HQ Player

[toddrhodes]

I had mentioned earlier that I was seeing an issue where playback stopped, but only when one track was ending and another was to begin. I was able to isolate it in the log file, in case it's helpful for finding the cause:

 

 2021/01/27 20:28:58 Teams: 1
  2021/01/27 20:28:58 Places: 1
  2021/01/27 20:28:58 Parallel threads: 16
  2021/01/27 20:28:58 Nested parallelism: 4
  2021/01/27 20:28:58 Parallel pipelines: 4
! 2021/01/27 20:28:58 NAA output clNetEngine::PushSDM(): clNetEngine::Start(): not connected to adapter
! 2021/01/27 20:28:58 clHQPlayerEngine::Execute(): push to FIFO failed
  2021/01/27 20:28:58 Stop request (reset)
& 2021/01/27 20:28:58 Stop...
- 2021/01/27 20:28:58 Playback engine stopped
& 2021/01/27 20:28:58 ...stopped
  2021/01/27 20:28:58 Set volume: -3 +
# 2021/01/27 20:29:07 NAA output network Audio Adapter keepalive: clNetEngine::RecvResponse(): clSocket::WaitRead()
  2021/01/27 20:29:08 NAA output network Audio IPv6 support enabled
  2021/01/27 20:29:08 NAA output discovery from 0.0.0.0
& 2021/01/27 20:29:08 NAA output discovered network audio: name='naa' version='Signalyst Network Audio Daemon 4.1.1'  @192.168.86.37:43210
& 2021/01/27 20:29:08 NAA output discovered network audio: name='naa' version='Signalyst Network Audio Daemon 4.1.1'  @192.168.86.37:43210
  2021/01/27 20:29:09 NAA output discovery from ::
& 2021/01/27 20:29:09 NAA output discovered network audio: name='naa' version='Signalyst Network Audio Daemon 4.1.1'  @[fe80::dea6:32ff:fe82:bff3%9]:43210
& 2021/01/27 20:29:09 NAA output discovered network audio: name='naa' version='Signalyst Network Audio Daemon 4.1.1'  @[2600:1700:8c30:de2f:dea6:32ff:fe82:bff3]:43210
  2021/01/27 20:29:10 NAA output connect to [2600:1700:8c30:de2f:dea6:32ff:fe82:bff3]:43210 [ipv6]
  2021/01/27 20:29:10 NAA output network endpoint: bcm2835 ALSA: bcm2835 ALSA (hw:CARD=ALSA,DEV=0)
  2021/01/27 20:29:10 NAA output network endpoint: bcm2835 ALSA: bcm2835 IEC958/HDMI (hw:CARD=ALSA,DEV=1)
  2021/01/27 20:29:10 NAA output network endpoint: bcm2835 ALSA: bcm2835 IEC958/HDMI1 (hw:CARD=ALSA,DEV=2)
  2021/01/27 20:29:10 NAA output network endpoint: USB HiRes Audio: USB Audio (hw:CARD=Audio,DEV=0)
  2021/01/27 20:29:10 NAA output discovered 1 Network Audio Adapters
+ 2021/01/27 20:29:10 NAA output connect to [2600:1700:8c30:de2f:dea6:32ff:fe82:bff3]:43210 [ipv6]
  2021/01/27 20:29:10 NAA output network format: 44100/32/2 [pcm] 
  2021/01/27 20:29:10 NAA output network format: 48000/32/2 [pcm] 
  2021/01/27 20:29:10 NAA output network format: 88200/32/2 [pcm] 
  2021/01/27 20:29:10 NAA output network format: 96000/32/2 [pcm] 
  2021/01/27 20:29:10 NAA output network format: 176400/32/2 [pcm] 
  2021/01/27 20:29:10 NAA output network format: 192000/32/2 [pcm] 
  2021/01/27 20:29:10 NAA output network format: 352800/32/2 [pcm] 
  2021/01/27 20:29:10 NAA output network format: 384000/32/2 [pcm] 
  2021/01/27 20:29:10 NAA output network format: 705600/32/2 [pcm] 
  2021/01/27 20:29:10 NAA output network format: 768000/32/2 [pcm] 
  2021/01/27 20:29:10 NAA output network format: 1411200/32/2 [pcm] 
  2021/01/27 20:29:10 NAA output network format: 1536000/32/2 [pcm] 
  2021/01/27 20:29:10 NAA output network format: 2822400/1/2 [dsd] 
  2021/01/27 20:29:10 NAA output network format: 3072000/1/2 [dsd] 
  2021/01/27 20:29:10 NAA output network format: 5644800/1/2 [dsd] 
  2021/01/27 20:29:10 NAA output network format: 6144000/1/2 [dsd] 
  2021/01/27 20:29:10 NAA output network format: 11289600/1/2 [dsd] 
  2021/01/27 20:29:10 NAA output network format: 12288000/1/2 [dsd] 
  2021/01/27 20:29:10 NAA output network format: 22579200/1/2 [dsd] 
  2021/01/27 20:29:10 NAA output network format: 24576000/1/2 [dsd] 
  2021/01/27 20:29:10 NAA output network format: 45158400/1/2 [dsd] 
  2021/01/27 20:29:10 NAA output network format: 49152000/1/2 [dsd] 

