New- MY Guide For Stereo to 5.1

[SurroundBoy]

Software Needed- Cool Edit Pro or Adobe Audtion
Besweet
Step 1
Front Left and Front Right
Open up your stereo sound file into cool edit pro or Adobe Audtion go to convert sample type or press F11 to get there. Where it says Channels, check Mono, then for the Left Mix put 100% and For the Right Mix put 0%, Click ok and now you have your Front Left Channel. Complete the process again and do the same For Front Right, except have the Left Mix at 0% and the right at 100% Save your tracks as "X-FL" and "X-FR"

Step 2
Center Channel for Cool Edit Pro
Open your Stereo file, and go to Effects, Filters, then Graphic Equalizer. Make sure it is set in 10 Bands and all levels are at 0db. Now Slide the <31, 62, 125, and 250 down to -18db, Click ok. Once the process is completed, go to Convert sample Type again and for Channels have 50% for the left mix, and 50% for the right, click ok. You should now have your center Channel, be sure to make sure each of your wave files are consistent and balanced with each other so one isn't louder than the other or lower than the other, The Front Left and Right Channels should be equal, the Center should be slightly lower or equal to the FL and FR Channels, Use the Effects> Amplitude>Amplify to Increase or Decrease the levels. Save your track as "X-C"

Center Channel For Adobe Audition (Recommended)
Open your Stereo file, and go to Effects, Filters, Center Channel Extractor
This is the setting I use for The Center Channel (http://members.aol.com/surroundfiles/images/c1.jpg)
Click ok. After that, go to Effects>Amplitude>Pan/Expand... and use the "Center on the Right Preset". Click Ok. Then hit F11 to convert the sample type to 16-Bit Mono, using %0 of the left channel and %100 of the right. Then you have your finished center channel with any tweaking that you think is neccessary for the sound you want. Save your file as "X-C"

Step 3
LFE
Open your stereo file, go to Effects>Filters>FFT Filter, Scroll down on the Presets list to "Only The Subwoofer", click ok, then once again go to Convert Sample Type and have 50% for left and 50% for right. The LFE Channel usually Depends on how much bass the original song has, if the song has strong bass and the waveform looks similiar to the song you need to Decrease the volume on the LFE track or else your final 5.1 track will be distored and overly bass heavy. Save your track as "X-LFE"

Step 4
Surround Left, Surround Right For Cool Edit Pro
Open your stereo file, Go to Effects>Amplitude>Channel Mixer then in the presets go to Vocal Cut, Click ok. Insert The finished Track into the Multitrack 2 Times. Once in the Multitrack session, pan one track 100% to the left and the other to the right. I usually right click the track thats panned to the right, go to wave block properties and change the time offset to 0:00.010 to give a little delay to the Surround tracks. The Surround left would have a Time Offset of 0:00.000 and the Right would have a Time Offset of 0:00.010. Once you have completed that, select the Surround Track you want to mix down first, click the solo button then go to Edit>Mix Down To File>All Waves Mono. This Mixes down your surround track to A mono track. Save your Left Surround Channel as "X-SL" then do the same for the Right Surround as "X-SR"

Surround Left, Surround Right For Adobe Audition (Recommended)
Open your Stereo file, and go to Effects, Filters, Center Channel Extractor
For the Surrounds I use this setting (http://members.aol.com/surroundfiles/images/c2.jpg)
This not only removes the center channel but it also gives you a stereo reproduction in the sound which isn't possible with Cool Edit's "Vocal Cut" Preset. This also makes it much easier to single out any certain sounds you may want to be portrayed in your mix. You can then adjust the levels using the amplitude function to decide how loud or low you want the surrounds in the mix. After this you simply use the same method used to produce the Front Left and Front Right Channels (Explained Above)

You Should now have 6 Mono Tracks that can be Muxed to a 5.1 Track in The Latest Version of Besweet.
Open Windows Notepad and type in the paths of your Mono Tracks-Example
C:\5.1 Mixes\X-FL.wav
C:\5.1 Mixes\X-C.wav
C:\5.1 Mixes\X-FR.wav
C:\5.1 Mixes\X-SL.wav
C:\5.1 Mixes\X-SR.wav
C:\5.1 Mixes\X-LFE.wav
Once you have that typed in, got to save and save it as "X.mux"

You can then Load the x.mux file into Besweet and convert to a 5.1 AC3

That's my Updated guide, happy Mixing!
-Matt, 17

[jorel]

:)

very cool SurroundBoy!

do you know "waves maxxbass"?
i use in soundforge4.5 fantastics results.
but don't know if work in cooledit.
will be very good to use in "step 3(LFE)"

;)

[SurroundBoy]

No I've never heard of it before, but thanks I'm gonna check it out

[jorel]

thanks for this hints Eye of Horus,really good!
:)


for Step 4 (Surround Left, Surround Right)
easy and great way:
using T-Racks 204(24 bits),load the wave and choose
Diff. !
it will remove the vocals without loose the "bright"
...and the best,give stereo for the back channels!
or use the pan like posted...your choice.

then back to CE and adjust the offset
to 0:00.010 for this channels.....!

;)

[Tantulus]

I love your technique and I have used it on a couple of my vocal CD's using Surcode DVD DTS to encode the 6 tracks. However, I am particularly interested in classical music and I was wondering if any one has suggestions for settings for the surround effects to mimic a concert hall ambience?

[jorel]

hy Tantulus,
welcome in forum!
:)
i was waiting SurroundBoy, Eye of Horus
or another developer to help you first.

i don't think that is a good idea change the "defaults"
cos "surround effects" depend of the ambient and the
home theater adjusts, it's personal...
you can choose the best when you are listen the "show".
effects is for play, not for encode...( only my opinion! )
encode with "changes" can give
artificial results and permanent regret!
think in the next home theater with great features when you got it!
:)

[DIggedy]

@ Jorel
What is this T-Racks 204(24 bits) you speak of? Software? Plug-in?
Thanx

[jorel]

T-Racks is one great software (tube simulator),
i really love it!
;)

take a look:
http://www.t-racks.com/Main.html?TRInfo

:)

[bobn4burton]

So I guess that we need Adobe Audition now instead of Cool Edit Pro to do this guide? I searched for Cool Edit in google and the page it took me to said that Adobe had bought Cool Edit and released Audition as basically the same product...

