Jump to content
Sign in to follow this  
sbgk

Visual studio 2012 c++ and wasapi minimalist player

Rate this topic

Recommended Posts

i do like the flabbiness, softer sound. easier on the ear with headphones. sometimes i find 2.17 a bit tough on the ears.
have uploaded 2.20 sse4 intel, hopefully not flabby and not lean like 2.17, I think it has a nice live sound to it.

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/

Share this post


Link to post
Share on other sites

not sure why, i am liking 2.19 more and more. it is presenting a piece of music very well like a good musician. i can really focus on the important stuffs. now i am listening to XA less and less because it tries to make everything sound important. maybe XA is more correct but that's not a musical way to play a piece of music.

Share this post


Link to post
Share on other sites
not sure why, i am liking 2.19 more and more. it is presenting a piece of music very well like a good musician. i can really focus on the important stuffs. now i am listening to XA less and less because it tries to make everything sound important. maybe XA is more correct but that's not a musical way to play a piece of music.

yes, there seems to be something about 2.19 that makes it sound different, seems to have more space around the instruments, can it be improved upon


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/

Share this post


Link to post
Share on other sites
yes, there seems to be something about 2.19 that makes it sound different, seems to have more space around the instruments, can it be improved upon

 

don't think I can improve on it, agree with JC - you can follow each instrument, very entertaining. so 2.19 is the winner for now. (maybe an avx chip would sound better with 256 bit registers and other improvements, for now mqn can't be built for avx instructions ). next is large memory pages - will they make a difference and another go at control.


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/

Share this post


Link to post
Share on other sites
not sure why, i am liking 2.19 more and more. it is presenting a piece of music very well like a good musician. i can really focus on the important stuffs. now i am listening to XA less and less because it tries to make everything sound important. maybe XA is more correct but that's not a musical way to play a piece of music.
failed quite badly on one of my test tracks where the vocals couldn't be heard. have uploaded 2.22 which has the alignment changed, but the rest is the same and the sound is better balanced.

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/

Share this post


Link to post
Share on other sites
failed quite badly on one of my test tracks where the vocals couldn't be heard. have uploaded 2.22 which has the alignment changed, but the rest is the same and the sound is better balanced.

 

prefer 2.19 because of stage. with headphones, 2.22 sounds like i am right there on the stage of a live performance. 2.19 sounds like i am sitting at the front row. with speakers, 2.22 sounds excellent.

Share this post


Link to post
Share on other sites
prefer 2.19 because of stage. with headphones, 2.22 sounds like i am right there on the stage of a live performance. 2.19 sounds like i am sitting at the front row. with speakers, 2.22 sounds excellent.

 

2.19 does have it's attractions, but some instruments were just too over emphasized compared to the vocals, there is one more thing I can try that is half way between the two.


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/

Share this post


Link to post
Share on other sites
2.19 does have it's attractions, but some instruments were just too over emphasized compared to the vocals, there is one more thing I can try that is half way between the two.

have uploaded 2.22 0, 2.22 4 and 2.22 8 versions to compare with 2.22. 2.22 would be called 2.22 16 if it was following the same naming convention. be interested to know if you can pick a favourite.


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/

Share this post


Link to post
Share on other sites
My favourite is 2.22 4 because voices are placed where I like. Details are just perfect.
thanks. I haven't listened to it long enough to work out the differences and make a decision. 0 seems free flowing and open, 4 adds a bit of control, 8 and 16 maybe start affecting the detail.

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/

Share this post


Link to post
Share on other sites
My favourite is 2.22 4 too. 4>0>8.

and presumably 4>0>8>16

 

Just a couple more things to try out, have tried them before, but wasn't convinced, sometimes doing the correct thing code wise doesn't sound the best eg when reserving a buffer that will contain 16 Byte data is it better to allocate it as Byte or 16 Byte ? Which one is more efficient for loading the data ? Does the load instruction need the data type to be explicitly stated ? All these have an effect on the sound. 2.22 uses a Byte buffer and no explicit data typing, but it's something else to try.