 

To reiterate, this has occurred back to 4.8 so I don't think it has anything to do with the 4.9 issue. The above was on 4.9, and I've also seen it on 4.91.

 

Chain is:

Win 10 PC with Roon Core

feeds Win 10 PC with HQPlayer running on it only

Both currently have their private network Windows Firewalls disabled, to see if that had any impact on this behavior

Running to a Pi4 4Gb running the latest HQP NAA image

 

And again, this is not the end of the world in terms of issues. I generally just have to tell Roon to skip a track or two (mind you it's on random/radio function at this point, I've not had it do this mid-album when I had a queue lined up or anything) and it starts back up without issue.

 

 

 

[toddrhodes]

Boy it's hard for me to explain how, but I'm hearing the difference in 4.9 vs 4.9.2 as well. 

 

The way I rigged up my test was to basically just install 4.9.2 next to 4.9.0, by giving it a different path to install at. I made shortcuts to each EXE on my HQP Win10 desktop, and just stop one instance, play a track; stop the track and quit that instance, then start the other version of HQP and replay the track or whatever I was listening to.

 

It's almost like 4.9.2 sounds a little more congested? It doesn't seem to layer as well, it doesn't seem to just hang in the room as if no speakers really exist. Meanwhile, 4.9.0 plays a masterful disappearing act. There are some tonality differences as well.

On Jesse Cook - Rattle and Burn, I could hear the vibrato of his strings and a little feedback from that, clear as a bell on 4.9.0. I initially listened to this track on 4.9.2 and didn't hear that. But then when I went back and listened to it again on 4.9.2 - I can tell the vibration is there, but it's because I know what to listen for.

 

On Lee Rittenour - Daddy Longlicks (6 String Theory), there's a section where I think he's palming the guitar and on 4.92 it certainly makes sound and is discernible but it just blends in with everything else he's doing. On 4.9.0 it's its own distinct sound in its own distinct space in between the speakers. It's a striking difference and not hard to discern.

 

The last track that really stood out was Dave Grusin - Gershwin Connection - Fascinating Rhythm. This is where I keep coming back to the word "cohesive" when thinking why 4.9.0 is better than 4.9.2 Neither sound bad, not by any means. But 4.9.0 sounds like a collective whole; 4.9.2 sounds like a bunch of stuff happening in the same space. This is where it's really hard to describe a sound - like getting a mechanic to diagnose an engine issue over the phone - but 4.9.0 is like when you have a really nicely calibrated TV and all looks right, you don't see Red/Green/Blue, you see a picture. The HDR is tuned right so the proper things pop and the blacks are dark as night and the colors are vivid without exaggeration. 4.9.2 is like seeing the same TV in a retail display. It certainly looks fine and if you had no calibrated set to judge from, you'd happily take it home and be happy. 

