Jump to content
IGNORED

HQ Player


Recommended Posts

  • 4 months later...

I use Roon for the library management and HQPLAYER control. I’m building a new computer based on the 8086K CPU and 2080ti GPU to run the EC modulators in HQPLAYER 4. My question is, can I leave my very stable current HQPLAYER 3 set-up, PCM=> DSD512x48/EXt2/ASDM7DSD, in my current computer alone/untouched to feed my T&A DAC8 DSD dac and simply add the new 8086K/2080ti computer to Roon as a new HQPLAYER  Zone to do PCM=>DSD256/EXT2/ADSM7EC to feed my Lampizator DSD-only dac? In other words, can Roon control separate HQPLAYER 3 and HQPLAYER 4 computers as different zones simultaneously? Can I keep my HQPLAYER 3 set-up untouch and simply add HQPLAYER 4 as a new Zone in Roon?

Link to comment
2 hours ago, Miska said:

 

Yes, that works fine. I have something like 10 different HQPlayer computers as zones in Roon. HQPlayer Desktop and Embedded. You can then easily choose which one you want to play to.

 


Thanks Jussi, that’s great. Since my Lampizator DSD DAC also supports native DSD512x48, I will be able to swap the T&A DAC8 DSD and the Lampizator DSD DAC back and forth between the HQPLAYER 3 and HQPLAYER 4 computers to check out which HQPLAYER setting configuration: DSD512x48/EXT2/ASDM7DSD or DSD256x48/EXT2/ASDM7EC, and with which DAC I like better and prefer for regular use in my reference system.

 

i may also revisit the EXT2 versus XTR filters. Do you still prefer or use the EXT2 as your go to or regular filter of choice as you have previously reported ? Or have you been using another filter, such as Closed Form 16M, with the EC modulators with HQPLAYER 4 Desktop?

Link to comment

Finally upgraded to HQP4 but I’m seeing a problems that others have encountered: I cannot get HQP4 to output DSD no matter what I try in the settings. It converts both PCM and DSD files to the sample rate in the PCM  setting although I clearly select DSM as the default output. Using the Amanero Combo 384 driver for my Lampizator DSD DAC. The AISO driver does not produce any sound, Roon gives a message that it lost control of the audio devise. With WASAPI it converts everything to PCM and cannot get a DSD output regardless of the output default setting being set to SDM. Please help as this is frustrating.

Link to comment
  • 2 months later...
  • 1 month later...
2 hours ago, Miska said:

 

We'll see how the i9-10900K will be, when it is available (and motherboards where you can put it)... 9900KS is still highest on all-core boost at the moment.

 

 

Jussi, My biggest interest here is if there are any benefit derived from the 9900KS as compared to the 8086K. I have the 8086K and I know that you also have the 8086K on one of your development PC’s. Based on your previous guidance on number of high frequency cores needed/used by HQPlayer: (# of channels multiplied by 2) + 2; it would indicate that for stereo (2 channel) HQPlayer users the 6 core 8086K would be ideal and that the 2 additional cores in the 9900KS would only add heat ( 32 Watts of  additional heat as compared to the 8086K, 127 Watts versus 95 Watts), which with the same thermal solution, CPU cooler, could lead to thermal throttling, which would wipe out any advantage of the 9900KS over the 8086K. So for me it would seem that going from your 8086K to 9900KS would be disadvantageous and a step backwards for stereo, 2-channel, HQPlayer users. Looking forward to your comments and the empirical results of your testing.

 

Link to comment
14 minutes ago, Miska said:

 

The only advantage you get is with filters, if you want some filter that doesn't run on 8086K, then you'd have two extra cores for running it. But otherwise, if that is not the case, as you noted there's not much point in going from 8086K to 9900KS. Since the two machines I have run different OS, it is hard to say if there is some efficiency benefit between the two, but at least now I have not noticed any.

 

If you want for example 8 channel output, then there is a definite gain from using 8-core CPU (with Nvidia GPU).

 


Thanks Jussi. For processing the filters I rely on the CUDA processing of the 2080-TI GPU; so in my particular case I would not gain, but could lose, with moving from the 8086K to 9900KS CPU, because of the extra heat-load of the two additional cores.

 

Thank you again for your always insightful responses.

Link to comment
  • 4 months later...
26 minutes ago, The Computer Audiophile said:

Smooth as butter at DSD256 and ASDM7EC :~)

 

Chris, have you tried single stage XTR filter at 48xDSD512 with your CAP Twenty? If the XTR filter works at 48xDSD512 rate, that would be a major and significant achievement. The follow up question would be does it work at 48xDSD512 with the ASDM7EC modulator?

 

If the answer to the XTR filter is negative, my next interested is if the EXT2 filter with ADM7EC modulator works at 48x512DSD on the CAP Twenty.

 

Eagerly awaiting your results. 

Link to comment
  • 8 months later...

