a Dolby PL signal is only 2 tracks...
A Dolby Digital 5.1 is 6 tracks...

Make sure you get them right...
I take it youre wanting to extract an ac3 into seperat PCM wav files.. be it 1 2 or 6 channels..??

Your best bet is a lil program called BeSweet
.
With this, there is a slight learning curve, but once you get your head around it, you can do virtually anything with an AC3 file.
One thing to ensure is to turn off the "Ac3 machines" DR processing.. you want a straight transcode

If vegas has created said AC3, Vegas will NOT reopen the file unless its been muxed into a VOB. That said, Vegas WILL allocate each appropriate channel to its respective allocation

BeSweet will do the job you wnat though..

"I take it youre wanting to extract an ac3 into seperat PCM wav files.. be it 1 2 or 6 channels..??"

No. I can do that pretty easily through Adobe Audition or even Vegas. I want to take a 2 channel PROLOGIC signal and break it into separate wav files

What I am looking for in the end is to be able to convert a prologic file into a AC3 file. But I need to be able to break the prologic file into individual wav files first in order to manipulate/edit the individual channels.

The whole idea here is to be able to use a surround mic like this one on my HC3 cam:
http://www.cascade-audio.com/index.html

This mike records as prologic onto the tape so you get surround sound on a HDV cam. But the entire deal is a "no go" if I can't figure out how to get prologic PROPERLY converted over to AC3

"PROLOGIC signal and break it into separate wav files'

Ok, a PL or PLII signal is still only stereo...

let me try to understand what youre trying to do here.. are you saying your wanting a stereo PL track, to convert into a 5.1 track?

OK, the other thing with "surround" mics is HOW they record the additional channels..
There are 2 ways..
Most times, the encoder is within the camera (obvioulsy), in turn, the rear is recorded on seperate channels.. (12bit audio)
Is THIS what youre tryin to extract?

Other times, the rear channels are recorded WITHIN the stereo track, BUT tagged with PLII metadata, being that its within the main stereo track (ie, 4 mic channels mixed into two FL RL./FR RR ), but the PLII metadata "positions' the RL and RR tracks to the rears.
Its still stereo, but it only carried PL metadata for decoders to read properly..

In any case, PL and PLII encodes are ONLY stereo...
U might need to check the camera and suss out how it works its audio magic

"Ok, a PL or PLII signal is still only stereo... "

Well... yes and no. Consider it the same as running 24p inside a 60i wrapper because that's pretty much what it is. PL is 4 (or 5 for PLll) separate channels running WITHIN a stereo channel, the other channels are hidden in terms of LEFT + RIGHT, LEFT - RIGHT, LEFT out of phase with RIGHT....etc...etc.

"let me try to understand what youre trying to do here.. are you saying your wanting a stereo PL track, to convert into a 5.1 track?' Yes... that's pretty much it.

While it is true that PL only encodes stereo... there are SEPARATE (not to be confused with the DISCRETE channels of AC3) channels laying within. What I need to to is get at those separate channels. I suppose I could run the file through a home audio prologic decoder and set up mics at each speaker output but there has to be an electronic or software solution to this. In the younger days for example I used to cheat a little and decode a rear channel without a PL decoder simply by taking a speaker and connecting one wire to the positive on the left channel and the other wire to the positive on the right channel. This would then decode the rear channel on a PL track.

Now what I am looking for is a software solution to decode the matrixed tracks within the original PL track. I am not interested in HEARING the matrixed tracks.... Instead I want to WRITE them as separate channels

The Vive ( http://www.cascade-audio.com/index.html ) looks very similar to the Holophone ( http://www.holophone.com/ ) both physically and technically.

Both use an analog matrix (a type of analog mux technique to convey in a somewhat risky/lossy process more final signals than the 2 channels that contain them would otherwise be expected to deliver). Sure, in an ideal world these would be flagged by the camcorder or delivery media in an out-of-band metadata channel. The HC3 won't have a clue about this. The concept of flagging will purely be the phase relationship of the difference/sum and filters used in the decoding software (or hardware). As I mentioned earlier these are analog filters with equations and co-efficients rather than some otherwise inaudible digital stream.

I can see why the manufacturers of these mics take the DPL analog route rather than the multi-channel SPDIF/AES-EBU digital route. At least with a suitable camcorder or field recorder you have an easy cost and convenience entry point if you want to readily sync this up with video footage.

