HQ Player discussion thread

When it’s ready… :wink:

For PCM outputs the extra attenuation doesn’t matter, because 32-bit integer cannot represent more than 192 dB (that’s where the spec came from). But the number of taps has effect. For DSD256 and higher outputs, also the attenuation begins to have some effect.

I personally use poly-sinc-short-mp for listening.

Number of taps doesn’t matter for that. For 16-bits quantization over 96 dB stop-band attenuation is enough.

Just tested a windowed Sinc with 2M taps, no hiccups processing 705.6kHz file in HQP! Not done any listening yet, apparently I cannot do flac at 705.6Hz and zero padded wav files are insanely huge! Will try to upsample a few tracks and listen later today.

@andybob - I just tried upsampling a few 44.1kHz tracks to 705.6kHz using 2M tap windowed Sinc filter in HQplayer, when listening DAVE I hear improved clarity and transparency, but not the day-and-night difference I was expecting. But then there are so many design choices and the parameters I used may not be the best, I chose a Blackman window and cut-off at 20kHz.

For those who are inclined to give it a go: https://goo.gl/gZF8Lc
I have Put two tracks (both original and upsampled by zero padding). And also the filter ‘sinc.wav’. The upsampled version is zipped. I tested original version with all processing disabled and the upsampled version with convolution enabled (CUDA + Pipline SDM). Would like some feedback from anyone else using DAVE too.

To pick up this thread. In the last few days I’ve been listening to Tidal upsampling in HQP to DSD 512 using poly-sinc-xtr-mp-2s filter and AMSDM7 512+fs modulater. It’s my new favourite setting.

1 Like

How does this compare to the other poly-sinc filters in terms of pre- and post-ringing?

Steep in the frequency domain, long in the time domain – lots of symmetrical ringing, Chord DAVE style.


1 Like

Am I correct in thinking that there is no way currently to select a different filter automatically depending on the input format?

Only with PCM/DSD granularity. The “Auto” setting in the Output menu on the main page (rightmost pull down menu) will use the PCM Output for PCM and the DSD Output for DSD.

Yeah, I thought so. Would really like to apply a different filter to RBCD and 2x material.

Let’s flag @jussi_laako and see if it’s feasible.

I think it should be doable without too much trouble. Rather than having a filter specified separately for every rate, I think the feasible way would be to have two filters, one for 1x rates and another for hires content. Having a filter for specific source rates is not feasible because there are no limitations for possible input rates (and not for output rates either), so it is not bounded set. But specifying boundaries what is considered “1x rate” is feasible.

By 1x, do you mean when upsampling 16/44.1?

Yes, both 44.1 kHz and 48 kHz, since they are the 1x bases of their respective rate set families.


Which filters would you use @joel and why ?

I need to listen, but my gut feeling is to use an apodizing filter for red book and a very short filter for 2x and above inputs.

I have been such a big fan of the poly-sinc-short-mp filter that I thought you had to be nuts. Then I listened to it. Boy, was I wrong! :smile:

For me it adds clarity without adding harshness or sibilance, which is a really neat trick. It may reduce the width and depth of the sound stage (using closed-back headphones), but I’m really splitting hairs here. I wouldn’t even bet a cup of coffee that tomorrow I’ll still think the sound stage is compacted. And I’m only listening to DSD256.

Thanks for posting your new favorite filter!


I’m using this setting with closed back headphones as well. It basically puts you in the front row but gives the best tone and impact.

I also found HQPlayer’s latest update seemed to lose a little depth overall on either of these 2 settings in favor of more weight. Was going back and forth between these 2 settings and latest update and previous one. Settled on the xtr-mp2 and latest.

For headphones I used to be Grado fan and liked rich tone and close setting, so these HQPlayer settings give me that. I’ve been using Audio Technica L3000 now and basically the closest you can get to Grado sound in a closed back headphone but with better bass and more space.

1 Like

Since my earlier post, I’ve come to prefer the DSD7 256+fs modulator over AMSDM7 512+fs. I find it preserves better attack whereas AMSDM7 feels a bit more “rounded”.

Just tried this change and do like better. Keeps the tone of the xtr filter but adds a little bit of depth.

1 Like

Hi there Derek - I know this is an old thread but it seems to address an item I have been contemplating. If you offer of advice is still on the table I would greatly appreciate your input. My current setup is ROON CORE on a NUC with internal SSD drive holding my modest (30,000 tracks) ripped music collection. This is hardwired (via my router) to a Cambridge 851N set as the ROON end point, DAC and pre-amp into a Cambridge 861W power amp. I am happy with the setup but like most here would love to improve the SQ if possible. :grinning:

So, I’ve been reading up a little on HQplayer and it seems like a reasonable price to pay for some SQ experimentation. Problem (for me) is that I have now idea where / how to introduce HQplayer into the system. I have a Windows 10 PC on the same network via Wifi. I also have a Windows 10 laptop also on the same network via wifi. Should I load HQplayer onto one of these and then connect that to the 851N? I could hard wire the laptop to the network although this would be awkward due to location of the router.

Will connecting the computer via wifi adversley effect SQ to the point of negating any benefit of HQplayer.

How do I setup the 851N to act as a DAC only - no ROON end point.

My apologies if my questions are basic - but I really am not a computer person.