HQ Player

[bibo01]

Miska,

 

can I please have a reply on this issue.

 

If in Settings I set 7.1 or 16, are the first 8ch equal?

Does LFE receive any particular treatment (like LP)?

[bibo01]

Setting 7.1 is like setting number of channels to 8. That setting only affects how many output channels are asked from the driver.

 

All channels are equal. If you want any particular treatment, you have to configure it yourself.

 

When source file channel mappings are known, the channels are routed automatically according to the SMPTE/ITU-R standard.

Thanks

[bibo01]

Miska,

 

a user friend of mine has a problem similar to Hammer's.

He has a Merging Hapi, but the same happens with a different DAC. OS is WS2012, but it happens with Win 10 too.

When DirectSDM is engaged, all PCM files strangely play at native resolution even though the setting in the main window indicates SDM. Nothing unusual happens in the main windows when playback is launched.

If DirectSDM is removed, conversion PCM->DSD goes back to normal.

 

This is an extract from the log file showing that a RB file plays at 44.1, although the output setting is DSD128:

 

* 2015/08/19 07:21:35 Starting...

2015/08/19 07:21:35 Signalyst HQPlayer Desktop v3.8.1

2015/08/19 07:21:35 Engine selected:

2015/08/19 07:21:35 Windows version 6.2.9200 (2),

* 2015/08/19 07:21:35 Control server started

.............

2015/08/19 07:21:53 libDSP version 14.0.1

2015/08/19 07:21:53 Audio engine: asio

2015/08/19 07:21:54 Set channels: 16 (16)

2015/08/19 07:21:54 DAC bits: 32

2015/08/19 07:21:54 ASIO drivers: 1

2015/08/19 07:21:54 Found ASIO driver: 'Merging Ravenna ASIO driver' (0)

2015/08/19 07:21:54 Period time set

2015/08/19 07:21:54 Loaded ASIO driver: Merging Ravenna ASIO driver

2015/08/19 07:21:56 ASIO device supports DSD

2015/08/19 07:21:56 ASIO device supports PCM

2015/08/19 07:21:56 ASIO default format is DSD

2015/08/19 07:21:56 Rate available: 2822400

2015/08/19 07:21:56 Rate available: 5644800

2015/08/19 07:21:56 Rate available: 11289600

2015/08/19 07:21:56 Default ASIO channels: 8 in / 8 out

2015/08/19 07:21:56 Default output channel 0: DSD64_1

2015/08/19 07:21:56 ASIO DSD format MSB1

2015/08/19 07:21:56 Default output channel 1: DSD64_2

2015/08/19 07:21:56 Default output channel 2: DSD64_3

2015/08/19 07:21:56 Default output channel 3: DSD64_4

2015/08/19 07:21:56 Default output channel 4: DSD64_5

2015/08/19 07:21:56 Default output channel 5: DSD64_6

2015/08/19 07:21:56 Default output channel 6: DSD64_7

2015/08/19 07:21:56 Default output channel 7: DSD64_8

2015/08/19 07:21:56 Volume limit: 0

2015/08/19 07:21:56 Volume control disabled

2015/08/19 07:21:56 DirectSDM enabled

2015/08/19 07:21:56 Pipeline SDM enabled

2015/08/19 07:21:56 IntegratorM: IIR

2015/08/19 07:21:56 Matrix pipeline 0: 0 -> 0/0

...

2015/08/19 07:21:56 Matrix processing enabled

2015/08/19 07:21:56 Initial parallel threads: 8

2015/08/19 07:21:56 Set oversampling: 1

2015/08/19 07:21:56 Set modulator: 6

2015/08/19 07:21:56 Set filter: 0

2015/08/19 07:21:56 Set oversampling: 1

2015/08/19 07:21:56 Set dither: 5

2015/08/19 07:21:56 Set modulator: 6

& 2015/08/19 07:21:56 Volume control disabled

& 2015/08/19 07:22:19 Playlist clear

2015/08/19 07:22:19 Drop event: file:///E:/Sample/Eric Clapton - Running On Faith .flac

& 2015/08/19 07:22:19 Playlist add file: E:\Sample\Eric Clapton - Running On Faith .flac

2015/08/19 07:22:19 Playlist add file: E:\Sample\Eric Clapton - Running On Faith .flac

& 2015/08/19 07:22:19 Set transport (240):

2015/08/19 07:22:19 Set sampling rate: 11289600 (11289600)

2015/08/19 07:22:26 Set sampling rate: 5644800 (5644800)

2015/08/19 07:22:30 Set oversampling: 7

2015/08/19 07:22:30 Set sampling rate: 5644800 (5644800)