If the surcode plug-in or the Nuendo (or dare I add ProTools) option isn't viable then I have seen some other software filters to achieve separate 2->4 , 2->5 or 2->6 channel expansions. These use a best guess algorithm to that of Dolby. I've since read that Nuendo 1.0 supported LRCS decode/encode into 2-channel deliverables. So I was wrong with the my earlier vers. 3 statement.

For clarity, the AC3, Dolby Digital, DTS and MLP techniques are quite different but would also be very difficult to record to spare channels on a pro camcorder.

Which takes me back to the HC3. This also requires something like a VMC-K100 to be able to connect up the Vive. Potentially more cost and clutter.

All in all I might be tempted to supplement an HC3 with a CX12. This might not give the same separation as the Vive or Holophone but the AVCHD format provides quite nicely for 5.1 audio within the file format on the MS and what goes over the USB2 umbilical. Sony do also provide a hot shoe device that provides a transducer to connect up a wireless bluetooth mic. The ECM-HW1, fwiw. Better suited to a camcorder that is already able to provide discrete digital surround (without any matrixed combining of compromised surround). Whether these are all too consumer to really be worth chasing with $$$ is hard to assess. As with all things, your production values and class of videography largely dictate how well these tools and utilities can be justified.

I have had a look for something written in accordance with common LCRS published matrices. The best freeware I've found has been:

http://www.savioursofsoul.de/Christian/VST/Quadrophonics.exe
and
http://www.digenis.co.uk/4Dec.zip
both VST plug-ins. (the latter being a derived work from CBS laboratories equations, found in the AES journal primarily for S/Q matrices rather than 5 or more separations.). You may need a VST host that is multi-channel capable for the plug-in and source routing itself. I missed whether VegasPro does everything that SoundForge does in this regard....?

"The HC3 won't have a clue about this."

Well that's the beauty of it.... the HC3 doesn't need to have a clue. You can record sound at 360 degrees into a simple stereo track.

Here are some reviews from B&H photo.... one in particular who is using on the HC3:

http://www.bhphotovideo.com/c/product/500287-REG/Cascade_Audio_VIV100_Vive_Surround_Sound.html

My problem would be converting that signal into something you can actually edit and convert over to AC3.

Thanks.... I'll have a look at those VST's.

"The HC3 won't have a clue about this."

Sorry Blink3times, this remark on the fact that the HC3 will continue unabated was more as a response to DJPadre's post. I think we are on the same sheet. Just that the Vive was a new mic to me when all I had really seen much of was the almost equivalent spec unit from Holophone.

Other than spending more than US$1000 on a solution to an otherwise rather small problem, I had two conclusions when I looked into this a couple of years ago:

1. Use http://matrix-mixer.sourceforge.net/#usage
(problem is you now need a directx host for this filter)
2. Look for a hardware decoder such as a soundcard with CPU-offload DPL decoding onboard with support, perhaps in the driver or the properties of one of the exposed codecs to write the bitstream output that would exist somewhere on the card to separate WAV.

Trouble is, the licensing of this older technology is such that any budget use of it must be linked or linkable with dishonest use. Well, if I was making a surround mic I would consider this seriously and look at an SPDIF or AES-EBU option. Which of course would give true discrete channels with certain separacy (in the format, not necessarily for the what the mic picks up with or without the casing reverberating).

I tried to use GraphEdit as the directx host for this filter. Perhaps there are better choices now and perhaps the DPL modes that http://matrix-mixer.sourceforge.net provides are as true as we can expect (and hopefully digitally simulated as filters with enough bits of precision in the binary digits). I think I was missing a multi-channel WAV writer or a single-channel multi-wav writer. Perhaps an application host would be better than a freebie tool from Microsoft!

If you get the workflow down, please share!

In the past when I've been asked to extract separated WAVs from Dolby Surround (inc. ProLogic) matrix encoded stereo files I've used Foobar2000 together with this free plugin.

Although the plugin is no longer being developed it is free and it works as well as anything else I've seen, with a number of modes that are designed to emulate the various flavours of Dolby Surround. I've never found any documentation for it though, so you just have to experiment a bit and use your technical knowledge to worrk out which mode will work best for any particular material. The output would normally be sent to a multichannel WAV which Vegas can open just fine.

As a possible alternative, a little while ago I noticed this suite of VST plugins that might be usable for this purpose. I haven't tried them myself so I don't know if they would be any good, but they look pretty.