What are the opinions on the sound quality of HQP release 4.10.3 as compared to 4.9.0? Curious as I was overseas and never got a chance to download and listen to 4.9.0 after reading the reports of its phenomenal sound presentation. I’m happy with what I’m hearing with 4.10.3 but wondering if 4.9.0 sounds markedly better. Not interested in startIng any controversy or a search for logic as that has been covered and discussed already just wondering what those that consider revision 4.9.0 special think of the sound quality of release 4.10.3 in comparison.

Link to comment
  • 3 months later...

I have been listening to new release version 4.12.0 of Desktop for Windows all night. I love the new Polysinc-XTR-Short-LP filter with the ADSMEC7 modulator. I’m transcoding everything to DSD256, using my LampizatOr DSD-Only DAC. The ADSMEC7 512FS modulator provides more air but at the expense of image focus and depth in my set-up. Great work Jussi, great new filters and modulators. In the end, I’m sticking with the EXT2 filter and ADSMEC7 modulator because in my system it provides the right balance of detail, air, depth and image focus. I’m hoping that the next generation of Intel CPU’s and NVIDIA GPU’s will provide the processing power to allow us to use these filters and modulators with DSD512. Has anyone with the latest CPU’s and GPU’s been successful in using any of the EC modulators and new filters with transcoding up to DSD512x48?

 

Link to comment
  • 3 months later...
On 10/1/2021 at 7:51 AM, Miska said:

I also like these, but depending on music genre I tend to go for different filters. Like multi-track mixed studio production vs live recording in a real acoustic space with minimal mics. So far, ext2 and gauss-long have been best all-rounders.

 


Jussi,

I have been so satisfied with the 7EC modulator + EXT2 filter combo that I had not done any experimentation with any of the other filters or modulators recently. Then last night I saw your above statement and I decided to give the Sinc-Gauss-Long filter a listen and WOW!!!! I now have a new filter. Beautiful space and presence without loss of details. I did find one song where the female singer’s voice sounded wider than natural but that was the only exception during a 4 hour listening session. There is no going back. Thank you, for again pointing me in the right direction. I’m transcoding all files to DSD256 x 48 with the 7EC modulator now with the Sinc-Gauss-Long filter. Simply awesome!

 

Can you provide any insight on the upcoming ASDM7ECv2 modulator and what we can expect from it, sonically, compared to the current ASDM7EC modulator?

Link to comment
  • 1 month later...
17 minutes ago, MagnusH said:

Seems a lot of people love the 7ECv2 + ext3 combo, its my current favorite as well.

That a great combination and the most detailed top to bottom. It is my second favorite combination, as I continue to like the 7ECv2 + Gauss-Long the best because of its spaciousness and spatial qualities.

Link to comment
  • 3 months later...
8 minutes ago, Mike Rubin said:

That is exactly what I figured, but just wanted to be sure.


Hqplayer is not bit-perfect, by design. Miska/Jussi has commented on this a number of times. Even when you set the output to the same word-bit-depth and same sample-rate as the input word-bit-depth and sample-rate, there is always some noise shaping that takes place that prevents the output from being bit perfect to the input. It is my understanding that HQPLAYER has no “bypass” mode.and Jussi/Miska can confirm if this is accurate.

Link to comment
  • 1 month later...
15 hours ago, bobparish said:

@Miska Was 4.18.0 sound quality improved or are my ears? Fuller and detailed come to my mind...

 

9 hours ago, MemoryPlayer said:

For me much better! Timbre more realistic, organic and natural...

Try this one: https://www.prestomusic.com/classical/products/9253095--silent-dreams


I couldn’t agree more, with 4.18.2 Desktop on Win 11 Pro doing DSD512x48 with Poly-sinc-gauss-XLA from HQPlayer and Roon the sound is raw, alive, organic, and very natural. Like listening directly from the mixing board on the cutting room’s monitors. Very special remastering to my ears with this combination.  Perhaps this combination of releases from HQPLAYER and ROON is optimal because of their synergies the sounds as good as I have heard from this combination. Like the sound finally achieved perfect focus. How long will this stay the same? I hope that Roon does not not update for awhile. Will obviously have a copy of HQP 4.18.2 but is there a way to keep or revert to a specific Roon release?
 

Even CPU utilization is down with 4.18.2, the cooling fans are quieter, running at the lower RPM’s. Great job Jussi/Miska! Enjoying sonic bliss as I type this.Need to capture this sound in a bottle as a reference for future releases from both HQPLAYER and ROON.

