HQ Player

[timeslip]

20 minutes ago, Miska said:

 

That error message is due to change in filter. But I suspect the issue may be that you get higher load because of the fixed output rate request? Does the situation change if you enable "Adaptive output rate" and use the filter you've used before?

 

 

The adaptive output rate fixed it and I have no idea why the stuttering also stopped, which was happening before I made any changes to HQP.  Will keep testing, but this is odd since I used it last week and for the past 2yrs with no issues.

[dericchan1]

@Miska I have a fitlet3 and a few raspberry pi 4 I can use as a NAA. Will they support pcm 1.536 or DSD1024?

 

Thanks

 

Deric

[Miska]

7 minutes ago, dericchan1 said:

@Miska I have a fitlet3 and a few raspberry pi 4 I can use as a NAA. Will they support pcm 1.536 or DSD1024?

 

Yes, both devices support those when running NAA OS or HQPlayer OS.

 

[Kalpesh]

On 11/17/2023 at 10:21 AM, Kalpesh said:

 

 

@MiskaA little up about HQPD control that ceased to see what's going on

the connection logo is (still) green though,

delete / reinstall on my iphone did not solve an eventual update issue

of course I checked via Client that network control is enable

@Ales Prochazka Am I the only one concerned ? then what might be wrong ?

[bogi]

1 hour ago, timeslip said:

 

The adaptive output rate fixed it and I have no idea why the stuttering also stopped, which was happening before I made any changes to HQP.  Will keep testing, but this is odd since I used it last week and for the past 2yrs with no issues.

If you have some favorite settings you are keeping for longer time, just save them from HQPlayer Desktop under a specific name. Then, any time you can return to that settings by loading the file.

[bogi]

15 hours ago, skipspence said:

Thanks so much, but doing from second table seems to be the hard tusk for me, since I can't get right config for 7.1 from 12.1 channel layout...

You can find more alternative suggestions in different discussions on the web, for example here: https://forum.doom9.org/showthread.php?p=1977955#post1977955

If suggestions are using decimal gain levels, you can convert them to dB with this online converter: http://www.sengpielaudio.com/calculator-FactorRatioLevelDecibel.htm

For example according to the table from the previous page decimal gain 0.707 should correspond to -3.01 dB

[Miska]

47 minutes ago, bogi said:

If suggestions are using decimal gain levels, you can convert them to dB with this online converter: http://www.sengpielaudio.com/calculator-FactorRatioLevelDecibel.htm

For example according to the table from the previous page decimal gain 0.707 should correspond to -3.01 dB

 

Remember that you can also use linear gains in HQPlayer matrix setup. In some cases it is even mandatory - if you want to do phase inversion (M/S decoding/encoding).

 

[Ales Prochazka]

10 hours ago, Kalpesh said:

@Ales Prochazka Am I the only one concerned ? then what might be wrong ?

 

Unfortunately, no idea. I have an iPad and an iPhone with iOS 17.1.2 and the latest version of HQPDcontrol (7.0.4-534) runs without problems. The lowest supported iOS version is 12.0.

[sworksone]

2 hours ago, Ales Prochazka said:

 

Unfortunately, no idea. I have an iPad and an iPhone with iOS 17.1.2 and the latest version of HQPDcontrol (7.0.4-534) runs without problems. The lowest supported iOS version is 12.0.

 

I had a first time situation yesterday of losing control of HQPlayer from HQPD control (MacBook and iPad both latest iOS updates).  I disabled and reenabled network control in HQPlayer and everything has been fine since.

[Kalpesh]

30 minutes ago, sworksone said:

 

I had a first time situation yesterday of losing control of HQPlayer from HQPD control (MacBook and iPad both latest iOS updates).  I disabled and reenabled network control in HQPlayer and everything has been fine since.

thank you;; unfortunately it does not work for me (Ventura 13.5 on Studio M1 and latest ios on iPhone 13)

and I have the green connected icon

btw, kudos and thank you @Ales Prochazka. Great nice to have when it works

[sworksone]

20 minutes ago, Kalpesh said:

thank you;; unfortunately it does not work for me (Ventura 13.5 on Studio M1 and latest ios on iPhone 13)

and I have the green connected icon

btw, kudos and thank you @Ales Prochazka. Great nice to have when it works

I’m just recalling that I also did an auto discovery for HQPlayer in HQPDcontrol.  I think this is the first time this worked for me as I previously had to manually enter the IP address.  Worth a try to see if this may fix your issue.  

