HQ Player

[wisnon]

Since the public demo version is dated 4/18/14, it apparently is the old one that requires a NAA client, not the new Beta Miska just announced.

 

Where do we DL the Beta version? What is the URL link?

Thx

[wisnon]

Thanks, Jussi told me yesterday.

 

Please reply to my PM...

[wisnon]

So what ARE your PCM/DSD settings?

[wisnon]

Seems to be same problem as Metric Halo has, now fixing it became much easier because I should be able to reproduce the problem here...

 

 

 

When will the commmercial Mac OSX NON-NAAA version be ready?

[wisnon]

Thank you very much for the information, frankly, I was ready to buy the SOTM for NAA, but when the OS X NAA software was released, it gave me the chance to prove to myself that this was the way to go, as opposed to directly connected DAC. I had a Mac Mini available, so it cost nothing and I only needed Dop to my DAC.

 

Im not advocating any OS or platform, I just was able to really improve my SQ without spending any money and it certainly seems logical that a more minimal OS/hardware platform would wow me even more. Now spending the money on that kind of solution is back on my mind. Thanks again.

Lionel

 

Can you elaborate on this Mac Mini setup you describe with OSX NAA? I am not sure what you are saying…

 

Why dont you need the SoTM?

[wisnon]

Im using the Mac Mini to run the OS X version of the NAA software that was recently released.

So you have 2 Macs, or is the HQP and the NAA software on the same Mac?

[wisnon]

Yes, HQP on MBP i7 and NAA on Mac Mini.

 

OK, understood now.

[wisnon]

Not yet, and unfortunately I am away for several weeks :-(

 

Whilst I am away, I will be dreaming about Fitlet ;-)

 

Cant find a price for the Fitlet. It looks potentially very good.

[wisnon]

James, I will email you the quick start guide that Geoff did that has settings for Exasound. You can use to extrapolate to Mytek. I used yesterday to test on ifi iDSD Micro and it worked. Kept crashing when I tried to load trees, so I only loaded actual folders.

 

AL, next version of AO (version 1.31) will automatically work with HQP in core mode, I believe.

[wisnon]

James...apologies. I mistook you for another audio pal named James in Melbourne. He reads, but never posts here...

 

How does Mytek do DSD128? Does it need 2 wires plugged in?

[wisnon]

Edward, with your exaSound DAC your USB is already galvanically isolated (assuming you haven't de-isolated it with a dual power supply). Why the Corning, then?

 

Ted, the Corning is NOT galv isolated (thin copper power wires in the cable) and no Dac is ...except after the USB receiver and before the Dac chip. See the except I posted today over at the 2Qute thread at headfi.

 

SuperDad specifically made a post about Exa Gal isolation and said its pretty standard stuff and its after the PHY...

 

The working theory about the Corning is that it increases the ground plane resistance to the Regen green and with the amber where the resistors are now embedded in the PCB, they think the need for Corning is now eliminated..

 

Excerpts:

SuperDad

Hi Miska.

Thanks for the reply. John can explain how some of both the GPN and the general spiking/switching noise on the chips makes it past isolators (it is not all just GPN). But I'll toss out a few comments from what he has taught me:

a) Choice and implementation of galvanic isolators varies quite a lot between the DACs that use them, and they always come after the main USB processor. For example, exaSound puts theirs a surprisingly great distance away from both their FPGA (for which they created a custom USB core--but do they also use it for filters and other things?) and the ESS DAC chip.

Or another example is the JLSounds I2SoverUSB (I just bought one because it can take an external clock and can output directly to a PCM1704 so we can simplify and improve my personal proto-reference PCM DAC): It uses the Silicon Labs Si8661BC isolator which uses an RF transmitter and receiver on either side of a barrier. Those have about 3.5 times the jitter (all isolators do of course, that's why a reclocking flop should always be used right afterwards) of John's preferred type and also may emit some RF, plus the way JLSounds uses its 6 channels and the connector pins, an external clock can not go through an isolator "clean" pin.

 

b) John told me that all the isolators are good at one part of the spectrum but not at all parts (I forget the ranges). So they are not a panacea.

 

c) For a variety of reasons, John is not convinced that the processors with integrated PHYs are the way to go. We have been avoiding them (e.g. the XMOS U versions).