[Jones_R]

Guys, you are so much behind with everything that relates to 5.1/7.1 from stereo. It's amazing. Nobody uses these outdated (yet smart, for their time) technics anymore. What we do today is duplicate the stereo signal to all the surround channels, and then use the convolution engine, to convolve each surround channel, with an impulse response (of a real hall/auditorium/recording studio) which corresponds to the place of your speaker in your room (At the end, when everything works right, you get the feeling like you are really listening to the song in the hall/auditorium/recording studio that the impulse response were recorded in).

One of the places to get such impulse responses, is www.altiverb.com.

Don't forget to remove the first 40ms of the impulse response, since it contains the direct sound, something which we do not want our reverb channels to contain.

Again, SurroundBoy, your guide is really good for what it is, but the technique is simply outdated.

[bobn4burton]

Originally posted by Jones_R


Again, SurroundBoy, your guide is really good for what it is, but the technique is simply outdated.

Your post and link might have some really good effects. Is this method useful for purely separating a stereo audio stream into a 5.1 audio stream? I don't necessarily want to put an effect on the audio, just "convert" to 5.1 audio...

Any clarification on your posted method would be appreciated!
Thanks

[Jones_R]

Eye of Horus,
Where exactly did you see me mention the word "Ambisonics" ?. Yes, I heard something about Ambisonics, but as far as I know, Ambisonics needs special encoded material to work with. It can't just work with any stereo recording. At least that's what I've heard...

Also, the technique I showed really needs a minimum of 4 surround speakers to sound good. 2 won't cut it. Also, the room should be acoustically dead.

Without those two factors, I'm affraid the results might not be satisfying.

In this light I found your remark about the "outdating" rather insulting and....... strange ! Because you point to a method that was developed in the 70's >>>>>>>>> Ambisonics !
Again, I'm not so sure what you mean by "Ambisonics", since I haven't taken a good look at it, yet (because it needs special recorded material to work with), but, the technique I've showed could not have done much help back in the seventies, since it requires a processors to convolve, in real time, several channels, with high accuracy. That level of processing power has only been available to us in recent years.


Your post and link might have some really good effects. Is this method useful for purely separating a stereo audio stream into a 5.1 audio stream? I don't necessarily want to put an effect on the audio, just "convert" to 5.1 audio...

Bob,
You can't convert 2 channels into 5.1 channels without adding any effects to the surround channels. The only way to do this, will be to duplicate the 2 channels singnals, into the surround speakers, and play it like that.

Consider for a moment, what do you want to recreate with your audio system ?. I, for one, would like to recreate a concert, like it has been heard in the first place, if I was really there. Meaning, the stage is in front of me, and I can hear the "hall reverb", all around me. Hearing a violin or a piano playing behind me, might sound cool, but is not how I hear it in a real concert, and it is not the way that the artist wanted you to hear it, when he recorded the 2 channels mix.

In order to achieve what I want, you first need the impulse response of the hall that you are trying to mimic in your own room. An Impulse Response, is the deacy of the sound which is played on stage. The decay, of course, is different between different halls/auditoriums, and is affected by their size, shape, and several other factors. The deacy is actually the sound which is reflected from the hall's walls. Of course, each wall of the hall reflect the sound in different ways. Good recordings of Impulse Response, record the decay, using several microphones, around the best seat in the hall. Then, at home, you attach each impulse response measurement, into its corresponding surround speaker.

The sound which is played through your main stereo speakers isn't changed at all. its signal is simply duplicated to all the surround speakers, and as I've said before, each surround speaker (channel) already has an impluse response attached to it. Then, this impulse response is convolved with the signal which was duplicated from the stereo pair, and what you get, is the same decay of a real hall, as if your speakers were playing in that hall.

In simple words, the only thing that this technique is trying to achieve, is replace the walls in your room, with the walls of a very good auditorium, by using your surround speakers.

Of course, in order for this to work well, your walls should be acoustically treated to the max, ie. lot's of foam on the wall, to kill any reverberation they might add to the sound.

The speakers and amps of the surround array, can be of low quality, since they only help us to mimic walls, so they don't have to be as accurate as the main stereo pair.

Also, the more surround speakers you have, the better. The guy who is pushing this way of listening, is using 26 surround speakers at home.

And one last thing, you don't have to pre process your 2 channel music, and save it as multichannel music. All you need is a multi channel convolver, where you'll attach each channel its corresponding Impulse Response, and then you simply feed the stereo signal, and listen to your multichannel music, in real time.

If you don't like what you hear, you can simply change the impulse response to a one from a different hall.

[Jones_R]

Eye of Horus,

I'll indeed read your guide. But in the meantime, answer me this:

1) Have you used more than two surrounds for the reverb channels ?.

2) Was your room acoustically dead ?. I'm not talking about an un-echoic room, since it is impossible to obtain realistically, but, I am talking about some significant wall treatments, to significantly reduce the room's own reverberation time and amplitude.

3) Which Impulse Response have you used ?. You do understand that this whole technique could fall, simply due to the fact that the Impulse Responses are not good enough. As of today, I know of NO impulse response, not even of one hall, which is free to download, and give you acceptable decay quality.

4) This is pretty crucial. When you add the surround reverb speakers, the phantom imaging of a usual 60 degree stereo setup, fades by some degree. So, it is kind of important to use a better speaker setup. A center speaker, or a stereo dipole setup, are two of the options.


Again, you talk about the method I'm showing, as it was invented in the seventies. What I don't understand is, how did they convolve the signals back then ?, even if they did it off line, with the Processing power that was available back then, it would have taken weeks to convolve each song... :confused:

[Jones_R]

Eye of Horus,
I've read your old guide. It's funny you've mentioned and used the Aurora filters there. The author of the Aurora filters, is Prof. Angelo Farina, of the university of Parma, Italy. His main research activity is in acoustics. Anyway, he is now in New York, at the 115th AES convention (Audio Engineering Socienty). He is there as part of the Ambiophonics team (notice, not "ambisonics"), which uses the technique I've just showed above, in order to derive its surround sound, from 2 channel sources.

If you think that your new guide use the surround channels in a more realistic way, I suggest that you contact Farina, or Ralph Glasgal (the originator who merged several known ideas into Ambiophonnics). They both are dedicated of using the surround speakers in the most realistic way possible, in order to recreate the experience of being in the original hall, while listening to your 2ch music.

