Jump to content
IGNORED

Visual studio 2012 c++ and wasapi minimalist player


sbgk

Recommended Posts

ONLY testing vocal:

 

MS vs intel (no cp vs cp)

-all 4 versions are good. no cp has more micro-details/better tune. cp is more airy/bright. prefer no cp.

-agree 8 4 seems better than 8 8. they equal in detail.

 

2.59 8 4 ms drv vs 2.56 8 8 16 16 8 eax add

-2.56 has more emotion. much more micro-details than all versions.

-this 2.56 is my farvorite vocal. unfortunate hardness built-in....

 

2.59 8 4 ms drv vs rax no zi

-2.59 has slightly more micro-details.

-rax no zi has more emotion.

 

2.59 8 4 ms drv vs 2.57

-2.57 probably has the best emotional focus of all versions. emotional and mellow and focus and decent details. lacks fullness. tire after a while.

 

that's all the best vocal versions i am keeping now.

 

thanks, does the SQ change if wireless is switched off/disabled ?

There is no harm in doubt and skepticism, for it is through these that new discoveries are made. Richard P Feynman

 

http://mqnplayer.blogspot.co.uk/

Link to comment
I use windows server 2012, trying to play wav files... Sorry for all the questions, I imagine you're busy with other things ;-)

would help if you answered the questions. ws 2012 core mode doesn't have wasapi, can you try wasapi_test, what format - 16/44 or other ? what is your dac ? what is the file name ? the more info the better.

There is no harm in doubt and skepticism, for it is through these that new discoveries are made. Richard P Feynman

 

http://mqnplayer.blogspot.co.uk/

Link to comment
little bit, but if your going to do that ethernet built in sound cards all unnecessary ports really should be disabled in bios

 

that's what I do

 

have uploaded mqnplay.exe 2.59 sse4 intel 8 4 16 16 4 8 which is an attempt at 2.56 8 8 1616 8 add eax with 2.59 settings

There is no harm in doubt and skepticism, for it is through these that new discoveries are made. Richard P Feynman

 

http://mqnplayer.blogspot.co.uk/

Link to comment
would help if you answered the questions. ws 2012 core mode doesn't have wasapi, can you try wasapi_test, what format - 16/44 or other ? what is your dac ? what is the file name ? the more info the better.

 

would help if you answered the questions. ws 2012 core mode doesn't have wasapi, can you try wasapi_test, what format - 16/44 or other ? what is your dac ? what is the file name ? the more info the better.

 

Thanks. Not sure if I understand all the questions correctly but here I go:

- I use gui mode, clean install (will try wasapi_test as soon as possible), have installed jplay now but it's not running when trying mqn

- I play 16/44 files

- my usb-spdif is jkenny mk3 (hiface based) to a metrum octave dac

- I'm not exactly sure what you mean with 'file name'

- I have to try the sse2 - 2 core combo still

Link to comment
1024/96000 = 0.01066666666 amount of time needed to fill 1 buffer.

1024*3/96000 = 0.03200 amount of time needed to fill 3 buffers.

i am guessing if you set clock rate at 0.032, it will fill 3 buffers as a batch with perfect timing.

i see XA does this too but doesn't explain. just guessing

 

I guess we are trying to synchronise the timing with the buffer loads, there are 2 x 16/8 x 44100 Bytes transferred per second for 16/44.1 = 176400 which is 176400/8192 = 21.54 buffers a second = 1/21.54 = timing setting of 46440 or in samples it is 2048/44100 = 46440, so my previous setting was wrong, oops

There is no harm in doubt and skepticism, for it is through these that new discoveries are made. Richard P Feynman

 

http://mqnplayer.blogspot.co.uk/

Link to comment
Thanks. Not sure if I understand all the questions correctly but here I go:

- I use gui mode, clean install (will try wasapi_test as soon as possible), have installed jplay now but it's not running when trying mqn

- I play 16/44 files

- my usb-spdif is jkenny mk3 (hiface based) to a metrum octave dac

- I'm not exactly sure what you mean with 'file name'

- I have to try the sse2 - 2 core combo still

 

there are problems with unicode characters in the file name of the wav file, try stopping the jplay service. don't expect 2 core control to change anything if sse2 doesn't work. have you optimised ws2012 you may have disabled the 2 audio services required for wasapi ?

