MQN

[jrling]

The outstanding questions are can these techniques be used in other players ? Given the popularity of KS would a KS MQn sound better ? What would a linux MQn sound like ?

It looks like with Win 8.1 that WASAPI is very much at the heart of the audio engine, so MQn is in a good place to benefit from any improvements Win 8.1 brings.

I have been on the fascinating journey with SBGK on MQn almost from the start. I continue to be amazed at what he has achieved in such a short time, in his spare time and am fantastically impressed with the current results.

I do feel though that SBGK is reaching the limits of what can be achieved within the limitations of running on Windows and WASAPI, even like me with an optimised Windows Server 2012 R2 OS.

Most here will know the downside of using Windows as the OS for playing music. The lack of a UAC Class 2 native audio driver means most of us are having to run proprietary drivers for our USB converters/DACs. Hardly ideal and another variable and possible reducer of SQ in the chain.

I used MPDpup based on an optimised Linux OS for a long time before MQn. ALSA (the equivalent to WASAPI in Linux) is good and UAC2 compliant so no proprietary drivers required. Linux distros can be made very small, stripped of unnecessary OS clutter and focussed to playing audio, as is MPDpup. BUT there are no really good Linux players of the same quality as MQn that I found for Linux. MPD is quite good, open source but not in the same league as MQn.

So I would be very interested in SBGK's question - What would a linux MQn sound like ? I have no idea what amount of work would be required to port MQn to Linux, but I for one would be very interested to hear what could be achieved in SQ.

Jonathan

[Sligolad]

have uploaded 2.59 sse4 intel 8 4 16 16 4 to compare with 8 4 16 16 8, haven't listened to it yet.

Liked 8 4 16 16 8 best listening to 24, 88 track Elvis Costello - Im in the mood again.
8 4 16 16 4 sounded like it was trying to hard to sound good.

My hearing is a bit funny tonight, but thought I liked what I heard with 8 4 16 16 4, never heard the bass and piano in my Oscar Peterson test track so well expressed and the bass was still there, but with added detail over 8 4 16 16 8, found the 8 version a bit hooded, with 4 it is a more open sound. Also, at low volumes the bass was dominant whereas normally the treble is more dominant at low volumes and then the bass extends as the volume goes up. the 4 version had a better balance at low volumes. Shall see how it fares over the next few days, hopefully my hearing will return to normal.

That was for 16/44.1, might be different for 24/88

I have a lot of Nordost Valhalla wire in my system and with the Meitner and lots of linear power maybe I am getting a lot more micro detail as the 8 4 16 16 4 sounded a bit too detailed and in your face, not much in it though and it was more a gut feeling listening with my eyes closed that the 8 4 16 16 8 sounded less forced to me.
I say keep both with 8 4 16 16 4 for the very and maybe too resolving systems and 8 4 16 16 8 for all the sane folk!!
Cheers,
Pearse.

[Peter Stockwell]

"I'm not yet convinced it's any better than JRiver MC19 when JRiver is fed the same files".

I am surprised Peter with this statement, without any desire to argue too much on it, we all hear things different.

I would like to know what DAC/Amplification/Speakers you have hooked up to your computer. I do have MC19 too, and Naim DAC-V1/Nap100/Harbeth P3ESR's.

To me the difference is obvious, and I'm not even thinking to go back and listen JRiver, but I do use GUI to locate music in it.

It may only happen for playing music in the background when I'm doing something else. It is easy to create long playlists, plus I can use remote control to skip back and forth between the tacks.
For any critical listening, MQn is the only choice from now on, until something better comes around.

My system is in the signature (Matrix mini-I dac, Chevron Paradox Pre-amp, Avondale S-100, Audium comp-5 speakers), but before using MQn, I had a Musical-Fidelelity V-LINK-II in the system, too. The v-link-II only works on 24 bit packets, so MQn won't work with it. I believe my dac is better on coax, than on usb, but connecting the computer via usb to the dac, I can compare MQn and JRiver, but not directly.

I'm not hearing what I consider a huge difference, I might be hearing more nuance, more sense of performance, but I'm not yet sure of it.

I'm going to switch dacs, I have a Fidelity Audio DA-150 I can press into service, and I think the usb and coax are on a par with this.

