Old thread for obsolete version of V.I bidule layout

[ursamtl]

Please Note, this original version of V.I has been replaced with a VST plugin and new Bidule layouts. This new setup improves on the original V.I by providing more flexible controls, the possibility of using the V.I algorithms in various programs besides Plogue Bidule, and professional solutions to avoid problems with file overloading, etc.

You can read all about it in
GUIDE: V.I Stereo to 5.1 Converter and II Surround Generator VST plugins (http://forum.doom9.org/showthread.php?s=&threadid=85446)

The rest of this thread contains messages about the old version, which is no longer available.

Regards,
Steve (UrsaMtl)

[daphy]

Hi folks,

I uploaded all the new files to needfulthings (http://www.needfulthings.webhop.org) this evening - so go on transcoding :D

[DSP8000]

Very pleased with the results from the bidule, I like it!!!
Simple and very effective layout.It gives nice surround for instrumental music.
ursamtl, can you try to filter the rears with vocal remover so the bidule can be suited for vocal music as well?
The LFE sounds real & not overdriven, every channel sounds very good & gives clean sound stage.
Keep up the good work, I'm following the other thread about stereo2surround in NUENDO as well.

DSP8000

[ursamtl]

Originally posted by DSP8000
Very pleased with the results from the bidule, I like it!!!
Simple and very effective layout.It gives nice surround for instrumental music.
ursamtl, can you try to filter the rears with vocal remover so the bidule can be suited for vocal music as well?
The LFE sounds real & not overdriven, every channel sounds very good & gives clean sound stage.
Keep up the good work, I'm following the other thread about stereo2surround in NUENDO as well.

DSP8000

Thanks for the feedback. Yes, I was really happy with the LFE as well. It seems to give just the right amount of "omph" to the bottom without sounding unnatural. I also noticed the same with the center channel. It added to or reinforced the soundstage without sticking out or sounding disjointed.

As for the vocal remover idea, I'll give it a try, but I didn't really run into a problem. Vocals are present in the rear, but on my system, they were much stronger from the front so there was no problem with the soundstage. For instance, on two of my test tracks, Elton John's "Tiny Dancer" and Robert Plant's "Heaven Knows", the lead vocals were clearly front and center, but the background vocals in the chorus were pleasantly spread right around the back and front with ambiance. I got similar results for the choral part of Beethoven's 9th, although some of the solo voices in the recording I have of 9th tend to stand out in a strange way, so they almost sound like they're coming from the back. When I compared with the stereo version, I discovered that they don't sit well in the stereo mix either.

So again, what's been said so many times applies here, the source material is the key. It's the method of recording and mixing that determines the outcome. Sounds that are hard-panned to either side of the soundstage can sometimes sound weird in an upmix. I found that running the stereo versions of some old Four Tops songs sounded amazing through V.I-just like having the singers plus the Funk Brothers playing in my living room! Still, some of those hard-panned 60s recordings give unpredictable results.

I'm reading some material during my vacation that might give some more clues. I also got some nice results last night from adding some third-order ambisonic calculations to the encoder section of V.I. The added harmonics tended to both reinforce the front center and the ambience in the rears. That's to be continued when I get back from vacation.

Regards,
Steve.

[DSP8000]

I agree totaly, it is the actual source that gives the final upmix sound.
However, the vocal remover(partial remover) idea sounds good to my ears & makes me "feel" real original 5.1
For now take a break & enjoy your vacation.

DSP8000

[old-hack]

I set this up originally according to the "new method" in the original post for converting stereo to 5.1 surround and then came across the "V.I." method. I got confused by the diagram included in this thread especially since the instructions said to set up bidule according to http://forum.doom9.org/showthread.p...2851#post362851.

Other visual examples in http://forum.doom9.org/showthread.p...2851#post362851 show straight line connections that contradict how pin connections are supposed to be from the Emigrator to the Recorder. For example, the poster that added LFE capability showed straight through pin connections. I also saw graphics of straight through pin connections by Eye of Horus after showing the 1>3, 2>1, 3>5, 4>6, 5>2, 6>4 pin connections. I just find it hard to be sure what will work between the Emigrator and the Recorder if the pin connections aren't consistently the same.

As another point, automatic links are created in the vi.bidule. The player-recorder link can become inaccurate. After adding and deleting parameters, bidule appends a number for the next instance of each parameter. The default vi.bidule link is audio file player to audio file recorder. When you add a player instance it becomes audio file player_0. If you leave the default vi.bidule link, your recording will not function. You have to make sure the audio file player instance you've added in the interface is the player used to link to the vi.bidule audio file recorder instance.

