HQ Player

[AnotherSpin]

On 3/29/2019 at 11:37 PM, Miska said:

 

Since the DAC is R2R PCM ladder, modulator doesn't have a play here like with SDM DACs, so this is quite a bit simpler case. You can then choose between different dithers/noise-shapers, in this case it pretty much boils down to selection between TPDF, Gauss1 and NS9.

 

After couple of weeks listening I found out 16 bits setting works better than others tried. I would confirm also, closed form or closed form M give very good result with good quality modern recordings, ext2 is excellent for so-so quality and 70-s rock remasters. Also TPDF gives more neutral sound, while NS9 seem to add some coloring.

 

[twluke]

Hi Miska, I have a question about the constant log messages from the hqplayer desktop running on my Ubuntu 18.04 machines (one with a 8700K CPU and the other with a 9700K CPU with a 4.15.0-47-generic kernel in each). The HQP (3.25.3) is working well in each machine, sending the quality sound of DSD512 sources to my B3SEpros from TPA (Twisted Pear Audio).

 

What appears weird to me is the continuous flow of log messages from the application in /var/log/syslog as shown below:

 

Machine 1:

  Apr 15 19:54:41 musik2 hqplayer.desktop[4052]: volumeRange
  Apr 15 19:54:41 musik2 hqplayer.desktop[4052]: state
  Apr 15 19:54:43 musik2 hqplayer.desktop[4052]: volumeRange
  Apr 15 19:54:43 musik2 hqplayer.desktop[4052]: state
  Apr 15 19:54:45 musik2 hqplayer.desktop[4052]: volumeRange
  Apr 15 19:54:45 musik2 hqplayer.desktop[4052]: state

Machine 2:

  Apr 15 19:55:20 musik3 hqplayer.desktop[1627]: volumeRange
  Apr 15 19:55:20 musik3 hqplayer.desktop[1627]: state
  Apr 15 19:55:22 musik3 hqplayer.desktop[1627]: volumeRange
  Apr 15 19:55:22 musik3 hqplayer.desktop[1627]: state
  Apr 15 19:55:24 musik3 hqplayer.desktop[1627]: volumeRange
  Apr 15 19:55:24 musik3 hqplayer.desktop[1627]: state

 

This set of messages including volumeRange and state is continuously repeated with a 2-sec interval while running the software. Usually it would be okay but occasionally it becomes annoying particularly when I look into the log files for other reasons.

 

Is there any tweaks to avoid this continuous flow of messages? BTW, I didn't find anything abnormal in the HQP log.

 

Thanks in advance.

 

[Miska]

2 hours ago, twluke said:

This set of messages including volumeRange and state is continuously repeated with a 2-sec interval while running the software. Usually it would be okay but occasionally it becomes annoying particularly when I look into the log files for other reasons.

 

Looks like Roon controlling HQPlayer, it wants to do that... This is just print out to stderr through g_message(), so there are various ways you can silence it down.

 

[twluke]

4 hours ago, Miska said:

 

Looks like Roon controlling HQPlayer, it wants to do that... This is just print out to stderr through g_message(), so there are various ways you can silence it down.

 

 

Thank you for this suggestion. Yes, Roon is running for HQP on each  machine. Later I'll figure out how I can silence it.

 

Regards,

[Miska]

11 minutes ago, twluke said:

Thank you for this suggestion. Yes, Roon is running for HQP on each  machine. Later I'll figure out how I can silence it.

 

One way is to edit hqplayerd.service and add line there

StandardError=null
StandardOutput=null

That will make all the printout disappear.

[itguy61]

Well I bought a Mac iMini six core and it is going back after two days with it.   I bought it to be able to use more filters at DSD 256 than my current iMac.   I immediately started having Bluetooth issues withe the mouse and keyboard.  Mouse would skip around and both items would disconnect.   Googled it and sure enough others are having the same issue with the new Mac mini as well.   No fix in site, only seems to happens with other usb devices attached.  

 

The second issue is audio dropouts are back.    No idea why, only Mac OS, Hqplayer and Roon on the computer.  Another google search and many hits on T2 chip audio glitches uses USB Audio devices.   No fix in sight.   Apparently time synch is causing this on computers with the T2 chip.   

 

Shame on you Apple!  Your  recent refreshes have been lousy, the MacBook Pro 2018 having all kinds of issues as well.

 

I guess it is time for me to go back to windows just building my own pc.   Apple stuff costs too much to have to put up with these quality issues.   

 

My advice is to avoid the new Mac mini until usb/bluetooth issue resolved.  Also, avoid T2 chip equipped models until fix is issued.

[Miska]

3 hours ago, itguy61 said:

The second issue is audio dropouts are back.    No idea why, only Mac OS, Hqplayer and Roon on the computer.  Another google search and many hits on T2 chip audio glitches uses USB Audio devices.   No fix in sight.   Apparently time synch is causing this on computers with the T2 chip.