Anyway, in your old guide, I didn't see any references to any impulse responses. Unless I'm missing something, you've simply transferred the original two channels signal into the B Format, and then spawned all the necessary channels out of it. I don't see how this can compare to the technique I've showed here.

Also, in the guide in this page, you simply use vocal erase for the surrounds, and you call this an advancement over the use of real IRs ?, I really don't think so. Don't get me wrong, it might sound really good, but it does not sound real. The surround channels will play direct sound out of the original two channel signal. This is not the way I want to listen to my stereo recordings. This is too artificial for me (but I know that some will like it).

What is your target in the 2ch to 5ch conversion ?, adding more amazement, or more realism ?.



Edit:
Eye of Horus, I've just noticed that at some point in this post I've confused you with SurroundBoy. Sorry for that. I've started reading your newer guide only now.

[Jones_R]

Ok, I've read the newer guide.

My thoughts are that we are simply after two different things, and each of us probably achieve his goal better with his own method.

First of all, I don't want distinct sounds behind me. For me, it is a turn off. It is the most artificial thing one can imagine.

My room's acoustics, like 99.9% of the rooms that people use in order to listen to music, wasn't built from the ground, in order to have perfect acoustics. Actually, it sucks. It has bad bass, flutter effect, un-even decay. You name it, it has it. Of course, I could have spent thousands of US dollars, in order to tame this room physically, to fit stereo reproduction (ie. strategic placements of all kind of absorbers, diffusers, bass traps etc.), but, this way, it wouldn't fit HT, because HT requires a room to be more dead, acoustically.

What I've decided to do, is kill my room's "signature" over the sound, by soaking it with thick absorbing foam, all around. So, the walls have practically no effect over the sound now. The decay is super fast, and the room is pretty much dead.

Now, instead of the dead walls, I'm using speakers, to digitally recreate any space I would desire to listen to my music in. This way, the reverb is perfect, I have no flutter effects, weird echoes, or other weird acoustical anomalies. I'm not just using the surround speakers (I have four, and no center, no need, since both the main speakers are at the center location) to enhance the sound, ie. to make it more immersive and amazing, I'm using the surround speakers, to accurately (given the numbers of surrounds I have) reproduce existing halls and other spaces. If Ambisonics does this (ie. record the signal with the reverb of the hall), then it should sound very good too, in this task of digital recreation of spaces. And I'm not talking about the use that you do with Ambisonics, I'm talking about the traditional use, which needs you to record the material in a speacial way, to include the room's signature over the sound.

So, I think that we are after different goals. But anyway, I respect yours, and I think that what you do is great for a lot of people :-).

Btw, speaking about guides, I've also written one, about using the PC as a Room Correction device (based on a code made by someone else). You know, correcting for the room's acoustical anomalies, after you've measured them, from the sweet spot in your room. I use it in conjunction to Ambiophonics (not ambisonics! :-)), and the results are superb. At least to my ears.

[Jones_R]

About contacting : it would be nice if any of these guys also reads his email and responds !
While I know Farina is a bit busy, and I didn't have much luck with him too, Glasgal is very patient, and answered numerous of my emails (except from the last one, but I think it is because he is in the AES now, because I keep getting an automated response that his email is full). I will be happy to direct him to your guide, if that's what you wish.

It seems like I have to spell it out for you ! No problem :-)
The W,X and Y files that are used to convolve are :................. roompulses !
Sorry!. Anyway, using this way is somewhat less accurate than using the technique I've showed before, at least if you don't want to compromise on the quality of your original 2ch source. Ok, you use the W,X and Y to create the B-Format, and then you spawn each speaker configuration you need, but, all of the channels are processed now. Also, a symmetrical speaker array will not use my hardware in the most efficient way (since people usually invest most of their money in the front two speakers, and this is exactly what Ambiophonics needs). Again, with ambiophonics, you do not touch the main two channels (which are used in a stereo dipole configuration), they are not processed in any way, you only process all the rest of the surround sperakers, to mimic the wall behaviour. I think that this is the main difference between Ambisonics and Ambiophonics.

A basic Ambisonics setup:

Click here (http://www.avsforum.com/avs-vb/attachment.php?s=&attachmentid=13159).

A basic Ambiophonics setup:

Click here (http://www.avsforum.com/avs-vb/attachment.php?s=&attachmentid=13160).
Notice there needs to be a barrier between the "Ambiopole" speakers (physical or digital), in order for the crosstalk cancellation to work. Also, the front two surrounds are the least important, since the Ambiopoles reproduce some of the ambience that was recorded in the original song in the first place. The side height speakers are next to go, if you are in a shrtage of speakers/Sound Card channels, but the lower sides and rear speakers are the bare minimum for this setup to work well.

So : record roompulses in your own room and use them as in the first method ! Perfect sound for your specific situation !
Why does using roompulses recoreded in my own room is perfect ?. I've already told you, my room (without the acoustical treatment) has horrible acoustics, nothing I want to recreate again. Remember, the whole use of the acoustical treatments was to kill my room's signature over the sound. Why on earth would I want to recreate it back with my speakers after I took the time to eliminate it ?. The thing is, that after the acoustical treatments, my room is dead. You can not play 2ch stereo in such a room at, since there is no decay, and it will sound dead.

Furthermore, The next quote was made by Anders Torger, who also work at the AES Ambiophonics team. He was trying to explain me about his Linux AlmusVCU software, which is used in order to take the roompulses and use them in an accurate way with the surround channels. He also mention the B-Foramt, as one of the options they've used at the beginning:
AlmusVCU has two types of reverb formats, Stereo B-Format, and traditional fixed direction impulse responses. In the first case, it is a simple mathematical operation to pick out an impulse response for any direction, since the recorded format is 3D. In the second case, AlmusVCU adapts the impulse responses if necessary, in an approximate fashion. The result is usually good anyway, it is not that sensitive that reverb is correctly reproduced, it is more important that decorrelation is good and parameters like that.

Anyway, we used Lopez' software back in 2001 (a real-time convolution software), and then we had B-format based impulse responses, unfortunately with quite poor sound quality.

Today they use traditional fixed direction impulse responses, of very high quality. they say that their results are better than the B-Format method they've used at first, by a large margin.

