[sendler2112]

HQPlayer is a very dangerous program for someone like me to get started with. I just sat and listened to the same 30 seconds of the same three songs for an hour and a half. Back and forth between LAN PCM and LAN DSD and USB PCM into a Gustard A26. And I haven't even started to experiment with any of the upsampler/ filter settings and have just been using the defaults as they come loaded . P-S-G long/ P-S-G-Hires-LP/ TPDF. Other than sample rate. I have to take it back that I said I thought USB sounded better than the LAN. I find the LAN is a bit better even with just cheap Cat6 cables linking the DAC and the laptop through a cheap TP-Link router. Contrary to theory, I find LAN PCM 384 sounds better than DSD 256 with the Gustard A26. And PCM 768 sounds better yet. Blacker, with more empty space between the central images and slightly deeper with better trailing edge decay. Some people who listen to nightclub small combo jazz singer stuff might prefer the DSD which has a smoother, warmer, but less detailed sound. More analog? DSD 512 gains back some of the advantage of fine detail from the PCM 768 but it had my cooling fan screaming for mercy using 25% of the processor. DSD256 only uses 13% and PCM 768 is 4%. And DSD 512 still didn't quite catch the transparency of PCM 768 which is apparently a breeze to process.

 

[EMINENT]

HQPlayer is a very dangerous program for someone like me to get started with. I just sat and listened to the same 30 seconds of the same three songs for an hour and a half. Back and forth between LAN PCM and LAN DSD and USB PCM into a Gustard A26. And I haven't even started to experiment with any of the upsampler/ filter settings and have just been using the defaults as they come loaded . P-S-G long/ P-S-G-Hires-LP/ TPDF. Other than sample rate. I have to take it back that I said I thought USB sounded better than the LAN. I find the LAN is a bit better even with just cheap Cat6 cables linking the DAC and the laptop through a cheap TP-Link router. Contrary to theory, I find LAN PCM 384 sounds better than DSD 256 with the Gustard A26. And PCM 768 sounds better yet. Blacker, with more empty space between the central images and slightly deeper with better trailing edge decay. Some people who listen to nightclub small combo jazz singer stuff might prefer the DSD which has a smoother, warmer, but less detailed sound. More analog? DSD 512 gains back some of the advantage of fine detail from the PCM 768 but it had my cooling fan screaming for mercy using 25% of the processor. DSD256 only uses 13% and PCM 768 is 4%. And DSD 512 still didn't quite catch the transparency of PCM 768 which is apparently a breeze to process.

Congrats.
Now do it again with the rest of the dithers and modulators.
Then repeat with each and every filter.
Then report.

 

[MLGrado]

I know nothing about ps audio dacs, is the direct stream actually dsd direct?

the output stage is.

Imagine HQPlayer implemented on FPGA inside the DAC for digital filtering and volume control. It won't be anywhere near as flexible or high fidelity as HQPlayer, but, the general idea is the same.

After the FPGA, all audio is modulated back to true 1-bit, 128fs. (Maybe 1 bit, 256fs in later models.. I am not sure about this). The output stage is similar to DSC DAC with multiple element, unary coded converter. The analog output uses output transformers.

So it is similar in some ways, quite different in others.

I do know they make constant 'improvements' via firmware. The firmware updates all seem to update the modulator parameters and maybe a few other things.

 

[sendler2112]

Congrats.
Now do it again with the rest of the dithers and modulators.
Then repeat with each and every filter.
Then report.

I'm sure that others here have already done that. So all I have to do now is look up the results that they posted. But then I might not agree with their preferences. In fact I am already disagreeing. Having HQPlayer convert WAV files to DSD 256 over LAN into the Gustard A26 with its AK4491EX chips is theoretically the best way to feed it since this format can skip the onboard conversion that PCM must undergo, and go straight to the analog conversion stage. But PCM 768 over LAN sounds better to me.
Unless there are better settings that I could be using in HQPlayer.

 

[Darion]

Friends, who uses what filters in hqplayer.
Somehow I can’t find the optimal one with chord hugo tt2.
Each variant of popular filter combinations sounds quite bland. I don't even know.
Of course, I’m still waiting for holo may kte, everything will be beautiful there. But with hugo tt I did not get satisfaction.

 

[sendler2112]

A couple more observations from a new user of HQPlayer and Roon with Qobuz: Streaming a file doesn't sound quite as good as playing the exact same 24/192 WAV file that I purchased onto my harddrive. And Dropping the WAV file from my harddrive into HQP and playing it from that screen sounds better than letting Roon send the exact same file from my harddrive to HQP. With the latter sounding more like the streamed file. I have enjoyed the added content that Roon scrapes up but would consider skipping the extra monthly charge. If HQPlayer wasn't prone to locking up when I change tracks from its screen. Which never happens when I change tracks in Roon.

