Topic: Buffer Issues over ASIO

I have mostly resolved my ASIO enigma by utilizing ASIO Bridge from VB-Audio, however, when in TIDAL exclusive mode since the swap over to FLAC files at 24/192 there is an awful buffer overrun that doesn't populate as errors in the MADIface USB console nor in the ASIO Bridge buffer (flashing red). This sometimes occurs with the first unfold MQA rate of 88.9? requiring me to pause playback, wait a few seconds, then resume... many times over to the point I'm fed up with exclusive mode and remove the option... thus defaulting all sample rates to windows resampling at 24/192 defeating the purpose of the platform.

I have the ASIO bridge set to an enormous buffer (max) but when MADIface is the output, it never goes over 8192 and we can not select a higher rate than the selectable 192 (i.e. 32768 @ 768Khz) when the Bridge is the system output allowing for clock to match the source automatically.

Is the work-around setting MADIface as the system output and jacking it up to 768 default (windows output is 24/192 resampling everything to match windows iTunes output) and allow exclusive to take control of the clock and tricking the buffer to max latency?

Is there another ASIO bridge you would recommend for use with Tidal Exclusive mode?
I'm running all this on my new Thinkpad X1 i7 32GB via Lightning 4 port to USB B. Heck, when my old Yoga 3 wanted to work with windows 10 it worked flawlessly. I have the latest MADIface and ADI2 drivers/firmware pushed (re-did them all last night)

I have no issues playing non ASIO files at 352KHZ, 758KHZ, or DSD 256 over Foobar.

Re: Buffer Issues over ASIO

I'll check back next decade.

3 (edited by KaiS 2024-02-15 05:56:09)

Re: Buffer Issues over ASIO

Four2One wrote:

I'll check back next decade.

Sorry if you feel left alone.

As you found out, the config seems to have a special quirk that is obviously related to something with your Thinkpad, as it works with your older laptop.

With Lightning you probably mean Thunderbolt?
My only tip: check and switch off the port’s power saving.

4 (edited by Four2One 2024-02-15 20:13:31)

Re: Buffer Issues over ASIO

KaiS wrote:
Four2One wrote:

I'll check back next decade.

Sorry if you feel left alone.

As you found out, the config seems to have a special quirk that is obviously related to something with your Thinkpad, as it works with your older laptop.

With Lightning you probably mean Thunderbolt?
My only tip: check and switch off the port’s power saving.

I'm running performance power profile on wall power. Only battery operation will cause power saving concerns.
The reason it didn't happen before was that Tidal's "MAX" did not support FLAC 24/192, they were beholden to MQA rates and without hardware "unfolding" that max rate was 88.2 with the software 1st "unfold".

The issue seems to be the MADIface utility locking the buffer size concurrent with the Windows default 24/192 regardless of "exclusive" mode driving the ASIO bridge with switching rates 44.1/48/88.2/96/192 16-24 bit depths. As I said, the ASIO bridge latency locks to the MADIface maximum regardless of the potential maximium far exceeding that of the defaulted to buffer size. I understand RME's driver automatically selects buffer size based on playback rate for optimal latency, however, with music streaming this is not an issue.

4096 is the default cap for 88.2, while 8192 appears to be the cap for 192 but not adequate with this passthrough.

Is there a way to "trick" MADIface into a higher bitrate buffer cap? i.e. set my windows rate at 364 or 768 to increase the RME buffer size?

EDIT: I just attempted to trick it... This only works with Apple Music/Spotify/Youtube aka system playback... It will increase buffer to accommodate 384 KHZ/16384 buffer size. However, as soon as you go to Tidal for exclusive ASIO control of the rate it returns to the inadequate buffer size for 24/192khz @ 8192 resulting in unlistenable chopped up playback.

5

Re: Buffer Issues over ASIO

8192 is the highest buffer size within the hardware. And that's why it can't be made bigger.

Regards
Matthias Carstens
RME

Re: Buffer Issues over ASIO

Understood. Are there any other ASIO bridge software solutions you can recommend I may try?

7

Re: Buffer Issues over ASIO

Sorry, no. Maybe someone else knows.

Regards
Matthias Carstens
RME