Mark

Blink, a less refine solution is simply to take an old computer, put a couple of sound cards to capture output from another computer producing an analog output of the Prolog data. Since most new sound cards have good D/A converters, the analog signal will probably be better that most software short cuts in converting prologic.

I'm impressed with what I've managed after following MarkWWW's ATSurround plug-in . (I used foobar2000, added the converter plug-in at install time and converted to WAV by right-mouse-clicking the source material after opening it in fb2k, receiving a 6 channel multi-channel.WAV).

One comment is that there isn't a DPL, DPL-II or SRS matrix to choose from. So this processing may be based on simulated surround. Yet this isn't much less inappropriate than using S/Q filters such as the free VST ones I pointed you to earlier. V.I Stereo to 5.1 converter has similar attributes. These cause the sound to be less dry and separate. The best feature (IMHO) being the LFE processing. That said, DPL/DPLII doesn't have LFE in the matrix, so this is definitely simulating what might have been created in the appliance had it been a better class of device than it is. These are all observations though as the claim from the creator is that ATSurround 'Reproduces surround sound information present in many stereo audio material'. So perhaps there is more stuff going on using a dynamically assessed algorithm for the separation.

The price is right and maybe there is further forum information about this subject on fb2k related topics.

I see that ProOptimizer has a wrapper for pl2xdll.dll that comes with the commercial DVD software from CyberLink ( PowerDVD Deluxe or higher). If you own this then the foo_dsp_pl2.dll and libfftw3f.dll from http://hosted.filefront.com/prooptimizer might be all you need to save having to go out via a DPL receiver only to have to recapture the analog channels again. I'm not sure if Cyberlink's license that they then delegate to you is intended to work in this way. It would seem unlikely for commercial purposes depending on the patent lifecycle and possibly which country you consider attempting this from.

Decoding the Prologic matrix is the least of your problems. I've looked at the Holophone, Sound Field and Tetramic. Luckily before parting with my hard earned I was fortunate enough to discuss this with one of the local audio gods who has tried them all. The concept itself is flawed.
Pretty much what he'd had to say agreed with what I'd heard listening to the surround sound from the Holophone. At first it kind of sounds inpressive. Yes, you're right there in the middle of it. Except pretty quickly you realise that's all there is to it, a gimmick!
Not to say that mics such as the Sound Field or Tetramic don't have their uses but you need to use them in conjunction with additional mics and more channels to control where the listener is sitting in the sound field.

Bob.

Thanks for that Bob. That really does help position where this is all at.

The so called 5.1 features of the Sony AVCHD camcorders are similarly laughable for their rendition. Although it is nevertheless comforting to see different envelopes when they arrive on the Vegas timeline (from these units) plus the benefit that there isn't any upmix requirements such as those described earlier. They just appear when you drop the tracks on the timeline in digital form, albeit compressed.

An awful lot of high street, consumer and prosumer kit would leave a professional short on features and quality. Perhaps the Vive is fitting for folks with HC3 camcorders who want more but not the next level up.

foobar2k and the ProOptimizer + CyberLink PowerDVD Deluxe (any version? the latest?) looks like the closest thing at a suitable expense compared to the equipment listed here to achieve a multi-channel.WAV that Dolby or Fosgate would find representative. Much better value than SurCode, Nuendo or ProTools given the cost of the other ingredients. I do worry that repurposing a player as a converter might require further licensing from the patent/IPR owners especially if the work is going to be used for commercial purposes but perhaps the licensing that granted the engine Vive or the Holophone would suffice for these self serving applications. Something for a lawyer to clear, not I.

This is an interesting topic and for me, the conclusion is that surround sound is still something that a crew of engineers ought to be supporting. Even if the video portion is handled by a single cameraman. If the recording is being handled by a crew, then the likelihood of it being recorded in anything much less than 96kHz 24bit per channel PCM seems also quite slim.

I also worry that the Holophone type of mic requires the right amount of impedance matching prior to reaching the camcorder you record it on. Similarly the calibration to hit those equations right might also deserve more than a solid state pod in the shape of an egg. As with white/black balance for a camcorder, a microphone characteristic should be tuned up and compared on a meter for incursion to stay within and not beyond full-scale to avoid peaking etc.

I enjoyed the exploration though. There ought to be a computer based solution for every signal processing problem. Even if it costs $$$$ or requires some hoop jumping and $$.

"Except pretty quickly you realise that's all there is to it, a gimmick!"

To a certain extent I agree with that Bob..... that's why it's important to be able to separate the matrixed channels into wav files so that they can be brought into an editor, worked on, cleaned up, and then converted over to AC3. The prologic approach is by no means entirely accurate on its own but it will serve as a base to work with.

