Jump to content
IGNORED

Visual studio 2012 c++ and wasapi minimalist player


sbgk

Recommended Posts

in your case, "sound is perfect and clearly", but "more digital, less analog", i believe you are having a nightmare. for me, tuning only improves sound clarity/stability. MQn is the only thing makes my machine sounds analog. or unless the recording is super good... i believe the next best player on earth is XA, it also sounds digital after listening to MQn.

 

I agree with you, I'm having a nightmare. Windows make me crazy, self-conflict. Difficult to satisfy with any windows tweaking.

Link to comment
thx lekt! just tested mqn.exenopgomemcpy sounds really simple/true/real/intense! compare to 2.30, hmm... difficult to say... one thing for sure, 2.30 can sound like it has ended before it is over. mqn.exenopgomemcpy makes everything sound to completion. that's the reason for older version being more real/true.

 

i mean 2.30 is all around better apart from being true/real.

 

makes no sense to me... 2.30 uses SSE4... and older versions don't.

 

I tried nopgomemcpy, played it for about 30 seconds. It has the characteristic digital edge of memcpy, so no go for me. I prefer 2.30 EAX to the others, EAX 64 is a bit sibilant. What do you think of the sse2 version ? That should be closer to nopgomemcpy in sound. I think where 2.30 sse4 comes into it's own on really complex pieces.

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 tried nopgomemcpy, played it for about 30 seconds. It has the characteristic digital edge of memcpy, so no go for me. I prefer 2.30 EAX to the others, EAX 64 is a bit sibilant. What do you think of the sse2 version ? That should be closer to nopgomemcpy in sound. I think where 2.30 sse4 comes into it's own on really complex pieces.

 

just deleted another chunk from the assembly code, bit of a big upgrade in sound quality. have uploaded 2.31 sse4 intel. listening to Art Pepper Modern Jazz Classics - the band is now in the room, can it get any 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
just deleted another chunk from the assembly code, bit of a big upgrade in sound quality. have uploaded 2.31 sse4 intel. listening to Art Pepper Modern Jazz Classics - the band is now in the room, can it get any better.

 

that version was a bit thin, see I said eax 64 counter sounded sibilant, the new version (also called 2.31 sse4 intel ) sounds better and uses eax counter. the old version has been moved to archive. Hopefully this version is better than nopgomemcpy, lol. I think it is spectacular.

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 version was a bit thin, see I said eax 64 counter sounded sibilant, the new version (also called 2.31 sse4 intel ) sounds better and uses eax counter. the old version has been moved to archive. Hopefully this version is better than nopgomemcpy, lol. I think it is spectacular.

 

what do you know, the counter eax loses detail in this version, have uploaded 2.31 sse4 intel rax version with more code removed and using a rax counter

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
what do you know, the counter eax loses detail in this version, have uploaded 2.31 sse4 intel rax version with more code removed and using a rax counter

 

as usual there is more than one to skin a cat, have uploaded 2 more versions 2.31 rax r9 and r11, am liking r9 at the moment.

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
2.29 no wc is great, should be named "Hi-End Audiophile".

BTW I'm wondering what is the fastest method convertting flac or mp3 to wav?

 

you'll find 2.31 sse4 intel rax r9 is quite a bit better, good to see the old time supporters still trying the new versions

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 tried nopgomemcpy, played it for about 30 seconds. It has the characteristic digital edge of memcpy, so no go for me. I prefer 2.30 EAX to the others, EAX 64 is a bit sibilant. What do you think of the sse2 version ? That should be closer to nopgomemcpy in sound. I think where 2.30 sse4 comes into it's own on really complex pieces.

 

yes. i picked EAX64 over EAX coz it has near zero earache. EAX is better otherwise.

Link to comment

What do you think of the sse2 version ? That should be closer to nopgomemcpy in sound....

 

that version was a bit thin, see I said eax 64 counter sounded sibilant, the new version (also called 2.31 sse4 intel ) sounds better and uses eax counter. the old version has been moved to archive. Hopefully this version is better than nopgomemcpy, lol. I think it is spectacular.

 

2.10 sse2 intel 256 don't sound like nopgomemcpy to me.

 

again 2.31 handles complex music better than all previous versions. eminem sounds better each version.

 

2.31 rax vs rax r9 vs rax r11 is a difficult one...

