Skip to main content
Topic: The best algorithm for upmixing stereo to 5.1? (Read 7310 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

The best algorithm for upmixing stereo to 5.1?

Hi all,

Can someone tell me, please, what is the best algorithm to:

1) upmix stereo to 5.1 (music or movie modes):
    * Dolby prologic II.
    * DTS NEO:6
    * V.I VST plugin.
    * ATSurround and fsurround plugins (foobar2000)
    * SRS
    * ....etc

2) downmix a 5.1 to stereo surround for headphone :
    * Dolby headphone
    * SRS headphone
    * ATsurround plugin and 5.1 to stereo plugins (foobar2000)
    * Cyberlink headphone
    * Ensonido
    * ....etc

Thank you

The best algorithm for upmixing stereo to 5.1?

Reply #1
I think that's a matter of a personal taste and preference. It depends on the type of music you listen and what you are expecting to hear after the upmix. I never liked those stereo widening switches on some of ye olde equipment  The same goes for stereo to 5.1. I like it subtle and sounding natural.

As for 1) I've tried all that you mentioned, and for me, DTS NEO:6 is the best for music upmixing.

I don't like those headphone effects, they change the sound too much for my taste. For the rare occasions when have to listen 5.1 through the headphones, I usually use plain 5.1 to stereo downmix and bs2b.
If age or weaknes doe prohibyte bloudletting you must use boxing

The best algorithm for upmixing stereo to 5.1?

Reply #2
Quote
1) upmix stereo to 5.1 (music or movie modes):

In most cases you are better-off leaving it stereo and letting your Pro Logic reciever "decode" it.  That way, you (or the listener) can choose the soundfield setting of their choice at playback time.

The exception is if you are trying to do something really specific...  like manually pan the sound, or add an echo or to the rear channels, etc.  In those cases six discrete channels is needed.

Quote
downmix a 5.1 to stereo surround for headphone :
If you are playing a DVD, there is usually a 5.1 track and a stereo track.  I don't think the stereo track is required by the DVD spec, but I don't think any of my DVDs have only a 5.1 track.  DVD players are required to downmix a Dolby 5.1 track, and there are metadata downmix instuctions in the 5.1 track for the DVD player.  Depending on the DVD player, you might have to use the analog-stereo outputs to get the downmix...  Now that I think about it, I'm not sure how/if my receiver is downmixing to the headphone output...  But I have a Dolby test DVD, and I can try it.

The best algorithm for upmixing stereo to 5.1?

Reply #3
I've tried all that you mentioned, and for me, DTS NEO:6 is the best for music upmixing.


I've tried none that you mentioned except for Pro Logic II and DTS NEO:6 (because they are on my reciever).  Usually I just like to listen to the stereo mix, as close to the source as I can get but the DTS NEO:6 Music algorithm actually makes music sound much better.  It seems clearer and the bass is more punchy.  I noticed an improvement in all kinds of music with this algorithm, I listen to rock, metal, classical, a bit of blues and some rap.

The best algorithm for upmixing stereo to 5.1?

Reply #4
Thanks all for your responses.

For you, then, NEO:6 is the best solution for upmixing music. How about movie?

I have no 5.1 system !! So I want to profit the surrund sounds through my headphones.

The best algorithm for upmixing stereo to 5.1?

Reply #5
Quote
I have no 5.1 system !! So I want to profit the surrund sounds through my headphones.
 

Your headphones only have 2 drivers and a stereo connection!  ...I know there are some "tricks" for surround-sound headphone listening, and 4-channel headphones have been made with front & rear drivers (I don't know if these are still made).  And, I've never heard Dolby headphone processing, so I can't say how well it works.

But, there is a major problem.  One of the most important factors in detecting the location/direction of a sound is head movement.  But, with headphones, the sound source moves with your head and this removes much of the directional perception.  When you move your head, your brain can "decode" the loudness changes, and maybe the phase changes.  For example, if your head was clamped-down, you could tell if the sound is coming from the left or right, but it would be very hard to tell if sound was coming from right-front or right-rear.  It would also be difficult to tell if a sound was coming from center-front, center-rear, or somewhere overhead-center!  (Any] center-sound is equal in both ears 'till you move your head.)