2015/08/19 07:22:35 Set convolution: 1

& 2015/08/19 07:22:35 GoTo

& 2015/08/19 07:22:35 Play

+ 2015/08/19 07:22:35 Playback engine running

2015/08/19 07:22:35 ASIO format set to PCM

2015/08/19 07:22:35 Rate available: 44100

2015/08/19 07:22:35 Rate available: 48000

2015/08/19 07:22:35 Rate available: 88200

2015/08/19 07:22:35 Rate available: 96000

2015/08/19 07:22:35 Rate available: 176400

2015/08/19 07:22:35 Rate available: 192000

2015/08/19 07:22:35 Rate available: 352800

2015/08/19 07:22:35 Rate available: 384000

2015/08/19 07:22:35 Default ASIO channels: 8 in / 8 out

2015/08/19 07:22:35 Default output channel 0: 1

2015/08/19 07:22:35 ASIO SampleFormat 17, 24-bit

2015/08/19 07:22:35 Default output channel 1: 2

2015/08/19 07:22:35 Default output channel 2: 3

2015/08/19 07:22:35 Default output channel 3: 4

2015/08/19 07:22:35 Default output channel 4: 5

2015/08/19 07:22:35 Default output channel 5: 6

2015/08/19 07:22:35 Default output channel 6: 7

2015/08/19 07:22:35 Default output channel 7: 8

2015/08/19 07:22:35 Set sampling rate: 44100 (44100)

2015/08/19 07:22:35 Rate or blocksize change triggered

2015/08/19 07:22:35 Rate: 44100, block size: 4704, frame size: 588

2015/08/19 07:22:35 Block size: 4704 (sample: 4), raw size: 0

2015/08/19 07:22:35 Playback engine ratio: 1

2015/08/19 07:22:35 Dither: triangular probability density function for 24-bit

2015/08/19 07:22:35 Initialization complete, starting audio engine

2015/08/19 07:22:35 ASIO channels: 64 in / 64 out

2015/08/19 07:22:35 Output channel 0: 1

...

2015/08/19 07:22:35 ASIO buffer sizes: 64/64/64 granularity: 0

2015/08/19 07:22:35 Using ASIO default buffer size

2015/08/19 07:22:35 Using ASIO buffer size: 64

! 2015/08/19 07:22:35 ASIO: unknown message

2015/08/19 07:22:35 Latencies: 64/64

2015/08/19 07:22:35 Not using ASIO output ready notifications

2015/08/19 07:22:36 ASIO engine started at 44.1 kHz, 16 channels, 64 sample buffer (16 channels)

+ 2015/08/19 07:22:36 ASIO engine running at: 44100

2015/08/19 07:22:36 Parallel threads: 8

2015/08/19 07:22:36 Parallel pipelines: 32

# 2015/08/19 07:22:36 clHQPlayerEngine::ThreadPoolInitThread(): AvSetMmThreadCharacteristics()

# 2015/08/19 07:22:36 clHQPlayerEngine::ThreadPoolInitThread(): AvSetMmThreadCharacteristics()

& 2015/08/19 07:22:49 Stop

- 2015/08/19 07:22:49 Playback engine stopped

- 2015/08/19 07:22:50 ASIO engine stopped

If output is changed to PCM, PCM can be upsampled even

[bibo01]

Miska,

 

Just to have a confirmation:

if I upconvert a stereo signal from RB to, let's say, DSD256 and map from 2ch to 8ch, is the upmix done after the upconvert or before?

Given that 2 stereo channel are simply repeated 4 times, it would seem strange in this situation to map channels before and process 8 channels together.

 

Although my friend has a rather powerful i7 CPU, the computer has problems to process 8ch DSD256. Eight DSD256 channels are about 10,8 MB/s. Could this be a problem for the LAN?

 

Thanks

[bibo01]

v3.8.2 is out

 

----------------------------------

Have you published new version for Embedded too?

 

In this new version it is possible to retain the playlist if Settings is opened.

Sometimes, however, if audio is playing, there can be a sound bump.

Furthermore, it would be good to get back to the last song playing, instead of the beginning.

[bibo01]

If we could have channel mixer stage after the upsampling stage, we could use our quad core i7 to do both upsampling to dsd256 and also channel mixer to replicate 2 channels to 8 channels.

Could we do such a request to you?

 

That makes things more complex to configure, and in addition you don't get as much performance gain. You would only save CPU time.

 

It would be quite a lot of work to implement too. This kind of feature would be best implemented in the DAC itself because there's no point in sending multiple copies of the same data, so you could ask Merging if they'd like to add it.

