HQ Player

[Jud]

1 hour ago, AnotherSpin said:

 

I use xtr 2s. Do you hear PCM filter selection makes difference even if you upsample to SDM?

 

I can't upsample from SDM to SDM because my old computer doesn't have the resources.  Upsampling from PCM and modulating with an SDM, I was assuming so, especially because when I changed from -sinc-2s to -xtr-2s with DSD256 output, I got occasional dropouts and had to switch back.  But I could easily be wrong; obviously Miska will know.

[Jud]

27 minutes ago, Bob Stern said:

 

Minimum phase filters have zero pre-ringing.

 

Apodizing has nothing to do with pre or post ringing. 

 

Apodizing minimizes both.

[Jud]

46 minutes ago, Bob Stern said:

 

You got it backwards.  Your Vandersteen speakers (and my Thiel speakers) are designed to be minimum phase.  Linear phase requires symmetrical pre-ringing and post-ringing.  A system with pre-ringing is called "non-causal" because it produces a output prior to the impulse that caused it.  Pre-ringing (a non-causal system) it is impossible to produce using only capacitors, inductors and resistors, i.e., without DSP.

 

From a Thiel review:

 

Quote

The benefit of a first-order crossover is that it maintains the phase of the original signal, as long as the speaker that it's used in has the acoustic centers of the drivers aligned in the vertical plane, often through the use of a sloping front baffle. As long as your head is positioned so that your ears are equidistant from all the drivers, the phase of the direct sound that first reaches your ears shouldn't vary by more than a few degrees. That's why THIEL and other manufacturers refer to this as a phase-coherent design.

 

A filter that maintains the phase of the original must be linear phase, I thought.  No?

[Jud]

40 minutes ago, Bob Stern said:

Whether the chosen upsampling filter also minimizes (2) is orthogonal.

 

Yes, I grant if you design a very bad apodizing filter that itself rings like hell, you will not get a result with minimized ringing.

[Jud]

2 hours ago, Jud said:

 

A filter that maintains the phase of the original must be linear phase, I thought.  No?

 

49 minutes ago, Bob Stern said:

 

Here's some hairy theory:

 

https://www.dsprelated.com/freebooks/filters/Zero_Phase_Filters_Even_Impulse.html

 

https://www.dsprelated.com/freebooks/filters/Filters_Preserving_Phase.html

 

https://www.dsprelated.com/freebooks/filters/Minimum_Phase_Filters.html

 

I've read the articles and note linear phase filters and the special case of them called a zero phase filter are classified as filters preserving phase (what I mentioned), while minimum phase filters are not so classified.

 

Is there something key here I'm missing?

[Jud]

Just now, EdmontonCanuck said:

 

@Dr Tone, @Jud thanks very much for the web resources. I think I get it now.

 

So, with respect to the Ayre MP white paper, which HQP filter combination comes the closest to the last "best of both worlds" digital filter they gave as being optimal?

 

Dave

 

I'm guessing there may not be one.  That "best of both worlds filter" actually hardly filters at all, a philosophy with which I don't think Miska agrees (but I am as always ready to be corrected).

[Jud]

2 hours ago, Bob Stern said:

"Apodizing" — Since there was some disagreement above as to its meaning, here's what Charles Hansen (Ayre owner/designer) posted in 2009 in the MeridianUnplugged web forum:

 

 

 

Right, but you can find other relatively early articles where Meridian figures out a slow roll-off apodizing filter.  I think maybe Charles wanted more credit for the slow roll-off idea, but it preexisted his filter by many years.

 

In kind of a turnabout, I get the feeling Meridian took a lesson from Ayre in giving the MQA filters even less cut.

[Jud]

2 hours ago, Bob Stern said:

 

When I said minimum phase distorts the waveform, I was merely describing it mathematically, not expressing an opinion as to whether this is audibly good or bad.

 

I strongly believe that people hear things differently and vary in their sensitivity to different types of imperfections because the mind is an integral part of hearing.  Therefore, I would never argue that one audio design approach is universally superior.

 

16 minutes ago, Miska said:

 

It doesn't any more than linear phase. There are many views and angles to linear vs minimum phase and there's no simple clear statement about the subject. If you look at step response of of the two, first in linear amplitude scale and then in logarithmic amplitude scale you understand what I'm talking about...