Most of the time, we are unaware of this slight head-movement, except we do sometimes turn to "see" where the sound is coming from, when we hear an unexpected sound.  And, when there is a short- intermittent sound or rattle (i.e. in your car) you usually need to turn your head and wait for a 2nd or 3rd "reading" to zoom-in on where the sound is coming from...  And, you've probably noticed that animals turn their head slightly when they hear a sudden-unexpected sound.

The best algorithm for upmixing stereo to 5.1?

Reply #6
Thanks for the tech eplanation DVDdoug.
I agree that HRTF technics can't replace a true 5.1 system. But they are better than classical stereo.
You can experiment ENSONIDO which is a dolby headphone like technology by lissening the radioclassic  mp3s (surround) stream using winamp with the fraunhofer mp3s decoder plugin configured to ensonido (5.1 sound over headphone).

http://www.radioclassique.fr/index.php?id=143

On the site you can download winamp and the plugin !!
enjoy

The best algorithm for upmixing stereo to 5.1?

Reply #7
Your headphones only have 2 drivers and a stereo connection!  ...I know there are some "tricks" for surround-sound headphone listening, and 4-channel headphones have been made with front & rear drivers (I don't know if these are still made).  And, I've never heard Dolby headphone processing, so I can't say how well it works.

Are you serious? Headphones with rear-speakers? Now that's a silly thing! What about 2 additional ears? 
But, there is a major problem.  One of the most important factors in detecting the location/direction of a sound is head movement.  But, with headphones, the sound source moves with your head and this removes much of the directional perception.  When you move your head, your brain can "decode" the loudness changes, and maybe the phase changes.  For example, if your head was clamped-down, you could tell if the sound is coming from the left or right, but it would be very hard to tell if sound was coming from right-front or right-rear.  It would also be difficult to tell if a sound was coming from center-front, center-rear, or somewhere overhead-center!  (Any] center-sound is equal in both ears 'till you move your head.)

Most of the time, we are unaware of this slight head-movement, except we do sometimes turn to "see" where the sound is coming from, when we hear an unexpected sound.  And, when there is a short- intermittent sound or rattle (i.e. in your car) you usually need to turn your head and wait for a 2nd or 3rd "reading" to zoom-in on where the sound is coming from...  And, you've probably noticed that animals turn their head slightly when they hear a sudden-unexpected sound.

I'm sure that head movement is a factor not to neglect - but it isn't that important as well. Since the 1930s(!) there's a technique called Kunstkopf which uses a dummy head ("Kunstkopf" in German) with microphones build into the ears. IMHO it gives an absolutely realistic surround impression when played back on headphones (at least if you doesn't shake you head all the time ). As far as I know, Dolby Headphone as well as some other DSPs try to emulate such a dummy head - but doesn't really succeed. Maybe it's just too complex to artificially shape distorsions introduced by the auditory canals of a human head into the sound.

See wikipedia for details: http://en.wikipedia.org/wiki/Dummy_head_recording

The best algorithm for upmixing stereo to 5.1?

Reply #8
another binaural recording (in addition to that from wikipedia article):
http://b3liar.wrzuta.pl/audio/j6FzvdG7Ws/niesamowity_efekt
(use only stereo headphones without postprocessing).

human brain detects the sound direction by analysing frequency bandwidth. sounds coming from rear usually have reduced high tones.

i think the only way for 100 % surround sound is to use stereo headphones with sound generated in realtime according to head position. it's nearly impossible to recreate very close sounds by 5.1 systems.


noob said

The best algorithm for upmixing stereo to 5.1?

Reply #9
Thanks donkey7, the surround sound in the link is great!! how can i download it? I don't understand the polish language !!
the binaural thecnology is the most easist way to expriment the surround sound.
another site with a lot of binaural samples:

http://freesound.iua.upf.edu/tagsViewSingle.php?id=1190

The best algorithm for upmixing stereo to 5.1?

Reply #10
binaural stereo.mp3

as you see it's only 128 kbps cbr mp3 so lots of important details are removed. i don't know the source for uncompressed sound data.

The best algorithm for upmixing stereo to 5.1?

Reply #11
Thanks donkey7.

 
SimplePortal 1.0.0 RC1 © 2008-2019