Visual studio 2012 c++ and wasapi minimalist player

[Clive]

I've been listening more to 2.40 SSE4 intel and 2.38 SSE4 intel. The only aspect that in my system & pc that's not superb is the treble balance. Vs jplay 5.1 and my two record decks I think the sound is just a little bright (just a db or two propably) which at the same time makes it seem there's less bass. On albums with little extended treble the sound is totally superb, great soundstaging and imaging, clarity and detail with great bass driver & timing. Are there other versions I should try for my perfect balance? Overall I prefer 2.38 to 2.40 as 2.38 is easier to live with.

 

For hi-res is all I should need to do is use either v5 or v6 mqncontrol?

 

I'd love to know how the measurists explain the differences in sound when it's all bit perfect. :-)

 

BTW I'm using an almost untweaked W8 laptop, it doesn't have the network connected, there's no anti-virus or firewall either (services stopped).

[sbgk]

I've been listening more to 2.40 SSE4 intel and 2.38 SSE4 intel. The only aspect that in my system & pc that's not superb is the treble balance. Vs jplay 5.1 and my two record decks I think the sound is just a little bright (just a db or two propably) which at the same time makes it seem there's less bass. On albums with little extended treble the sound is totally superb, great soundstaging and imaging, clarity and detail with great bass driver & timing. Are there other versions I should try for my perfect balance?

 

For hi-res is all I should need to do is use either v5 or v6 mqncontrol?

 

I'd love to know how the measurists explain the differences in sound when it's all bit perfect. :-)

 

BTW I'm using an almost untweaked W8 laptop, it doesn't have the network connected, there's no anti-virus or firewall either (services stopped).

 

if you are using the laptop and experiencing treble issues then you should look at the power supply (I use a linear supply, does it improve when using the battery ?) also the external disk power supply (I use linear) also the spdif convertor power supply (I splice a linear 5v into the usb cable at the receiving end) also set the power setting to 0% for min and max in the power plan. Are you using jplay on the same machine ? Also important to disable the usb power setting - allow windows to switch off power to this device setting etc.

 

24 bit dacs try v5 or v6 (v6 has an attempt at 16 bit in 24 as well), but uses an old version of MQn

[Clive]

if you are using the laptop and experiencing treble issues then you should look at the power supply (I use a linear supply, does it improve when using the battery ?) also the external disk power supply (I use linear) also the spdif convertor power supply (I splice a linear 5v into the usb cable at the receiving end) also set the power setting to 0% for min and max in the power plan. Are you using jplay on the same machine ? Also important to disable the usb power setting - allow windows to switch off power to this device setting etc.

I had already turned off USB power saving.

 

The M2Tech usb board in the Octave I power via an iFi iUSBPower with a spliced cable.

 

I uninstalled jplay from this machine to ensure it doesn't interfere at all.

 

There is a small improvement in treble when using battery but in truth I suspect it's more the recordings/mastering, the sort of album with the issue is Annie Lennox / Diva. It sounds a little "juiced up" and over-impressive. More natural recordings are rather better and probably I don't have an issue with those. It seems like I'm suffering more from the transparency & truthfulness of MQn more than anything else.

 

24 bit dacs try v5 or v6 (v6 has an attempt at 16 bit in 24 as well), but uses an old version of MQn

So it should work with 2.38 and 2.40 or do I have to use an older version of mqnplay?

[lekt]

... don't get discouraged by the noise. listen to the vibration, bouncing sound- it adds realism.

 

hi jesuscheung,

i found in archive MQn.exe2012, mqn.exe2.2vs2012 very like as mqn.exenopgomemcpy. sounds of them like is one, proportion of bass/mid/treble like as one in all 3 these versions. but bright (sharpness) is difference: (1) mqn.exe2.2vs2012, (2) mqn.exenopgomemcpy, (3) MQn.exe2012. they are correct in technical norms and have pretty bass and drum sound and also non-musical sound, that i difficult have found in last versions (2.3x, 2.4x).

 

try with mqn.exe2.2vs2012, piano sound is OK, bouncy and weight in norms, have micro-vibration. i feel true piano sound. i'm happy with this version, but it is 2.xx for USB DAC, not for me.