 

4.9.0 seems like everything just works in harmony. It's really effing strange - because I'm not one to get too caught up in software code changes, outside of specific filtering and sampling updates, impacting what I hear. I just did this for fun, and I can say I'll just stick with 4.9.0, even if it does have the occasional issue. 

[toddrhodes]

6 hours ago, Confused said:

A question to those of you reporting that 4.9.0 has superior sound quality, (or indeed those who do not find this to be the case) what is your use case for HQPlayer?  That is, are you upsampling PCM or performing PCM to DSD conversion?  Similarly, do you have a PC + endpoint scenario or a "one box" system?

 

I guess I am just thinking aloud here, but I am just wondering if there is any correlation between those finding (or not finding) 4.9.0 to be superior and how HQPlayer is being implemented.

 

Good question.

 

My chain is:

Roon core on Win10 PC

HQPlayer on separate Win10 PC, Intel i9900KF CPU

Pi4 NAA using 4212 NAA image

USB out from Pi4 to DF Venus II DAC

upsampling all to DSD256, ASDM7EC, EXT2 filter

[toddrhodes]

I'm essentially running a 3-device, Roon-to-HQPlayer setup. And it sounds absolutely beguiling to me. I've had occasion in the past to dismiss file-based audio, going from a Roon machine to a network endpoint and out to a Schiit DAC via USB. It was ok, but rather easily bested by a good CD transport on like material, into the same DAC.

 

Now, though... with a dedicated machine running HQP at DSD256, ASDM7EC, the HQP NAA Image feeding my DAC from a completely stock, basic Pi4 with no manner of magic USB cable in sight, it's stunning.

 

I've spent most of my music-listening life as a devout metalhead, an armchair rocker and air drum aficionado. But since this new digital rig and the addition of Dyn Confidence C2 Platinums, well, I find myself right now as I'm typing this listening to Concert Allegro Op. 134, Busoni works for Piano. This setup does just fine with Tool and Death and Dire Straits, too, but the level of transparency and refinement I now get from digital files is unmistakable. 

 

The only other time I had this much fun in this hobby was when I discovered vinyl. Sadly, my table hasn't spun a record in several weeks. I need to fix that, but have I mentioned how amazing this piano concerto sounds?!? 

[toddrhodes]

I went back to 4.9.2 yesterday. It still sounds amazing to me.

 

That said, I also took some time to use REW to create convolution files which I've loaded into HQP now as well. I find the difference to be pretty subtle, but quite effective for the limited time I've been able to listen so far. I'm not sure if the image depth and space should be changing as a result but I do feel like even more than before, sounds in the soundstage hang and float more convincingly than before, and it was already something I was quite happy with. Bass, too, is more refined and taut. And given the actual vs predicted lines in REW, it didn't make massive changes to anything, but it did smooth response somewhat.

 

I saved my .wav files with the highest sampling rate REW allows and saved them as mono files, one each for left and right, and 32-bit. HQPlayer never complained when I loaded them, so I assume I did that right? I convert everything to SDM output, if that matters.

 

Thanks again @Miska for this amazing software.

 

I also gave the built-in Roon upsampler a go yesterday. I was using Smooth, MP and 7th order CLANS noise shaping, DSD512. It worked my Ryzen 3900x pretty well on my Core PC but it played without a hitch. And I did like it - just not quite as much as HQP. Another buddy tried the same experiment yesterday and also found the same result. Roon is pretty good, HQP is just better, more refined, more transparent and resolving. 

[toddrhodes]

Hey @Miska, maybe something to be on the lookout for if other users have issues on your NAA firmware. I had this issue before but got a new Pi4 which worked fine. Tonight I was futzing around with something and put the "bad" Pi4 back in the system so I thought I'd do a little investigation, at least to my technical ability. Didn't find much, but found two things:

 

Using RoPieee to grab board information, and also used it to make sure both boards are up to the latest USB firmware, just in case (they are, no change in behavior):

Non-working board specs:

 

 

Working board specs:

 

 

Only thing I see of note is the board revision - 1.2 works, 1.4 is not working for me.