Another observation. While I like faster processing of files, the vi.bidule grabs 100% of the processor time and has side affects. The cpu & disk utilization indicators at the bottom of the screen turn into a counter instead of cpu utilization and is empty for the disc utilization.

These aren't complaints, they are observations that I haven't seen posted by anyone since this thread started.

Edit: When I add Microsoft Sound Mapper, the cpu & disk utilization values appear. Must be because it has to play the audio at a normal rate of speed.

Here's something odd. The audio is speeding up during processing. The player shows what appears to be the normal playback time while the recorder is 11 seconds behind.

[ursamtl]

Originally posted by old-hack
I set this up originally according to the "new method" in the original post for converting stereo to 5.1 surround and then came across the "V.I." method. I got confused by the diagram included in this thread especially since the instructions said to set up bidule according to http://forum.doom9.org/showthread.p...2851#post362851.

Other visual examples in http://forum.doom9.org/showthread.p...2851#post362851 show straight line connections that contradict how pin connections are supposed to be from the Emigrator to the Recorder. For example, the poster that added LFE capability showed straight through pin connections. I also saw graphics of straight through pin connections by Eye of Horus after showing the 1>3, 2>1, 3>5, 4>6, 5>2, 6>4 pin connections. I just find it hard to be sure what will work between the Emigrator and the Recorder if the pin connections aren't consistently the same.

As another point, automatic links are created in the vi.bidule. The player-recorder link can become inaccurate. After adding and deleting parameters, bidule appends a number for the next instance of each parameter. The default vi.bidule link is audio file player to audio file recorder. When you add a player instance it becomes audio file player_0. If you leave the default vi.bidule link, your recording will not function. You have to make sure the audio file player instance you've added in the interface is the player used to link to the vi.bidule audio file recorder instance.

Another observation. While I like faster processing of files, the vi.bidule grabs 100% of the processor time and has side affects. The cpu & disk utilization indicators at the bottom of the screen turn into a counter instead of cpu utilization and is empty for the disc utilization.

These aren't complaints, they are observations that I haven't seen posted by anyone since this thread started.

Hi old-hack,

Sorry if the instructions caused some confusion. When I said to follow the instructions from the other thread, I was referring mainly to what to do with the 6-channel file, as in how to run it through Besweet. I was going on vacation and someone asked me to post what I'd developed so far. I still intend to write a step-by-step guide once I clear a backlog of work that piled up while I was away. For now, I'll edit my original post to clarify the statement.

Nonetheless, I should point out that V.I is not meant to be used with Emigrator at all. The only VST plugin required is as stated at the beginning of this thread, the mda dither plugin. The pin connections for V.I follow the standard ITU spec for 5.1, Left Front, Left Right, Center, LFE, Left Surround, Right Surround. The pin connections in EoH's original guide are different because he is using the Pentagon speaker layout and then rotating it 36° to the left so that the #1 speaker serves as a center, #2 as left front, #5 as right front, etc. If you check other bidules uploaded here that use different speaker layouts either in Emigrator or elsewhere, you'll find that the pin connections change according to the setup. Just always remember that the target output connections (whether they be a 6-channel file recorder or a multichannel ASIO device) should be the ITU 5.1 layout L, R, C, LFE, sL, sR.

Yes, if you delete any objects in bidule, parameter links associated with them will also be deleted. Likewise, if you add a device, it won't automatically be linked. The choice here was to provide the V.I file in a "ready to go" state so that all someone needs to do is load a 2-channel file and choose 16- or 32-bit output format, and then create the 6-channel file. I left the bidule set to a 44.1 sampling rate, since the main interest in the threads here seems to be music and as far as I know, surround CDs need to be at a 44.1 sampling rate to work with DVD players.

Depending on your computer system, offline processing can take up a lot of processing power. The indicator changes at the bottom of the screen are part of the bidule program so there's no way to change that behavior in the bidule.

Thanks for your feedback. These are good points that I'll certainly take into account when I write a full guide.

Regards,
Steve.

[old-hack]

That certainly clears up some of my confusion. However, when I load the vi.bidule, I can only see the audio file recorder with line running up and out of the screen. I can't see any connections that resemble your graphic in this thread.

[ursamtl]

Originally posted by old-hack
That certainly clears up some of my confusion. However, when I load the vi.bidule, I can only see the audio file recorder with line running up and out of the screen. I can't see any connections that resemble your graphic in this thread.

