Jump to content
IGNORED

HQPlayer Pro


Recommended Posts

46 minutes ago, bmoura said:

 

I haven't tried DSD 512 Stereo files with Roon, but the ones from NativeDSD work with JRiver Media Center 26 for Windows (64 bit version) as well as HQ Player 4 Desktop here.  I've also successfully played the NativeDSD Stereo DSD 512 files with Tune Browser for Windows. 

 

As for Audirvana for Windows, I've had it crash on files at lower bit rates than DSD 512.  So I stopped using Audirvana here.

 

Thanks for the response. Yes I don’t know what the difference could be between the NativeDSD 512 files and the ones I’m producing with HQPlayer Pro on my MacBook Pro, unless it is an unintended byproduct of the trial version that I’m using. 
 

I was only using Audirvana as a front end via UPNP to my HQPlayer OS server when I wanted to play files on my MacBook that I hadn’t transferred to the NAS that Roon and HQPlayer use as a library. I don’t use Windows.
 

My main reason for trialling HQPlayer Pro was to see if I could produce files to use with Roon without the HQPlayer server in the chain. 

Link to comment
9 hours ago, craighartley said:

Has anyone worked out why 512DSD files produced with HQPlayer Pro won't import into Roon or why they crash Audirvana (which shows them as 557DSD)? I've tried editing the metadata to make sure that is behaving properly, but it doesn't seem to stem from that. They play with HQPlayer, but that's not why I wanted to make the files.

 

Based on that display, by any chance the files are with 48k-base sampling rate (48k x512)? I'm surprised if such would confuse those applications so much, but that is one possibility and the 557DSD hints towards exactly that.

 

(48000*512)/44100=557

 

In addition to many DACs not supporting 48k-base DSD, could be that there is also software that doesn't.

 

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to comment
33 minutes ago, Miska said:

 

Based on that display, by any chance the files are with 48k-base sampling rate (48k x512)? I'm surprised if such would confuse those applications so much, but that is one possibility and the 557DSD hints towards exactly that.

 

(48000*512)/44100=557

 

In addition to many DACs not supporting 48k-base DSD, could be that there is also software that doesn't.

 

Brilliant! Thanks! - that’s exactly what is happening. The errors were all with 48k based 512 DSD. I’ve just tried two 44.1k based files and they work perfectly with both Roon and Audirvana.
 

I’ll raise it on the Roon forum. Easy to work around, except of course it rules out using non-apodising filters with 48k based source files. 
 

The NativeDSD files must be limited to 44.1k-512 (even though they employ an apodising filter). Sensible decision. 

Link to comment
2 hours ago, craighartley said:

Brilliant! Thanks! - that’s exactly what is happening. The errors were all with 48k based 512 DSD. I’ve just tried two 44.1k based files and they work perfectly with both Roon and Audirvana.
 

I’ll raise it on the Roon forum. Easy to work around, except of course it rules out using non-apodising filters with 48k based source files. 

 

There are couple of non-apodizing filters that can do conversion between rate families.. For example poly-sinc-xtr and poly-sinc-hb.

 

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to comment
50 minutes ago, Miska said:

 

There are couple of non-apodizing filters that can do conversion between rate families.. For example poly-sinc-xtr and poly-sinc-hb.

 

Thanks, I’d forgotten that. Actually one reason for exploring HQPlayer Pro was to try the effect of the two filters I can’t do in real time with HQP OS - sinc-L and closed-form-16M. I can try them with 44.1x files though. 

Link to comment
  • 1 month later...
2 minutes ago, ASRMichael said:

@Miska surely the other benefit is not have your CPU running at high rates/temps creating noise? 

One of the best upgrades you can make for hardly any money is to move your DAC/audio chain away from being directly connected to your PC.

Even just grabbing a $40 raspberry pi, putting ropieeeXL (which supports roon, DNLA, and HQPlayer NAA) on it and streaming to that rather than a local connection, makes a very nice difference for not much money at all.

Higher performance streamers are available of course but for the price hard to argue with a pi. And that way you'll get a better result no matter how busy/idle your PC is

https://youtube.com/goldensound

Roon -> HQPlayer -> SMS200 Ultra/SPS500 -> Holo Audio May (Wildism Edition) -> Holo Audio Serene (Wildism Edition) -> Benchmark AHB2 -> Hifiman Susvara

Link to comment
8 minutes ago, GoldenOne said:

One of the best upgrades you can make for hardly any money is to move your DAC/audio chain away from being directly connected to your PC.