 

I also took a picture of the actual boards side by side - there are differences for sure but beyond me to explain what they mean. Diffs are by the USB 2.0 ports and the micro HDMI ports on the side:

1.2 is on top (with heatsinks)

1.4 is bottom

 

 

I'm up and running just fine on the 1.2 board with 4112 NAA image. Again, this is just in case there's something different about Rev 1.4 that may cause issues for you with other customers. IIRC, I bought the 1.4 board in December 2020, I think.

 

Lastly - both boards work and are discoverable by HQP Desktop using RoPieee, 1.4 just not for 4112 NAA image.

[toddrhodes]

Not sure if this is the best place to ask this but I recently built a digital-digital converter using an Amanero Combo384 USB streamer and some Ian Canada goods.

 

I am able to play Native DSD512 to it via direct connection to my PC but from the 4112 NAA Image, all I can get it to do is DSD128 via DoP. I'm assuming/guessing this is a driver issue or possibly a 384 FW issue, but my results on Windows at least are encouraging in that respect.

 

Any suggestions on how to enable Native DSD to at least 256 via the NAA image? NAA does pick it up as "Combo384" as the output device.

 

Thanks!

[toddrhodes]

Thank you both! Admittedly it was late last night when I was playing with the Firmware tool, but I was having a bit of a hard time even figuring out what the "newest" options were through their tool. 

 

But, the good news is it is not hard to do, so I just kinda need to figure out what versions of CPLD and CPU to flash and go from there.

 

To be clear - I couldn't care less about DSD512 from Windows as I'll never use it with Windows. I was having other issues on top of the firmware side of things so that was just my way of making sure all was working as it should, even if the scenario I tested isn't one I intend to use - it was just the easiest one to mock up at the time.

 

Hope that helps, and thank you.

If anyone knows what versions of FW to run off the top of your heads, I'd be indebted to you. Otherwise, I'll start digging :D

[toddrhodes]

3 hours ago, Solstice380 said:

Just find the T+A thread.   The numbering is confusing... 11 or 12be is the one.  I think, maybe, could be.... 😜

Turns out this was the secret sauce in my setup:

 

Amanero Combo384 Native DSD512 for Linux:

 

Not much of a secret really, but for posterity's sake, here she is :)

 

And just for grins, here is the setup I'm running this in (it's crude for now):

 

Stuff in that pic:

LifePO4 mkiii on left, powered by a Dell laptop 19V/300W power supply
On the right, bottom to top:
Optical Ethernet converter and
Pi4 hardwired, both converter and Pi are running fine off LifePO4 5V DC
Amanero Combo384
BridgePi
FifoPi Q3 with AccuSilicon 90/95 oscillators - both "sides" are run off 3.3V sections of LifePO4
HDMIPi Mk I i2S/LVDS transmitter

Vinyl boxsets I haven't touched since playing with HQPlayer

[toddrhodes]

I saw that option, but haven't had a chance to figure out if they correct yet. Good call - thank you!

[toddrhodes]

So I started typing this out last night and figured I'd see if I was on the latest version of HQP for Windows - I wasn't. So I upgraded to 4.10.3 and that fixed things at least for one night. But now tonight, the same issue is back so...

 

For some reason, my HQPlayer endpoint out of Roon seems to be ending songs early? With 3-4 seconds remaining in a track, it just stops playing and queues up the next song. I know it doesn't do the same thing on my upstairs endpoint which is basically a PC connected to an Anthem receiver via HDMI cable. I just haven't had time to try other "zones" or what have you to see if it's a Roon problem, or an HQ Player issue. And since I'm running the HQP NAA image on my Pi, I haven't taken a chance to swap back to like a RoPieeeXL image to see if just running it as a RoonBridge has the same issue.