There is no harm in doubt and skepticism, for it is through these that new discoveries are made. Richard P Feynman

 

http://mqnplayer.blogspot.co.uk/

Link to comment

MQn.png

 

The MQn logo, would include it in the program, but it affected the sound. It is made up of the letters MQN placed within a square so that it reflects the ancestry of the turntable and the transition to computer audio with the two triangles in the upper right representing the optimised transfer of data from memory to the end audio device which defines the MQn sound signature. It also communicates the bare bones nature of MQn and squaring the circle of getting analogue turntable like sound out of a computer.

 

There, how did you like that marketing mumbo jumbo ? Could put in a bit about ying and yang, but that would be too pretentious.

There is no harm in doubt and skepticism, for it is through these that new discoveries are made. Richard P Feynman

 

http://mqnplayer.blogspot.co.uk/

Link to comment
that's what I do

 

have uploaded mqnplay.exe 2.59 sse4 intel 8 4 16 16 4 8 which is an attempt at 2.56 8 8 1616 8 add eax with 2.59 settings

 

removes the hard edges of 2.59 sse4 intel ms drv 8 8, but still plenty detail, sounds very good.

There is no harm in doubt and skepticism, for it is through these that new discoveries are made. Richard P Feynman

 

http://mqnplayer.blogspot.co.uk/

Link to comment
[ATTACH=CONFIG]8228[/ATTACH]

 

The MQn logo, would include it in the program, but it affected the sound. It is made up of the letters MQN placed within a square so that it reflects the ancestry of the turntable and the transition to computer audio with the two triangles in the upper right representing the optimised transfer of data from memory to the end audio device which defines the MQn sound signature. It also communicates the bare bones nature of MQn and squaring the circle of getting analogue turntable like sound out of a computer.

 

There, how did you like that marketing mumbo jumbo ? Could put in a bit about ying and yang, but that would be too pretentious.

 

how does it affect the sound?

that weird

Link to comment
I guess we are trying to synchronise the timing with the buffer loads, there are 2 x 16/8 x 44100 Bytes transferred per second for 16/44.1 = 176400 which is 176400/8192 = 21.54 buffers a second = 1/21.54 = timing setting of 46440 or in samples it is 2048/44100 = 46440, so my previous setting was wrong, oops

 

I am a bit confused. In a previous posting I thought that you recommended 46440 as the best setting? Or have I missed an alternative setting along the road?

Link to comment

Hi SBGK,

 

Question About: Tasks\P ro Audio registry settings.

 

Are these settings meant to be running ‘mqnplayer’ and ‘mqncontrol’ in high priority?

 

[HKEY_LOCAL_MACHINE\SOFTWARE\Microsoft\Windows NT\CurrentVersion\Multimedia\SystemProfile\Tasks\P ro Audio]

"Scheduling Category"="High"

"GPU Priority"=dword:00000008

"Affinity"=dword:00000000

"Clock Rate"=dword:0000b568

"SFIO Priority"="Normal"

"Priority"=dword:00000008

"Background Only"="False"

 

I applied these registry changes, re-started server 2012, checked if registry changes where applied and than checked in task manager in what priority mqnplayer and mqncontrol where running. They ran in normal priority. Also audiodg.exe ran in normal priority.

 

Are these registry settings meant too change the priority-level of these processes?

 

Other questions related too this:

What audio-quality related processes should preferably run in ‘high priority’?

In task manager I can also set processes too ‘Real time’. Would this be better than ‘high priority’?

 

 

Mark

Cable Modem/Router < Cat 7 -> Netgear Switch GS108Ev3 -> Cat 7 (25 meters) -> 2nd NIC on Thunderbolt -> Mac Mini (HD-plex -> Uptone DC-Conversion / Linear Fan Controller Kit (MMK), OSX on SD-card, Wifi-module physically removed, SSD unplugged from power and SATA, Audirvana, Sonarworks Room EQ).
* Mac Mini -> Cat 7 -> Merging Hapi -> Vovox Mucolink D-sub 25 Direct SD 100 -> Vovox Direct SD XLR -> PMC TwoTwo.8
* Mac Mini -> Supra USB 2.0 -> Crane Song Solaris -> Vovox Direct S XLR -> PMC TwoTwo.8
All LAN cable: ISTP Belden Cat.7 1885ENH with Telegartner MFP8 Cat 6a RJ45 plugs.
LAN shield connected at both ends to plug. Modem & switch powered with Linear PSU.
All interconnections: Vovox.