 

2.31 vs nopgomemcpy/old mqn.exe

-given CPU=full speed, when each word of a song is long and deep, like "i will always love you", "edge of glory", older versions clearly win in terms of true/real. again, new version doesn't complete each word of a song as well as the old. lacks the depth/finishing.

-given CPU=min speed, the difference in true/real is much much smaller.

Link to comment
just deleted another chunk from the assembly code, bit of a big upgrade in sound quality. have uploaded 2.31 sse4 intel. listening to Art Pepper Modern Jazz Classics - the band is now in the room, can it get any better.

 

I have test by headphone 2.31 sse4 rax, rax9, rax11:

- treble, mid OK, approximate to true sound, analog as nopgomemcpy. Congratulation!! I'm very happy with this sound.

- bass (including drum) power and vibration less than nopgomemcpy, not natural. nopgomemcpy is better, full effect of guitar bass + drum

Maybe reason is you used eax counter, and rax (what is this, I don't know).

Link to comment
I have test by headphone 2.31 sse4 rax, rax9, rax11:

- treble, mid OK, approximate to true sound, analog as nopgomemcpy. Congratulation!! I'm very happy with this sound.

- bass (including drum) power and vibration less than nopgomemcpy, not natural. nopgomemcpy is better, full effect of guitar bass + drum

Maybe reason is you used eax counter, and rax (what is this, I don't know).

 

mostly agree. as true/analog/real as nopgomemcpy is, i still believe 2.10 SSE4 and mqn.exe3.82 reproduce much more(the most) emotions especially in vocal. it is easy to forgive their flaws when there is so much emotions. 2.31 is best in a lot of things, except for real/true/emotion. i have heard grade 2/3 kids plays an instrument better than a grade 8 adult, because these kids play with emotions wherelse the adult plays with good techniques without soul.

Link to comment
mostly agree. as true/analog/real as nopgomemcpy is, i still believe 2.10 SSE4 and mqn.exe3.82 reproduce much more(the most) emotions especially in vocal. it is easy to forgive their flaws when there is so much emotions. 2.31 is best in a lot of things, except for real/true/emotion. i have heard grade 2/3 kids plays an instrument better than a grade 8 adult, because these kids play with emotions wherelse the adult plays with good techniques without soul.

 

did usual trick of leaving an incorrect setting in. have uploaded a rax and eax version with the setting removed, the old 2.31 versions are in the archive. Think I prefer the eax version. just uploaded 2.32 sse4 intel eax rel which shaves a few more instructions off the assembly code

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
did usual trick of leaving an incorrect setting in. have uploaded a rax and eax version with the setting removed, the old 2.31 versions are in the archive. Think I prefer the eax version. just uploaded 2.32 sse4 intel eax rel which shaves a few more instructions off the assembly code

 

and an al version of 2.32 for comparison with eax

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
and an al version of 2.32 for comparison with eax

 

lol. 2.31>2.32. 2.32 is a different style! think there is a layer on top of the original sound.

 

need better bass for 2.31 as lekt said. need a bit softer vocal/treble which seems to sound a bit hard maybe that's why i can't feel as much emotions like that of 2.10.

Link to comment
lol. 2.31>2.32. 2.32 is a different style! think there is a layer on top of the original sound.

 

need better bass for 2.31 as lekt said. need a bit softer vocal/treble which seems to sound a bit hard maybe that's why i can't feel as much emotions like that of 2.10.

I quite like the sound of 2.32 eax rel, thought it was easier to listen to than 2.31, sometimes think that the perceived bass is going down because you can hear every instrument so instead of hearing a blob of bass that is now split into different constituent parts. Whether 2.32 has a problem with bass shall find out after a longer listen.

 

The cycle of development seems to be discover something that might make a difference, implement it, release, then tune.

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 quite like the sound of 2.32 eax rel, thought it was easier to listen to than 2.31, sometimes think that the perceived bass is going down because you can hear every instrument so instead of hearing a blob of bass that is now split into different constituent parts. Whether 2.32 has a problem with bass shall find out after a longer listen.

 

The cycle of development seems to be discover something that might make a difference, implement it, release, then tune.

 

yea. 2.32 is very very easy to listen to. think that coz the sounds are kind of blurred(like a samsung tv, pictures are blurred a bit). i don't feel emotional with 2.32 when a song is emotional. yea, eax > al in 2.32.