Please, if you do not mind, allow me to insist.

I have spoken to Merging (Dominique Brulhart) about this feature which I agree should be added by them. However, whereas their upcoming multi-channel NADAC is going to double up 1ch to 3/5/7 and 2ch to 4/6/8 automatically in stereo mode, they do not intend to add this feature for their Hapi DAC at driver level.

So I am here to kindly ask again if you could implement in Channel Matrix an option to process after DSP. Your effort would be VERY appreciated by Hapi users, knowing that is a lot of work involved.

[bibo01]

From a business perspective, how many Hapi users are there? Given you have a contact at Merging, that should be an easy number to get, at least a ballpark number.

 

As you mentioned "knowing that is a lot of work involved", don't you think knowing the size of the market this feature would potentially benefit would be helpful to know? If I was Miska, I would like to know this given the laundry list of other development requests/plans?

 

Sorry, not trying to shoot this down out of hand, but as you also mention - "allow me to insist" - on an item that I am guessing would benefit a handful of people but feel free to prove me wrong. For all I know, there could be 1,000's of Hapi users that also use HQP.

 

If we are "insisting", I would like to see Miska finish the DSC2 design.

The contact I have at Merging is simply "occasional". I was contacted by him directly through this forum.

Therefore it would be difficult to put a number for Hapi users and I doubt he would reveal it anyway.

By saying "a lot of work involved", I used Miska's words and thus trust his judgement. I also know that Miska himself would like to get one of these DAC down the line, so I was hoping there was a chance, perhaps slim, to convince him to add this feature, even for a handful of users.

 

I have already "insisted" with Miska - and he knows it - to finish DSC2 design too!

[bibo01]

You get best performance using current approach, but that requires quite a bit of processing power. Alternatively copying the same data to multiple channels is not nearly as good, but still better than doing nothing...

So we could say that current approach is the best sounding (channel mixer before upsampling engine) at the cost of high processing power and that copying the same data to multiple channels (channel mixer after upsampling engine) is not as good as the current approach but lesser processing power... so we could hope it could be implemented in the future in hqplayer?

Miska,

when you produced "-2s" filter version, you helped people who did not have enough power and were not able to use the top choice poly-sinc version.

[bibo01]

Miska,

 

have you tested Channel Matrix with Windows 10?

It seems to work in straight WinServer 2012, but not in Win10.

Also WinServer slimmed with AO does not work.

These problems refer to a Hapi DAC in both PCM and DSD.

[bibo01]

...just to add to the above message.

The problem in CM is with upsampling at the same time, at any resolution.

[bibo01]

Miska (another question),

 

if I insert a convolution filter obtained with Acourate in Channel Mixer instead of Convolution Setup, the filter is processed and convolution obtained.

Is it normal?

[bibo01]

HQPlayer Desktop and HQPlayer Embedded 3.9.0 released!

Modulator tuning for ASDM7 and one new filter. Track number metadata is now used for sorting tracks. Mac OS X version has now equivalent performance features as other versions (now requires OS X 10.9+). Plus usual small fixes and improvements.

[bibo01]

...

BTW, the previously mentioned Muso plugin, probably isn't out of alpha testing yet, as it doesn't recognise foobar2000 or HQPlayer on my system.

HQPlayer with Muso works pretty well. Have you tried to set it up under Tools/Options/Players?

Be sure to use latest build 01 of 2.3.22

[bibo01]

...

HQPlayer supports .m3u, .m3u8 and .pls playlist formats, .m3u8 being recommended format. HQPlayer doesn't support XML playlists (.xspf) at the moment. If your library scan folders include playlist files in recognized format, they are also added to the library, without metadata so they appear as unnamed folder in the main window view. You can also open playlist files using the corresponding toolbar button, or simply by just dropping the playlist file on HQPlayer window.

Is HQPlayer going to support .xspf playlist anytime soon? I was thinking in terms of integration with Tomahawk

[bibo01]

Miska,

 

I discovered probably a bug.

When single repeat is inserted, clicking on Next Track does not work but Previous Track does.

It's one way or the other, I suppose.

[bibo01]

Miska,

 

what are the characteristics of close form interpolation?

Most of all, when is it recommended? What hardware, what recording?

[bibo01]

Just downloaded 3.10 (Mac OSK 10.11) and tried out the new Closed Form Interpolator with both DSD 7 and ASDM 7, up sampling Red Book CD to 2xDSD. There's an underlying low-level popping noise that's continuously there every second or so. Anyone else hearing that? (Feeding Geek Pulse via USB)

I, instead, have a problem after about 1min of playback - sound stops, time counter still moving but sound hiccups. Unusable.