[John Dot]

This is my first post here : I followed the herd from the CA forum to here, because I do like the MQn player.
On the CA forum I posted earlier (2 oct. post #1175) a link to my Autohotkey script to make it easier to use MQn with automatic conversion to .flac files. I use Flac.exe in my script, because I could not get Sox.exe to work properly in the scripts published on the MQn page.
Here is my second version : 3 text files in a .zip file in my dropbox with an extensive Readme included for the more inexperienced computer users. It is really not so difficult! (says I, a real computer nerd) Give it try!
https://dl.dropboxusercontent.com/u/907 ... 20V1.2.zip

Peter_T
thank you so much. Was playing with your prev version with no success. This one is working just great now, MQn is almost finished for me.
(would be great to have fully working 24/96 version, but I can live with what I have now).
Thanks!

[DaveF]

I intend to try this out over the weekend sometime. Now my PC is built for high end graphics so from an electrical point of view it's noisy but as it's all watercooled, it runs virtually silent.

Given my background in embedded systems, ASIC & FPGA design I'm very sceptical as to how the SW sounds different if its bit perfect. But I'm usually quite open minded so I'm willing to give this a go.

What other player should I compare it to? JRiver?

[jkeny]

My system is in the signature (Matrix mini-I dac, Chevron Paradox Pre-amp, Avondale S-100, Audium comp-5 speakers), but before using MQn, I had a Musical-Fidelelity V-LINK-II in the system, too. The v-link-II only works on 24 bit packets, so MQn won't work with it. I believe my dac is better on coax, than on usb, but connecting the computer via usb to the dac, I can compare MQn and JRiver, but not directly.

I'm not hearing what I consider a huge difference, I might be hearing more nuance, more sense of performance, but I'm not yet sure of it.

I'm going to switch dacs, I have a Fidelity Audio DA-150 I can press into service, and I think the usb and coax are on a par with this.

Peter, the qualities of MQN are both subtle when described individually & also important when taken as a whole i.e how the presentation of the music is greatly changed by these subtleties - it's the old adage, the whole is greater than the sum of the parts.

[cvrle59]

I just want to encourage people as we still haven't reached the target of £250.00 here.
Some people may not know about it, so these are links.
http://www.justgiving.com/MQn-Just-Good-Music
http://mqnplayer.blogspot.co.uk/

[LowOrbit]

I intend to try this out over the weekend sometime. Now my PC is built for high end graphics so from an electrical point of view it's noisy but as it's all watercooled, it runs virtually silent.

Given my background in embedded systems, ASIC & FPGA design I'm very sceptical as to how the SW sounds different if its bit perfect. But I'm usually quite open minded so I'm willing to give this a go.

What other player should I compare it to? JRiver?

Hi Dave

JRiver is a fair benchmark for comparison. It's widely used and known to offer "bit-perfect" playback. That or Foobar offer a reasonable baseline. Many people have been happy with either and the "bits is bits brigade" regard them as "as good as it can get" as they are uncontestedly bit-perfect when set up properly.

I am an eternal sceptic, have an engineering background and a fair understanding of digital audio processes and techniques. The bits are what they are but they are only one aspect of the complex chain of processes and entities involved in reproducing music from digital files. My main objection is that they only define the multiple instant measurements of amplitude. They do not mean anything without a temporal context. Getting a consistent, non-varying timebase from software to DAC I/V is where it gets difficult and where the "bits are bits" view seems to oversimplify the problem. And without an accurate, non-varying timebase accurately turning the bits into waveforms is impossible.

Getting as close to an absolute end to end time reference seems to be the main engineering challenge and from the experiments that have given us MQn, the software feeding the bits through the contended resource domain inside the PC seems to be a significant factor in establishing that reference.

[nige2000]

Claus
installed the ethernet driver on r2 the usual way without any issues
you must have forgotten something

have you got my core server install pack
all you need is in there

[sbgk]

So, to recap, where are we ? There has been a fair amount of non debate generated, most of it smelling of fish.

I think we are sometimes guilty of not seeing the wood for the trees, the objectivist/bits are bits fundamentalists are good at point by point retorts, but lets look at their position from a wider viewpoint.

To summarise their position

All bit perfect players sound the same foobar2000 is the same as jrmc is the same as squeezebox touch, itunes, CICs etc

All bit perfect audio engines sound the same Asio, wasapi, ks, alsa etc

Memory play sounds the same as non memory play

Different versions of the players sound the same

Different buffer settings sound the same

All O/Ss sound the same - xp, win 7, win 8, ws 2012, linux, osx, all sound the same

optimisers such as fidelizer don't make any difference

If you look at the slim forums , hydrogen audio and jrmc forums are there threads discussing buffer settings, best sounding versions, best audio engine settings ? Of course there are, these are the logical fallacies that blow their arguments out of the water. People in glass houses shouldn't throw stones.

So I think there is a tiny vocal minority of people arguing these points and a vast majority who hear differences and use those differences to increase their enjoyment of their hobby.