 

i think you are so close to combining strengths of each version into one best thing.

Link to comment
yea. 2.32 is very very easy to listen to. think that coz the sounds are kind of blurred(like a samsung tv, pictures are blurred a bit). i don't feel emotional with 2.32 when a song is emotional. yea, eax > al in 2.32.

 

i think you are so close to combining strengths of each version into one best thing.

 

recently, i optimized my win2012 even more by disabling even more things like kernel drivers. now even 2.26 which is supposed to be hard to listen to becomes much easier. i guess i trying to say is that maybe sometimes MQn is hard to listen is reflecting the external factors.

Link to comment
yea. 2.32 is very very easy to listen to. think that coz the sounds are kind of blurred(like a samsung tv, pictures are blurred a bit). i don't feel emotional with 2.32 when a song is emotional. yea, eax > al in 2.32.

 

i think you are so close to combining strengths of each version into one best thing.

 

If you look on J Kenny's site there is a section on rf attenuators

RF Attenuators - Isolated Power Devices and a link to users feedback. It appears the lack of digital noise can be interpreted as lack of dynamics, don't know if that is a factor in 2.32, but I can assure you that all I am doing is removing instructions. Whether it can be configured to have a better sound we shall see. I bought a 15db rf attenuator and it does work, though don't use it with the NAD as it has aes/ebu input instead of BNC which the Benchmark had.

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
If you look on J Kenny's site there is a section on rf attenuators

RF Attenuators - Isolated Power Devices and a link to users feedback. It appears the lack of digital noise can be interpreted as lack of dynamics, don't know if that is a factor in 2.32, but I can assure you that all I am doing is removing instructions. Whether it can be configured to have a better sound we shall see. I bought a 15db rf attenuator and it does work, though don't use it with the NAD as it has aes/ebu input instead of BNC which the Benchmark had.

 

very interesting! is this what you did?? my DAC is PCIE, so....

7.jpg

Link to comment
....It appears the lack of digital noise can be interpreted as lack of dynamics, don't know if that is a factor in 2.32, but I can assure you that all I am doing is removing instructions...

 

ok! are you sure that those instructions were useless, hence removable? this means i have been confusing jitters with real sound? damn! let me listen more on 2.32.

Link to comment

i find myself turning up volume over the last few versions

the last few versions have more clarity and detail but require more volume i guess that is a result of having less noise?

was unwanted noise was making the sound louder as well as distorting it?

ive seen this before when introducing cleaner power into the pc

Link to comment
i find myself turning up volume over the last few versions

the last few versions have more clarity and detail but require more volume i guess that is a result of having less noise?

was unwanted noise was making the sound louder as well as distorting it?

ive seen this before when introducing cleaner power into the pc

 

i find myself having to tune my ram-timing to enjoy 2.32. 2.32 is now truly amazing. prior to 2.32, i had a different ram-timing, which has the effect of cleaning noises at the expense of clarity. this 2.32 is so clean such that the old ram-timing is damaging to SQ. wow... 2.32 is truly a big step forward in my system. i need more listening to confirm this finding! wow...

Link to comment
did usual trick of leaving an incorrect setting in. have uploaded a rax and eax version with the setting removed, the old 2.31 versions are in the archive. Think I prefer the eax version. just uploaded 2.32 sse4 intel eax rel which shaves a few more instructions off the assembly code

 

last 2 version 2.31 & first 2 version 2.32: real/true/clearly sounds, but emotion not enough as jesuscheung said. Why is that?

I think that. As in orchesrta, all instruments and vocals are OK but placed in the wrong position or not optimization. Vocal need to nearer, bass need so far (little) at listeners, drum need to more power and pressure (but don't turn up volume). Ah, difficult for you, sbgk.

Drum+bass sound effect is very important in pop/rock and plays almost all time in musical track.

I recommend test drum beats of Phil Collins - ...But Seriously/03.Do You Remember, first seconds of track. nopgomemcpy version give me excellent sound effect, pressure really enough. Vote to nopgomemcpy once again.

mqnplay.exe 2.31 sse4 intel eax r9 no wc better than others, 2.32 wrong volume.

Now all version give very analog and true sound (99%, I think that), sound of each individual instrument is excellent, bass vibration OK,...

Congratulation, sbgk & jesuscheung.

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...