Setting: closed-form, dsd7, dsd128 on ArchLinux.

Back to poly-sinc it's fine.

[bibo01]

What CPU do you have? It is certainly heavy affair for high oversampling ratios. Doing 8x to 352.8/384k PCM is light, but doing something like 256x is heavy.

 

My Xeon E5v3 (Haswell-EP) can do DSD256 to iFi iDSD Micro (on Ubuntu Studio) without problems, but DSD512 is out of question.

 

I could make a "light" version of the conversion by relaxing quality requirements (now the target is 32-bit PCM worth of resolution). Relaxing it to 24-bit level or so would already make it quite a bit faster.

i7-4790

I should not have upsampling problems with such cpu in this condition. I think the problem resides in the filter itself.

 

Are you saying that closed-form is more demanding than poly-sinc, with which I have no problem, for doing 44.1>DSD128?!

[bibo01]

I think it should be fine. Not sure where the problem could be, since it is working fine for me on Linux too. Do you have pipeline SDM enabled? I do... I also have Buffer Time set to 100 ms.

 

Could be some incompatibility with ArchLinux, especially since I moved to C++11 in 3.9.

 

 

 

I didn't compare it to that one, but it is heavier than poly-shrt-mp-2s I'm using normally for going to DSD512, so I had to drop to DSD256.

With no NAA my buffer time it's fine at 5 ms with poly-sinc/short/mp filter (pipeline enabled) RB->DSD128.

With closed-form filter increasing buffer time to 50 ms (it may work with less) was enough to playback fine RB->DSD128.

[bibo01]

Miska,

 

there are various reports that v3.10 (Windows, OSX) has improved sonically compared to the past and it is also a touch more optimized in terms of resources.

Have you made any changes that could explain it? Different compiler perhaps???

I noticed that 3.10 is slightly smaller in terms of KB compared to the previous three versions.

[bibo01]

A picture is worth 1,000 words.

 

It's at the bottom.

 

 

[ATTACH=CONFIG]21458[/ATTACH]

 

Exactly! It is "closed-form", not "closed loop"

[bibo01]

Miska, please see also this - http://www.computeraudiophile.com/f22-networking-networked-audio-and-streaming/hqplayers-network-audio-adapter-13892/index40.html#post478951

[bibo01]

Are there any differences in the trial version of HQplayer and the paid version ?

 

As far as I know, the only difference is that the trial version is limited to 30min per session.

[bibo01]

Hello Miksa,

 

I am not sure if this has been discussed elsewhere but I am trying to go beyond DSD256 with my HQPlayer-Amanero setup on Win7 running on a core i7 PC. There is stuttering but CPU load is about 30% only so not being stressed at all. Tried different settings and even from DxD - thought this will be easier than all the way from 44k. Not sure if this an issue with HQPlayer or Amanero ASIO drivers. No problems up to DSD256 and sounds fantastic.

What DAC do you have that supports DSD512 through an Amanero board?

Sometimes DAC companies using Sabre chip and Amanero claim DSD512 capability, but in reality it is not supported. In this case stuttering does not depend on your system or HQPlayer.

[bibo01]

Hello. I am still with the trial version, and pretty impressed with the sound quality of HQP. But before going definitively onboard I would want to know more about some issues.

I have a PS Audio Direct Stream with Bridge II, a Mac Mini mid 2010, a Lenovo NAS and a Surface book. I am now using either Jriver thru Ethernet and the PSA bridge, or Roon via USB. I would prefer to keep on using the bridge. So my main question is if I would be able to use HQP with a NAA seated on the mac mini and the main HQP program managed by the Surface book, everything hooked up via RJ45. The other option could imply to "install" the NAA on the bridge and the HQ player on the Surface, avoiding therefore the use of the mac mini computer. Are either or those options feasible? If yes, how? Or if not, which would the alternative for keep on using the bridge of the Direct Stream? Regards, Ricardo.

Perhaps you do not get replies because there is not much experience with DirectStream with Bridge II, at least relative to HQPlayer.

You can use HQP with a NAA seated on the mac mini. Having HQP on a Surface book, it's possible but it may limit your usability. It depends on what CPU requirements you have. You can connect directly, but it is usually easier to connect HQP PC and NAA through a switch.

I do not have experience with Bridge II, but I doubt you can install networkaudiodeamon, the program necessary for NAA, on your Bridge II. I believe the Bridge is a UPnP device.

HQP offers a solution for UPnP - it's called HQPlayer Embedded and it works on Linux only. It is free for owners of Mac/Linux license.