LINK ??????
Access to the official (and LONG) thread at AVSForum.com, click here (http://www.avsforum.com/avs-vb/showthread.php?s=&threadid=283878&highlight=drc)

Access to a direct link for the guide download, click here (http://www.mooneyass.com/DRC/DRC.html)

[Jones_R]

Correct, but... so what ?
What are you trying to prove or say ?
I was just trying to prove that Ambisonics alters the original signal, so the end result will be less of what the song's creator intended for you to hear, and more of what Ambisonics sporadically did. Not to say it will sound less good (in fact, some people might even perceive it as better sounding), just less accurate.

Basically I only hear you say about your situation.
Why is it only for my situation ?. How many people do you know have the same level of quality speakers and amps, all around ?. Maybe you have, and that's great for you. But my economic level currently does not allow me to spend so much money on this hobby. This is why I've said that Ambiophonics works great for me in this regard, since it does not require you to have quality hardware, except for the front stereo pair. I think a lot of people can relate to this, so your comment that it only serves my interest, is uncalled for.

And when I take a look at your guide, I think it will merely stay that way !
My guide has nothing to do with Ambiophonics.

We made a software only method (3 !). Kpex made a software only method and Surroundboy did. You want us to buy extra hardware ? Well, that way we can make everything better....... ! Go for a decent Ambisonics decoder..... (hardware ofcourse !) LOL !!!!
Well, believe me, I wish I didn't have to use any special hardware for the guide. It just, that when you need to correct frequency and time (ie. decay) anomalies inside a room, you first need to measure the room, and let the software know what kind of anomalies it needs to correct. There is no way around measuring the room.

Anyway, as I've aready said, since I don't have much money to spare (university, second year...) I did my best to search for the least expensive hardware which will still do a great job for measuring the roo's acoustics. I bought the Behringer ECM8000 mike ($40) and UB802 mixer ($50) which serves as a pre amp for the mike, since the mic pre amps of sound cards, really lacks quality. Anyway, if $90 is too much for you, you can just rent the necessary hardware for one day (in order to create the correction filters for your sweet spot). It shouldn't cost you more than $20-30. Yes, you still need to spend money, but belive me when I say (or you can read the thread I've mentioned, and see other people say it too), it will be money well spent.

Dedicated hardware which does what my Guide let you do for $90, cost several thousand dollars. (Tact Audio and Perpetual technologies are offering such room corrections devices. You are welcomed to check their prices). This guide has brought Room Correction to the average people, for the first time ever. The creator of the Room Correction code for the PC, should get all of the credit anyway (Denis Sbragion), I was merely writing the guide.

I begin to get a feeling of comparing apples with pears.
No offense intended, but I think you should make it more interesting if you want to reach a large audience
It doesn't need to be "interesting", only to sound good :-).

And You really should keep an open mind for the differences between all these forums !
I do. The subject naturaly rised from the discussion in this thread. You're quite a nice guy to talk too :-) and I hope others read what we write too.

this is more the "average Joe" forum for audio (no offense ! I am on it too ). With not extreme expensive sound setups. When they could afford that, they wouldn't subscribe to a forum about copying DVD or Audio encoding. They just would buy an extra piece of hardware or an extra copy of their software.
Me too. But as I've already said, I'm limited by economic boundaries...

That's why I repeat this : if Surroundboy made his method and is satisfied with the results : Perfect ! No matter if it is outdated or not. That has nothing to do with it. The only thing that counts is the fact that he and others, like the results.
You are right, I stand corrected about this one. I just didn't realize at the beginning we are after different goals.

A lot like my method and I am getting a bit tired of those theoretical attacks.
I'm sure you're guide is useful to many. I wasn't questioning that at all.

Take a stereo piece of music. Convert it with all methods (Software only !!!) and then judge the results ! Then we talk again. OK ?
I already know that the results will sound amazing with some CDs, but again, this is not what I'm looking for. I'm not just after amazing surround effects, I'm after accurate reconstruction of known acoustical spaces, without touching the original signal for the main stereo pair.

Small correction :
a. this is not "basic" . You know you can use a speaker rig with as much speakers as you wish. This is a picture of the Pentagon rig. You could also showed us a Surround rig or a 8.1 rig.
By "basic", I meant more or less the minimum amount of speakers needed to make the method work (If I had a pentagon image, I would have used it instead of the hexagon, which is showen in the image). As I've mentioned before, people (who have the $$$) are running the Ambiophonics setup, with dozens of speaker at the same time. This is just a "basic" setup in the image, like in the Ambisonics image.

To get the best results from music in your room is to play it back as if it is recorded in your room. That's what I meant. Wouldn't the two eliminate each other ??
I'm afraid I still not quite sure of what you mean... Even if the mateiral was recorded in my room, it would have sound terrible, since the recording is also affected by the acoustic of the room.
My guide to Room Correction, let you play the song, as if your room had better acoustics, but for that, you first need to accurately know what the acoustics of you room are, and for this, you must use measurement hardware... (sorry, there are no freebies after a certain point).

You want to sound your room like a giant hall > add giant hall pulses.
You want your room to sound like a studio > add studio pulses.
You want your room to sound like your room > add your room pulses.
Sure, but in order to achieve this accurately, you first need to take your room's acoustics out of the equation. Ie. use foam on the walls to cancel reverberations at higher frequencies, and use a room correction device (like the one portrayed in my guide) in order to effectively deal with low frequencies anomalies (bass modes etc.), which the foam can't deal with. Then, you can chose to use the B-Format, or use dedicated Impulse Responses, which measured the acoustical space from several positions, so each Impulse Response position will be fed to a different speaker in your surround array, with no B-Format interpolations needed. Less processing, more accuracy.

Let me analyse these lines :
Is their method now sounding better because of the high quality impulses or is their B-format now sounding better than the B-format they've used at first ?
The former.

Anyway : our roompulses are high quality !
Which are they ?. You only gave me a link to Farina's IRs. These are not high quality, by absolute terms. I think that these are the ones Farina used at 2001's AES, with much less success in regard to what he uses today. I'll ask Anders about this issue.

[jorel]

excuse me!
:o

this thread "was" a simple guide for newbys like me
got some effects in surround in easy way!
after read this long discussion with tons of "quotes" i got:
"Warning: mysql_connect(): Can't connect to MySQL server on '192.168.0.1' (111) in /home/avsforum/www.avsforum.com/avs-vb/admin/db_mysql.php on line 40"
in the Jones_R's last links posted .