Even just grabbing a $40 raspberry pi, putting ropieeeXL (which supports roon, DNLA, and HQPlayer NAA) on it and streaming to that rather than a local connection, makes a very nice difference for not much money at all.

Higher performance streamers are available of course but for the price hard to argue with a pi. And that way you'll get a better result no matter how busy/idle your PC is

Been there, done that! One box solution works better for me, only once you address clocks and power. Not to say never, as my next project will probably be dual CPU with HQPlayer desktop on 1 CPU and NAA on another. It never ends!!

Link to comment
30 minutes ago, ASRMichael said:

Been there, done that! One box solution works better for me, only once you address clocks and power. Not to say never, as my next project will probably be dual CPU with HQPlayer desktop on 1 CPU and NAA on another. It never ends!!

Im looking to build a Roon core/HQP processing dedicated machine at the moment, but will definitely always use a good streamer for the endpoint.

Currently using the SMS200 Ultra which I just cannot get the same level of quality out of any actual direct-to-machine connection from. 

https://youtube.com/goldensound

Roon -> HQPlayer -> SMS200 Ultra/SPS500 -> Holo Audio May (Wildism Edition) -> Holo Audio Serene (Wildism Edition) -> Benchmark AHB2 -> Hifiman Susvara

Link to comment
18 minutes ago, GoldenOne said:

Im looking to build a Roon core/HQP processing dedicated machine at the moment, but will definitely always use a good streamer for the endpoint.

Currently using the SMS200 Ultra which I just cannot get the same level of quality out of any actual direct-to-machine connection from. 

I would look through music server builds thread. Things have moved alone very much since SMS200 Ultra. 

Link to comment
1 hour ago, ASRMichael said:

@Miska surely the other benefit is not have your CPU running at high rates/temps creating noise? 

 

If you use a NAA it becomes irrelevant. That's one of the points behind a NAA.

 

Otherwise it is very system dependent. For example I'm running the EVGA NuAudio card inside a powerful i7-8086K server running filters and ASDM7EC modulator and still the analog output is very clean (measured). It still depends on the PSU and motherboard too, and such. In this case running from a high power Seasonic Prime Titanium (TX) PSU.

 

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to comment
16 minutes ago, Miska said:

 

If you use a NAA it becomes irrelevant. That's one of the points behind a NAA.

 

Otherwise it is very system dependent. For example I'm running the EVGA NuAudio card inside a powerful i7-8086K server running filters and ASDM7EC modulator and still the analog output is very clean (measured). It still depends on the PSU and motherboard too, and such. In this case running from a high power Seasonic Prime Titanium (TX) PSU.

 

I've been through the NAA route and much prefer single server, probably before the quality of the server. I would maybe agree if both my server and NAA were of similar quality, otherwise I don't see the point.  

Link to comment
30 minutes ago, Miska said:

 

If you use a NAA it becomes irrelevant. That's one of the points behind a NAA.

 

Otherwise it is very system dependent. For example I'm running the EVGA NuAudio card inside a powerful i7-8086K server running filters and ASDM7EC modulator and still the analog output is very clean (measured). It still depends on the PSU and motherboard too, and such. In this case running from a high power Seasonic Prime Titanium (TX) PSU.

 

Just to add what I have said above, I have a NUC with separate OCXO clocks, main board and OCXO board powered by SJ DC3 LPS's, I have used this as for NAA. It doesn't sound as good as my Server. Nowhere as good. As I said above the NAA might sounds as good if the server and NAA where the same. 

Link to comment
49 minutes ago, ASRMichael said:

I would look through music server builds thread. Things have moved alone very much since SMS200 Ultra. 

I have tried a few server and streamer options. Dedicated streamers were always better to my ears, and my two favourites were the pi2aes (if using i2s) and the SMS200 Ultra (with a good PSU).

I'm not sure what the argument for using a server rather than a streamer would be other than convenience. Certainly putting more powerful/noisy components in the endpoint would not be a GOOD idea.

Though my main dac has galvanic isolation on the USB and a PLL that attenuates jitter to a ridiculous degree. So the source device with my main dac in particular makes almost no difference. But on some other dacs I've got like the ADI-2 the source device can be a very noticeable change

https://youtube.com/goldensound

Roon -> HQPlayer -> SMS200 Ultra/SPS500 -> Holo Audio May (Wildism Edition) -> Holo Audio Serene (Wildism Edition) -> Benchmark AHB2 -> Hifiman Susvara

Link to comment
3 hours ago, ASRMichael said:

I've been through the NAA route and much prefer single server, probably before the quality of the server. I would maybe agree if both my server and NAA were of similar quality, otherwise I don't see the point.  