[sbgk]

I had already turned off USB power saving.

 

The M2Tech usb board in the Octave I power via an iFi iUSBPower with a spliced cable.

 

I uninstalled jplay from this machine to ensure it doesn't interfere at all.

 

There is a small improvement in treble when using battery but in truth I suspect it's more the recordings/mastering, the sort of album with the issue is Annie Lennox / Diva. It sounds a little "juiced up" and over-impressive. More natural recordings are rather better and probably I don't have an issue with those. It seems like I'm suffering more from the transparency & truthfulness of MQn more than anything else.

 

 

So it should work with 2.38 and 2.40 or do I have to use an older version of mqnplay?

 

mqncontrol.exe v5 doesn't use mqnplay as it is self contained. The latest versions are a bit harsh because I have removed the loop optimisation, but at least it highlights the really harsh versions

[Clive]

mqncontrol.exe v5 doesn't use mqnplay as it is self contained. The latest versions are a bit harsh because I have removed the loop optimisation, but at least it highlights the really harsh versions

I have v5 running fine now thanks and I now understand that's it's self contained!

 

Will the loop optimisations be coming back? I know you aren't looking to produce a totally finished product but the way this is going it should be the only player I need...

[sbgk]

I have v5 running fine now thanks and I now understand that's it's self contained!

 

Will the loop optimisations be coming back? I know you aren't looking to produce a totally finished product but the way this is going it should be the only player I need...

am trying to put the whole loop in assembly, so yes one way or another it will get less harsh

[sbgk]

hi jesuscheung,

i found in archive MQn.exe2012, mqn.exe2.2vs2012 very like as mqn.exenopgomemcpy. sounds of them like is one, proportion of bass/mid/treble like as one in all 3 these versions. but bright (sharpness) is difference: (1) mqn.exe2.2vs2012, (2) mqn.exenopgomemcpy, (3) MQn.exe2012. they are correct in technical norms and have pretty bass and drum sound and also non-musical sound, that i difficult have found in last versions (2.3x, 2.4x).

 

try with mqn.exe2.2vs2012, piano sound is OK, bouncy and weight in norms, have micro-vibration. i feel true piano sound. i'm happy with this version, but it is 2.xx for USB DAC, not for me.

 

you can download visual studio express and have a go yourself if you like noisy versions, it's easier than you think

[sbgk]

uploaded a rebuilt version of 2.40 sse4 intel using what I thought were the original settings and I'm sure it sounds a bit mellower

 

2.43 sse4 intel

[Clive]

it's a feature of mqn, shall fix it one day

 

uploaded a rebuilt version of 2.40 sse4 intel using what I thought were the original settings and I'm sure it sounds a bit mellower

 

2.43 sse4 intel

I'll test it tomorrow morning, sounds interesting!

[lekt]

uploaded a rebuilt version of 2.40 sse4 intel using what I thought were the original settings and I'm sure it sounds a bit mellower

 

2.43 sse4 intel

 

2.43 sse4 intel is good, good! sbgk.

sounstage/sharpness/clearly/clarity= (nopgomemcpy + 2.40) / 2

other criteria seem correct (total weight, vibration, bouncy)...

i love this version. need to careful test for details. i'm happy, happy!

 

jesuscheung: there piano sound is more true then in 2.41 movq, but 2.41 movq give pretty, romantic piano sound. try it.

[jesuscheung]

2.43 sse4 intel is good, good! sbgk.

sounstage/sharpness/clearly/clarity= (nopgomemcpy + 2.40) / 2

other criteria seem correct (total weight, vibration, bouncy)...

i love this version. need to careful test for details. i'm happy, happy!

 

jesuscheung: there piano sound is more true then in 2.41 movq, but 2.41 movq give pretty, romantic piano sound. try it.

 

no, i disagree. 2.43 is only good with piano. 2.43's issues with piano is a difficult one. notes lack weight/finishing here and there, just weird, sounds like the pianist is still practicing to master the piece. most people won't notice that, so it is ok. 2.41 movq has great vibration + bouncing effects + finishing on each note- these are essential elements for an expensive piano. 2.41 movq is able to present them quite properly. 2.41 movq's issues are obvious. it has noise and lacks weight especially in bass notes.