Link to comment
  • 4 weeks later...
On 2/13/2022 at 12:50 PM, Bertel said:


I bought the SMSL M500 MkII and received it on Friday, but it has strange issues with occasional electrical noise (low level stuff that sounds like quantisation noise or so, plus high pitched squeal at low volume at the end of tracks at fade out), although I feed it DSD512 only - guess that‘s what you mean with „it misbehaves“… 

 

@Miska have you tested and measured any ESS9038Pro based DAC yet which works without issues and quirks and plays DSD512 without any flaws? I like the very clean and transparent sound a lot and definitely want such a DAC in my collection, but haven‘t found properly engineered units yet it seems…


This is exactly what I’m hearing with my LampizatOr DSD-Only (chipless) DAC with Amanero Combo384 USB interface card, exactly the same noise that Bertel describes in his post. The noise is most noticeable at the end of track and the noise goes away when HQPLAYER goes in active, either through a stop or pause. This issue is not there at DSD256. My dac has played flawlessly at DSD512 with the exact same settings for months and all of the sudden I’m getting this quantization swirling noise every time HQPLAYER opens the USB port and then it stops to a dead silence when HQPLAYER releases the USB port. I reverted back to older HQP Desktop versions and the issue persists.
 

I thought that it was GPU noise from the PC coupling through the USB cable or ground-loop induced noise because of lack of galvanic isolation but after spending money on the various iFi digital enhancer and ground-loop breaking products the issue remains

 

Any ideas? Was Bertel’s issue resolved? What was the root-cause or fix for it? I’m having the exact same issue. As I said before it worked flawlessly for months but all of the sudden this noise is driving me crazy. One other data point, if I don’t use my setup for a few days the issue is not there when I first start playing music but after about 15 minutes of playing it shows up and does not go away unless I shut everything down and de-energize the dac and disconnect the  USBcable.

 

Miska/Jussi and others, all help resolving this issue is greatly appreciated.

 

Could some software have enabled a Spread-Spectrum setting and that is what is causing this noise at DSD512? How can I check if any Spread-Spectrum settings are enabled? Or does HQPLAYER inhibit those types of noises from getting into the DSD stream?

Link to comment
57 minutes ago, Miska said:

 

That happens with various DACs when the clocking is going wrong. I would say it is likely related to the Amanero USB interface.

 

Have you tried with another USB cable? 

 

Please also pay attention if this is related to 44.1k base vs 48k base output rate?

 

 

No, even if it would, it wouldn't have such implications because computer clocking is not related to DAC clocking in any way.

 


Miska my settings are for 48K base output rate. This morning I started to narrow down the root cause and made great progress towards resolving the issue.
 

I disabled CUDA offloading and it made no difference. This told me that my previous theory about GPU borne noise is not the culprit.

 

I then followed the Spread-Spectrum theory and looked for the CPU Host Clock Control] setting to change it to [Manual] to disable "Spread Spectrum", but could not find the setting on my Z690 Xtreme BIOS.

 

I then noticed that the noise correlated with the fans speed and CPU temperature going up at start of HQPLAYER playback and coming down after HQPLAYER playback is halted, when the tracks start and then goes away when the tracks stop, after the noise swirls down to complete silence. I went into the Gigabyte BIOS to see if I could find the performance mode switching setting for the CPU and found that Gigabyte has a setting in the Z690 BIOS called “CPU Upgrade” with setting 3 options: 1. Default, which is what it was set to. 2. Gaming Mode, which disables the E-Cores, and 3. MAX-Performace, which boosts the P-Cores and retains the E-Cores on the Intel 12900-KF CPU. I changed this setting from Default to MAX-Performance and HQPlayer and the DAC have been playing DSD512 x48, with 7ECv2 modulator and Gauss-XLA filter, for over 2 hours without issues. I hope this issue has been resolved by keeping the CPU from actively transitioning between speeds during HQPLAYER playback.

Link to comment
  • 1 month later...
1 hour ago, Schafheide said:

What you say makes sense.  But this has only started happening since I started fiddling with EQ.  BTW Matrix function is not enabled.

 


Why don’t you use the “Clear Data” command in HqPlayer to reset all parameters and start over with default settings? It’s under the help menu and it can delete a setting that might be set inadvertently.

Link to comment
  • 3 weeks later...
20 minutes ago, Miska said:

 

I use "loudgain" on Linux for processing ReplayGain tags to files when/if I want such. Not sure if any of the hires stores supply such tags.

 

But Qobuz supplies this information and Adaptive Gain works when streaming Qobuz using HQPlayer Client.

 


Jussi have you reviewed “Leedh Processing” digital volume control principle? Any thoughts or comments you care to share on it or how it compares to digital volume control in HQPlayer 4? Curious what your take on it is. Marketing hype or technical advancement?

Link to comment
19 minutes ago, Miska said:

 

Yes, nothing interesting there.

 

 

Marketing hype. It is dead simple thing implemented on top of limited resolution integer PCM samples and only useful when you don't perform any other processing on the data, such as digital oversampling filters, delta-sigma modulators or such.

 

I wouldn't want such thing on my system, since there are much better ways to do things. And without all the limitations of Leedh.

 


Thank you for your feedback. As alway, very much appreciate your knowledgeable and valuable insight. 

Link to comment
  • 9 months later...
  • 1 month later...

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...