[babarrun]

Glad to hear it works!

Just for clarification, your Roon screenshot said Spring 2 so I assumed you have a Spring 2 not a Spring 3, is that correct?
Spring 3 is accurate up to 20 bits (instead of 16) and can take 1.536 Mhz instead of 384 Khz.

works really well. now need to fine tune

Is there a page where I can find simple language explanation of what popular filters do. without going into longwave / shortwave and other technical discussions. more like 'this is good for rock' 'or classical' etc. and what other things do.

currently I feel i am getting mild popping noises in playback of some tracks. eg the bird song by anette askvik

 

[bogi]

Is there a page where I can find simple language explanation of what popular filters do. without going into longwave / shortwave and other technical discussions. more like 'this is good for rock' 'or classical' etc. and what other things do.

Look at the filter table in HQPlayer Desktop manual, which is installed together with application (not available online).

 

[plumpudding2]

works really well. now need to fine tune

Is there a page where I can find simple language explanation of what popular filters do. without going into longwave / shortwave and other technical discussions. more like 'this is good for rock' 'or classical' etc. and what other things do.

currently I feel i am getting mild popping noises in playback of some tracks. eg the bird song by anette askvik

The popping can be due to clipping (intersample overs). The dev recommends to either turn on Roon volume normalization or set the HQplayer volume to -3db.

Next to the HQplayer executable in the installation folder theres a manual that describes the filter and what genres they're recommended for.

 

[babarrun]

The popping can be due to clipping (intersample overs). The dev recommends to either turn on Roon volume normalization or set the HQplayer volume to -3db.

Next to the HQplayer executable in the installation folder theres a manual that describes the filter and what genres they're recommended for.

thanks.

I think I have it set to -3.0 , and have attached a picture. please let me know if that is ok.

I have also attached a picture from Roon which shows -3.0

there is another picture from Roon as well showing setting of HQ Player.
also on this what do you suggest I do for MQA? I tend to use TIDAL a lot though I also have qobuz and others.

I am happy ok not to use the volume on either as I am am accustomed to using the volume on the Bliss.

currently it's a bit annoying as I am hearing mild clips/pip sounds in the music.

 

[genefruit]

thanks.

I think I have it set to -3.0 , and have attached a picture. please let me know if that is ok.

I have also attached a picture from Roon which shows -3.0

there is another picture from Roon as well showing setting of HQ Player.
also on this what do you suggest I do for MQA? I tend to use TIDAL a lot though I also have qobuz and others.

I am happy ok not to use the volume on either as I am am accustomed to using the volume on the Bliss.

currently it's a bit annoying as I am hearing mild clips/pip sounds in the music.

if you repeat play a file with the noise, does it happen in the same location each time? If so, turn the volume to -4.0 and test again. Reduce until noise goes away as an option. If the noise doesn't happen in the same place, it could be a limitation of your compute. Try a different filter or shaper and test again or reduce the output to something like 768.

 

[EMINENT]

Hailo.

Pops can be from buffer too. Try changing from default to 100 as stated on main HQP page or you can try one of the other figures.
Sometimes I hear it in one or another and change it around.

 

[EMINENT]

For Tidal, leave it at no MQA support, then click the show advanced in Roon, enable MQA Core Decoder so it handles the first unfold. From there on HQP handles the filtering with your selection as you are replacing MQA's filter.

 

[Louisiana]

works really well. now need to fine tune

Glad to read this, HQPlayer is great

Is there a page where I can find simple language explanation of what popular filters do. without going into longwave / shortwave and other technical discussions. more like 'this is good for rock' 'or classical' etc. and what other things do.

There are two threads with a lot of information about all the filters, and for what pc power xou need to run them:

https://community.roonlabs.com/t/which-hqp-filter-are-you-using-2023/244185/863
https://audiophilestyle.com/forums/topic/19715-hq-player/page/1198/#comment-1258703

On both threads, Miska, the founder from HQPlayer is always present, and very helpfull.

currently I feel i am getting mild popping noises in playback of some tracks. eg the bird song by anette askvik

that is the resulution from this album?
44,1 oder 48 oder 96, or xx...

 

[bogi]

currently it's a bit annoying as I am hearing mild clips/pip sounds in the music.

Does popping happen regardless on target sample rate and filter/modulator selection?
Is your target sample rate in the same family (44.1x or 48x) as source content played? If not, try if changing target rate to source rate integer multiple helps.