 

tried MQn.exe2012 and mqn.exe2.2vs2012. prefer mqn.exe2.2vs2012. true and balance.

 

2.43 is a difficult one comment! hmm... mixed feelings. it sounds great in one song, i am thinking this is a breakthrough, and then the next song sounds worse than previous version. wtf. 2.40 is well-defined. 2.43 is good and bad.

[lekt]

no, i disagree. 2.43 is only good with piano. 2.43's issues with piano is a difficult one. notes lack weight/finishing here and there, just weird, sounds like the pianist is still practicing to master the piece. most people won't notice that, so it is ok. 2.41 movq has great vibration + bouncing effects + finishing on each note- these are essential elements for an expensive piano. 2.41 movq is able to present them quite properly. 2.41 movq's issues are obvious. it has noise and lacks weight especially in bass notes.

 

tried MQn.exe2012 and mqn.exe2.2vs2012. prefer mqn.exe2.2vs2012. true and balance.

 

2.43 is a difficult one comment! hmm... mixed feelings. it sounds great in one song, i am thinking this is a breakthrough, and then the next song sounds worse than previous version. wtf. 2.40 is well-defined. 2.43 is good and bad.

 

thanks, i'll try more test, i was probably wrong.

maybe technical stabilization? as you said "..good and bad.". i think true/natural sound usualy is not too clearly, not too pretty, have noise, as in vinyl sound, as a girl without makeup. it is the true/truth/reality. we can accept it or not...

[jesuscheung]

thanks, i'll try more test, i was probably wrong.

maybe technical stabilization? as you said "..good and bad.". i think true/natural sound usualy is not too clearly, not too pretty, have noise, as in vinyl sound, as a girl without makeup. it is the true/truth/reality. we can accept it or not...

 

forgot to mention bass! in general, bass is nopgomemcpy>2.41 movq. i don't listen to bass a lot so i don't really use nopgomemcpy

 

yes. nopgomemcpy, 2.41 movq, mqn.exe2.2vs2012 are not clean but very real. 2.40-2.43 are very clean and only good realism in comparison. dirty sound is not a musical death sentence. dirty vinyl can sound great.

[lekt]

forgot to mention bass! in general, bass is nopgomemcpy>2.41 movq. i don't listen to bass a lot so i don't really use nopgomemcpy

 

yes. nopgomemcpy, 2.41 movq, mqn.exe2.2vs2012 are not clean but very real. 2.40-2.43 are very clean and only good realism in comparison. dirty sound is not a musical death sentence. dirty vinyl can sound great.

 

piano sound have some frequency ranges (mid, bass, and also non-musical beats/knocks as raw/rustic sound as when fingers touch/beat on keys). so if player give good sounds on these frequency ranges then piano sound will come true. non-musical raw/rustic sound in noise frequency range, so be careful when reducing noise, very easy to lose the precious sound.

[sbgk]

2.43 sse4 intel is good, good! sbgk.

sounstage/sharpness/clearly/clarity= (nopgomemcpy + 2.40) / 2

other criteria seem correct (total weight, vibration, bouncy)...

i love this version. need to careful test for details. i'm happy, happy!

 

jesuscheung: there piano sound is more true then in 2.41 movq, but 2.41 movq give pretty, romantic piano sound. try it.

 

2.43 lost some of the treble, have uploaded 2.43 -128 which brings it back,but whether it is too much, time will tell.

 

too much and not enough bass, 2.43 add xmm has better balance

[lekt]

2.43 lost some of the treble, have uploaded 2.43 -128 which brings it back,but whether it is too much, time will tell.

 

right as you respected about treble, sbgk, "but whether it is too much". i see that.

bass is not better. 2.3x - 2.4x bass vibration less than nopgomemcpy, although other sounds having become so much better.

[jesuscheung]

2.43 lost some of the treble, have uploaded 2.43 -128 which brings it back,but whether it is too much, time will tell.

