Jump to content
IGNORED

HQ Player


Recommended Posts

9 minutes ago, Miska said:

Have you tried using ASDM7EC or ASDM5EC modulator?

Yes I tried them but discovered they are very power demanding, just a lot of click, pops and hiccups therefore with those modulators is definitely a no go ... in my setup.

... could you please elaborate “depends also on the configured TDP figure” as I cannot understand (my computer competency is very low ... )

... during my tests I noticed a variable behavior playing gapless tracks, mainly Opera recordings, Carmen, Falstaff ...., I can say that almost 3/4 times on 10 gapless fails ... upsampling wave 16/44.1 to 4xDSD/ASDM7 and ASDM5/poly-sinc-ext2, is this a known issue or could be due to my setup?

To be noted that wave files are served by Minimserver which transcode Alac and flac to wave without modifying bit depth nor sample rate

thank you

 

Stefano

 

My audio system

Link to post
Share on other sites
7 minutes ago, stefano_mbp said:

... could you please elaborate “depends also on the configured TDP figure” as I cannot understand (my computer competency is very low ... )

 

TDP is Thermal Design Power, these mobile CPUs can be configured for different thermal designs and if necessary efficiently throttle the performance to stay within thermal boundaries of lighter cooling systems.

 

9 minutes ago, stefano_mbp said:

... during my tests I noticed a variable behavior playing gapless tracks, mainly Opera recordings, Carmen, Falstaff ...., I can say that almost 3/4 times on 10 gapless fails ... upsampling wave 16/44.1 to 4xDSD/ASDM7 and ASDM5/poly-sinc-ext2, is this a known issue or could be due to my setup?

To be noted that wave files are served by Minimserver which transcode Alac and flac to wave without modifying bit depth nor sample rate

 

If you are using UPnP with MinimServer, many times it is a bit too sluggish to serve a new track, likely because of the Java runtime.

 

It would be better to play those natively on HQPlayer if possible. Since it is local content, maybe from NAS, mounting the NAS file share for HQPlayer and using HQPlayer's library would be more efficient than UPnP.

 

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to post
Share on other sites
11 minutes ago, stefano_mbp said:

... unluckily I have many Alac files ... ( iTunes inheritance ...) that’s why I use Minimserver to transcode to wave ... they aren’t supported by HQPlayer (?)

Thank you

Convert 'em, with something like db poweramp.  You can do whole batches.  And I like db poweramp's FLAC uncompressed (which isn't compression level 0) cuz it supports metadata/tags better than wav, but has no theoretical compression artifacts.

Link to post
Share on other sites
2 minutes ago, ted_b said:

Convert 'em, with something like db poweramp

.... I could .... they are more or less 8k albums ... my library is quite huge ....

and I doubt, that’s to say I’m sure, that hqplayerd can manage metadata like Minimserver ...

this gapless issue is really a bad surprise, never had any issue using MPD/upmpdcli 

Stefano

 

My audio system

Link to post
Share on other sites
1 hour ago, stefano_mbp said:

... unluckily I have many Alac files ... ( iTunes inheritance ...) that’s why I use Minimserver to transcode to wave ... they aren’t supported by HQPlayer (?)

 

That transcoding is likely not making MinimServer particularly faster when switching tracks...

 

You could just convert ALAC permanently to some more generic format?

 

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to post
Share on other sites
1 hour ago, ted_b said:

Convert 'em, with something like db poweramp.  You can do whole batches.  And I like db poweramp's FLAC uncompressed (which isn't compression level 0) cuz it supports metadata/tags better than wav, but has no theoretical compression artifacts.

 

What are the theoretical compression artifacts? Because lossless compression theoretically doesn't have any?

 

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to post
Share on other sites
44 minutes ago, Miska said:

That transcoding is likely not making MinimServer particularly faster when switching tracks...

it could be true but the issue do not exist using mpd/upmpdcli nor using JRiver as a dlna renderer ... in JRiver I can see that there are alway two tracks in the cue and this helps to avoid the issue .... using in both scenarios using Minimserver as upnp server.

If it can help my setup is as follow:

Lumin app as control point, Minimserver as upnp server, hqplayerd as upnp renderer, rpi4/ropiee xl as naa

Stefano

 

My audio system

Link to post
Share on other sites
1 hour ago, Miska said:

 

What are the theoretical compression artifacts? Because lossless compression theoretically doesn't have any?

