HQ Player

[jbparrish]

2 hours ago, ericuco said:

Are you sure your DAC supports 48k rates?

 

I think it does, but I tried 48K DSD unchecked, and still no change.

[Yviena]

@Miska is it possible to prevent HQPlayer from crashing when doing custom cpu core affinites when starting playback something with  a console OMP window pops up, i can still only get my 5900x to play EC7 DSD256 this way. 

[Yviena]

Is phase  in the matrix pipeline supposed to be identical to power, or different currently i have this in the response plot?

 

[chips666]

11 hours ago, jbparrish said:

Oops, my apologies, I grabbed the preferences screen shot when the DAC was off. Here it is again. I have not emailed Piero at audiolinux. The NUC/NAA and DAC work very nicely, actually, just not ASDM7EC.

 

I think multicore DSP grayed will do

Enjoy...

;)

[Miska]

10 hours ago, Yviena said:

Is phase  in the matrix pipeline supposed to be identical to power, or different currently i have this in the response plot?

 

No, it is not supposed to be identical. Magnitude and phase responses are separate. How the responses look like depend on the filter(s) in question.

 

[Crom]

Hi Jussi,

Thanks as ever for a fabulous product. Not a priority but would you consider making a small adjustment to the convolution ui in the web front end for HQ Player embedded? Could you make the "set" (see below) more prominent as it catches me out everytime....all I see is the No files selected. Perhaps remove the "No files selected" green text when a channel is set or put "set" in bold+ some other colour?

 

Many thanks,
Crom

 

[Miska]

1 hour ago, Crom said:

Perhaps remove the "No files selected" green text when a channel is set or put "set" in bold+ some other colour?

 

I have now made it bold for future release. The "No files selected" comes from the browser, not from my web page.

 

[AudioDoctor]

@Miska Can you expand on what this means from your website re: the embedded HQP?

"Preliminary support for AMD GPUs on Ubuntu build. "

 

Does it mean that you are working on using AMD GPUs in the same way you are using CUDA with Nvidia GPUs currently?

[Miska]

12 minutes ago, AudioDoctor said:

"Preliminary support for AMD GPUs on Ubuntu build. "

 

Does it mean that you are working on using AMD GPUs in the same way you are using CUDA with Nvidia GPUs currently?

 

Yes, it should already work, but not really tested since I'm still waiting for a new Radeon RX 6800 XT to arrive...

[AudioDoctor]

11 minutes ago, Miska said:

 

Yes, it should already work, but not really tested since I'm still waiting for a new Radeon RX 6800 XT to arrive...

 

Ok... I have a 6900XT and a 5950X. Is it possible to install the embedded version on my desktop running Ubuntu 20.04 and test that?

[Miska]

8 hours ago, AudioDoctor said:

Ok... I have a 6900XT and a 5950X. Is it possible to install the embedded version on my desktop running Ubuntu 20.04 and test that?

 

Yes it is, if it's a desktop version of Ubuntu, just disable the hqplayerd system service and run it under your user session instead. Otherwise it will likely fail to start at boot because network is managed in a different way on desktop systems.

 

You need to have "rocm-dkms" package installed (unless you have the "amdgpu-pro" driver installed) in addition to the dependencies hqplayerd will automatically pull in from the AMD repository. See installation instructions here:

https://rocmdocs.amd.com/en/latest/Installation_Guide/Installation-Guide.html

 

And then check that you have "rocm" enabled in your hqplayerd.xml

 

[AudioDoctor]

It looks like rocm-dkms will be it because the amdgpu-pro has not been released for 20.04 yet. Give me a day or two to get this running and i'll report back.

[SwissBear]

Hi Jussi,

 

I have a difficulty with a new configuration (interface + DAC), where I am not able to play DSD512 with HQP.

 

Config is:

Mac Mini M1 -> NAA ->(usb)-> Singxer SU-2 ->(i2s)-> Topping D90

1. When using Roon only and replacing NAA with Roon endpoint, I am able to play DSD512
2. When using Roon+HQP, and using NAA, I am able to play DSD256. When switching to DSD512 (obviously non EC modulators), I either ear white noise, or music with interference
3. In the same situation, with a different DAC (x-Sabre), I have no problem playing DSD512 either with Roon alone or with Roon+HQP.