Link to comment

some questions and answers for those who question why MQn would make a difference

 

1. all players sound the same ?

 

a. this is patently not true and yet is the premise of some peoples beliefs. If yes, then either your hearing/system is poor or immune to player differences, even reclocker manufacturers say the player makes a difference, try changing the buffer size on your player, you should hear a difference.

 

The question should be - given the thousands of variables in computer audio playback how can two different players sound the same on the same system ?

 

2. how can MQn make a difference when it's so small ?

 

a. MQn optimises the only bit of code that is running during music playback and has cut out the extraneous instructions that would normally run, up until MQn the requirement for this code was to do it fast enough so that it worked, without attention being paid to the impact on SQ. This code functionality is common to all players ie moving data from one device to another and is the most critical to SQ as demonstrated by the various versions of MQn.

 

for a glimpse into the depraved thoughts of the modern day internet troll look at the MQn thread on pink fish media, getting laughs out of children helped by Unicef, hope karma is a bitch.

There is no harm in doubt and skepticism, for it is through these that new discoveries are made. Richard P Feynman

 

http://mqnplayer.blogspot.co.uk/

Link to comment
some questions and answers for those who question why MQn would make a difference

 

1. all players sound the same ?

 

a. this is patently not true and yet is the premise of some peoples beliefs. If yes, then either your hearing/system is poor or immune to player differences, even reclocker manufacturers say the player makes a difference, try changing the buffer size on your player, you should hear a difference.

 

The question should be - given the thousands of variables in computer audio playback how can two different players sound the same on the same system ?

 

2. how can MQn make a difference when it's so small ?

 

a. MQn optimises the only bit of code that is running during music playback and has cut out the extraneous instructions that would normally run, up until MQn the requirement for this code was to do it fast enough so that it worked, without attention being paid to the impact on SQ. This code functionality is common to all players ie moving data from one device to another and is the most critical to SQ as demonstrated by the various versions of MQn.

 

for a glimpse into the depraved thoughts of the modern day internet troll look at the MQn thread on pink fish media, getting laughs out of children helped by Unicef, hope karma is a bitch.

 

I've been a member of Pinkfishmedia for ten years or more. I rarely bother with the Audio room, it's full of twonks. But I couldn't resist a riposte on the MQn thread. No one will listen, but that's just it - too many of them look at the logo and the price tag before making up their mind's how things sound.

 

Just been listening to2.59 SSE4 Intel ms drv 8 4 - Now that is sweet.

 

 

 

MarkW

Link to comment

"or a glimpse into the depraved thoughts of the modern day internet troll look at the MQn thread on pink fish media, getting laughs out of children helped by Unicef, hope karma is a bitch."

SBGK, I would seriously advice you not spend any second of your live reading such ignorance, that I have never read in my entire life.

You've got more important, and more exciting things to deal with.

To talk such way about something that you've never tried goes too far to negative side on "my stupidity scale". Those people never heard an old saying "DO NOT SPIT ON A HORSE, YOU'VE NEVER RIDDEN"

I'm listening Diana Krall right now, and all I can say is, she is sitting in my living room singing just for me.

Yes, I still use JRiver, it helps me to nicely locate my music and pool it into MQn, nice GUI.

JRiver/CTRL-C>>>>>>>MQn.bat/2xClick's>>>>>>So much fan!

Link to comment
1. all players sound the same ?

 

a. this is patently not true and yet is the premise of some peoples beliefs. If yes, then either your hearing/system is poor or immune to player differences, even reclocker manufacturers say the player makes a difference, try changing the buffer size on your player, you should hear a difference.

 

Are you saying this is true independent of the type of DAC, type of interface (USB, s/pdif, etc), and type of clocking/buffering architecture in the DAC?

 

The question should be - given the thousands of variables in computer audio playback how can two different players sound the same on the same system ?

 

Because none of those variables actually affect the data being sent to the DAC?

 

a. MQn optimises the only bit of code that is running during music playback and has cut out the extraneous instructions that would normally run, up until MQn the requirement for this code was to do it fast enough so that it worked, without attention being paid to the impact on SQ. This code functionality is common to all players ie moving data from one device to another and is the most critical to SQ as demonstrated by the various versions of MQn.

 

And how much of the actual code that is involved in playing the music does MQn represent, and how much is Windows kernel, device drivers, and WASAPI layers?