 

 

Am I the only one with this issue? 

 

I'm using 4.10.3 on a W10 PC to a Pi4 with the HQP NAA image on it, USB out to my DAC. I listened to a few different zones through Roon today which are all routed through standard RoonBridges and had no similar issues.

 

[toddrhodes]

Trying out the new Sinc-Mx and using it on PCM384/LNS15 (also tried NS5 but LNS15 seems more clear and more well-defined to me?) and really digging it. 

 

Reference material is none other than the original CD of Megadeath - Rust in Peace - Tornado of Souls.

[toddrhodes]

5 minutes ago, AudioDoctor said:

 

Have you previously tried Sinc-L and LNS15? If so, what are the differences with Sinc-Mx?

 

I have used that combo before, yes, but IIRC I was using it on DSD256 so Sinc L was doing a lot more taps than even Sinc-M would have done at that point. It was a very nice sound but was a bit, polite? 

 

I recently switched to battery power for my Pi and my D2D converter (Ian Canada parts). Part of that switch was using an Amanero Combo384 USB input and, to my ears, I prefer the muscle and immediacy of PCM384 on this setup, so it's a bit of an apples and oranges comparison I think. Because even converting 44.1 to 384, Sinc L isn't doing a whole heck of a lot different from Sinc-M at that point in terms of taps and such. And since I believe that lends to the overall sound - I switched to using Sinc M for both PCM options, on 4.10.3.

 

To me, and this is completely unscientific, just trying to pull out sonic memories from the brain jelly, Mx + LNS15 seems to be a bit stronger sounding versus Sinc-M + NS5. Stronger in the upper bass and lower midbass, which is where I find a lot of the meat and foundation of the music I prefer. And since it seems to marry that up with the detail and imaging I really enjoyed from the DSD256 Sinc-L days - it seems like a pretty nice combo to me.

 

Listening now to Art Blakey & the Jazz Messengers - Mosaic. This was a track I really grew to love through that DSD256/Sinc L combo. But man does it tax even a very good PC - but I recall how vivid the title track sounded, especially when Art pivots to his solo. It's one of those moments where you can be chatting with buddies or reading liner notes or something and that part comes on and you just kinda snap to attention and stay that way til it's done. Maybe it's just a really good recording. But with this Mx + LNS15 combo, I kinda have that same feeling throughout the whole track, not just the solo. 

 

 

[toddrhodes]

1 minute ago, AudioDoctor said:

 

Interesting. I am listening now and I can almost see the polite-ness in the sound, or how someone could get there. But god its beautiful on the right music. And to me, the most natural. Perhaps I will give Mx a try this weekend. Thanks for the reply.

 

That's the thing - if I could, I'd will it to be that Roon would allow integration with HQP to a level that some genres and styles use this format and filter/shaper setting, and others use different settings. For example - metal and prog and hard rock? PCM all day. Jazz, Classical and Acoustic music? DSD. 

 

As it is, I just have to kind of go with what I listen to most. But I do occasionally stop and change over the settings and go on a classical/jazz/classical guitar binge.

 

But as I listen to some Tool right now? I just prefer the immediacy of PCM that I get in my room/system, and I don't necessarily need the space and 3D feel that DSD adds at the expense of some of that leading edge muscle. 

 

But that's also why I really like this whole setup - it can adapt and it's not that inconvenient to do.

 

Cheers!

[toddrhodes]

Quick question - I've been using, and really enjoying, gauss-long. Upsampling to DSD256, 7EC modulator (W10, Intel CPU). But, on version 4.12.0, occasionally I would get a high pitched distortion occurring. If anyone is familiar with the end of "Have a Cigar" by Pink Floyd, Wish You Were Here - it sounded a little bit like the end of that track.

 

I switched back to Sinc-Mx for 4-5 tracks - no issues.

 

I updated to 4.13.1 and went back to gauss-long and so far, no issues. Was there anything that changed from 4.12 to 4.13.1 that may have corrected an issue with that filter?