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/

Share this post


Link to post
Share on other sites
and presumably 4>0>8>16

 

Just a couple more things to try out, have tried them before, but wasn't convinced, sometimes doing the correct thing code wise doesn't sound the best eg when reserving a buffer that will contain 16 Byte data is it better to allocate it as Byte or 16 Byte ? Which one is more efficient for loading the data ? Does the load instruction need the data type to be explicitly stated ? All these have an effect on the sound. 2.22 uses a Byte buffer and no explicit data typing, but it's something else to try.

building on the 2.2 4 version I tried a number of different ideas and hopefully made some progress. Have uploaded mqnplay.exe 2.23 8 4 and 2.23 4 4. More detail, better bass and tighter staging, think I like the 8 4 version best - the 4 4 is light and airy but a bit unfocused.

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/

Share this post


Link to post
Share on other sites
building on the 2.2 4 version I tried a number of different ideas and hopefully made some progress. Have uploaded mqnplay.exe 2.23 8 4 and 2.23 4 4. More detail, better bass and tighter staging, think I like the 8 4 version best - the 4 4 is light and airy but a bit unfocused.

 

i like 4 4 better. just as you said, airy and light. i value airy very highly. have not heard any other player like it. JIE, HQ sounds tight and control. in fact, i can easily tune the bios to sound tight and control. i think airy/free/open is difficult to achieve- it is a good feature of a good CD player that is usually missing on a PC hifi.

 

4 8 is great too. think it is a bit bright.

Share this post


Link to post
Share on other sites
loving 2.23 4 4. thx! i almost have no reason to use XA (my previous favorite player).

have uploaded a 2.24 8 4 and 4 4 version which may remove a bit more of the brightness, just changes one instruction - resetting the counter between loops by doing a register to register move instead of moving 64 directly into the register. my thinking is that moving the value 64 into a 64 bit register causes more noise/is slower than moving a value from a 64 bit register to another 64 bit register because with 64 the datatypes are not aligned ie 1 Byte into 8 Bytes. the only issue is that there is an extra mov instruction at the start to initialise the extra register with 64, which is why I though just having the mov rax, 64 was the best method, but it does seem to introduce some brightness/sibilance.


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/

Share this post


Link to post
Share on other sites
loving 2.23 4 4. thx! i almost have no reason to use XA (my previous favorite player).

 

may have overlooked a good version from a few days ago, have uploaded it as 2.25


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/

Share this post


Link to post
Share on other sites

sbgk, I just want to say thank you for making the player public.

 

I'm test driving it (2.22 sse4) and find the results really interesting. Bass clarity, detail/texture and timing are the best I've heard and soundstage likewise, it's not pinned around the speakers. Soundstage is more like a good vinyl setup (yes that is a compliment!). I'm using W8 with an i5 machine and the 4 core affinity version of mqncontrol.

Share this post


Link to post
Share on other sites

wrong reply

Edited by markvandepas
wrong reply

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.

Share this post


Link to post
Share on other sites

tc electronic level pilot

Just what you need

 

Hi... can you do some sort of digital volume parameters?? i´m one like many folks that don´t have preamp or analog volume...

just dac to powered speakers...

tks


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.

Share this post


Link to post
Share on other sites
sbgk, I just want to say thank you for making the player public.

 

I'm test driving it (2.22 sse4) and find the results really interesting. Bass clarity, detail/texture and timing are the best I've heard and soundstage likewise, it's not pinned around the speakers. Soundstage is more like a good vinyl setup (yes that is a compliment!). I'm using W8 with an i5 machine and the 4 core affinity version of mqncontrol.

Clive, thanks for the feedback, it is gaining quite a bit of interest now and has been a bit of a journey to get to this stage. I believe it is nearly finished and all the factors are known about, just teasing out the final subtleties. I'm also trying to get the driving rythm of vinyl. It's now a question of what is good sound, an expansive soundstage or a tighter controlled sound, I was hoping that it would be obvious when I heard it, but can be difficult to decide.


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/

Share this post


Link to post
Share on other sites

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
Sign in to follow this  

×
×
  • Create New...