HQPlayer v4 released

Vinyl, things like that, no. However a slimmer playback engine. Maybe.

Are you saying when installing you have the option to not install the GUI? I assume the settings section would be there in order to setup filtering and up-sampling, etc.

That would interest me. When Roon separated to Server install, later to be called Roon Core and having GUI and controller only on your controlling device it made a huge difference.

You see, this topic is very personal. It is hard to come up with description relevant to oneā€™s prioritiesā€¦

Desktop always has some amount of GUI, if you want to get rid of it, thereā€™s the Embedded. But the part dealing with browsing and controlling playback from HQPlayerā€™s library has been split out and made look and feel better while allowing remote control.

Now there are two things slightly like Roon Core, HQPlayer 4 Desktop and even more so HQPlayer 4 Embedded.

While HQPlayer Client (bundled as a separate item with HQPlayer 4 Desktop) works with either HQPlayer 4 Desktop or HQPlayer 4 Embedded. So you can decide to use either Roon as source, or even more optimal, HQPlayer 4 Desktop or Embedded playing a source directly (one extra layer cut off).

Hi Katun

Give HQP poly-sinc-mp (not short, which you already tried) a try also, if you can?

And can you share a screenshot of your complete HQP settings. Just to check what else you can try.

I know you posted some other screenshots when troubleshooting but I donā€™t know which settings you ended up with, for doing your comparisons with Roon.

Embedded is only for Linux, correct? Anyway, it does sound like you are moving it forward which is a good thing. I am not sure that it would give me an improvement in my current configuration.

My configuration is a Falcon Northwest Tiki Windows 10 Pro PC that I have configured and dedicated as a music server. It runs Roon and HQPlayer and nothing else. I have only internal SSDs in the machine (which I load all my music on) and connect via USB to a W4S Recovery and on to my DAC. Both USB cables are AudioQuest Coffeeā€™s. My DAC appears on the top of your list of recommended DACs, the T+A DAC 8 DSD which I feel is absolutely amazing.

I up-sample everything to DSD512/48, 24.6Mhz. Comparing back and forth, using HQPlayer v3 as my playback engine sounds better to me then Roon straight.

I appreciate you taking the time to answer my concerns. While I do not agree with the $200+ pricing for the new version after having v3 for only about a year, I am not sure it would benefit me anyway. I do usually prefer having the latest iteration of all the software I run; on this one I will have to think on it a bit.

Hi Sean,
I used the same filters for comparison in Roon and in HQPe. For comparison I chose the filter that I prefer in Roon. Of course, it was possible to go through a lot of options with various combinations of filters and modulators in Roon and HQPe, but this does not allow a very limited testing time of 30 minutes to be done - it is simply ridiculous. To conduct a full-fledged testing of anything in audio, you need to be able to listen to it for at least an hour or two in a row without haste for several days and not thinking about the fact that now the music will soon break off and you should still have time to listen to something. It distracts from normal listening without letting you relax.

Hi Katun

I donā€™t know if they are the same filters but the reason I asked to try poly-sinc-mp is that maybe it is closer (or better) to the Roonā€™s smooth + min phase you likeā€¦ Maybe, maybe not but I thought maybe itā€™s something to try.

Also try that with ASDM7 modulatorā€¦

I see your fixed volume at the top is 0dB. The recommended is minimum -3dB, to help avoid clipping (depending on tracks played). -6dB should be safer.

-3dB is fine 99% of the time for ā€œFixed Volumeā€. I would set ā€œStartup Volumeā€ to 0dB.

Itā€™s not possible to say this without seeing/knowing the details of a personā€™s music libraryā€¦

For me this is unacceptable since I listen to DSD on my DAC in Direct DSD mode without intermediate transformations including changing the bit depth to change the volume.
And when testing I used the Mapleshade labelā€™s audio record without compression and artificially raising the volume to the maximum.

You can setup output mode to ā€˜autoā€™. DSD playback should then always be in ā€˜Direct DSD modeā€™ and your PCM content will be up-sampled to DSD256 per your current settingsā€¦ I think - @jussi_laako can double check if Iā€™m wrong/right.

No problems. I have too many recordings that clip with -3dB so I now just leave it at -6dB and donā€™t worry about it. I donā€™t want to be checking track to track, or album to album.

Clipping levels can vary from person to person, since we have different music libraries.

The criterion for technical correctness in audio is hearing. The naturalness of the sound is estimated not by formulas but by hearing.

No. Hearing perception is purely subjective. The criteria for technical correctness in audio are found in measurements. What sounds good to you may be nothing like what the artist recorded in the studio and put onto vinyl or CD. As Jussi suggests, you canā€™t have correct and great sound without fidelity.

No, that would play PCM as as PCM. But if you enable Direct SDM and select SDM as output mode, then DSD sources are played unprocessed and PCM is upsampled to DSD.

Technical correctness is verified by analysis and measurements.

1 Like

Yes, for various reasons. One being that only Linux allows me to build fully custom OS tailored for the single purpose.

Windows and macOS come with a GUI anyway. Linux is best suited for systems that donā€™t have any graphical output. Linux is also allows running HQPlayer on various architectures, like ARM64.

You cannot do such, HQPlayer and Roon donā€™t have any filters in commonā€¦ Thereā€™s absolutely nothing in common between the two.

1 Like

Ah yes, silly me.

This was your response to a question about specific sound quality improvements made to HQP 4. Is it fair to say that the filters themselves are the same in HQP 3 vs 4? For instance, poly-sinc in HQP 3 is the same as in HQP 4? And the Gauss1 dither is the same in both HQP 3 & 4?

And that the only difference, on a sound quality front, is that HQP 4 has ā€œsome efficiencyā€ improvements which may make a difference in oneā€™s individual sound system?

Do I understand you correctly?

Thanks

By ā€œsound qualityā€, people usually mean something subjective that covers entire system. Not just mathematical formula of the DSP processing.

Mathematically, same settings on both produce same results (with some caveats). v3 just doesnā€™t have all the same DSP features v4 has.

Possibly maybe, because the result as whole in the system as entirety is a little bit more than just the mathematical part constrained to digital domain. Assuming using exactly same DSP settings.

If you for example go for the new modulators or something, then the results are objectively different also mathematically in digital domain.

1 Like

Thank you for the thoughtful reply. What specific DSP features does v4 have the v3 does not?

Thanks

These are described on the Signalyst website and include a real time input, DSD 1024 upsampling and three new high precision (EC) modulators.

I havenā€™t upgraded yet as I understand the current implementation of the new modulators wonā€™t run on my i7 server at DSD 512. But Jussi has spoken about future optimisations and that may bring them within reach.

1 Like