HQ Player

[Miska]

1 minute ago, dericchan1 said:

Is there anything negative to have multiple matrix pipeline settings and swap them on the fly in client?

 

No, you just have multiple configurations you can switch.

 

1 minute ago, dericchan1 said:

say I have set different convolution filters with different bass boost settings and I would just load the one with higher bass boost when I like a bit more bass in this track and switch back to another matrix setting on other tracks?

 

That's perfectly fine to do.

 

[Lukasluis]

For a almost a week now I have been trying to download the latest Hqp-embedded using the Audiolinux menu but i'm getting timeouts and/or errors. I have raised the issue in the Audiolinux thread but was advised that the build is hosted at the Signalyst server (see below quoted reply). Is you limiting the upload speeds of your servers. I hate to state the obvious but my downloads from other sites/servers are blisteringly fast!

 

"HQplayer is downloaded from Signalyst server with Audiolinux PKGBUILD."

 

[stefano_mbp]

7 minutes ago, Lukasluis said:

but i'm getting timeouts and/or errors.

Just downloaded (in Italy): start 18:39 end 18:42 … not lighting but fast enough 

[Lukasluis]

3 minutes ago, stefano_mbp said:

Just downloaded (in Italy): start 18:39 end 18:42 … not lighting but fast enough 

It looks like there is a filtering in place when the connection is coming from Singapore

[Mops911]

2 hours ago, Miska said:

Do you hear any differences between any digital filters?

Yes, but the filters also react how the DAC works, doesnt it?

ok, when we accept to hear more than 20kHz, then of course it makes a difference. I guess the science is not settled ;-) 

Last, the article brings up the question if a audio chain can reproduce ultrasonics correctly or if it adds other artifacts in the analogue stages. And if we rather concentrate our efforts on the actual audio band or not. And no, I cant answer these questions. But thanks @Miska for your thoughts

 

 

[Mops911]

3 hours ago, woshifeng3627 said:

can't see this video website, I need you to roughly say what content!

It says the "Mistery of Music I Mysticism" A Indian Yogi and professor talks how fundemantal music is, much more than the word. A universal truth...

From the description

Hazrat Inayat Khan (1882 –1927) was an Indian professor of musicology, singer, exponent of the saraswati vina, poet, philosopher, and pioneer of the transmission of Sufism to the West. He initially came to the West as a representative of several traditions of classical Indian music, and established an order of Sufism (the Sufi Order) in London in 1914. By the time of his death in 1927, centers had been established throughout Europe and North America, and multiple volumes of his teachings had been published. Inayat Khan's teaching emphasized the oneness of God (tawhid) and the underlying harmony of the revelations communicated by the prophets of all the world's great religions. His discourses treated such varied subjects as religion, art, music, ethics, philosophy, psychology, and health and healing.
 

[Jud]

3 hours ago, Miska said:

 

In addition, it is too simplistic to view transient response from plain frequency response point of view. Limiting signal bandwidth to 20 kHz has much wider repercussions than just some disappearing frequency components.

Beyond the bone conduction information you cited with regard to frequency, there is also the matter of perception of timing. The majority of experiments find we can perceive timing differences as brief as 10 microseconds, which would correspond to a frequency of 100kHz. (Kunchur’s paper gave a result of 4 microseconds, but since this is controversial with many people, let’s stay with the majority.)

[1laraz]

On 4/21/2024 at 10:41 PM, Miska said:

It says Intel ARC graphics, so integrated graphics in the CPU. So no CUDA offload.

 

Technically it would be possible to run some offload on that GPU too.

If only you could add some offload on Intel ARC / Iris Xe graphics, please! That will be so much appreciated by us, the NUC users! 🙏🙏🙏

[bogi]

8 hours ago, Mops911 said:

But once something went through a ADC and is cleaned up digitally there should not be any artifacts left. The reason digital is so much better than analog as filters work as they should ;-) 

 

But once something went through a ADC and is cleaned
How is cleaned ??

The reason digital is so much better than analog as filters work as they should ;-) 
Yes, brickwall filters, required for downsampling to 44.1k,  produce ringing as they should. Then that ringing interacts with valid audio signal and intermodulates into audio band as distortion - as it should.

[liyin123]

2 hours ago, 1laraz said:

如果您可以在 Intel ARC / Iris Xe 显卡上添加一些卸载功能就好了！我们 NUC 用户将非常感激！🙏🙏🙏

我也非常同意这个不错的建议，如果技术上可行的话，会把英特尔的集成显卡充分利用起来，这样英特尔NUC在audiolinux系统下也真正发挥了更多作用，也会有更多的用户使用HQPLAYER这个美妙的软件。期待啊...

[Mops911]

2 hours ago, bogi said:

Yes, brickwall filters, required for downsampling to 44.1k, 

Why would digital brickwall filers create aliases or ringing? thats analog filters...

 

 

[Miska]

Server hosting HQPlayer Embedded is on 500 Mbps internet connection.

 

I have one server also in Hong Kong for some of the things, and likewise my connection to that one is rather slow.

 

Hard to say what kind of things are going on on the routes the traffic happens to take for long distances.

 

[Miska]

11 minutes ago, Mops911 said:

Why would digital brickwall filers create aliases or ringing? thats analog filters...

 

Higher the filter order, longer it's impulse response is and thus more it rings. So steeper the filter, more it rings (a bit simplified).

 

[Mops911]

shoot, the moment I think I understood a bit more, back to square one.... ;-)

[woshifeng3627]

15 hours ago, Mops911 said:

It says the "Mistery of Music I Mysticism" A Indian Yogi and professor talks how fundemantal music is, much more than the word. A universal truth...