@Jones_R
you wrote:
" speaking about guides, I've also written one,"
"and the results are superb. At least to my ears."
great.
i will that your guide can use less words
than this posts and good results!
i'm against some of your opinions but they are too big to "quote".
oh yes, Farina is a bit busy,and answered numerous of your emails
but i listen my musics for more than 42 years > every day <
...seems that he don't have time to listen musics?!?!?
and i can write:
no matter what you use, you can't recreate a concert stage
back to the original sound. (you don't have wind in your room)!
:p

@Eye of Horus
your guides are great and i have it saved in my hd
but for newbys like me they are so complicated!
you don't have another simple way?
...and yes, the method was developed in the 70's !
and the "2 cents" in your post seems "pure gold"!


@bobn4burton
yes,you can use Adobe Audition,read this thread:
http://forum.doom9.org/showthread.php?s=&threadid=61757

sorry my poor english!
oops...i forgot:
i have ~50.000 musics,play guitar(more than 40 years)
piano(just a little),drums,bass,work with audio&video
more than 30 years,don't work no more(thanks GOD)
(again) i listen my musics every day (all day long)
....but don't sing cos it will break my speakers!
:p
for this reason i'm against of some opinions!

thanks!
:)

edited
i found the link to bobn4burton!

[Umma]

I think EoH's bidule method is as simple as it gets, and the other methods (KPex, Surroundboy, DSP8000...) vary with growing levels of complexity and time requirements...but ALL the methods, just as EoH said from the get-go, depend on the listener, room environment, and type of music. He has always promoted that the students of surround ( :D ) experiment, read, and share.

I use several different methods, depending on the music. Funny thing is, my very first conversion was almost identical to Surroundboy's method in this thread because I couldn't get the Aurora plugins to work! And I was trying to go by EoH's first guide. (just a moot FYI: it was Sheep from the PF Animals album).

I even found out that Samplitude, with a more just a little patience, can isolate certain frequencies/instruments\sounds for emphasis or de-emphasis. But I didn't use it simply because I didn't have the time to plan it out carefully. MX51 even works great for some songs, primarily Beatles, I think, though I haven't drawn my final conclusions, yet.

I love each one of these threads on converting 2-channel to surround! I have learned a LOT just from the dialogues, and it would be interesting to hear input from the ambisonic / ambiphonic academia such as D. Farina, et al.

And I'm glad there was no flame war. The cold print on a screen cannot convey facial expressions and vocal inflections that keep a discussion a discussion and not an argument.

How about a different thread, a sticky, for discussions on surround techniques and results, and let the other threads stick for the dialogues to focus on those particular techniques?

Thanks, eveyone, for the education!

[Jones_R]

Just one comment (and I'll quote with context now ;-))

When you say the stereo is altered by Ambi, I agree 100 % but at the same time : every method alters it. Yours too ! Leaving only the 2 front speakers unaltered makes no difference. You hear your stereo different from how it was intended in the first place ! Accurate is only the original !!

I agree, accurate is only the original. But, what is the "original" ?. Definitely not how I hear it in my room with a plain stereo setup.

I think that in this context (or any other, for this matter), the "original" is what the guy who did the mix, heard in the mixing room (if it's a studio album). Or, what the people who sat at the theater, while the band was playing, and the recording was recorded, heard.

So, in order to reproduce something closer to the original, you need, like you say, only to play the recording in plain stereo, but, in the original room that it was recorded in. That's the catch. And that's what Ambiophonics is trying to solve.

[echooff]

I have been reading this thread with a great deal of interest. What most of you seem not to realize is how unusual you are. You probably have better than average pitch. You can actually hear the difference in a audio file by equipment or space. Most people can't. My wife drives me to distraction because she can't hear what I consider to be major recording imperfections and wants to keep playing terrible recordings and even turns up the volumn:eek: Any two of you probably would not hear the same tune, on the same stero, in the same room, at the same time and be in agreement with what you hear. I guess the bottom line is it's all subjective. Not trying to flame anyone. Just pointing out something most people don't even know about, much less think about.

[echooff]

Sorry, didn't mean it to be addressed to anyone. Just making the general observation you just stated in a more round about manner. Everyone hears it differently. BTW, Once I managed to overcome my codec difficulties, your ambisonic method sounds great.

[Jones_R]

But.....uh.....the guy in the mixing room hears thru speakers what is played in the studio. What brand of speakers ? How do they affect the sound ? Who tells me the mixer has perfect ears ? And the mikes in the studio ? What's their effect ? And so on..... And so on....

First of all, I want to make it clear that I don't think that 100% accuracy in reproduction is possible. But, with this said, I do think that it is possible to drastically increase the level of accuracy. It won't be 100% identical to the original experience, but it will be much closer than using the usual reproduction means.

But.....uh.....the guy in the mixing room hears thru speakers what is played in the studio. What brand of speakers ? How do they affect the sound ?
Well, any good mixing room today contain a pair of monitors that are practically flat from at least 30hz to 20khz. Apart from this, the speaker's non linear distortion is low, as it should be for a good pair of mixing monitors.

If you use a decent pair of speakers at home (not "Bose" and alike), and on top of that you use a room correction device, like the one portrayed in my guide, you'll get a much closer representation to what the mixer heard himself, in terms of the direct sound (of course, it won't be "identical", but it will be much closer than letting your speakers act as a "loose canon").

I'm talking from my own personal experience. With my own ears I've heard how a bunch of different (but good) speakers, become almost undistinguishable from one to the other, after being corrected with a room correction software. I guess you need to hear it for yourself, in order to truly believe it.

True, it cost $90 USD (again, still way cheap in comparison to what you need to pay for a dedicated device which does Room Correction). In my country, you need to add 100% for this price, since my country is very expensive. That's why I've bought the parts abroad. And since they are so cheap, I didn't have to pay tax on them. Furthermore, you can also buy the hardware with your friends, since the hardware only needs to be used once (in order to measure the acoustics of your room one time, and that's it). You can also rent it for a day, as I've said before.