 

For classical music I use poly-sinc (linear phase) and for everything else I use poly-sinc-short-mp (minimum phase). But I want to emphasize that's only my personal subjective preference...

 

 

Perhaps rather than distorting the waveform it is better to say phase differences move one to a different place on the waveform?

 

I agree with both of you that one approach isn't necessarily universally superior.

[Jud]

1 hour ago, Bob Stern said:

 

Scientists say the ear consists of thousands of very narrow band filters.  So perhaps dispersion (phase delay versus frequency) actually is a more helpful way to estimate the audible effect of a filter rather than the time-domain waveform.

 

I recall reading a long time ago that the brain's perception of the timbre of a note (e.g., whether the note sounds like a flute or a clarinet) depends entirely on the frequency spectrum during the initial attack of the note.  (I wish I remembered the supposed duration of that period.)  The spectrum of the sustained note after the attack contributes almost nothing to the brain's perception of timbre.

 

I'm not sure I've seen it put quite that way.  But there are some old well-established research results asking college music students to identify instruments with the initial transient attack portion of the sound removed, with very entertaining results regarding which instruments were mistaken for each other by these people who'd been listening to various orchestra instruments on pretty much a daily basis for years and years.  I may even be able to locate some audio.

 

("Timbre" tends to become ambiguous in usage.  It seems to have this connotation of just fundamental plus harmonics, but in fact it's *everything* that distinguishes, say, the trumpet note middle C from the piano note middle C, and that includes everything to do with attack, sustain and decay.)

[Jud]

5 minutes ago, Ralf11 said:

does it need that much hp for upsampling?

 

 

 

Some folks like to use it for multichannel upsampling to DSD512 with convolved room equalization or crossover filters.  That can eat up a few CPU cycles.

[Jud]

9 hours ago, mfsoa said:

Miska, will a linux-based NAA pass 512?

Thanks

 

If the DAC works at DSD512 with the Linux driver, yes.

[Jud]

4 minutes ago, Confused said:

It is my second day of experimenting with HQPlayer and I have simply been trying to see what works and what does not.  I have actually been surprised at how straight forward this has been, and I have been successfully running my mR Mutec MC3+USB at DSD256, DoP128, PCM upsamped, etc.  Everything appears to work fine.  One thing puzzles me though.  In the settings GUI pull down menu Bit Rate (/Limit) if I select 48k x128, then the Mutec indicates 256.  If I then select 44.1k x128, the Mutec shows 128.  This I do not understand!  I have taken a look though the manual which has not clarified this point.  Can anyone explain the specifics of these settings and why the Mutec shows 256 when 48x128 is selected.  So far I have been upsampling with mainly 16/44.1 material.

 

It is a little untraditional to offer the 48x rates, so the Mutec software running the sample rate display chooses its next rate above (44.1x) 128.

[Jud]

10 hours ago, Confused said:

I posted this a while back in the 'other' HQ Player thread, but with no response, possibly because it was not the right thread for the query!

 

Anyway, Is there a list of HQ Player NAA's anywhere?  (Meaning commercially available 'End point' type products)  I note that the Signalyst website has a list of 'recommended hardware', but this does not include kit such as the sMS-200 or Antipodes Edge.  In the absence of a list anywhere, if looking for commercially available HQPlayer NAA 'Network Endpoints', is there anything to add beyond Sonore, SOtM and Antipodes products?

 

I think beyond your list (there may be others, but I don’t know them; a Roon forum may be a place to look), many people like to DIY things by installing the NAA software on small, inexpensive Linux or Windows minicomputers. So that may be why you’re not seeing a lot of responses.

[Jud]

2 hours ago, Confused said:

I am sure those guys treated my query with the disdain it rightly deserved! 

 

I’m sure it neither deserved nor was treated with disdain.

[Jud]

41 minutes ago, ted_b said:

+1. As is life.

 

As are we.   (Unless you believe my mom.  She said to me once, “Jud, people don’t change, they just get more like themselves.”)

 

OK, back to HQPlayer...

[Jud]

13 minutes ago, GUTB said:

The truth can’t be stopped.

 

Unfortunately it appears inanity can’t be either.  But it can be ignored, so on my very short list you go.