Link to comment
I am a bit confused. In a previous posting I thought that you recommended 46440 as the best setting? Or have I missed an alternative setting along the road?

yup, thought I said a lower number previously, still don't understand how it works, how do the timer events and buffer fill events gets synchronised ? maybe it's a compromise between too much noise being produced if it was set to 1 46440, maybe dividing 46440 by 2 until the optimum SQ is achieved would be a way of finding the best value, don't think higher values would have anything to offer.

There is no harm in doubt and skepticism, for it is through these that new discoveries are made. Richard P Feynman

 

http://mqnplayer.blogspot.co.uk/

Link to comment
Are you saying this is true independent of the type of DAC, type of interface (USB, s/pdif, etc), and type of clocking/buffering architecture in the DAC?

 

 

 

Because none of those variables actually affect the data being sent to the DAC?

 

 

 

And how much of the actual code that is involved in playing the music does MQn represent, and how much is Windows kernel, device drivers, and WASAPI layers?

 

i admire your insight, julf. please just don't troll here.. no one feed julf pls. otherwise he won't stop. i have learnt my lesson a while back.

Link to comment
"or a glimpse into the depraved thoughts of the modern day internet troll look at the MQn thread on pink fish media, getting laughs out of children helped by Unicef, hope karma is a bitch."

SBGK, I would seriously advice you not spend any second of your live reading such ignorance, that I have never read in my entire life.

You've got more important, and more exciting things to deal with.

To talk such way about something that you've never tried goes too far to negative side on "my stupidity scale". Those people never heard an old saying "DO NOT SPIT ON A HORSE, YOU'VE NEVER RIDDEN"

I'm listening Diana Krall right now, and all I can say is, she is sitting in my living room singing just for me.

Yes, I still use JRiver, it helps me to nicely locate my music and pool it into MQn, nice GUI.

JRiver/CTRL-C>>>>>>>MQn.bat/2xClick's>>>>>>So much fan!

 

you can do that with JRiver, that's useful, is that what peter_t was trying to emulate.

There is no harm in doubt and skepticism, for it is through these that new discoveries are made. Richard P Feynman

 

http://mqnplayer.blogspot.co.uk/

Link to comment
yup, thought I said a lower number previously, still don't understand how it works, how do the timer events and buffer fill events gets synchronised ? maybe it's a compromise between too much noise being produced if it was set to 1 46440, maybe dividing 46440 by 2 until the optimum SQ is achieved would be a way of finding the best value, don't think higher values would have anything to offer.

 

according to XA, just don't go over more than 5 digits clock rate(e.g. 464000 is bad). 4 digits is good. 5 digits is best, apparently.

Link to comment
you can do that with JRiver, that's useful, is that what peter_t was trying to emulate.

I'm not sure what peter_t said, I missed it, sorry.

Once you locate an album or a play list, you just highlight what you want to listen than copy into clipboard. The rest is the same, just two clicks on MQn.bat.

I love it because I don't need to mess around through folders by Windows Explorer. I can take advantage of JRiver's GUI.

Link to comment
i admire your insight, julf. please just don't troll here.. no one feed julf pls. otherwise he won't stop. i have learnt my lesson a while back.

obviously MQn has just the right amount of code, how much code is optimal ? it would be funny except for the fact a grown man is holding up the squeezebox touch as being the pinnacle of computer audio reproduction. wonder if he ever steps out of his mum's cupboard.

There is no harm in doubt and skepticism, for it is through these that new discoveries are made. Richard P Feynman

 

http://mqnplayer.blogspot.co.uk/

Link to comment
that's what I do

 

have uploaded mqnplay.exe 2.59 sse4 intel 8 4 16 16 4 8 which is an attempt at 2.56 8 8 1616 8 add eax with 2.59 settings

 

2.59 sse4 intel 8 4 16 16 4 8

-tried it. loved it. but i shall kept this version for vocal only. don't think this is an all-around solution.

-no offense to newer versions, i still occasionally use the old 2.82, 1.2 and 2.41 movq. i just feel that MQn is yet to reach its full potential.

-for new versions, my farvorites are rax no zi, tot align and this 2.59 sse4 intel 8 4 16 16 4 8.

-for tuning OS/bios, i still use XA, coz i have used it the longest, know it the best. i don't know another player that is more sensitive to tuning than XA.

Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...