Would you have any clue about what is happening ?

Thanks,

[ted_b]

Pierre,

Which ALSA driver is your NAA finding, the Su-2 or the Topping?  Also, have you tried bypassing Roon and testing with HQPlayer only (w/ NAA of course)?  Two random questions but....who knows.

[SwissBear]

21 minutes ago, ted_b said:

Pierre,

Which ALSA driver is your NAA finding, the Su-2 or the Topping?  Also, have you tried bypassing Roon and testing with HQPlayer only (w/ NAA of course)?  Two random questions but....who knows.

Thanks Ted,

Not sure what you mean by ALSA Driver as HQPlayer is running on a Mac which has no such drivers. I am selecting the SU-2 as a device visible from the NAA though.

I tried to send some files directly from HQPlayer without Roon and the results are the same: with DSD512, I have the music, attenuated, with noise. This is random though ; I could also have just white noise...

[Miska]

2 hours ago, SwissBear said:

Mac Mini M1 -> NAA ->(usb)-> Singxer SU-2 ->(i2s)-> Topping D90

1. When using Roon only and replacing NAA with Roon endpoint, I am able to play DSD512
2. When using Roon+HQP, and using NAA, I am able to play DSD256. When switching to DSD512 (obviously non EC modulators), I either ear white noise, or music with interference
3. In the same situation, with a different DAC (x-Sabre), I have no problem playing DSD512 either with Roon alone or with Roon+HQP.

 

Have you tried omitting the SU-2? Those I2S interfaces can be really problematic.

 

What kind of NAA is it?

[ted_b]

So Jussi, where in Linux workflow does it identify the dac if, in this case, a DDC comes first and NAA announces it?  I assume the SU-2 is seeing something different between the Topping (no DSD512) and the X-Sabre (DSD512), but what is it in a "driverless" Linux that tells the SU-2 what IT is connected to, especially when it's not USB?  For example, in my ASIO world I choose the SU-1 driver when connected through it (to Holo) and the Holo driver when going straight into it via USB.  But I have no idea how the SU-1 knows it's connected to a Holo (via I2S and an HDMI cable).  So there are two drivers hanging around, and I chose. Note: it used to be that I could leave the older Holo driver selected, regardless of SU-1 or direct...not anymore since I updated both of them yesterday.  Now they act normal.

[bibo01]

@MiskaYou have implemented Ravenna protocol. Is that in your HQP OS/NAA ? I read that Ravenna is an extension of Dante protocol, is that true? Does it mean that small Dante converters, like Digimedia or Tascam, can be seen by your OS?

Any experience in general with Dante converters?

[SwissBear]

7 hours ago, Miska said:

Have you tried omitting the SU-2? Those I2S interfaces can be really problematic.

Hi Jussi,

 

Yes I tried through the USB interface of the DAC, it works flawlessly with DSD512.

 

Having read you extensively on this subject, I appreciate that you do not have a lot of consideration for those interfaces 😉.

Having said that, I have to acknowledge that I am very sensitive to the presentation of music which is achieved with high precision clocks (Mutec REF10 in this case). And those I2S interfaces are the only way I found to feed the DAC with a higher precision clock than their own, when they are not equipped with a master clock input.

 

And those interfaces are just another input for the DAC. So I would be interested to make it work if I can.

 

To summarise, it is the conjonction of HQPlayer and this interface which causes DSD512 not to be processed correctly by this D90.

 

Quote

 

What kind of NAA is it?

This is the UpBoard with your USB thumb drive.

 

Thanks

[Miska]

6 hours ago, bibo01 said:

@MiskaYou have implemented Ravenna protocol. Is that in your HQP OS/NAA ?

 

There is RAVENNA version of HQPlayer OS.

 

6 hours ago, bibo01 said:

I read that Ravenna is an extension of Dante protocol, is that true?

 

No, both are based on AES67.

 

6 hours ago, bibo01 said:

Does it mean that small Dante converters, like Digimedia or Tascam, can be seen by your OS?

 

I don't know, maybe to a limited extent. RAVENNA has extensions to support higher sampling rates and DSD.