Just hold down your up arrow key to scroll up in bidule, or CTRL+the up or down arrow keys to zoom in and out in bidule.

[old-hack]

Whew! I was going nuts trying to find a window refresh command or something that would center the graphics. Thanks.

[ursamtl]

Thanks to everyone for the encouraging remarks and feedback concerning V.I. Rather than answer your PMs and emails with the same response over and over, let me say that I'm working on the complete V.I guide. I've been experimenting with ideas I got from my vacation reading, so I may even implement a couple of these. In the meantime, V.I in its present form still provides excellent surround results, so enjoy!

Steve.

[daphy]

Hi Steve,

after a really long 'abstinent multichannel periode' I tried your new bidule :cool:
Sounds cool to my ears - maybe I have to play with the LFE a little. Using Teufel speaker system M200/M6000 on a Yamaha A2/Luxman M02 (only for LFE) is always a little critically because this system is too 'honest' - if there is no bass in the source they won´t invent one! :sly:
On the other hand real good sources will be played fine :D

Back to the LFE - is it possible to integrat two controller for high bass f.e. (80-250HZ) and low (80-20HZ) ones? This should give possiblities to turn over the gap between some satilite + subwoofer systems in frequenzy process (hope you´ll understand what I mean -
my English... :( )


Finally I noticed again -> transcoding to DTS gives the whole thing the final kick! So I would always prefer DTS not AC3 ;)

Thx

[LigH]

Not that I misunderstand something:

Is there a real ProLogic Surround decoder involved? Or does it just add some kind of reverb to rear channels, to give these speakers anything to output?

[ursamtl]

Originally posted by LigH
Not that I misunderstand something:

Is there a real ProLogic Surround decoder involved? Or does it just add some kind of reverb to rear channels, to give these speakers anything to output?

Hi LigH,

V.I is based on Ambisonic algorithms that are found around the internet. Although these are designed for material recorded using specific miking techniques, when they are applied to ordinary stereo signals, they redistribute harmonics and ambience already in the recording. In other words, no reverb is added. Reverb that already exists in the recording is spread out around the listener. If a recording is very dry with no reverb, then the dry sounds will tend to spread around the listener. ProLogic tends to put ambience only in the rears, whereas V.I distributes it around the full soundstage.

The best thing would be to give it a try. Some recordings will sound better than others due to how they're originally mixed, but so far I haven't found any music that didn't benefit somewhat from V.I.

Steve.

[ursamtl]

Originally posted by daphy
Hi Steve,

after a really long 'abstinent multichannel periode' I tried your new bidule :cool:
Sounds cool to my ears - maybe I have to play with the LFE a little. Using Teufel speaker system M200/M6000 on a Yamaha A2/Luxman M02 (only for LFE) is always a little critically because this system is too 'honest' - if there is no bass in the source they won´t invent one! :sly:
On the other hand real good sources will be played fine :D

Back to the LFE - is it possible to integrat two controller for high bass f.e. (80-250HZ) and low (80-20HZ) ones? This should give possiblities to turn over the gap between some satilite + subwoofer systems in frequenzy process (hope you´ll understand what I mean -
my English... :( )


Finally I noticed again -> transcoding to DTS gives the whole thing the final kick! So I would always prefer DTS not AC3 ;)

Thx

Hi daphy,

Glad to hear your period of abstinence is over. ;) You might try adding a bit of equalization in the bidule between the Audio File Player and the V.I. group. Boosting the bass frequencies you seem to lack might do the trick. Don't try more than 3dB at a time because what seems to be a small boost at a given moment, often sounds too much the next time you listen to it. Our ears tend to get to accustomed to the modified sound. There are several good freeware EQ VST plugins out there. Probably one of the nicest I've seen is at http://www.kjaerhusaudio.com/ although Voxengo also offers a free basic EQ VST, the Voxengo EssEQ VST at http://www.voxengo.com/freevst/. You could also go into the Outputs group inside the V.I group and change the LFE crossover frequency.

For testing, you could try using CDRWs if your DVD player reads them ok. Mine does and I find this a great way to test a DTS CD before burning the final copy.

You're right, DTS really does seem to sound nicer for music. Depending on the encoder used, AC3 sounds good, but DTS seems to have that extra kick or sheen.

Regards,
Steve.

[ursamtl]

I might also add that tons of free VSTs can be found at http://kvr-vst.com. Their search engine takes a bit of getting used to. If you do a search and get 0 results, chances are you clicked something wrong somewhere, but keep trying. I've found a lot of gems there!

