Audirvana.

In other words, Audirvana trusts what the driver (wrongly) reports as its capabilities. A way to override and manually set sampling rates, as I suggested before, would be awesome.

Thanks!

Alright, I took advantage of Roon’s Black Friday offer ($1 for 3 months) and I installed Roon Core on my Windows 10 PC (Intel i7-4770S CPU @ 3.10GHz, fanless, no hard drive, 8GB RAM) connected via USB to my Emm Labs DAC2X. As expected, Roon let me override the sampling rates reported by the Wasapi XMOS driver and I was able to play DSD128 without issue. ASIO plays fine too, without a gap between DSF tracks, as experienced using Audirvana and the same ASIO driver.

However, the most interesting result is that Roon does not sound as good as Audirvana! I know that sceptics out there will claim that bits are bits and that an audible difference is not possible. I disagree. My girlfriend hears the same thing as me. I would say that Audirvana is best, JRiver second, and Roon third.

Now I really wish Audirvana would allow me to override sampling rates…

Cheers!

You should be able to override this. Do you see the DSD section in the DAC settings?

Here is the screenshot from the settings on my system. My DAC is not even DSD capable.

Thank you for your suggestion, but I do not want to apply any sample rate or format conversion. I want to send DSD128 over PCM without conversion (bit perfect), which requires a PCM sampling rate of 352.8Khz/24-bit, encapsulating the DSD data.

My DAC supports 352.8Khz/24-bit via Wasapi (and ASIO), but Audirvana thinks it does not, most probably because the driver does not report it correctly to Windows. This is a common problem addressed both by JRiver and Roon, which both allow the user to select sampling rates not reported by the driver.

Just select DSD over PCM 1.1 and try.

Please, read the whole thread.
DSD over PCM 1.1 is selected, but since Audirvana thinks DSD128 is not supported in Wasapi, it converts DSD128 to PCM at 176.4 Khz, not what I want. It works using ASIO.

Once again, what is missing in Audirvana is a way to override the reported capabilities in Wasapi.
See the link above to the Roon article explaining the feature, which also exists in JRiver, but not (yet) in Audirvana.

Quick update that may be useful to Windows users and owners of XMOS-based DACs:

The limited sampling rates that Audirvana uses (and which cannot be overriden as they can in Roon or JRiver) are reported by the drivers made available by the DAC manufacturers. I experimented with several XMOS drivers provided by different manufacturers (Gustard, EmmLabs, Meitner, which all repackage Thesycon drivers). None of them reported the correct maximum sample rates. However, at some point during my driver shuffling, no manufacturer driver was compatible and a Microsoft XMOS driver got automatically loaded. To my surprise, the correct sample rates showed up in Audirvana under Wasapi !!

I am not sure when Microsoft started to provide XMOS drivers for USB Audio 2.0. My version is Windows 10 Pro 20H2.

So this problem is solved for me. The key is to uninstall the manufacturer provided driver and let Windows install its own. Of course, you’ll lose the ASIO driver doing so, but at least the Wasapi driver will work properly with all correct sampling rates under Audirvana.

Cheers!