Who tells me the mixer has perfect ears ?
Let's assume that even if he doesn't have good ears (ie. he has fast roll off above 16khz, which is very common for those who work at the music industry, even if they are young guys), still, he is professional enough to know how to compensate for this, in the mix. If he isn't doing his first mix, and has credentials, which is usually the case when mixing for popular musicians, then you can be sure it's not going to be the mixer's ears which are going to ruin the song. Usually this part is reserved only for the band itself :-).

And the mikes in the studio ?
If the mikes are not good, it is going to affect both you and the guy who do the mix (and he will try to compensate for this in the mix, especially if it is a linear distortion), so you are still going to hear what he heard.

If you are talking about a live concert, well, in this case, the mikes should be indeed very good. But even if they don't, it doesn't means that your audio system should help the bad mikes do their bad job.


In conclusion:
Again, I'm not saying that 100% autheticity is achievable, but, we can still drastically improve the authenticity level of our audio system, and the results should be heared in order to believe.


One last thing,
I own a pair of two-way speakers, which are considered, at least by several audio magazines, to be one of the best (if not the best) two-way speakers in the world. Here is a pic:

http://www.norh.com/products/norh9/black9_0_7.jpg

Anyway, with the Room Correction software applied, and the speakers at Ambiophonics setup, my cheap, $150 USD Aiwa speakers, sound better in more than several categories than these expensive speakers by themselves. There is no question that in my room, I prefer to listen to the Aiwa speakers with Room Correction and Ambiophonics, than to these expensive speakers, by themeselves (or in any 2ch to 5ch expansion that my Denon receiver offers).

How is that for a $90 tweak ?.

[jorel]

Originally posted by Eye of Horus
I have this idea :



@ Jorel : I thought they were as easy as possible..... actually you're one of the first who call my guides difficult :-)

kind regards,

EoH

sorry my friend,!
i can't follow your guide with more than 164 replies!
is my english limitations.
:o
i download and save everything....all the thread include
but i loose the idea when reading it all.
for this reason i ask for a "simple guide".
i have some doubts and don't want to post with (more)
inexpressives questions to encrease the size of the thread.

i need a "resume" of it all.
thanks for atention.
:)

[jorel]

;)

ok, i will try this weekend!

thanks for atention and for your (great) guide EoH!
if i have doubts...i ask you!
:)

[kempfand]

Great threat to read :)

@ SurroundBoy : Thanks for posting your method. I always wanted to try what you did, as the big advantage IMHO is the 32-bit processing of the intermediate steps.

@ Jones_R : It looks to me that the set-up & method you worked out and outlined in the paper is very close to EoH's Ambisonics (original method with the impulses). I'd love to give it a listening-test, but my wife would probably jump at me if I started to foam up the living room :eek:

Regarding the discussion of 'getting close to the listeng-experience of the original recording': This is a question of taste and can be discussed for hours, so I won't trigger it again. For me personally, there is also more to it than just the music (i.e. Pink Floyd's Pompeii, even with perfect pulses of the arena near Naples, will be far from how I listen to it when I know the place and how 'it feels').

Kind regards,

Andreas

[Tantulus]

Originally posted by jorel
hy Tantulus,
welcome in forum!
:)
i was waiting SurroundBoy, Eye of Horus
or another developer to help you first.

i don't think that is a good idea change the "defaults"
cos "surround effects" depend of the ambient and the
home theater adjusts, it's personal...
you can choose the best when you are listen the "show".
effects is for play, not for encode...( only my opinion! )
encode with "changes" can give
artificial results and permanent regret!
think in the next home theater with great features when you got it!
:)

THanks so much for your reply. I've switched to the Ambisonics method because I found SurroundBoy's method too time consuming (no offence).

I don't know if this is out of thread but I found that after encoding to DTS the surround channels gave me the impression that the instruments were coming from the back of the room. I decreased the gain on the surround channels and use the concert hall setting on the Surround receiver (Onkyo TX-DS797) to provide a little extra ambience. Seems to work OK.

[YZ_-Freak]

Well i've tried all the other methods and this method works the best for me. I like the fact that you can control/adjust every channel as needed. Since I do all of my audio editting in Cool Edit, no need to open yet another audio tool to convert the audio. Ambisonics always seemed to lose the vocals in the mix. I'm glad I found this thread.

[arturjose]

in the step 4 , have this phrase:
"pan one track 100% to the left and other to the right."
how?

[arturjose]

thank you, for me your guide is the best,
it`s possible to make your Guide with Pictures

[nFury8]

This is the first time I'm doing this, I tried this guide and was successful for a few test tracks. Now I need some clarifications regarding the 32-bit conversion and processing since I decided to heed the experts advice and go 32-bits through the whole editing. I'm a bit stumped as to where in each individual step do I switch back to 16-bits. For example:


Step 1 (Front Left and Front Right):
Open up your stereo sound file into cool edit pro, go to convert sample type or press F11 to get there. Where it says Channels, check Mono, then for the Left Mix put 100% and For the Right Mix put 0%, Click ok and now you have your Front Left Channel. Complete the process again and do the same For Front Right, except have the Left Mix at 0% and the right at 100% Save your tracks as "X-FL" and "X-FR"

Please bear with me if i'm wrong somewhere here, so here goes.

The very first thing is to convert the 16-bit source wav to 32 bits by going to 'Convert Sample Type' and saving this new 32-bit file as the source, yes? Or no need to save and just work on it directly? I'm not sure on this.

Next go back to 'Convert Sample Type' again? And then check Mono, setup Left Mix to 100% and 0% for Right Mix, now here is where I'm stumped, when, before you check Ok, is this the part where we switch back to 16-bits, or stay 32-bits? If not, then I presume its in the part where we save each individual tracks as "X-FL" and "X-FR"? But when I tried to 'Save As' I couldnt find an option to convert the file back to 16-bits in the 'Options' button.
Or is there? :confused:

These questions also apply to Steps 2 and 3 as well.
But for Step 4, since it's a different process from the first three as outlined in the guide, how do I check it stays 32-bits all the way? Hope I'm making sense here. :D First time, first time .... :D

[CobraVerde]

Originally posted by SurroundBoy

Center Channel For Adobe Audition (Recommended)
Open your Stereo file, and go to Effects, Filters, Center Channel Extractor
This is the setting I use for The Center Channel (http://members.aol.com/surroundfiles/images/c1.jpg)
......
Surround Left, Surround Right For Adobe Audition (Recommended)
Open your Stereo file, and go to Effects, Filters, Center Channel Extractor
For the Surrounds I use this setting (http://members.aol.com/surroundfiles/images/c2.jpg)


Hi SurroundBoy
this images not exist :( where are now?
please, send me via mail: andrzej@cobraverde.net or publish here...