No matter how good the foundation of management is not as good as their own ears! Sometimes what you hear is different from what you describe!

[Fredc]

8 hours ago, Mops911 said:

Why would digital brickwall filers create aliases or ringing? thats analog filters...

 

In any case, it makes no sense technically that downsampling will improve the audio. 

[wolft]

Hello, Fedora upgrade to Version 40 does uninstall the latest HQPlayer5Desktop. Reinstall under Fedora 40 not supported.

Message: -nothing provides libQtNetwork.so.6(Qt_6.5_PRIVATE_API)(64bit) needed by hqplayer5desktop-5.6.1.fc39...

[stefano_mbp]

31 minutes ago, wolft said:

Hello, Fedora upgrade to Version 40 does uninstall the latest HQPlayer5Desktop. Reinstall under Fedora 40 not supported.

Message: -nothing provides libQtNetwork.so.6(Qt_6.5_PRIVATE_API)(64bit) needed by hqplayer5desktop-5.6.1.fc39...

 

[Mops911]

2 hours ago, Fredc said:

11 hours ago, Mops911 said:

Why would digital brickwall filers create aliases or ringing? thats analog filters...

 

In any case, it makes no sense technically that downsampling will improve the audio. 

 

read the article and watch the videos. You might learn something ;-) Obviously when you lower sampling frequency you give something up. But by theory (Nyquist) all you give up is bandwidth. And the debate is apparently ongoing how much we hear. Scientifically, many agree that 20kHz is the upper bound. so why sample at 96kHz which gives you an upper limit of 48kHz. So if you downsample from 48 to 24 you dont loose anything you would hear..

[bogi]

3 hours ago, Mops911 said:

So if you downsample from 48 to 24 you dont loose anything you would hear..

 

You don't seem reflect what was written above ... I'm not sure how much such kind of "discussion" can be helpful for you.

 

 

 

 

 

[Miska]

6 hours ago, stefano_mbp said:

 

 

Good that you reminded me to update that! Download buttons are now for Fedora 39 since Nvidia finally moved forward. Embedded is still supported on both 37 and 39 though.

 

But your point is of course still correct, if one wants to use HQPlayer on Linux, I would recommend not to upgrade to the fresh new release until HQPlayer has moved forward. And I can do that earliest when Nvidia does so (if we want to keep CUDA offload capability on Fedora as well).

 

[woodroww]

Miska,

 

I have re-ripped from ISO to DSF a number of files from multiple SACD's, Both with sacd_extract and with ConverteR 48x44.

 

I pretty sure as that these rips are fine. When I play these DSF's fine as DoP with Squeezelite they're find They are send the the same 1 bit DAC in my T+A HA200 as the HQPlayer DSF's are. But when I play them with HQPlayer, as DirectSDM, there is a small, but annoying, squeek in between. I believe that it occurs at the start of the track.

Both from the same Holo Audio Red.

 

Can you please guide me in how to solve this?

 

Many thanks!

 

P.S. If you need more information on my system etc. please let me know. For you troubleshooting will be a cinch, for me it's not possible.

[Miska]

15 minutes ago, woodroww said:

I have re-ripped from ISO to DSF a number of files from multiple SACD's, Both with sacd_extract and with ConverteR 48x44.

 

The older sacd_extract is known to set DSF sizes incorrectly, so that the tails contain unnecessary data. Which results in snap-crackle-and-pop. I don't know about the other one.

 

15 minutes ago, woodroww said:

I pretty sure as that these rips are fine. When I play these DSF's fine as DoP with Squeezelite they're find They are send the the same 1 bit DAC in my T+A HA200 as the HQPlayer DSF's are. But when I play them with HQPlayer, as DirectSDM, there is a small, but annoying, squeek in between. I believe that it occurs at the start of the track.

 

Are you able to also reproduce this with most commercially available DSD content, for example from nativedsd.com?

 

Are you playing in album mode or in playlist mode? Pop suppression algorithm in playlist mode can deal with minor pops, but not large ones.

 

[Fredc]

12 hours ago, Mops911 said:

 

read the article and watch the videos. You might learn something ;-) Obviously when you lower sampling frequency you give something up. But by theory (Nyquist) all you give up is bandwidth. And the debate is apparently ongoing how much we hear. Scientifically, many agree that 20kHz is the upper bound. so why sample at 96kHz which gives you an upper limit of 48kHz. So if you downsample from 48 to 24 you dont loose anything you would hear..

Even if you don't lose anything, you are not gaining anything other than some disk space which is irrelevant. Why hang your hat on one article when there are so many contrary ones. 

[Gato]

Under Windows :

• With hqp5-control.exe (delivered with HQPlayer5Desktop-setup-561.exe):
  • When used from the cmd console (... or a .bat), the response to C:\b>hqp5-control.exe --discover is:
    • stdout is empty
    • stderr is : [::ffff:192.168.1.51] 'HQPlayerEmbedded' - Signalyst HQPlayer Embedded 5
  • When used from a program (tested with c# via "new Process" and electron via "spawn"), the response to "C:\b>hqp5-control.exe —discover” is always empty and no error is reported:
    • stdout is empty
    • stderr is empty
  • When used from a program (tested with c# via "new Process" and electron via "spawn"), the response to "C:\b>hqp5-control.exe -causing-error-discover” the response report an error:
    • stdout is empty
    • stderr is: "error: Host not found"
• So, under Windows, why are stdout and stderr always empty when, a program creates a sub-process that executes "C:\b>hqp5-control.exe --discover"?
• Have I missed something or is there something to look for in the hqp5-control .exe?

 

PS:

Under OSX and under Linux there is no problem, but to be noted that the response is always present in stderr never in stdout.

 

Thanks for helping!