 

 

and

 

 

Originally Posted by EuroDriver

 

Wow it's the first time I have seen those noise spectrum graphs, the GI seems to be very well implemented. So how come on the E22 an optimized OS can make such a difference to SQ when the CPU is doing intensive computation ?

 

Because galvanic isolation is not a cure-all. It is just one of many elements in a good design.

 

Edit: you were in these threads and would know all this, so perhaps I am not addressing your real concern?

[wisnon]

Thanks.

 

Here is the post on the E28 Galv Isol. Post #45...

http://www.computeraudiophile.com/f8-general-forum/schiit-wyrd-universal-serial-bus-industry-standard-cables-connectors-and-communications-protocols-between-computers-and-electronic-devices-decrappifier-22085/index2.html#post361698

 

Having looked closely at the e28 PCB, I can tell you that he places the isolators AFTER the FPGA (the same one that runs the filters and everything else), just before the ESS DAC chip (though kind of far from it). That's fine and pretty standard, but not quite what we think of when we talk about the USB input being galvanically isolated.

[wisnon]

Back to our regular program in a minute, but just had to interject something:

I just got out of a wonderfully musical hour+ of listening to two of my new SACD rips (Roxy Music Stranded and Roxy Music For Your Pleasure...thanks PS), two SHM analog-to-DSD transfers, played back at convolved DSD64 to DSD256 (11.2Mhz). WOW! An amazingly musical and time-machine experience. It was a massive reminder of why we do this stuff.

 

Thanks Jussi and the HQplayer gang here.

 

Hijack over.

 

Good hijack Ted.

 

I can only imagine hyow good it will sound when you deploy the full arsenal with AO style Win10 and Regen upconverting with HQP. I know your server is already tricked out.

 

Please be kind enough to do a Win10 conversion tutorial, as you are great at making those.

[wisnon]

Here is a theoretical idea being discussed a bit over on the T+A thread. These are my takes on what I am calling a proposed option called "HQP Offline".

 

Since DSD512 realtime upsampling is so cpu intensive and requires some major hardware commitment, and since many have asked for other GUI front ends for HQPlayer, what about this crazy idea: Jussi develops an offline capability to upsample (using same choices of filters, modulators, etc) to, say, DSD512 (or whatever). Yes, the storage requirements are massive, but it's not like we all listen to ALL of our library anyway. Offline upsampling cold be done on a need to hear basis, of course.

 

My stupid brain says this might free up the cpu needs somewhat (i.e batch process the upsampling overnight and let lower powered cpus take their time doing the work since this is not required realtime) and maybe free up playback to most any player or control point/renderer (assuming the final playback stage of HQP is not anything required....I do not know cuz I've never played HQPlayer in "passthrough" mode). Just spit-balling here.

 

My $.02

 

Me likey!

[wisnon]

Idea is good

Someone with low powered pcs can enjoy dsd upsampling

 

But then how can you choose one day before what you want to listen?

Personally choice of music depends on how the day went, how is the weather, etc

 

Also if you see someone who can afford a dsd 512 capable dac can obviously afford an i5 based pc

In recent times we are getting pocket rockets like ifi micro idsd, but not sure how good they are as have not heard them

 

Recent times? I have owned that Dac for over a year now!

 

I though to avoid stutter in real time 512 upsampling you need i6700k SKYLake, not an i5.

[wisnon]

I hear what you say, but the ifi existed before there was any practical means to deliver DSD512 to it. Recent times for me is the last few months since SGM and Skylake CPU based computer have been available to do the job.

 

I want it all. Offline and on the fly upsampling capabilities. suppose I want to take my ifi to a pal with an i5 computer and show him/her what DSD512 sounds like. I could capure a few demo tracks at 512 and take them along to their place with the idsd micro and impress the hell outta them. No way would I be lugging a beastly 'Puter. LoL

[wisnon]

I agree with what you say and reasons for it

 

But I am not sure if Miska will be so keen to develop offline upsampling.

 

Unfortunately, I think you are correct! Jussi may be concerned about his IP.

I think similarly Rob Watts of Chord does a trick to shield filter data from being captured in his Watts Transient Aligned (WTA) filter module in the FPGA of his Chord Dacs. HQP is software, so perhaps is more easily reverse engineered?