 

6 hours ago, bibo01 said:

Any experience in general with Dante converters?

 

So far no, since they are limited to 192k PCM and no DSD...

 

[Miska]

8 hours ago, ted_b said:

I assume the SU-2 is seeing something different between the Topping (no DSD512) and the X-Sabre (DSD512)

 

I don't think it sees anything different, since I2S over cable is like S/PDIF, unidirectional and clock at the wrong side.

 

8 hours ago, ted_b said:

But I have no idea how the SU-1 knows it's connected to a Holo (via I2S and an HDMI cable).

 

It doesn't...

 

8 hours ago, ted_b said:

For example, in my ASIO world I choose the SU-1 driver when connected through it (to Holo) and the Holo driver when going straight into it via USB.

 

On Windows, these drivers detect the device based on the USB vendor and product ID, so usually the drivers are locked to be product specific and don't allow the driver to be used with someone else's product. This is unlike the generic USB Audio Class driver built into Windows 10, macOS or Linux.

 

On ASIO, you have layer of two drivers. You have the hardware device driver and then the ASIO driver DLL which hooks to the device driver. This is a pair, and if it gets mixed for some reason you would usually have functional problems.

 

8 hours ago, ted_b said:

Note: it used to be that I could leave the older Holo driver selected, regardless of SU-1 or direct...not anymore since I updated both of them yesterday.

 

Means that older Holo driver wasn't so pedantic about which product(s) it hooks up to. Either at device driver level, or between device driver and ASIO driver DLL. You can have only one driver attached to a product. And the two layers shouldn't get mixed between two driver packages.

 

I guess both drivers are sourced from Thesycon, so they have largely same codebase.

 

[Miska]

1 hour ago, SwissBear said:

Having said that, I have to acknowledge that I am very sensitive to the presentation of music which is achieved with high precision clocks (Mutec REF10 in this case). And those I2S interfaces are the only way I found to feed the DAC with a higher precision clock than their own, when they are not equipped with a master clock input.

 

Usually using external clocks will actually make the DAC perform worse. For example using 10 MHz external clock means the device needs to use PLL to generate clocks it needs from the wrong clock frequency, instead of using crystal of correct frequency. I2S has additional problem of clock being at the wrong side like in S/PDIF too, at the source side, instead of being where it should be, close to the conversion stage inside the DAC.

 

So the higher precision usually gets lost on the way.

 

[SwissBear]

39 minutes ago, Miska said:

 

Usually using external clocks will actually make the DAC perform worse. For example using 10 MHz external clock means the device needs to use PLL to generate clocks it needs from the wrong clock frequency, instead of using crystal of correct frequency. I2S has additional problem of clock being at the wrong side like in S/PDIF too, at the source side, instead of being where it should be, close to the conversion stage inside the DAC.

 

So the higher precision usually gets lost on the way.

 

Thank you for your explanations.

 

It is the interface which does the translation of the 10 MHz clock signal and the clock signal to be used by the DAC. Singxer has a good expertise in this field.

 

The x-Sabre has a mode, called synchronous, which can be selected through the interface, where the DAC would follow the clock signal received together with the music flow;  the PLL can be disconnected as well. 

In the case of the D90, the technical support of Topping mentioned that the DAC would have the same behaviour if fed through the i2s interface.

 

My understanding is that i2s has brought a lower level of interface to the DACs which allows better use of external clocks. And I have to say that my listening experience matches this understanding.

[Ipoci]

On 1/16/2021 at 12:22 AM, Miska said:

Why!? It has not been designed to be used that way...

Direct connection could be useful when source HQPD has two separate Ethernet interfaces one for "normal" network access (cable or wifi with DHCP server somewhere) and the other as a straight cable to NAA (catX or fiber optical). In my case I use a MacBookPro that reads audio files locally on SSD or from NAS by accessing home network via WiFi, then I use a straight Fiber Optical cable to send audio stream to NAA, Mac and NAA use very tiny 100-X media converters (3M Volition) that have the beauty to be powered by 5V provided by any USB port ... 
 

[bibo01]

4 hours ago, Miska said:

There is RAVENNA version of HQPlayer OS.

I haven't seen/used it yet. Do you implement a GUI in order to use Aneman?