[LigH]

Understood.

I was just afraid, because many people want to prefer 5.1 over 2.0 just "because 5.1 is cooler", not because there are real 3D positioning parameters available at all (they tend to waste bitrate by uselessly blowing up audio streams).

So I'm probably not quite your target: Ambisonic might be fine for chill-out music, but not suitable for giving audio sources (like speech in movie audio) a more exact position in a sound space, or even reverse-engineer 5.1 sound from ProLogic encodes.

[ursamtl]

Originally posted by LigH
Understood.

I was just afraid, because many people want to prefer 5.1 over 2.0 just "because 5.1 is cooler", not because there are real 3D positioning parameters available at all (they tend to waste bitrate by uselessly blowing up audio streams).

So I'm probably not quite your target: Ambisonic might be fine for chill-out music, but not suitable for giving audio sources (like speech in movie audio) a more exact position in a sound space, or even reverse-engineer 5.1 sound from ProLogic encodes.

All I can say it to give it a try. You might actually be quite surprised with the results. No one is "targeting" you. We're simply sharing this info because it's fun and cool.

V.I is primarily designed for music, but I imagine it could be useful for movie soundtracks as well, depending on the quality of their stereo mix. If you just want a big movie-style surround, you're probably better off just running it through ProLogic II. for music, however, I'm much more satisfied with the results from V.I than from my receiver's ProLogic II circuit. The sense of realism is better and more of the natural ambience present in the original recordings comes through. I've tested V.I on a variety of music. Yes, it sounds good on "chill out" music, but anything with ambience benefits greatly. Here are some examples:

Led Zeppelin - Gallows Pole: acoustic guitar and vocal intro appear as in the stereo version mixed left and right respectively but with recordings natural reverberation in the surrounds. Song buildup spreads the instrumentation out nicely all around. Repeated background vocals near end are much more distinct than in stereo version.

Pink Floyd - Time: dry clock sounds appear everywhere. Deep reverberating drum sounds in intro come from front but with big reverberation sound spreading all around. Good separation of keyboards.

Beethoven's 9th (fairly generic recording): Solo vocals appear in front with choral parts spreading all around. Music in front with reverb all around.

Four Tops - Bernadette (stereo mix): Bass and lead vocals appear in front center. Music spreads across fronts with some wrap around to the side and reverb in the rear.

Rush - A Farewell to Kings: Classical guitar has uncanny sense of realism as it moves from front left to right side. At the same time vibes sound spread out. Chirping bird sounds come from right and sound as if they were outside a nearby open window.

Pink Floyd - High Hopes: Bell sound from front right with a great sense of distance and reverberation. Buzzing fly sounds incredibly realistic.

John Williams - Concierto de Aranjuez: Tremendous sense of realism on guitar with natural reverb all around. Orchestra sounds natural and big.

Led Zeppelin - Black Country Woman (remastered): Plane at beginning sounds very real. Guitar and vocals bright and real with front imaging good and reinforced by very slight reverb in rears. When drums come in they sound as if they're right in front of you.

Roxy Music - Avalon: Percussion comes from front but is reinforced by natural reverb spread out all around.

Seal - Love's Divine: Thunder claps and rain at beginning come from all around. Vocal is dry in the front center. Reverb on vocal appears gradually. Synth strings fade in nicely and sound really spacious behind dry instrumentation and tight bass.

Deep Purple - Lazy (remastered 2-channel version): Incredible recording to start with, but the sense of the room is amazing when the guitar riff starts and the drum stabs happen. This is a great example of how a good recording can shine through a good surround upmix (it's even more amazing given the circumstances of the original recording :eek: :)

Anyway, I could go on. The point is, you have to try it. Test a track or two, burn them to CDRW using the instructions in this forum and see what it sounds like when played through your DVD player.

Regards,
Steve.

[LigH]

Indeed - I never shall argue without testing first.

I just wanted to state "This is not a Dolby Surround Software Decoder". ;)

[ursamtl]

Originally posted by LigH
Indeed - I never shall argue without testing first.

I just wanted to state "This is not a Dolby Surround Software Decoder". ;)

Ok sorry, I didn't get this from your original message or follow-up.

To decode a Dolby source it has to first have surround information encoded. What we've been doing here is create a surround mix from an original two-track stereo source. We have two choices, either add some sort of artificial reverb--whether it be by a synthetic processor or convolution--or else extract the natural ambience already present in the stereo recording. V.I uses the latter approach. Dolby ProLogic II also extracts some natural ambience from a stereo recording , but it tends to route all the ambience to the rears. V.I and many of the other methods here attempt to distribute the sound a bit better than this.