 

too much and not enough bass, 2.43 add xmm has better balance

 

2.43 add xmm

much better than 2.43! very airy. lacks some bass, so treble feels cold. very good vibration in some instruments. yet to see vocal like that in 2.41 movq which has rich vocal that's composed of vibration. i give up commenting on the stage.

[lekt]

2.43 add xmm

much better than 2.43! very airy. lacks some bass, so treble feels cold. very good vibration in some instruments. yet to see vocal like that in 2.41 movq which has rich vocal that's composed of vibration. i give up commenting on the stage.

 

i just only tested bass & drum:

bass vibaration better, better than other 2.4x. pretty vibration! => many other sounds having bass layer (piano, guitar,...) must be better (i will test about this)

"very airy" => slightly, must be richer (a bit) and bass weight will enough.

drum and beat/knock sound is more difficult on testing, because they are dumb, non-musical. nopgomemcpy sounds it better.

[Clive]

2.43 strong on vocals and bass, probably due a lack of treble - could be a great option for listening when tired.

 

2.43 -128 has a more correct treble level which gives the effect of there being a little less bass (more treble can make bass seem less or less treble can seem like more bass). Vocals aren't as spotlighted as there's more happening in the treble.

 

2.43 xmm bass seems reduced at first but I feel that's due to the bass notes being more detailed rather being a simple thump. Treble seems well balanced. Vocals have great micro-dynamics leading to excellent intonation / expression and flow. The central soundstage is better too with more more instruments occupying space around the centre vs being stuck near the speakers.

[Clive]

I've been dipping in a out of the MQn versions for the last 3 weeks or so. 2.43 xmm is very good indeed, I don't hear any obvious vices whereas some versions are great in some areas but flawed in others. 2.43 xmm is very balanced and very good indeed, great presence, unforced yet detailed, instruments and vocals have great tunefulness and character. If development stopped here for 16/44, I would not be sad - I'm sure though the experiments won't stop here!

[jrling]

I've been dipping in a out of the MQn versions for the last 3 weeks or so. 2.43 xmm is very good indeed, I don't hear any obvious vices whereas some versions are great in some areas but flawed in others. 2.43 xmm is very balanced and very good indeed, great presence, unforced yet detailed, instruments and vocals have great tunefulness and character. If development stopped here for 16/44, I would not be sad - I'm sure though the experiments won't stop here!

 

Clive - I have heard SBGK saying that's the final version quite a few times!!

Thankfully for us he keeps up the quest for perfection and there a number of technical avenues he has yet to explore,and which are holding out promise of making MQn better and better.

 

Jonathan

[nige2000]

hi Jonathan hows R2 settling in?

[sbgk]

I've been dipping in a out of the MQn versions for the last 3 weeks or so. 2.43 xmm is very good indeed, I don't hear any obvious vices whereas some versions are great in some areas but flawed in others. 2.43 xmm is very balanced and very good indeed, great presence, unforced yet detailed, instruments and vocals have great tunefulness and character. If development stopped here for 16/44, I would not be sad - I'm sure though the experiments won't stop here!

 

sounds promising, thanks for the feedback, hopefully getting close to the end in terms of SQ.

 

How are you testing it ? I thought you could only play 24 bit and only mqncontrol v5 can do that and it is standalone ie doesn't use mqnplay.

[Clive]

sounds promising, thanks for the feedback, hopefully getting close to the end in terms of SQ.

 

How are you testing it ? I thought you could only play 24 bit and only mqncontrol v5 can do that and it is standalone ie doesn't use mqnplay.

I must have unintentionally given you the impression I could only play 24 bit when we were discussing HD playback. I am playing 16/44.1 no problem. I'm using the standard mqncontrol.exe

 

The the system is i5 HP laptop, W8, Metrum Octave DAC with iUSBPower, John Chapman AVC-1 / Dave Slagle AVC preamp into 300B SE amps + XTZ Sub Amp 1 DSP feeding some big open baffle speakers which are Bastanis / Trans-Fi hybrids (10" wideband with two 15 bass dipoles per side).