[wisnon]

HQPlayer keeps ASIO driver loaded as long as it is running or until you go to the Settings dialog...

 

Hey Jussi,

 

I have an Intel i7 4700T (2.7ghz) quad core Server with an MSI MS-7851 motherboard and 8gb Ram.

I use Win2012 in A.O GUI mode and HQP shell to do direct playback to Amanero.

 

I cant get DSD256 upsampling to work reliably without stutter. After 10-15 seconds, it starts to stutter. DSD128 is no problem at all. I tried all filter combinations and increased buffer size to no avail. Could it be a BIOS setting or some other computer setup issue? Indeed, I cant get DSD256 upsampling to work at all with JRiver! Even native 256 was a problem.

 

Native 256 playback is not a problem with HQP at all.

 

I would imagine that an i7 4700T should have no problem doing DSD256 upsampling, especially with -2 filters! I dont think I need to upgrade to Skylake 6700K unless I want to do DSD512 upsampling.

 

Any tips appreciated from you or anyone.

[wisnon]

I have an i7-3770S (3.1ghz, quad) also with 8gb RAM running AO and I can do DSD256 upsampling from PCM. A couple of things that come to mind:

 

Are all 4 cores active?

Have you tried adjusting the buffer settings in your DAC driver?

 

Thanks Tboooe

 

No I have not. I am more a Mac guy and not sure how to do either of these things. Any pointers...especially for activating all 4 cores?

 

I have an Amanero driver...not sure how to adjust buffer settings, other than doing it in HQP as I described in my original post. Max was either 100ms or 250ms.

[wisnon]

Make sure that Pipeline is checked under the Setting Menu. This spreads out the workload across 4 cores.

Thnks

 

That was already done.

[wisnon]

With Amanero ASIO driver it should work even if you leave "Buffer time" setting in HQPlayer to "Default". With your CPU, with -2s filters and Pipeline SDM enabled it should work fine.

 

I would recommend trying out with stock Win 10 Pro and HQPlayer, nothing else and no "optimizations" (all my Windows installations are stock, just with the necessary device drivers but nothing else).

 

Thanks fellows,

 

I will try the suggestions and report back.

Jusii, I only have win2012...but will try in GUI or minimal server mode...

[wisnon]

@ted_b, it would probably be easier to write a virtual dac driver that just sends the stream to hqplayer, a la roon.

 

 

Sent from my iPhone using Computer Audiophile

 

Who can do that?

[wisnon]

I usually just pick a filter and listen to it for at least one full session, unless there is stuff going on I don't like then I switch. As to what filter I like best for song x, thats way to anal for me. I find in general I prefer MP filters better than linear filters, I think that means I am less offended by post ringing than pre-ringing. Some of my favorite filters are poly-sinc-shrt-mp, poly-sinc-shrt?-xtr-mp 2s (the non-2s take forever to initialize, I haven't tried beta 15.2 yet), and for some reason I liked closed form and even mining FIR. If I had to pick only 2 then poly-sinc-shrt-mp and the XTR (2s) version of it. Really been digging the XTR lately. I have found that this xtr filter does not like 48K based files (with my dac), my dac will only do 22m6 (even multiple of 44.1) and does not support 24m6 (even multiple of 48). I am getting a T+A dac 8 dsd this week and that will solve the 44 vs 48k issue as it supports both 44 and 48 multiple rates up to 512x. Since August I have been up sampling to DSD512.

 

Quadman, What Dac do U have now?

[wisnon]

Currently I am running a Gustard X20u (non-pro version) heavily modified with EVS level 1 mods, a OCXO clock added near the dac chips and a DIYinHK USB DSD512 capable board running directly into the I2S lines. For the $ it is a killer dac.

Thanks for the reply. What did you have before?

[wisnon]

Of course on further inspection nmedium is for Windows phones only. But the bigger question is how, as a JRiver remote app, it remotely controls the handoff from JRiver to HQPlayer. The only way I've seen it done is using the JRIver GUI to drag and drop, as a sort of better-than-Windows-Explorer GUI. The WIndows on Top youtube tutorial is where I started to do that. The problem there is that RDC doesn't always like Windows on Top, and there is no remote functionality otherwise.

Would a Windows tablet running Win10 sub for a Windows phone?