[Miska]

12 hours ago, sworksone said:

I had a first time situation yesterday of losing control of HQPlayer from HQPD control (MacBook and iPad both latest iOS updates).  I disabled and reenabled network control in HQPlayer and everything has been fine since.

 

DHCP lease expired and the IP changed? Or otherwise momentarily HQPlayer machine lost IP? If DHCP lease expires before it is successfully renewed, OS drops the IP and the network interface may be momentarily without IP. This will cause dropped connection.

 

[satshanti]

I have two questions:

 

1) I really like the sinc-short/medium/long filters for classical and acoustic music, but also for some other well-recorded material. They're the ones I always try first. I use various other filters for pop, rock and other generally less well recorded material.

 

At 44.1k I can easily run sinc-medium, but for a lot of things I do prefer sinc-long. The problem is that it stutters about every 30 seconds, which is not bad and led me to the conclusion that not much more would be needed to make it work. So, I overclocked my CPU from 4000 to 4150 MHz, and initially this did seem to do the trick. I had prefetch on, and tracks would often stutter a few seconds before the end. I set it to off and things seemed fine, at least for my short test tracks. But when I started playing longer tracks, after 4 or 5 minutes, a stutter would suddenly happen, and then continue to happen every 5 seconds.

 

Now for my question: From these symptoms I deduce that it's not the CPU that's unable to keep up, but that something else is going on. I'm thinking RAM, maybe?. Any ideas what it could be?

 

2) Because of the above, and also that I'm very close to being able to run sinc-medium on hires files, but even the overclock didn't manage to get me there, I was wondering if it would technically be possible (and if any others would be interested in this) to introduce two additional "intermediate" filters: one in between sinc-short and -medium, and one in between sinc-medium and -long.   

[erekola]

15 minutes ago, satshanti said:

I have two questions:

 

1) I really like the sinc-short/medium/long filters for classical and acoustic music, but also for some other well-recorded material. They're the ones I always try first. I use various other filters for pop, rock and other generally less well recorded material.

 

At 44.1k I can easily run sinc-medium, but for a lot of things I do prefer sinc-long. The problem is that it stutters about every 30 seconds, which is not bad and led me to the conclusion that not much more would be needed to make it work. So, I overclocked my CPU from 4000 to 4150 MHz, and initially this did seem to do the trick. I had prefetch on, and tracks would often stutter a few seconds before the end. I set it to off and things seemed fine, at least for my short test tracks. But when I started playing longer tracks, after 4 or 5 minutes, a stutter would suddenly happen, and then continue to happen every 5 seconds.

 

Now for my question: From these symptoms I deduce that it's not the CPU that's unable to keep up, but that something else is going on. I'm thinking RAM, maybe?. Any ideas what it could be?

 

2) Because of the above, and also that I'm very close to being able to run sinc-medium on hires files, but even the overclock didn't manage to get me there, I was wondering if it would technically be possible (and if any others would be interested in this) to introduce two additional "intermediate" filters: one in between sinc-short and -medium, and one in between sinc-medium and -long.   

 

96k - 512×48 asdmec7 512+fs sinc-long with 14900K, RTX4090 and 48gb 7000mzh. It just need 4090.

 

[Kalpesh]

44 minutes ago, Miska said:

 

DHCP lease expired and the IP changed? Or otherwise momentarily HQPlayer machine lost IP? If DHCP lease expires before it is successfully renewed, OS drops the IP and the network interface may be momentarily without IP. This will cause dropped connection.

 

as far as I'm concerned, even auto discovery works. And I have the green connected icon. So I don't see how I can make my network the culprit.

Yet, no info is sent to HQPDcontrol by HQP Desktop.

Are you sure there can't be a API issue with 5.3.0 for Apple silicon ?

[Miska]

30 minutes ago, satshanti said:

Now for my question: From these symptoms I deduce that it's not the CPU that's unable to keep up, but that something else is going on. I'm thinking RAM, maybe?. Any ideas what it could be?

 

Cooling maybe? Thermal throttling would at least cause similar behaviour.

 

[erekola]

I can't edit the text. Yes, I had a 4080 and yes it could run any combination of dsd256. The wall was met only with 512 biggest combinations and only 4090 helps.

[Zauurx]

@Miska

It's been years since I've used a dac directly on my laptop with headphones.
I'm reconnecting my iBasso Bushmaster to my Asus (Win10).
I launch HQPlayer (trial) to listen a flac > DSD256...
Everything is played in PCM 384... HQPlayer configured normally (SDM output).
I try with my Hidizs AP80Pro, same thing >> PCM 384.
A wasapi problem under win10.
I'm looking... but can't find it !