Up until now I have been taking a pure stereo sound track from the HC3, bringing it into Adobe Audition and physically chopping it up and dispersing the various sounds to new tracks labeled Front, Center, rear...etc. (ie: the cam operators voice is more rear than front so that voice goes to the rear track with just a slight mix to the front) It's an EXTREMELY time consuming affair... it takes weeks to convert 20 minutes of stereo over to discrete surround channels.

I could physically set up 5 recorders on the 5 edges of the room, and record 5 separate channels and then mix them for AC3.... this is probably the easiest solution but then it requires a certain amount of set up time and is not a very portable situation.

I figure the Vive although not entirely accurate, will give me a base to work with.... as long as I can get separate tracks to work with.

APIT... thanks.... I know that there are a few hardware solutions to this and I may experiment with that if all else fails.

I will say.... Foobar is looking pretty good... thanks.

"This is an interesting topic and for me, the conclusion is that surround sound is still something that a crew of engineers ought to be supporting."

I quite agree RB..... surround sound for some odd reason has always fascinated me. The things you can do with it are pretty amazing.

its my staple diet :)~
Sadly not many are licensed for DD, which si good for me, coz it drums up busines, but in the end, sound is 80% of what you see..

as for PLII, the way these ENCODES work is that it "muxes" the multicahnnel recording into a stereo stream, the uses metadata to allocate those frequencies to their respenctive channels., based on frequency range, and placement.
Its STILL stereo, and within that track, there is NO way of "extracting" the other channels because they do not physically exist

check out Dolby.com

You kinda miss the point.
Even starting with say a Tetra Mic and recording the 4 channels you don't get 4 discreet channels that you'd use for a true surround mix. You can take those 4 channels and through a matrix extract a stereo signal. It's a rather good way to get an ambient stereo recording as you can control the separation by basically steering the phased array using controls in the matrix plugin.

4 or 5 shotguns in the corners of the room would be just tragic.

Most surround recording seem to be normal stereo with added mics to feed the ambience into the rear channels. This avoids the problem with true ambient recordings where the listener hears sound that is far away from the source and perhaps muddied.

Bob.

"This avoids the problem with true ambient recordings where the listener hears sound that is far away from the source and perhaps muddied."

Yes... but having things a bit "muddled" is a good thing. For me anyway it seems to be more accurate.... so long as the mic at the source (usually the center mic) picks up clearly. Corner mics would in turn pick up the ambient noise (including the echo from front and center). Of course in the final mix gain becomes quite important since you don't want the "muddled" behavior from a far mic over powering the front and center mic..... so in the end what you have in a proper mix is the muddled sound from a corner mic working WITH the front and center mic to ENHANCE and add depth.

Albeit... some of the corner action is in fact NOT muddled.... an interview in a large room for example. A question is asked from the right rear section of the room while the cam stays stationary front and center. The situation becomes reversed That corner mic is heard pretty clearly and the front center mic becomes the muddled one

When I first started physically chopping up stereo tracks in Adobe Audition and dispersing the sounds to new tracks (front, center...etc) I would place the cam operator's voice on the rear track and leave it at that. On play back it sounded absolutely fake and it took me a while to figure out why. The answer is so simple really..... the cam operator's voice is not JUST on the rear track. What it is, is that the cam operator's voice is MOST POWERFUL on the rear track and LESS POWERFUL (or more muddled) the other tracks...... and that's what takes HUGE swaths time.... getting the mix in the proper proportions. I suppose the best comparison for this in video is Depth Of Field. The subject is in focus while the background is still there..... but rather out of focus (or "muddled").

"......STILL stereo, and within that track, there is NO way of "extracting" the other channels because they do not physically exist"

Well, I'm not sure I would go so far as to say "NO way" since there is a simple hardware solution.... (Play back a prologic sound track and hook the rear channels to a recording device instead of speakers). I'm just a little surprise to see so little in the way of software solutions.

RB is quite correct in pointing out that Prologic is almost purely an analogue ordeal and that I think makes all the difference in the world

Sure you can decode what would come out of each speaker and that might be what went into each mic. This is not what you want at all for what you're trying to do. You need to grasp the importance of phase in sound recording and playback. If it was a simple case of 'louder in one speaker / mic' then we'd never be able to tell the difference between sounds from in front of us and behind us, after all we only have two ears.

Bob.