[Sycho]

Originally posted by ursamtl
You're right, DTS really does seem to sound nicer for music. Depending on the encoder used, AC3 sounds good, but DTS seems to have that extra kick or sheen. but how well does dts reflect the orginal recording dts encoders, hardware ones, have been known to modifiy the sound, LFE higher, or surrounds higher

[ursamtl]

Originally posted by sycho
but how well does dts reflect the orginal recording dts encoders, hardware ones, have been known to modifiy the sound, LFE higher, or surrounds higher

this may well be...I've never heard it before. All I know is that on my own surround system, switching between AC3 and DTS seems to have very little effect on the levels of any speakers. DTS does seem to have a slightly more open, transparent sound that makes the AC3 tracks sound somewhat more "boxy." It's a very subtle difference.

[Sycho]

what did you encode the AC3 with?
originaly posted by me in another forum
this comparisen was taken from the 'Germerica' DVD-Audio Disc it show's the LFE channels of each track http://www.highfidelityreview.com/tech/germerica.asp, it's funny though each of those was encoded using the same master, look at this (http://www.highfidelityreview.com/reviews/review.asp?reviewnumber=523387) about 2/3 the way down

[ursamtl]

Originally posted by sycho
what did you encode the AC3 with?

Originally posted by somebody, somewhere :)
this comparisen was taken from the 'Germerica' DVD-Audio Disc it show's the LFE channels of each track http://www.highfidelityreview.com/tech/germerica.asp, it's funny though each of those was encoded using the same master, look at this about 2/3 the way downEr, this quote is from another thread. Did you get this mixed up somehow?

[Sycho]

it was a quote of my self from dvdfile forums, it links to a revieew of a dvdaudio in which all tracks were encoded from the same master yet the dts sounds different,sorry about the confusion


what did you use to encode the "boxy" AC3 track?

[ursamtl]

Ok, that makes more sense. :)

I wasn't talking about a specific encode but in general. This includes both movies and convert DVDs. The AC3 tracks tend to have a very subtle boxy sound compared to the DTS tracks. Depending on the particular DVD, it's almost inperceptible, plus it may be partly a psychological thing, i.e., I'm expecting the DTS to sound better, so it does. After reading the "battle of the specs" technical documents on both the Dolby and DTS web sites, I'd have to say that the DTS docs seem less influenced by marketing hacks.

[Sycho]

don't compare tracks that sound different, as they are different.

[ursamtl]

Originally posted by sycho
don't compare tracks that sound different, as they are different.

The whole point of the exercise is to switch between a given DVD's AC3 5.1 and DTS 5.1 soundtrack to compare their sound. When I do, DTS sounds slightly better.

[Sycho]

Originally posted by ursamtl
The whole point of the exercise is to switch between a given DVD's AC3 5.1 and DTS 5.1 soundtrack to compare their sound. When I do, DTS sounds slightly better. but if their encoded with different masters, they will not sound the same, you may prefer the mix they used, it does not make dts better than dolby digital.

[ursamtl]

Originally posted by sycho
but if their encoded with different masters, they will not sound the same, you may prefer the mix they used, it does not make dts better than dolby digital.

A valid point, although given the extra costs for producing two different masters, I imagine that most use the same masters for both.

What does make DTS better is a series of tests done by independent third-parties. This info is readily available if you check out the DTS site.

Anyway, this is a topic best addressed elsewhere. This is a thread about V.I. Have you tried it?

[Sycho]

Originally posted by ursamtl
A valid point, although given the extra costs for producing two different masters, I imagine that most use the same masters for both. the same masters are rarily used.

What does make DTS better is a series of tests done by independent third-parties.LOL thats really funny as many coporations have stoped using dts after conducting there own test, Warner, the Producer of that DVD-Audio, do a search at the AVS forum and digg deep, you will fing many members love dts some of the most respected know that it is unnessary

Anyway, this is a topic best addressed elsewhere. This is a thread about V.I. Have you tried it? [/B] not at all,what kind of decoding method is used?

[ursamtl]

Originally posted by sycho
not at all,what kind of decoding method is used? [/B]

Read the thread, then give it a try.