[Jud]

Hi @Miska - I updated Lubuntu to 17.10 and it did not want to boot on my machine for some reason.  Since I have decided to reinstall anyway, I wondered whether you have an idea if HQPlayer will work with Lubuntu 18.04 (I believe it has moved to LXQt?).

 

Thanks.

[Jud]

On 11/11/2017 at 5:49 PM, Miska said:

 

Safest is to just go with currently supported 16.04 LTS. Since 18.04 is the next LTS version, I will move to it once it becomes upgradeable from the current 16.04 LTS on the first point release, meaning 18.04.1.

 

I'm using and recommending Ubuntu Studio 16.04 which uses XFCE desktop and has some tunings for audio use in general. But other desktops are fine too (LXDE, Unity, KDE...)

 

I don't recommend non-LTS releases because of their very short life/support span.

 

 

Thanks, Miska.

 

It turns out Lubuntu wasn’t failing to boot; it’s that I somehow managed to bork the GPU driver, so nothing shows on the screen for Lubuntu or any of the other Linuxen I’ve tried to install in its place since, not even terminal output.  (Windows is perfectly fine.) So looks like I’ll have to reinstall the driver over SSH.  As my wife and I are moving across country in less than two weeks, it will have to wait until we’re settled in the new place.

[Jud]

Is there an iFi DAC, or any other, that does DSD512 on MacOS today?

[Jud]

1 hour ago, lmitche said:

I am at home and now have the user manual for the IDSD pro. I can confirm that the IFI can be set to "DSD remastering" mode where all inputs, USB, aes, spdif, etc... and formats pcm and dsd, except dsd512 from the USB port, can be upsampled to either DSD512 or DSD1024 internally.

 

So if your source is a PC or Mac, as long as you send it anything but dsd512, it will upsample to dsd512 or dsd1024 if you configure it to do so.

 

Interesting. I am wondering how that works with MacOS drivers.

[Jud]

1 minute ago, lmitche said:

Jud, it is not clear what you are wondering about.

 

Hardware usually must work with OS drivers. But I'm assuming in order to supersede the usual DSD256 limit on MacOS, the iFi would have to be using internal hardware and firmware that are entirely separate from whatever computer source may be feeding them. I personally haven't encountered such an arrangement before other than units that weren't capable of anything like DSD512 or 1024.

[Jud]

9 minutes ago, lmitche said:

The IFI idsd pro doesn't supersede the dsd256 limit on Mac osx. If you play dsd256, or anything else the Mac is capable of, and you set the output format and upsampling rate in the idsd pro to dsd512 or dsd1024, whatever is input will be played at the dsd512 or dsd1024 setting. Format conversion and upsampling are all done internally in the DAC.

 

It's as though Hqplayer SDM upsampling functionality moved into the DAC. As Rajiv says above, SQ is unknown until we give it a listen.

 

We're saying the same thing in different ways.

[Jud]

Graphics card and mobo deal on Massdrop (I have not followed the computer hardware discussions especially closely, so I don't know and make no representations about the quality of the components or their performance with HQPlayer):

 

https://www.massdrop.com/buy/msi-gtx-1060-1070-1080-1080ti-gaming-x-bundles?1=1&utm_placement=0&referer=9QFFBA&mode=guest_open&utm_campaign=Automated Daily Promotional 2018-02-14&utm_source=SparkPost&utm_medium=email&utm_term=Daily Promotional&utm_content=1518597021392.805224564397184751610170

[Jud]

11 hours ago, austinpop said:

 

I assume HQPlayer spends the bulk of its time in compute-intensive functions or methods in user-mode.

 

The Meltdown/Spectre fix penalizes code that that has a high frequency of system calls. I doubt HQPlayer fits this profile.

 

Thanks - makes sense, and I like the more specific discussion (high frequency of system calls) vs. the general (without "speculative execution," some stuff will slow down).

 

I think the new power profile available on Win systems probably operates on a different level than speculative execution/branch prediction anyway. It works with stuff like when/whether the computer goes into sleep mode.

[Jud]

Well, I selected the power management profile so my computer wouldn't sleep while running a long backup task. But no, not with HQPlayer, which I assume is what we're interested in.