 

NAA should be a low power device, the ones I use have Intel Atom CPU and consume around 5W or less power.

 

While most servers I have are high power devices consuming over 200W and PSU capacity ranging from 500 to 1000W. Another point is that it doesn't matter if it has loud cooling fans and such, since it can be located far from the listening room, in a basement or something.

 

For servers directly connected to DACs I use Gigabyte motherboards that have their DAC-UP USB ports.

 

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to comment
3 hours ago, edwardsean said:

I just downloaded 4.8 to try out the new Sinc-L filter. It sounds fantastic but now the converted files seem to have a few seconds of silence at the start of the track. I've not experienced this before. Is this intentional or is something wrong?

 

That is delay of the filter. Since the filter is about two seconds long.

 

If you convert an album instead of single files, you get the delay only at the beginning and at the end of the album, but at track transitions. If you convert individual tracks, then the delay is with each track.

 

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to comment

sorry wrong thread!!🙄

Mac Mini with JRMC26 or Audirvana  / Raspberry4B_4GB(GentooPlayer_LMS) / Raspberry Rpi3B+: Allo DigiOne(GentooPlayer) - M2Tech Evo DAC Two Plus/iPurifier2 - Schiit Vali 2 - Densen DM20pre/30pwr amps - Spendor SP2/3E, Sennheiser HD600 & HD25Aluminum - Audeze Sine

Cables: Vovox, DIY, Furutech. 

Portable sources: iPad, DELL Laptop with JRiver MC26

Link to comment
  • 6 months later...
5 hours ago, bibo01 said:

@MiskaI suppose that, if you wanted to make very long filters available in Pro, a-la-PGGB, you could do it easily, couldn't you?! 

And...are you?! :)

 

Why would I? I could do it for Desktop and Embedded just as well. I see it as pointless exercise.

 

Problem is that either your songs get longer with longer head and tail. Or you truncate your transients except the one in the middle of the song. Which totally ruins the performance. Such long filters actually work better for gapless realtime playback where you only have a lot of delay, but not problems regarding beginning and end.

 

If PGGB uses a billion tap filter, but doesn't make your song billion samples longer, it is doing it wrong and truncating your transients, making performance much worse than with shorter filters and thus more than making point of using longer filters moot.

 

I've said this before, but the whole thing with those very long filters is just silly and pointless. It just screws up your time domain completely. A billion poor taps cannot beat thousand good ones.

 

For me, the interesting thing is to make filter as short as possible while still having as perfect frequency domain performance as possible. IOW, making filter kind of short and long simultaneously, getting as close as possible to the mathematically impossible - at the sweet spot in the middle. So reaching goals of both MQA and Chord (totally opposite), but better in both respects.

 

There has been an undocumented option to adjust length of sinc-M in HQPlayer, but because of the side effects I don't want to make such things as official option.

 

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to comment
On 4/21/2021 at 3:37 PM, Miska said:

Why would I?

Thanks for your reply.

What I was inferring as well, however, that the Pro version can have scope/filters beyond the online capabilities of Desktop and Embedded. I mean, with Pro you can push the boundaries as much as you like - what it's not practical for online, it might be for offline. ...Of course if you see the point :) 

Link to comment
15 hours ago, bibo01 said:

What I was inferring as well, however, that the Pro version can have scope/filters beyond the online capabilities of Desktop and Embedded. I mean, with Pro you can push the boundaries as much as you like - what it's not practical for online, it might be for offline.

 

Only thing where there is advantage with Pro over Desktop at the moment is running EC modulators to higher DSD rates such as DSD512 and DSD1024 which you cannot do in realtime at the moment.

 

Anything else is not limited by current hardware, any PCM upsampling the least... So there are no particular boundaries to push elsewhere.

 

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to comment
  • 1 month later...

Hi, @Miska,

it is really impressive how good red book cd rips sound just applying filter and shaping at  x1, x2 !

 

I I'm running HQP on a mid 2012 dual core i5 mbp, and my dacs are Mojo and a topping Dx3 pro.

Which setting would you suggest to explore?

 

Thank you for your work.

Link to comment

poly-sync-sinc-short-mp, DSD5, X256, Mojo, ATH-R70X. Yellowjackets first album, a 1981 digital recording and..

Robben Ford guitar is flying in the darkness...

... and my mbp too: 103° peak then fan kicks in in vacuum mode to get cores back to 93°.

I 've got a long headphone cable.

Bello, grande, grazie.

Stefano.

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...