[ursamtl]

Originally posted by daphy
Sounds cool to my ears - maybe I have to play with the LFE a little. This is just to elaborate a bit more on what I said about raising the crossover frequency. I set it to 60Hz in the V.I since this is way down out of the way below any bass management crossover and allows just the really low transients from drums or well-recorded bass guitar for music, or of course lots of effects from movie soundtracks. One option might be raise the crossover to above the range of typical bass management crossover freq. to say 250-300Hz. Then use the LFE slider to boost the entire bass region. To do this, just open the V.I group, and then in turn the Outputs group. Inside, you'll find the constant 60.00000 in the upper middle section. Try changing this to whatever value you want. Depending on how you've set up your system's bass management, this might give better results. I'll provide some sort of controls for this type of thing in a new "Super V.I" I'm working on now (initial tests are underway and the results are very promising).

Some people recommend avoiding using the LFE channel at all, whereas others use it. The technical documentation on Dolby's site reflects both sides of the argument. The one thing that is clear in their documents and elsewhere: when encoding using an LFE, it's important to be able to monitor your mix through a system with bass management. This allows you to hear what's happening to your bass.

As I mentioned in another thread, I corresponded a few weeks back with Kevin Shirley, the engineer who worked wonders with the surround on the Led Zeppelin DVD last year. He told me he used the LFE on it. Kevin is a professional who has won awards for his work on this release, which has become the best-selling music DVD of all time, so obviously he knows his stuff!

Finally since you're doing this upmixing for your own enjoyment and not to illegally distribute to others, have fun with it, and do what sounds good to you!

Regards,
Steve.

[ursamtl]

As I detailed in another thread, if you'd like to compare V.I's accuracy with the original source file and your computer is equipped with 5.1 speakers and soundcard, try the following:

1. Right-clicking on the box in the middle called "V.I" and switch the Processing Mode menu item to "Bypass." This switches the effect off and gives you the original two-channel stereo input.

2. After listening for awhile, switch the Processing Mode menu item back to "Processing."

As you switch back and forth, you'll hear the soundfield grow and shrink. The fronts will sound pretty much the same, but the overall sense of ambience truly does expand and contract quite naturally.

Regards,
Steve.

[krbo]

Hi Steve,

first attemp with your V.I. and latest downloaded Bidule (0.662)

no go - I don't have connected blocks after opening your layout,
just opened V.I and audio recorder windows (according to included
PDF I should click on VI module to have that)


Am I doing something wrong or your layout isn't compatible with
latest Bidule ??

(other layouts opens correctly)

regards,

[ursamtl]

Originally posted by krbo
Hi Steve,

first attemp with your V.I. and latest downloaded Bidule (0.662)

no go - I don't have connected blocks after opening your layout,
just opened V.I and audio recorder windows (according to included
PDF I should click on VI module to have that)


Am I doing something wrong or your layout isn't compatible with
latest Bidule ??

(other layouts opens correctly)

regards,

Hi krbo,

I've never heard of this problem before. V.I was created in 0.6602 so it certainly works with it. I just checked the files from both download links in this thread and they both opened successfully on my machine.

If you look at the following image of the default V.I layout, you'll see what it should look like when you open it: the blue Audio File Player box with two connections to the top of the blue V.I box, which in turn has six connections to the blue Audio File Recorder.

http://www.vdn.ca/~ursamtl/VI_default_layout.gif

If you didn't install the mda dither VST plugin, then the V.I module may appear in a different color to indicate that there's a problem. However, I've never seen a layout open without any connections. One problem you might have is that on some PC screens, the whole layout is shifted off the top left corned of the screen. In this case, simply click on the black screen background to switch focus to the layout screen, then use the left and up arrow keys to scroll the layout to where you can see it.

Anyway, give this a try and let me know if there's anything else I can help you with.

Regards,
Steve.

[krbo]

Originally posted by ursamtl
One problem you might have is that on some PC screens, the whole layout is shifted off the top left corned of the screen. In this case, simply click on the black screen background to switch focus to the layout screen, then use the left and up arrow keys to scroll the layout to where you can see it.


Hi Steve,

but of course quoted was a problem :)

green square should be driven by arrows to almost the same position as
black square - voila - layout is here !

this worth some mentioning in PDF

thanks Steve !

small "wish list" - if it's possible some setting in VI module to
reroute channels to be directly compatible with Softencode

[krbo]

Another question to everyone in this thread:

If you are using Softencode, do you keep default parameters for 3/2 +lfe encoding (center mix level,surr. mix level,DC/high/low filters)

???