 

Purely a comment I categorized as "theoretical" to appease the thousands of posters on any wav vs flac debate.  :)

Link to post
Share on other sites
29 minutes ago, stefano_mbp said:

More on my tests, I removed transcoding from Minimserver but the gapless issue is still there with flac files ... this is really a pity as the SQ is really fantastic but MinimServer and gapless are ... indispensable.

 

I had some gapless issues with MinimServer as well. Playing for example from Audirvana was working better. Or using Rygel as UPnP server. mConnect Player on iOS is snappier as control point than BubbleUPnP (for some unknown reason).

 

But just use HQPlayer library facilities and it is better. I use UPnP only for playing Tidal.

 

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to post
Share on other sites
1 hour ago, ted_b said:

Purely a comment I categorized as "theoretical" to appease the thousands of posters on any wav vs flac debate.  :)

 

I prefer objective data... ;)

 

For the record I'm using FLACs with maximum compression level to get smallest files and thus least I/O bandwidth usage.

 

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to post
Share on other sites

I have a technical difficulty with an SOtM sMS-200 Ultra, when it is configured as NAA and fed with DSD 256: it is producing clicks and pops, and white noise from time to time. This device is working well when configured as a Roon endpoint and running the same resolution files. And when running another NAA from the same HQP server, things are working smoothly as well.

Does anybody know if this is a recurrent problem linked to the implementation of the NAA client on the sMS-200 ?

Link to post
Share on other sites
1 hour ago, SwissBear said:

I have a technical difficulty with an SOtM sMS-200 Ultra, when it is configured as NAA and fed with DSD 256: it is producing clicks and pops, and white noise from time to time. This device is working well when configured as a Roon endpoint and running the same resolution files. And when running another NAA from the same HQP server, things are working smoothly as well.

Does anybody know if this is a recurrent problem linked to the implementation of the NAA client on the sMS-200 ?

 

These small devices require network to support 802.3x Flow Control. Otherwise they get overwhelmed by sustained gigabit traffic and their hardware buffers overflow resulting in packet losses and re-sends. Which in turn increases amount of total traffic resulting in network stalls.

 

So in case you use smart switches on your network, I'd first check their configuration that the 802.3x is enabled on all ports.

 

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to post
Share on other sites
9 hours ago, Miska said:

 

These small devices require network to support 802.3x Flow Control. Otherwise they get overwhelmed by sustained gigabit traffic and their hardware buffers overflow resulting in packet losses and re-sends. Which in turn increases amount of total traffic resulting in network stalls.

 

So in case you use smart switches on your network, I'd first check their configuration that the 802.3x is enabled on all ports.

 

Two questions I would like to ask there:

- can buffer ajustements inside HQPlayer settings help with this kind of difficulties ? I played with them but not in a structured way and did not notice any improvement. Are there recommended values ?

- speaking of smart switches, there are a lot of audiophile switches on the market which are offering high precision reclocking to reduce phase noise of the Ethernet data stream. Have you been able to measure the jitter reduction of the output signal of the DAC which was linked to such devices ? Do you have any such switches in your own network ?

Thank you for the time you are taking to help us all. Your expertise and dedication are very much appreciated :-)

Link to post
Share on other sites
9 hours ago, Miska said:

These small devices require network to support 802.3x Flow Control. Otherwise they get overwhelmed by sustained gigabit traffic and their hardware buffers overflow resulting in packet losses and re-sends. Which in turn increases amount of total traffic resulting in network stalls.

 

So in case you use smart switches on your network, I'd first check their configuration that the 802.3x is enabled on all ports.

 

I'm confused.  I thought a switch was supposed to forward to a device only the traffic addressed to that device's MAC address.  So why would the micro-computer hosting NAA receive a lot of traffic in addition to the audio data being streamed by HQP?

 

Would the answer be different if the NAA host was connected to a router instead of a switch?

Mac Mini (2012 i7) > HQPlayer > RME ADI-2 v2 > Benchmark AHB-2 > Thiel 3.7

Link to post
Share on other sites
7 hours ago, Outlaw said:

My PC has a 8086K CPU.Has drop outs with Sinc-L filters.Have a chance to buy a asus strix 1080 ti at a very good price,Would this solve issue of drop outs ?