 

I believe I've read that Apple has fixed this in the latest macOS updates... But I've only read about the bug in context of MacBook Pro and iMac, but didn't realize that they've put T2 also in Mac Mini...

 

I just cannot understand how this kind of bug can possibly have passed their QA.

 

[Bob Stern]

3 minutes ago, Miska said:

didn't realize that they've put T2 also in Mac Mini.

 

Macs with T2 chip:

https://support.apple.com/en-us/HT208862

[Miska]

Just now, Bob Stern said:

Macs with T2 chip:

https://support.apple.com/en-us/HT208862

 

OK, interesting, so only the iMac Pro has it, but not the regular iMac... I believe I've seen some rumors that Apple would be refreshing the iMac line soon, so that'll likely also change soon.

 

As a very computer security conscious person I like that they introduce new security measures, but I also know that it doesn't have to screw up other things like this audio problem...

 

[Miska]

I know that as a work around people have been using a Thunderbolt-connected dock for the Mac. These have a separate USB controller that doesn't get messed up by the communication latency introduced by T2. Unfortunately that is not very cheap solution and adds unnecessary dongles...

 

With HQPlayer, using a NAA is possibly cheaper option.

[Bob Stern]

4 minutes ago, Miska said:

rumors that Apple would be refreshing the iMac line soon

 

iMac range got a "speed bump" one month ago.

 

https://www.macrumors.com/roundup/imac/

[twluke]

13 hours ago, Miska said:

 

One way is to edit hqplayerd.service and add line there

StandardError=null
StandardOutput=null

That will make all the printout disappear.

Thanks for this suggestion. Unfortunately I'm using a desktop version of HQP on linux and hqplayerd.service is not available. Maybe living with these messages for now until I can find the other way.

 

Regards,

[Sevenfeet]

4 hours ago, Miska said:

 

OK, interesting, so only the iMac Pro has it, but not the regular iMac... I believe I've seen some rumors that Apple would be refreshing the iMac line soon, so that'll likely also change soon.

 

As a very computer security conscious person I like that they introduce new security measures, but I also know that it doesn't have to screw up other things like this audio problem...

 

 

The iMac Pro and the Mac Mini both have the T2 chip.  But the recently refreshed iMac did not get it.  Speculation was that the T2 required a more thorough refresh which Apple didn't want to do right now for some reason.  I imagine this will be the last iMac that won't have it.

[itguy61]

I have not see anything in the audio forums that show this has been resolved.   It happened like clockwork at around 20-30 minutes.   Supposedly timed daemon causing issue.  I am not spending 200-300 to buy a thunderbolt hub to fix this.   

 

The other issue that may be bigger and harder for them to fix is the bluetooth issue.   The minute I plugged in the DAC the mouse started acting flaky, slow, and disconnects.    Did not matter which port and the DAC is several feet away.    This is getting to be a well known issue also.

 

The 2018 MacBook pros are having all kinds of issues too, keyboard is better but not fixed.    Apple is not doing great on their design and qc job. 

 

I will just build a windows pc, I can build one with an i7-9700 8 core processor for less than the mini.   I have been a loyal apple user for years but may be time to move on.

 

Both of these issues point back to the usb bus,   I have read that the bluetooth radio is on the regular usb bus as well.  Sounds like hardware design issue.

[Miska]

6 hours ago, twluke said:

Thanks for this suggestion. Unfortunately I'm using a desktop version of HQP on linux and hqplayerd.service is not available. Maybe living with these messages for now until I can find the other way.

 

Regards,

 

Oh, what distro are you using? I don't remember qDebug() printouts ending up in syslog on my systems. By default it goes to stderr too. However, you should be able to control output either through environment variable or by creating suitable configuration file.

 

Documented here. If you create ~/.config/QtProject/qtlogging.ini and set it up like

[Rules]
*.debug=false

 

Or alternatively set same rule through 

QT_LOGGING_RULES="*.debug=false"

environment variable.

 

[itguy61]

#203

I just returned my i7 Mac Mini. Very disappointed as it was perfect for my home theater. The first irritating issue I could not resolve was the flakiness and disconnects with my bluetooth keyboard and mouse. They work fine until anything is plugged into the USB ports and then they flake out.

The second issue is audio dropouts with my iFi Idsd dac. I would get a dropout about every 20-30 minutes. 10.14.4 did not resolve this although it was less often. 

I am just going back to my 21.5 IMac until these issues are resolved. Very disappointed because the 6 core processor would run all the filters in HQplayer at DSD 256. 

Very disappointing. Can't believe they did not catch these in development.  I would avoid any Mac with the T2 in it until this is resolved.   