[ursamtl]

Originally posted by krbo
Hi Steve,

but of course quoted was a problem :)

green square should be driven by arrows to almost the same position as
black square - voila - layout is here !

this worth some mentioning in PDF

thanks Steve !

small "wish list" - if it's possible some setting in VI module to
reroute channels to be directly compatible with Softencode

Glad you found the layout. We'll update the PDF once daphy gets back. He manages web site where the download with the PDF is located. He's on vacation right now but I think he mentioned returning next week. In the meantime, I've updated the download from my personal web site.

As for the output channels, they follow the ITU standard of L,R,C,LFE,sL,sR. This is compatible with Besweet as well. SoftEncode is an older program that's no longer available, so perhaps it followed an older standard. In any case, SoftEncode does have the layout graphic to the left of each channel, so just check the channel position once you load the 6-channel file.

Alternatively, you could reroute the outputs between the V.I object and the Audio File Recorder for your own copy of V.I. The signals coming out of the V.I object are from left to right L,R,C,LFE,sL,sR.

Happy Upmixing!

[ursamtl]

Originally posted by krbo
Another question to everyone in this thread:

If you are using Softencode, do you keep default parameters for 3/2 +lfe encoding (center mix level,surr. mix level,DC/high/low filters)

??? I've personally had good results from the default settings, but depending on your source file, tweaking the settings may produce a better result. You have to experiment. If you have a DVD player that accepts CDRW if you're burning Dolby Digital wave files for surround CDs, or DVD-RW if you're encoding DVD soundtracks, you can do lots of testing to get things right.

As for the settings themselves, no need to address the technical details here. In addition to the SoftEncode help file, check the excellent thread in this forum dedicated to encoding AC3. That would be the best place to post any questions about details of AC3 settings. For some other technical information, you can also check the Dolby web site for PDFs covering many aspects of AC3 encoding.

Here are the links:

Sticky:GUIDE: How To Properly Encode Dolby Digital Audio (AC3) (http://forum.doom9.org/showthread.php?s=&threadid=56020)

Dolby Technical Information (http://www.dolby.com/resources/tech_library/index.cfm)

Regards,
Steve.

[krbo]

Originally posted by ursamtl
In any case, SoftEncode does have the layout graphic to the left of each channel, so just check the channel position once you load the 6-channel file.

Alternatively, you could reroute the outputs between the V.I object and the Audio File Recorder for your own copy of V.I. The signals coming out of the V.I object are from left to right L,R,C,LFE,sL,sR.


it was easy to remap channels in Softencode, I was a little confused at first as surround channels had larger amplitude then front one.

For now I must examine one problem in reproduction, looks some channel are too boosted so clipping is audible.

Test material was from JMJarre, ripped goes straight to 0dB so if there are any amplification in VI module there will be problem.

Must demux bidule output to check each wave.

[krbo]

Originally posted by ursamtl
I've personally had good results from the default settings, but depending on your source file, tweaking the settings may produce a better result. You have to experiment. If you have a DVD player that accepts CDRW if you're burning Dolby Digital wave files for surround CDs, or DVD-RW if you're encoding DVD soundtracks, you can do lots of testing to get things right.



yeah, whole set of equipment is on disposal.Unfortunatelly my
Yamaha RXV-596 is a little picky, for now it can't recognize DD 5.1
from CD/RW (stays on ProLogic) , had to burn sample as DVD video with dummy video track to hear it properly.

But there is room for more test.
Only one trial so far - burned ac3 file as audio CD with CopyToDVD

[ursamtl]

Originally posted by krbo
it was easy to remap channels in Softencode, I was a little confused at first as surround channels had larger amplitude then front one.

For now I must examine one problem in reproduction, looks some channel are too boosted so clipping is audible.

Test material was from JMJarre, ripped goes straight to 0dB so if there are any amplification in VI module there will be problem.

Must demux bidule output to check each wave.

I tested the output levels fairly thoroughly before posting V.I plus I had a friend who owns a pro studio check it as well. I'll recheck the bidules that are available for download again on the weekend, but yours is the first report I've had of clipping out of about 20 people who've given me feedback on this version of V.I. Try it again with different material, and if the problem re-occurs, try adding a bidule Gain object between the file player and V.I object.

Steve.

[ursamtl]

Originally posted by krbo
yeah, whole set of equipment is on disposal.Unfortunatelly my
Yamaha RXV-596 is a little picky, for now it can't recognize DD 5.1
from CD/RW (stays on ProLogic) , had to burn sample as DVD video with dummy video track to hear it properly.