THX :)

[itaarn]

[QUOTE=SurroundBoy;361559]Software Needed- Cool Edit Pro or Adobe Audtion
Center Channel For Adobe Audition (Recommended)
Open your Stereo file, and go to Effects, Filters, Center Channel Extractor
This is the setting I use for The Center Channel (http://members.aol.com/surroundfiles/images/c1.jpg)

Surround Left, Surround Right For Adobe Audition (Recommended)
Open your Stereo file, and go to Effects, Filters, Center Channel Extractor
For the Surrounds I use this setting (http://members.aol.com/surroundfiles/images/c2.jpg)

Where can I found those images ? Or does anybody have a complete guide ?

[journstyx]

Hi Surroundboy,

Could you possibly re-post or send me the settings as described in your stereo to 5.1 mix using Adobe Audition 1.5? I refer to Center Channel settings and Surround Left and Surround Right. Unfortunately the links don't work.
I've used Cool Edit pro sofar, but think your settings will create a better result. I appreciate hearing from you. Thanks!
Bets regards,
Journstyx.

[ursamtl]

Hi journstyx,

I don't think surroundboy is a member of the forum any longer. If you check his profile, it says that he hasn't been on the forum since May 2005.

[journstyx]

Hi Ursamtl,

Thanks for pointing this out. If anyone does have the settings that Surroundboy is referring to, please post it or let me know.
Re: Usamtl: thanks for your insightful guides. They're very useful to try and create 5.1 from a stereo source. I'll be doing some stereo to 5.1 mixes in the near future using different set ups. If I come across something useful to add to the already excisting guides, I'm sure to post it.
Look forward to hearing from anyone re: possible seetings for Adobe Audition especially the Center hannel and Surround left and right channels.
Best regards,
Journstyx.

[pc_speak]

Hello journstyx

I've just finished writing a batch file to upmix multiple stereo ac3 files to 5.1. You'll find it here. http://www.users.on.net/~pcspeak/files/ac3upmix.zip

Please, download and open the batchfile in notepad or something. The instructions are commented in the batchfile.
Haven't done a lot of testing but it has worked so far. Will really appreciate you trying it out and giving me some feedback.
Regards. pc_speak

[journstyx]

Hi PC Speak,

Thanks for your response. I'll try and have a go at it in the next few days. I noticed some addtional stuff that needs to be downloaded, which I'll do. If I have questions, I'm sure to let you know.
Best regards,
Journstyx.

[journstyx]

Hi PC Speak,

Just a quick question. I've downloaded and installed the BeHappy and Avisynth. Can you describe what to do next? This is different then working other set-ups I've done, so I appreciate your help. Once I have a ac3 folder including a track I'd like to turn into 5.1 what would be the working/running order to get this done? Do I start up avisynth or Be Happy?
Look forward to hearing from you again.
Best regards,
Journstyx

[tebasuna51]

Haven't done a lot of testing but it has worked so far. Will really appreciate you trying it out and giving me some feedback.
Seems you use the Farina/Sursound upmix method added by NorthPole to BeHappy (http://forum.doom9.org/showthread.php?p=787216#post787216).

I have some comments to the generated avs. Here is a simplified version:
NicAc3Source("St.ac3")
Stereo = last # (1)
function farina_upmix(clip a) { # (1)
# front = a.soxfilter("filter 20-20000") # (2)
la = GetLeftChannel(a) # (3)
ra = GetRightChannel(a) # (3)
back = a.soxfilter("filter 100-7000")
fl = mixaudio(la, ra, 1.0, -0.1875) # (4), (6)
fr = mixaudio(la, ra, -0.1875, 1.0) # (4), (6)
cc = mixaudio(la, ra, 0.375,0.375)
lfe = ConvertToMono(a).SoxFilter("lowpass 120","vol -0.5") # (9)
sl = mixaudio(back.GetLeftChannel(),back.GetRightChannel(),0.668,-0.668)
sr = mixaudio(back.GetRightChannel(),back.GetLeftChannel(),0.668,-0.668)
sl = DelayAudio(sl,0.02)
sr = DelayAudio(sr,0.02)
return MergeChannels( fl, fr, cc, lfe, sl, sr)
}
2==AudioChannels(Stereo)?farina_upmix(Stereo):Stereo # (5)
# AmplifyDb(2) # (7)
# Normalize() # (8)

1) Random names are not needed here. Generated by BeHappy to avoid variable name conflict with possible avs input.
2) This filter maybe is not needed with ac3 input already filtered.
3) Only to make more readable the next
4) To make fast and more understandable. With some maths you can see equivalent to:
fl = mixaudio(
mixaudio(front.GetLeftChannel(),front.GetRightChannel(),0.500,0.500),
mixaudio(front.GetLeftChannel(),front.GetRightChannel(),0.500,-0.500),
0.8125,1.1875)
5) Not necessary ConvertAudioToFloat(), NicAc3Source output is already float.
6) If audio input is already normalized we can get here clips, to avoid clips maybe need:
fl = mixaudio(la, ra, 0,8421, -0,1579)
7) With normalized input we can't amplify without clip danger.
8) We can use, instead Amplify, Normalize(). Seems the filter function of SoxFilter support the Normalize().
9) Maybe we can let this work to audio equipment and use a empty lfe like this:
lfe=Tone(Audiolength(a),440,Audiorate(a),1,"silence")

[tebasuna51]

Hi PC Speak,

Just a quick question. I've downloaded and installed the BeHappy and Avisynth. Can you describe what to do next? This is different then working other set-ups I've done, so I appreciate your help. Once I have a ac3 folder including a track I'd like to turn into 5.1 what would be the working/running order to get this done? Do I start up avisynth or Be Happy?
Look forward to hearing from you again.
Best regards,
Journstyx
To work with BeHappy or BePipe you Need AviSynth and .NET Framework 2.0.

- To decode ac3 you need NicAudio.dll (http://nic.dnsalias.com/NicAudio.zip) in AviSynth\plugins folder.