I am pretty sure my Ryzen 6 5600X/1989 Ti can do Sinc-L with DSD256/ASDM7 (no EC)

I tested this after finding that my i9-9900KS could not do Sinc-L with DSD256/ASDM7EC but only DSD128/ASDM7EC.

The 1080 TI can do Sinc-L DSD256 because it has 11GB memory. Think >8GB is required, so rules out many of the newer faster GPU

Link to post
Share on other sites
17 hours ago, Miska said:

 

I had some gapless issues with MinimServer as well. Playing for example from Audirvana was working better. Or using Rygel as UPnP server. mConnect Player on iOS is snappier as control point than BubbleUPnP (for some unknown reason).

 

But just use HQPlayer library facilities and it is better. I use UPnP only for playing Tidal.

 

To narrow down the cause of the issue I’ve just tried to play same files transcoded to wave by Minimserver with different resampling rates: 16/44.1 files can be played gapless resampling them in hqplayerd up to 24/176.400, beyond this gapless is broken
I tried using MConnect too but the issue is there anyway

Stefano

 

My audio system

Link to post
Share on other sites
10 hours ago, Outlaw said:

My PC has a 8086K CPU.Has drop outs with Sinc-L filters.Have a chance to buy a asus strix 1080 ti at a very good price,Would this solve issue of drop outs ?

 

Possibly yes, at DSD256 the amount of RAM on GPU can become close to being issue though. I don't have any 1080Ti model to test and the amount of RAM needed depends on the GPU model.

 

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to post
Share on other sites
2 minutes ago, stefano_mbp said:

To narrow down the cause of the issue I’ve just tried to play same files transcoded to wave by Minimserver with different resampling rates: 16/44.1 files can be played gapless resampling them in hqplayerd up to 24/176.400, beyond this gapless is broken
I tried using MConnect too but the issue is there anyway

 

This confirms the lag that MinimServer has when requested for higher bandwidth tracks...

 

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to post
Share on other sites
2 hours ago, Bob Stern said:

I'm confused.  I thought a switch was supposed to forward to a device only the traffic addressed to that device's MAC address.  So why would the micro-computer hosting NAA receive a lot of traffic in addition to the audio data being streamed by HQP?

 

Traffic is packet based, on multiple layers. There is no "stream", it is only virtual. Same for USB, when you use USB Audio Class, there is packet sent every 125 µs (8 kHz rate).

 

These small devices have only few ethernet MTUs (1500 bytes) worth of hardware buffer. At higher level, single TCP or UDP packet consists of multiple MTUs. When that hardware buffer comes full, the ethernet interface sends 802.3x pause frame towards switch asking it to stop sending. Switch handles this in it's own buffering and sends another 802.3x pause frame towards HQPlayer computer as needed.

 

Ethernet hubs used to be such dumb devices, but modern switches can be relatively smart, with multiple priority queues (buffers) for each port and such. So they know how to handle traffic of different priorities, like HQPlayer uses 802.1p/802.1Q.

 

Now the challenge is that the connection from the ethernet interface to the CPU and the CPU itself cannot handle full gigabit speed traffic. If the 802.3x flow control packet is getting ignored by the upstream devices, the hardware buffer will overflow and the remaining arriving packets are lost. This in turn triggers re-send at the higher level protocol when it is noticed. Since the traffic needs to be constant rate to transfer realtime audio, lost packets mean increased amount of traffic. Which is not making things any better...

 

3 hours ago, SwissBear said:

- can buffer ajustements inside HQPlayer settings help with this kind of difficulties ? I played with them but not in a structured way and did not notice any improvement. Are there recommended values ?

 

Not much, it is recommended to leave buffer setting at Default for NAAs.

 

3 hours ago, SwissBear said:

- speaking of smart switches, there are a lot of audiophile switches on the market which are offering high precision reclocking to reduce phase noise of the Ethernet data stream. Have you been able to measure the jitter reduction of the output signal of the DAC which was linked to such devices ? Do you have any such switches in your own network ?

 

I don't have any of such devices. Why would phase noise of ethernet matter in this context? There are no audio clocks being transmitted over the network when using NAA.

 

Using NAA is not so much different than playing something from Tidal or Qobuz servers, where the traffic passes through number of networking devices on the way from some data center to the player application. Using a VPN would allow you to have HQPlayer and NAA located in different countries and have the traffic passing internet. And it would still likely work fine given enough internet bandwidth.

 

 

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


×
×
  • Create New...