 

Log:

... 
2023/12/09 18:12:38 SDM conversion parameters (8, 0)
  2023/12/09 18:12:38 DSD 6 dB gain disabled
& 2023/12/09 18:12:38 Play
  2023/12/09 18:12:38 Offload: resampler=disabled convolution=disabled
+ 2023/12/09 18:12:38 Playback engine running
  2023/12/09 18:12:38 Offload: resampler=disabled convolution=disabled
  2023/12/09 18:12:38 SDM conversion parameters (8, 0)
  2023/12/09 18:12:38 DSD 6 dB gain disabled
  2023/12/09 18:12:38 WASAPI output set sampling rate: 384000 (384000)
  2023/12/09 18:12:38 Automatic output rate: 384000
  2023/12/09 18:12:38 Rate or blocksize change triggered
  2023/12/09 18:12:38 Rate: 48000, block size: 5120, frame size: 640
  2023/12/09 18:12:38 Block size: 5120 (sample: 4)
  2023/12/09 18:12:38 Analysis initialized
  2023/12/09 18:12:38 Playback engine ratio: 8
  2023/12/09 18:12:38 Filter: long Gaussian poly sinc
  2023/12/09 18:12:38 Dither: 15th order linear noise shaping for 32-bit
  2023/12/09 18:12:38 Set volume: -9 +
...

 

[bogi]

2 minutes ago, Zauurx said:

A wasapi problem under win10.

Do you attempt native DSD or DoP?

Does ASIO driver exist for your DAC?

[LoryWiv]

7 hours ago, satshanti said:

I have two questions:

 

1) I really like the sinc-short/medium/long filters for classical and acoustic music, but also for some other well-recorded material. They're the ones I always try first. I use various other filters for pop, rock and other generally less well recorded material.

 

At 44.1k I can easily run sinc-medium, but for a lot of things I do prefer sinc-long. The problem is that it stutters about every 30 seconds, which is not bad and led me to the conclusion that not much more would be needed to make it work. So, I overclocked my CPU from 4000 to 4150 MHz, and initially this did seem to do the trick. I had prefetch on, and tracks would often stutter a few seconds before the end. I set it to off and things seemed fine, at least for my short test tracks. But when I started playing longer tracks, after 4 or 5 minutes, a stutter would suddenly happen, and then continue to happen every 5 seconds.

 

Now for my question: From these symptoms I deduce that it's not the CPU that's unable to keep up, but that something else is going on. I'm thinking RAM, maybe?. Any ideas what it could be?

 

2) Because of the above, and also that I'm very close to being able to run sinc-medium on hires files, but even the overclock didn't manage to get me there, I was wondering if it would technically be possible (and if any others would be interested in this) to introduce two additional "intermediate" filters: one in between sinc-short and -medium, and one in between sinc-medium and -long.   

Perhaps a CUDA capable video card would help if you are not already using one.

[satshanti]

1 hour ago, LoryWiv said:

Perhaps a CUDA capable video card would help if you are not already using one.

I am! Not a 4090 unfortunately, but a 3060 Ti. I have the CUDA offload set to greyed out, so only the convolution is offloaded to the GPU. I've tried full offload, and it does free up the CPU and allows me to run some of the filters I would otherwise not be able to run, but the thing is that on my system, half-engaged CUDA sounds better than fully engaged, at least with my previous DAC. Now that I've got a new one, I will revisit this. Thanks for the reminder! 

[bogi]

from fb page:

Quote

 

HQPlayer 5 Desktop 5.3.1 released.

IIR to FIR delay compensation. Support for changing filter/dither/modulator on the fly. Add support for changing input rate through control API. Support for LosslessSwitcher on macOS.

 

I like the ability to switch modulator in a fraction of second!
Filter switching is also much easier.

It makes finding the best combination much easier to fit the recording played. 🙂

[bogi]

Miska, could you please show a hqp5-control example for "Add support for changing input rate through control API" ? Does it address also adding input URI into transport?

[Sunny_Player]

Dear all, good evening. 

 

@Miska

Can you explain to me how the sound changes when I adjust the FFT filter length? Thank you!

 

Have a nice evening. 

 

[bogi]

21 minutes ago, Sunny_Player said:

Can you explain to me how the sound changes when I adjust the FFT filter length?

 

See