This is a repost of the following old thread:
I recently moved from Audirvana+ to Audirvana Studio.
The issue is still the same as before - when using upscaling, 1-2 seconds of sound is trucated from the end of the track. This does NOT happen with native DSD tracks, so it is clearly not the fault of the DAC.
I was hoping this was solved in the recent versions but sadly it seems to be a permanent issue. Am I the only one experiencing this?
Using a 2012 Quad Core Mac Mini.
Are these files purchased or being streamed from a legitimate music streaming service subscription or ripped from your personal CD collection?
Are you running the Mac Mini headless?
Have you tried increasing the playback buffer RAM to 1GB?
Note that I moved the exact same Mac Mini to Windows 10 with Boot Camp, and running Audirvana Studio on Windows with the same hardware works well. So this is specifically a MacOS issue, either with the app or with the Direct Mode kernel.
Please post (copy and paste) your debug info here so folks can help as best possible… This is found in the menu bar at the top left of the Audiirvana UI…
There is an issue with running a Mac Mini headless where the graphics sub-system is not being engaged and the CPU handles all of the UI graphics rendering and other activities associated to the GPU and in concert with all system level activities… These devices have helped Audirvana users here on this forum…
Mac and PC computers can pack phenomenal power into a compact form factor, but when they’re operated without a display attached, the GPU isn’t activated — meaning it’s not pulling its weight. So when you access your Mac or PC remotely, the CPU does all the work in rendering the user interface. As a result, the interface lags; OS animation, cursor movement, menu navigation, and typing all suffer, making remote operations a pain.
https://www.newertech.com/products/hdmi_headless_video_adapter.php
So, this speaks to the reality of a 4GB system memory… I believe that MacOS requires approximately 4GB to run efficiently… the limited memory forces the CPU to use virtual storage memory available on the HDD or SDD, this in concert with running headless, is taxing the CPU clock-cycles, along with the overhead of up-sampling/modulation of PCMxxx to DSDxxx in Audirvana, MacOS + Audirvana requires DSDxxx to be transferred to the output bus controller/interface via DoP protocol (DSD over PCM) and this is CPU intensive and memory intensive… You have to ask yourself why users like myself, that up-convert/modulate all PCMxxx to DSD128 in Audirvana via r8Brain on MacOS, are not experiencing the behavior you are exhibiting.
My system:
Mine works flawlessly too,
Audirvana Origin 2.4
MacOS 10.15.7
2012 Macbook Pro, 2.5 GHz Dual-Core Intel Core i5 16GB memory
USB - DAC, no streaming.
I have the same problem with Audirvana 3.5.50 (final version) on my 2020 M1 mac mini with 16 GB. It happens with my own ripped CD library and with Qobuz. I recently upgraded my DAC to a Denafrips but the problem persists. I have had SysOptimizer turned off for about the past 3 years because it always crashes. MQA is off. Native DSD streaming method is set to DSD over PCM 1.1.
Apparently not many of us experience this issue but it is real and is very frustrating. It even affects gapless playback, sometimes causing a short glitch between songs. It sounds like current Audirvana software can have the same problem. So if no one has any other ideas I might start looking for something else.
I may have solved this problem! I turned off oversampling to see how my new DAC sounded when doing the oversampling on its own. Surprise! Since then there has not been a single truncated song or a messed up gapless transition.
It makes no sense to me that oversampling in Audirvana 3.5 would truncate songs but here we are. Turn it off. Enjoy the music.
This topic was automatically closed 90 days after the last reply. New replies are no longer allowed.
