bogi

How old is the PC? Does BIOS remember other settings after powering off? Maybe your CMOS battery reached end of life.

IMO it is a digital correction filter, which aims to compensate DAC frequency and phase response based on measurements. At least in audible range there is not much what to correct in frequency domain, so I think it may do here some corrections in time domain. But I'm convinced the corrections are not restricted to 20k but go higher depending on sample rate used. I expect there is a set of correction filters for different sample rates for each supported DAC. Miska can confirm or deny my speculations if he wants. :).

@LewinskiH01 Is your USB disk the #4 Kingston? If so, when you moved #4 to appear on top (to be #1), didn't you forget to exit with saving settings? If you did that correctly, try disabling fast boot - maybe it affects USB boot device detection.

@liyin123 please use Google or Microsoft translator or something similar to post here in English. For example @woshifeng3627 also started to post here in Chinese but then he successfully started to use a translator. When a forum page contains text in other language in many parts, so it is mixed English and other language, one needs to select every foreign language part separately and let it translate to see the content. But other parts remain without translation. It is not easy to read such a page.

I'm sorry but you still didn't explain what the general word "correction" means in this case. Now I at least understand that the correction is not an automatic correction to set a working or optimal sample rate. To get at least some basic idea, please explain the areas where the correction is applied. If you would have a working D300 DAC in hands (I know yours is bricked), would you be able also to generate something like correction for it?

There are more ways how high frequency content can become audible. In order to understand my answer and the subsequent Miska's answer you need to self study the following topics: - sampling and spectral images - anti aliasing requirement during downsampling - digital filters and ringing - intermodulation distortion These topics still don't cover all ways how ultrasonic content can influence what we hear.

Just the explanation is missing ... the link contains only PCM / DSD rates table for supported DACs. So I can only guess what that correction could mean. Does it mean that only the supported rates are provided for rate selection when one of those DACs is detected? Or something other? I see Gustard A26 lists only DSD rates where direct DSD works.

I guess you make some programming error. Try also with some other program producing both stdout and srderr (or write one). C:\Users\bogi>python Python 3.9.13 (tags/v3.9.13:6de2ca5, May 17 2022, 16:36:42) [MSC v.1929 64 bit (AMD64)] on win32 Type "help", "copyright", "credits" or "license" for more information. >>> from subprocess import run >>> import os >>> os.chdir('c:\Program Files\Signalyst\HQPlayer 5 Desktop') >>> os.getcwd() 'c:\\Program Files\\Signalyst\\HQPlayer 5 Desktop' >>> run(["hqp5-control", "--discover"], capture_output=True).stdout.decode() '' >>> run(["hqp5-control", "--discover"], capture_output=True).stderr.decode() "[::ffff:192.168.0.26] 'md1ep5ac' - Signalyst HQPlayer Desktop 5\r\n[2a01:c844:2507:e100:90cb:9367:9dd1:cdfa] 'md1ep5ac' - Signalyst HQPlayer Desktop 5\r\n" >>> It's like you observed in Command Prompt...

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

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.

The main difference is caused by downsampling to 44.1k. That requires steep filter to be used since it needs to operate in very narrow transition band 22.05 - 20 = 2.05 kHz. Recordings which were processed by such a filter are more prone to ringing (transient smear) and that can be audible, depending on music content. Recording which were downsampled to 44.1k using not enough steep filter are then prone to aliasing of out of band content from the area between Nyquist rate and sample rate into audio band. We can hear intermodulation products which are result of out of band ultrasonic content. So yes, we can experience existence of a content above 20kHz, but not directly, only indirectly, when it intermodulates into audio band, or as result of aliasing. Content above 20kHz may exist as result of insufficient filtering (my previous paragraph). Also ringing becomes audible by means of intermodulation distortion. So the things are not so easy as it may look. Higher sample rates allow to use gentler filtering, since the filter transition band gets much wider. That lowers ringing and makes it easier for a filter to reach full attenuation at Nyquist rate, so no aliasing products appear in audio band.

With --get-transport I got transport: 240 "" for 3 different cases: playlist, HTTP stream and device input. Only for album mode I got something different: transport: 2 "G:\AUDIO\..."

Use hqp5-control <HQP-server-host> --playlist-get You get different king of outputs for network stream, device input, playlist and album inputs: bogi@md1ep5ac MINGW64 ~ $ "c:/Program Files/Signalyst/HQPlayer 5 Desktop/hqp5-control.exe" localhost --playlist-get [1] (88200/24/2/4233600) http://192.168.0.178:56923/content/psc.wav {audio/wav} ////HTTP stream 8116.000 (0) "" bogi@md1ep5ac MINGW64 ~ $ "c:/Program Files/Signalyst/HQPlayer 5 Desktop/hqp5-control.exe" localhost --playlist-get [1] (0/0/0/0) {} ////audio:default/44100/2 0.000 (0) "" bogi@md1ep5ac MINGW64 ~ $ "c:/Program Files/Signalyst/HQPlayer 5 Desktop/hqp5-control.exe" localhost --playlist-get [1] (44100/16/2/1411200) file://G:\AUDIO\Zenit\Zenit - Milujem.flac {} ////G:\AUDIO\Zenit\Zenit - Milujem.flac 233.000 (0) "" bogi@md1ep5ac MINGW64 ~ $ "c:/Program Files/Signalyst/HQPlayer 5 Desktop/hqp5-control.exe" localhost --playlist-get [1] (192000/24/2/9216000) file://G:\AUDIO\Electric Light Orchestra\Eldorado (1974) (2015 HDtracks 24-196)\01-Eldorado Overture.flac {} Electric Light Orchestra///Eldorado (Hi-Res)/Eldorado Overture 132.000 (0) "" [2] (192000/24/2/9216000) file://G:\AUDIO\Electric Light Orchestra\Eldorado (1974) (2015 HDtracks 24-196)\02-Can't Get It out of My Head.flac {} Electric Light Orchestra///Eldorado (Hi-Res)/Can't Get It out of My Head 262.000 (0) "" ... Edit: I see album and playlist mode is not distinguished this way but in that case 'hqp5-control localhost --get-transport' makes a difference. bogi@md1ep5ac MINGW64 ~ $ "c:/Program Files/Signalyst/HQPlayer 5 Desktop/hqp5-control.exe" localhost --get-transport transport: 2 "G:\AUDIO\Electric Light Orchestra\Eldorado (1974) (2015 HDtracks 24-196)" bogi@md1ep5ac MINGW64 ~ $ "c:/Program Files/Signalyst/HQPlayer 5 Desktop/hqp5-control.exe" localhost --get-transport transport: 240 ""

Yes, it converts DSD to PCM. Best to feed Chord DACs with 705.6k from HQPlayer. Maybe iFI Neo iDSD 2.

I implemented such a thing named ISO2DSF with help of Miska's free DFF to DSF command line converter about 10 years ago. But already more years ago sacd_extract tool was updated and fixed the truncation bug. ISO2DSD tool (GUI for sacd_extract from Sonore) was also updated. I didn't update ISO2DSF for the newer sacd_extract since it works with the old one as expected. Who is on Windows, foobar2000 with DSD converter component is also easy to use. My 4 years old post with pictures how to use it: https://audiophilestyle.com/forums/topic/60298-older-32-bit-version-of-iso2dsd-for-windows-pc/?do=findComment&comment=1086208 DSD Converter works with 64bit foobar2000 too.