Hi I’ve my Bartok since its launch, and have been using its network and AES input most of the time.
Recently i try the USB 1 type B input, and find it can’t render MQA files / Tidal for all Roon device settings combination of MQA capability + MQA core decode options.
All combination of options will play music for ~1 sec then the audio device lost and reappear after a few sec, looks like lost control and then resume.
The only option i can get music playback is “no MQA support”, then Roon do the decoding itself and send the first unfolded music to Bartok to playback, and there is no more renderer unfold happening.
My Roon core is QNAP NAS, connect to Bartok USB1 type B by a typical A to B USB cable. By the way, I’ve a ifi iSilencer3 at the type A end, and a ifi iPurifier3 at the type B end, will they break the MQA signal flag?
Here is a suggestion. Take a look at the dCS web site. On the subject of MQA it says “Full decoding and rendering of MQA data from the Network and USB2 inputs. Final rendering of unfolded MQA data only from the other inputs.”. Apparently the input you are using is the issue.
Are you sure about this? My understanding is the first step - MQA unfolding - can only be performed by the network card, which would mean USB is not going to work. MQA rendering however works with any input - requires pre-unfolding which either Roon or TIDAL can do.
If I’m wrong can you point where in the dCS website this is stated?
Am I sure about this? Yes, if what I read on the dCS web site is accurate. Where does it say this? On the dCS web site under the specifications of the Bartok in the section entitled MQA. It’s right there in black and white. I am not an advocate for dCS. I just did five seconds of research to find this. I think it’s nuts or maybe a misprint but given that the OP can’t get it to work I guess I’'d have to say it’s true.
Notice it says “USB2” - a type-A connector to which you can connect an external hard drive - whereas the OP said USB1 - a type-B connector used to connect to a computer.
So what dCS is saying is the Bartok can unfold MQA files sourced from the network or a USB drive, but cannot do this over the USB interface to a computer. Both the network and USB2 drive inputs use the network card whereas the USB1 input does not and thus cannot unfold MQA, but can render it (all inputs can render).
So if connecting to a computer via USB, you should have Roon or TIDAL unfold the data before sending.
BTW, this is the same in both the Rossini (which I own) and the Vivaldi stack.
I assume you tried “Render Only” in the audio device setup? I think that should work fine, but I guess you’re getting the USB dropoff. I wonder if it is possible the issue is your QNAP’s USB interface.
Any chance you can try connecting to a laptop and stream to the laptop->DAC via USB?
I don’t think the silencers would mess up the data. But you should try removing them to see if they do.
Actually I just tried it. I connected the Rossini over USB to my mini (which is my Roon core), set it up as “Renderer Only”, and enabled “MQA Core Decoding”.
The Rossini will not light the MQA indicator, so I presume it is NOT rendering. But it doesn’t drop out, plays fine.
PS: Tried again based on a suggestion in the dCS forum (dcs.community). Reloading defaults (and then changing the Volume to Device Volume rather than max) allowed it to render MQA properly.
I own a Bartok and John is right about it. I did email dCS and they confirmed. A strange design choice. I use the Network input so it doesn’t bother me.
It is not strange - the first unfold requires some compute power and dCS decided to use the network card for that.
On the flipside, dCS MQA implementation was one that MQA Co. took very seriously and worked diligently to get it right. So it’s a top-notch implementation.
Additionally, the first unfold is a standard - same regardless of the platform or taget DAC - so it can really be performed by any capable device. Rendering is more for DAC-specific.
So the first unfold processing by the network card is shared with the USB2 (USB A) input. Since USB1 (USB B) input is more common for most USB DACs, why didn’t they use USB1 instead of USB2? That’s what I found strange.
I agree that the dCS Ring DAC as well as all other implementation is brilliant. I enjoy the Bartok very much, playing musical sound.
Because USB1 does not go to the network card - it feeds the DAC directly. There’s no way to reroute the data through the network card.
Having said this, consider that the first unfold is common to all implementations, it is the rendering that is DAC-dependant, and rendering does work on USB1. So you can unfold in any program (Roon, Audirvana, Tidal, possibly others) and feed the unfolded and tagged stream over USB1 and have the DAC render.
Isn’t the cabling weird in this case? My streamer output is a Type A connector. The input into my DAC is Type B. In the dCS case I would need a Type A to Type A USB cable. Don’t see those too often.
Like I said above, USB1 (type B connector) is used EXCLUSIVELY for connecting to a computer or streamer. USB2 (type A connector) is used EXCLUSIVELY to attach an external hard drive and use dCS Mosaic to browse and play music.
You CANNOT attach a computer to the USB2 port. Clear?
Sorry for late update, I have figure out the unplug and plug into the USB input again and the problem is gone, even with the ifi iPurifier3 in the chain.
I guess the MQA capability detection by Roon core on the USB end may have goes wrong, due to my previous plug in sequence.