[twluke]

4 hours ago, Miska said:

Oh, what distro are you using? I don't remember qDebug() printouts ending up in syslog on my systems. By default it goes to stderr too. 

lsb_release shows the info below:

  Distributor ID: Ubuntu
  Description:    Ubuntu 18.04.2 LTS
  Release:        18.04
  Codename:       bionic

4 hours ago, Miska said:

Documented here. If you create ~/.config/QtProject/qtlogging.ini and set it up like

[Rules]
*.debug=false

 

Or alternatively set same rule through 

QT_LOGGING_RULES="*.debug=false"

environment variable. 

Thank you for this kind suggestion; really appreciated it. I tried both at the user level but they didn't work and the log messages are still flowing. Maybe I need more homework.

 

Regards,

[Miska]

13 minutes ago, twluke said:

lsb_release shows the info below:

  Distributor ID: Ubuntu
  Description:    Ubuntu 18.04.2 LTS
  Release:        18.04
  Codename:       bionic

Thank you for this kind suggestion; really appreciated it. I tried both at the user level but they didn't work and the log messages are still flowing. Maybe I need more homework.

 

OK, strange, I just checked and I don't get anything in /var/log/syslog or "journalctl -a" on Ubuntu Studio 18.04. In fact, normal userid shouldn't be able to spam journal/syslog in first place due to access right restrictions...

 

I have mostly AppArmor access control messages and HDD/SSD temperature change messages from smartctl...

 

[Yviena]

@Miska is it possible to get a option to load the track in memory, as it seems that when there is disk activity on the same hdd that the music is on usually leads to skipping/dropouts.

[Miska]

6 hours ago, Yviena said:

@Miska is it possible to get a option to load the track in memory, as it seems that when there is disk activity on the same hdd that the music is on usually leads to skipping/dropouts.

 

No, because then that would ruin gapless playback... Disk access in HQPlayer is highly tuned to work as intended. If such happens due to disk IO, then either the OS is bad, has bad I/O scheduling settings or the storage hardware is bad... USB connected disks are usually among the bad hardware category, M.2, SATA, Thunderbolt, Firewire and network (NAS) usually not.

 

[Bob Stern]

27 minutes ago, Miska said:

USB connected disks are usually among the bad hardware category, M.2, SATA, Thunderbolt, Firewire and network (NAS) usually not.

 

Does the problem with USB disks include USB 3?  I've been using a USB 3 spinning disk for my music files because I was under the impression it was less likely to incur I/O interruptions than a NAS.

[Yviena]

10 hours ago, Miska said:

 

No, because then that would ruin gapless playback... Disk access in HQPlayer is highly tuned to work as intended. If such happens due to disk IO, then either the OS is bad, has bad I/O scheduling settings or the storage hardware is bad... USB connected disks are usually among the bad hardware category, M.2, SATA, Thunderbolt, Firewire and network (NAS) usually not.

 

Hmm i see...

 

on another note I'm thinking about acquiring a RME-ADI-2 DAC, would you only recommend to use the direct DSD path if upsampling to DSD, or will it be okay to use it with for example upsampled 705/768khz material if possible, And if Direct path is possible to use with upsampled PCM should DAC Bits also be set?

 

 

[Miska]

4 hours ago, Bob Stern said:

Does the problem with USB disks include USB 3?  I've been using a USB 3 spinning disk for my music files because I was under the impression it was less likely to incur I/O interruptions than a NAS.

 

Problem with USB is that CPU is heavily involved in USB protocol packet handling. (e)SATA, M.2 and Thunderbolt (usually SATA) can do lower overhead bus-master DMA. Cheap network adapters are not better than USB, but better ones include pretty much full TCP/IP stack offloading.

 

I have three means I access my content, I have external Thunderbolt HDD on Mac Mini. Then another copy of content is on my Linux workstation's internal SATA HDD from where it is also shared over Samba (SMB) on the network, so that is accessed both locally and remotely.

 

I use external USB 3 HDDs for backup purposes. But I have one small portable USB 3 HDD I use for content when I'm traveling with laptop, because the laptop's internal M.2 SSD doesn't have enough space.

 

[Miska]

1 hour ago, Yviena said:

on another note I'm thinking about acquiring a RME-ADI-2 DAC, would you only recommend to use the direct DSD path if upsampling to DSD, or will it be okay to use it with for example upsampled 705/768khz material if possible, And if Direct path is possible to use with upsampled PCM should DAC Bits also be set?

 

It performs pretty nicely also at 705.6/768k, but then you have it's digital volume control and modulator on the path. Since it is SDM DAC, there's no direct path for PCM.

 

No need to set DAC Bits, it can accept 32-bit and that is best input for it's internal DSP.

 

[Bob Stern]

2 hours ago, Miska said:

Cheap network adapters are not better than USB, but better ones include pretty much full TCP/IP stack offloading.

 

I infer full TCP/IP stack offloading is implemented in the ethernet controllers used in all Macs and in modern Synology and QNAP NAS's?