- To filter in AviSynth you need SoxFilter (http://forum.doom9.org/showthread.php?p=761154#post761154) in AviSynth\plugins folder.

- To re-encode to ac3 you need the last Aften 0.07 (http://forum.doom9.org/showthread.php?p=996683#post996683) in working/BeHappy folder

Now you can use BeHappy or Bepipe/pc_speak batch method:

1) BeHappy method, one ac3 by time. You need last Shon3i BeHappy package (http://www.box.net/shared/nkihizx1dh) (NicAudio and SoxFilter included)

2) Bepipe/pc_speak batch method. You need Bepipe.exe from BePipe alpha (http://workspaces.gotdotnet.com/behappy) at working folder (like Aften.exe). And now execute (double click) the ac3upmix.cmd batch file.

[pc_speak]

Edit:- oops! Missed the last post from tebasuna51 before I posted this. I forgot about the soxfilter.dll AND aften. Really messed that up. Thanks tebasuna51.

Journstyx.
No need to run avisynth. It is automatically associated with AVS script files. It all happens in the background. Basically there is nothing to see.
You can just uninstall BeHappy if you wish, though there are some great things happening with it. All we wanted were the two files.
If you put ALL the files, except soxfilter.dll as tebasuna51 mentioned, including the audio files, in a folder by themselves and double click my batch file, it will do the upmix. Folder contents will be:

ac3upmix.cmd
aften.exe
AudioLimiter.dll
BePipe.exe
NicAudio.dll
audio file1.ac3
audio file2.ac3
audio file3.ac3

tebasuna51.
Yes. You're right. Farina/Sursound is used.
I only know batch files. When I saw that an AVS script was created by BeHappy I figured I could stuff it into a batch file to upmix all my 2 channel audio.
Partly understand, but over my head with regards to the rest of your post & bow to your greater knowledge on audio.
Please. If the AVS script has redundant lines, take them out. All improvements will be a great benefit.

Ahem. OK, I admit that all I did was a cut and paste from the BeHappy script. Just needed a simple way to upmix. :D

[journstyx]

Hi PC Speak, Tebasuna 51,

Thanks for the further info. I'm kinda new at these scripts as my 5.1 mixes are usually done by manually working various programs such as Adobe Audition and Nero Soundtrax.
So I've put the following content in one folder:

ac3 file
Nicaudio.dll
Soxfilter.dll
Bepipe.exe
Aften.exe

I couldn't find the audiolimiter.dll in the BeHappy zip file.
Also, which Aften.exe do I use? There seem to be many .exe in the zip folder.
In short; I'm still puzzled. Please be so kind to take through this step by step. I appreciate your help.
Best regards,
Journstyx

[pc_speak]

Rethink & Refresh.....
Let's try it this way. I include most of the files, except your audio, in the zip. :)

Upmixes all 2 channel AC3 files in a folder. Creates 6 channel (5.1) files @ 448 bitrate. Audio can be increased (open AC3upmix.cmd in notepad to change volume).

Install avisynth v2.5.7 (http://downloads.sourceforge.net/avisynth2/Avisynth_257.exe?modtime=1168199840&big_mirror=0)

Unzip soxfilter.dll (http://forum.doom9.org/attachment.php?attachmentid=5193&d=1136242927) into the, usually, C:\Program Files\AviSynth 2.5\plugins folder.

Download and unzip AC3upmix.zip (http://www.users.on.net/~pcspeak/files/ac3upmix.zip) to a folder. Put the AC3 files in the same folder.

Double click on AC3upmix.cmd. Thats it!

[tebasuna51]

So I've put the following content in one folder:

ac3 file
Nicaudio.dll
Soxfilter.dll
Bepipe.exe
Aften.exe
NicAudio.dll and SoxFilter.dll can be better in ...\AviSynth 2.5\plugins folder and now can be used by any AvySinth app, not only by BePipe in your working folder.
I couldn't find the audiolimiter.dll in the BeHappy zip file.Really, Audiolimiter is a Dimzon's experimental plugin not released in the official page. But don't worry is not used here.
Also, which Aften.exe do I use? There seem to be many .exe in the zip folder.
Choose the adequate to your hard/OS. See the wisodev comment (http://forum.doom9.org/showthread.php?p=998384#post998384):
"The Win32 builds work under Win64 operating systems (for example they work without any problems under Windows XP x64, Windows Server 2003 x64 and Vista x64). The Win64 binaries are native binaries so they only work under Win64 OS.

The most generic binary is placed in exe_pgo directory (win32 binaries archive) and it is the most compatible build to include in any software pack."

[journstyx]

Hi guys,

I appreciate your guidance to get me started! I will once again give it a go and let you know if I can get it to work.
Best regards,
Journstyx

[journstyx]

Hi guys,

So a avisynth script has been generated. I guess that's step one....now what do I do with this script?
How do I run avisynth? I've got it installed, but what do I do next?
Please advise,

Best regards,
Journstyx

[tebasuna51]

So a avisynth script has been generated. I guess that's step one....now what do I do with this script?
How do I run avisynth? I've got it installed, but what do I do next?
This command line, included in ac3upmix.cmd, run Avisynth and send the output to Aften to be ac3 encoded:
bepipe.exe --script "import(^your.avs^)" | aften.exe -v 0 -b 448 -m 0 -readtoeof 1 -cmix 0 -smix 0 -dsur 0 -dnorm 31 -dynrng 5 - "your-Upmixed.ac3"

If all is ok you must have a new xxx-Upmixed.ac3

[journstyx]

Hi Tebasuna 51,

This may sound strange but how do I run Avisynth? I really need some guidance with this, as it's quite different from the way I normally work. I don't see any .exe file.
Please advise,

Thanks
Journstyx

[tebasuna51]

Yes don't exist a exe file, only avisynth.dll at windows\system32.

When an app. (than support this procedure) open an .avs file, avisynth.dll is activated and work like a intermediate (frameserver) between the physical files and the app.

Avisynth can open audio and/or video files, modify the data and send after to the app. For instance, when VirtualDub open an .avs works like if the file is an .avi.

Here we use Bepipe to open an .avs and, with avisynth.dll, output uncompresed audio to STDOUT, then is 'piped' (|) to Aften STDIN and encoded.

Media Player Classic can play .avs files like other multimedia files and so on.