The big advantage of running Roon is that 80% of my listening is streamed from Qobuz so I would need the way to link it to HQP that Roon provides. If Qobuz was smart they would add the HQP as a choice to their player selection at the bottom right hand corner tab and skip the middle man. I don't see why Qobuz couldn't easily do everything streaming its files that Roon now does for it eventually including the added content scraping.

 

[Failed Engineer]

Qobuz is integrated into HQP Client so try that.

 

[GoldenSound]

So I did a bit of looking into it. This is a comparison of no correction vs the Holo May correction using 705.6khz 32 bit output from HQP (Sinc-Mx and TDPF just in case but shouldn't matter for comparing this test). This is the comparison of the actual digital output, so no potential inaccuracy induced by a DA/AD loop.

First an FFT of both files:

No clear differences generally, though a very slight difference at the highest freqs. If we look at the actual delta of spectra we can see there is a slight rise in freq, upto just over 0.5dB at 20khz.

This also then creates a very small phase shift at high freqs (as would be expected of a minimum phase adjustment I suppose though I've not yet checked to see if this corresponds perfectly/has no excess phase difference). But it's less than half a degree regardless

So it seems that basically the DAC corrections are small adjustments to compensate for whatever analog filter the DAC may be using that would otherwise cause a very slightly non-flat FR?

 

[1laraz]

No clear differences generally, though a very slight difference at the highest freqs.

I am hearing more emphasis on highs with the correction enabled

So it seems that basically the DAC corrections are small adjustments

Small adjustments although substantial RAM usage when both the first "matrix pipeline" and the DAC correction are checked. But I am not sure if this is the only right way to enable the new function. Will it also work with the "matrix pipeline" unchecked?

UPD: OK, need to dig into the log file - Miska said elsewhere that there will be a confirmation entry "DAC correction post-process active"

 

[markkr]

New to HQP, I have 5.5 embedded deployed and functioning but would like to update to 5.6. I'm struggling to find any documentation on the process to get to 5.6, can anyone point me in the direction?

 

[bogi]

First an FFT of both files

What for file did you use? What length in seconds?
Try an impulse or a short music passage with one strong transient.
I would try higher sample rate input content. With 44.1k content you don't see the response at higher frequencies.

 

[GoldenSound]

What for file did you use? What length in seconds?
Try an impulse or a short music passage with one strong transient.
I would try higher sample rate input content. With 44.1k content you don't see the response at higher frequencies.

File was about a minute long. Was intending to get a better look than an IR itself would give in case there was some other stuff happening

I'm less interested with what happens with high sample rate input given as most people are usually not doing that, and 44.1khz shows you up to 22.05khz already.
I'll check with some 192khz input stuff at somepoint though

 

[ra990]

So I did a bit of looking into it. This is a comparison of no correction vs the Holo May correction using 705.6khz 32 bit output from HQP (Sinc-Mx and TDPF just in case but shouldn't matter for comparing this test). This is the comparison of the actual digital output, so no potential inaccuracy induced by a DA/AD loop.

First an FFT of both files:



No clear differences generally, though a very slight difference at the highest freqs. If we look at the actual delta of spectra we can see there is a slight rise in freq, upto just over 0.5dB at 20khz.



This also then creates a very small phase shift at high freqs (as would be expected of a minimum phase adjustment I suppose though I've not yet checked to see if this corresponds perfectly/has no excess phase difference). But it's less than half a degree regardless



So it seems that basically the DAC corrections are small adjustments to compensate for whatever analog filter the DAC may be using that would otherwise cause a very slightly non-flat FR?

Very interesting. Whatever it's doing it's definitely using resources. My gaming PC that I've turn into a dedicated HQPlayer server is now choking on anything over 44/48k rates at DSD512 for the first time.

 

[GoldenSound]

Very interesting. Whatever it's doing it's definitely using resources. My gaming PC that I've turn into a dedicated HQPlayer server is now choking on anything over 44/48k rates at DSD512 for the first time.

I've not looked into DSD yet just pcm.

 

[GoldenSound]

Not to sound crass but why would someone have HQplayer and not input high sample rates to these dacs.
I think a proper test would be at minimum 352khz /DSD256.

I mean into hqp. Most people are feeding 44.1khz into hqp then 384/768khz out into the DAC.

The tests above were with 44.1khz into hqp and 705.6khz out