But there is room for more test.
Only one trial so far - burned ac3 file as audio CD with CopyToDVD

Are you running at 44.1kHz or 48kHz sampling? For CDRWs, the whole project has to be done at 44.1kHz, whereas for an audio track for a DVD video, the project has to be at 48kHz.

Steve.

[krbo]

Originally posted by krbo
For now I must examine one problem in reproduction, looks some channels are too boosted so clipping is audible.


confirming, this time used much softer material , gone through VI
and surround channels are too much amplified - a lot of clipping

soundforge data:

original:
max value -1.75dB
RMS power -21.04dB

SL from VI:
RMS power -18.31dB

[ursamtl]

I've edited the original V.I miniguide to reflect the new guide links.

If you're having any trouble getting V.I to work, try following the steps in 6-wave guide Shockwave animation of bidule-based process (http://www.freewebtown.com/andiz/audio/guides/first-6WAV-guide.exe) by Daphy (http://forum.doom9.org/member.php?s=&action=getinfo&userid=33004). For a detailed, step-by-step guide, see Step-by-Step Instructions for Plogue Bidule-based Stereo-to-Surround methods (http://forum.doom9.org/showthread.php?s=&threadid=83844).

As for the new version of V.I that I've mentioned in earlier posts, it's nearing completion and should soon be available.

Regards,
Steve.

[MaroonMike]

Steve,

What will be changed in the new V.I version? Just curious since I have a bunch of files I've encoded with V.I, but yet to play on the surround system (tying to do a final compare vs. a couple other methods.)

[ursamtl]

Hi Mike,

The basic V.I formula combination remains intact. Therefore, your files should be ok. Most of the changes are enhancements that help optimize the sounds or provide more flexible usage options. I found that the first version of V.I works really well with a good, well-recorded source file. However, if the source is less than optimal, the results aren't as good. I've found some ways to compensate for a variety of problems, so these will be part of the new release, thereby extending its usefulness

New features include: Multiple ambiance controls.
Dialog/vocal clarity.
Improving bass response without overload.
Compensating for overly wide or overly narrow source.
Better level management.
A couple other changes that increase flexibility.
I'm just working the kinks out of a couple of these and doing some final testing and then it'll be ready.

Regards,
Steve.

[MaroonMike]

I had a chance to test out the V.I bidule and it is good....nice work.

However, after testing a bunch of different methods lately, I still think the best sounding 2ch to 5.1 method is still the Preservation_5_Speaker_v2_REC (modified to reduced rear channel volume + a slight addition of reverb to the rears.)

Isn't V.I built using some of the same algorithms as Gerzon's 1997 ambisonic method? Just curious in the differences. Thanks.

[ursamtl]

Originally posted by MaroonMike
I had a chance to test out the V.I bidule and it is good....nice work.

However, after testing a bunch of different methods lately, I still think the best sounding 2ch to 5.1 method is still the Preservation_5_Speaker_v2_REC (modified to reduced rear channel volume + a slight addition of reverb to the rears.)

Isn't V.I built using some of the same algorithms as Gerzon's 1997 ambisonic method? Just curious in the differences. Thanks.

My initial conclusions on the Preservation_5_Speaker_v2_REC were very much the same as yours. I also reduced the rear channel and added a bit of reverb. However, I recently revisited it and although the results still sound great, I'm concerned about its "portability." What I mean by this is that if you listen to the front left and right speakers without the center, much of the bass is gone. Many if not most center speakers on big home theatre 5.1 systems are not full-range speakers, so chances are that the fidelity will be compromised. On the right system, it does sound very good. On my PC, it produces stunning sound, probably because I'm using a 4.1 speaker system and simply doing a phantom center, but when I play these back on my big 5.1 system, the stereo image seems strained, with a lot of sound coming from the center speaker but only ambience from the front left and right. V.I through the big system seems more balanced. The same goes for upmixes I've done with methods such as the two DPL, SurroundBoy's or Kpex's Upmix; these all produce a fairly balanced front image on both my systems. I'm planning to upgrade the PC soundcard and speakers to proper 5.1 by the end of the year, so I'll see again.

As for V.I, it's based on Ambisonic algorithms developed by Furse, Malham and Farina, with along with delay compensation suggested by Wendy Carlos. The initial design is fairly limited, but the new version is coming well. As I mentioned in my last message, I've come up with a couple of interesting controls. I fixed a final glitch with it last night, so after some final testing, I'll upload it, probably next week.

Thanks for the feedback.

Steve.