 

[Louisiana]

Does popping happen regardless on target sample rate and filter/modulator selection?
Is your target sample rate in the same family (44.1x or 48x)…

That was what I was going to write above, no idea how I came up with Resolution.
Maybe because I was thinking of "High-Res" audio.

 

[rkt31]

That's not so easy. Long filters are accurate in frequency domain but have poorer performance in time domain. And short filters the opposite.
Long filter response is affected by transients which occurred longer time before and after an event and thus introduce a transient smear.

These posts explain difference between long and short filters and also disadvantage of very long filters - at least theoretically:

https://audiophilestyle.com/forums/topic/19715-hq-player/page/1082/
"When the filter gets longer (more taps) in time domain, it gets steeper in frequency domain and vice versa. This is because frequency and time have 1/x relationship. Longer the filter, longer it "rings" or "smears" in time domain. While if it gets too short, it begins to roll-off early which in turn means it reduces the high frequency components needed to make quick time domain details, so it "slows down" the transients. Too short filter can also begin to "leak" that means the reconstruction accuracy is lost as result, and as result produces imaging distortion.

Making a filter that is perfect in both domains simultaneously is mathematical impossibility. Only unlimited bandwidth, like you practically get with high rate DSD recording (minimum DSD128, but optimally DSD256+), can give you both at the same time. Band limited PCM has the problem though."

https://audiophilestyle.com/forums/...rdinary-person/?do=findComment&comment=175928
"Longer/steeper filters change faster from passing frequencies to not passing frequencies as function of frequency. Shorter/gentler filters transition more slowly or "gently" from pass to stop as function of frequency. More accurately the filter wants to detect frequencies and transition pass/stop faster, longer time the filter has to "look" at the signal. This has side effect called "ringing" or rather "time blur". On the other hand, extremely short filter like a one that looks only at single moment cannot filter anything at all, because it sees only single point of time at once without any history or future (so it cannot detect any frequencies as those are a change over time)."

About "irrespective on genre" - that's also not so easy,
"Thus the rough starting point recommendations go along the lines of:
- Long linear phase filters for classical in acoustic spaces
- Short minimum phase filters for multitracked studio rock
- Relatively short linear phase filters for jazz clubs"
read the whole post about it: https://audiophilestyle.com/forums/topic/19715-hq-player/page/1086/

I am not an electronics engineer and not an expert too. But theoretically infinitely long filter with sinc function will recreate the original band widh limited signal back. If longer filters are poorer in time domain then at infinitely long filter should have worst time domain accuracy rather than recreating original signal back. Isn't it? I think time domain vs frequency domain accuracy is more to do with windowing function. Again I would say I sm not expert. But to me sinc medium had the best in everything, right from timing, depth, soundstage, separation, realism and musicality. I could test sinc long due to stuttter. May be the hqplayer developer can shed more light on it.

 

[Failed Engineer]

Jussi (the developer) isn't going to post here or read the thread. Better to ask in the Roon forums or AudiophileStyle. He will respond and answer these types of questions there.

 

[bogi]

infinitely long filter should have worst time domain accuracy

According to Jussi this issue exists with band limited signal only.
I am also not a pro. It's really better to ask more technical questions on AS forum.

But to me sinc medium had the best in everything, right from timing, depth, soundstage, separation, realism and musicality.

That's amazing! That's more important than all theoretical debates.

I only wanted to point out, that $$$$ investment into ability to run longest and most computer resources consuming filter need not to be technically the best choice, particularly for lowest sample rate content with fast transients.

 

[rkt31]

Jussi (the developer) isn't going to post here or read the thread. Better to ask in the Roon forums or AudiophileStyle. He will respond and answer these types of questions there.

According to Jussi this issue exists with band limited signal only.
I am also not a pro. It's really better to ask more technical questions on AS forum.


That's amazing! That's more important than all theoretical debates.

I only wanted to point out, that $$$$ investment into ability to run longest and most computer resources consuming filter need not to be technically the best choice, particularly for lowest sample rate content with fast transients.

in real world Audio Signal has to have some band width limit in any case. Ie the adc which records has this built in.

 

[babarrun]

Glad to hear it works!

Just for clarification, your Roon screenshot said Spring 2 so I assumed you have a Spring 2 not a Spring 3, is that correct?
Spring 3 is accurate up to 20 bits (instead of 16) and can take 1.536 Mhz instead of 384 Khz.

sorry for the mistake. I have a Spring3KTE and it showed 1.536 Mhz.