GUIDE: Stereo to 5-Channel Surround

[specise_8472]

About my Mono to Stereo conversion groups.

Had nothing but trouble with the new version of plogue.

So reverted back to the old version 6502. Works fine in this version. Will have to look into why. But I think all the fancy DSP preference settings etc are going to be nothing but trouble. So will stick with a known good version (6502).

[kempfand]

@ ursamtl: Thanks for sharing your bidule, I am keen to try it.

This was all done @ 44.1k, 16-bits. I haven't had time to experiment with higher levels. One thing I am curious about is that no one has mentioned dithering if you convert to 32 bits for processing and then write 16-bit files.

I am just looking at this topic too. There are various commercial options (such as Wave's IDR which uses the Gerzon approach, or Wave Art's FinalPlug).

There is also a nice free version from maxim digital audio (mda) (http://www.abel.net.uk/~maxim/vst/help/mdaplugs.htm) called mda dither (http://www.abel.net.uk/~maxim/vst/help/dither.htm) .

Cheers,

Andreas

[Eye of Horus]

Originally posted by specise_8472
About my Mono to Stereo conversion groups.

Had nothing but trouble with the new version of plogue.

So reverted back to the old version 6502. Works fine in this version. Will have to look into why. But I think all the fancy DSP preference settings etc are going to be nothing but trouble. So will stick with a known good version (6502).

My old version said : trial version ended, new version available for download.
I tried the new version for only a brief moment with some new ideas, but didn't find any problem...... (I must admit however that I didn't use the filerecorder.....)

EoH

[Tantulus]

Thanks Andreas for your help. I was starting to add some SIR to the SAD5.1 bidule and forgot about the dummy SIRs. You saved me some frustration.

Ursamtl: Can't wait to try your bidule!
Species_8472: Your's too. It will be a while though.

There are so many bidules out there that I am becoming overwhelmed. I've been so busy testing what you all construct that I haven't had time to work on some myself. Once again, I am idebted!

Regards,
Scott

[ursamtl]

Originally posted by Eye of Horus
When you convert in Soundforge from 16 bits to 32 bits, there is no dither option.
The 6 mono 32 bits files are directly used in Surcode.
So there is no dithering involved at all.......
Of course it would be different if you do the 32 to 16 bits in SF or CE !

No I meant dithering when converting back to 16-bit files. My understanding is that to do surround audio CDs, the files have to be 44.1khz and 16 bits. I know the 44.1 part is right because when I tried one at 48KHz, my DVD player just played digital noise. I haven't had time yet to try 32-bit files.

Good job and perhaps you can consider a newbie manual ?
Thanks. Actually I wouldn't mind but it will take some time as I'm very busy at work. Plus, I want to investigate this whole field more. I'm fascinated by the potential.

[Tantulus]

As promised, I just uploaded a zip file containing the bidule I created in an attempt to emulate the equations for 5.1 I found in the MatrixMixer DirectX filter available on SourceForge.net. I've also included a couple of screen grabs to show the basic layout I use (it includes the kX ASIO outputs, which you may not have)

IS KX ASIO necessary for your bidule and if so what does it do? I've downloaded the drivers but I'll wait for your respose before adding it.

Regards,
Scott

[ursamtl]

Originally posted by Tantulus
IS KX ASIO necessary for your bidule and if so what does it do? I've downloaded the drivers but I'll wait for your respose before adding it.

Regards,
Scott

No not at all. Sorry I should have made that clearer. As I mentioned in a previous post, I use the kX drivers with my Soundblaster Live. These are free drivers available at www.kxproject.com and they provide an incredible increase in usability over the standard Creative drivers. For instance, you can have 15 channels of ASIO with almost no latency.

The bidule I uploaded to the ftp site has no kX output in it.

[Eye of Horus]

Originally posted by ursamtl
[B]No I meant dithering when converting back to 16-bit files. My understanding is that to do surround audio CDs, the files have to be 44.1khz and 16 bits. I know the 44.1 part is right because when I tried one at 48KHz, my DVD player just played digital noise. I haven't had time yet to try 32-bit files.

But I did mention that :D !!
The 6 mono 32 bits files can be used in Surcode to create an audio CD. There is no need to convert them to 16 bits first ! Surcode itself handles the conversion back to 16 bits.

EoH

[Eye of Horus]

BTW under which name did you upload them ?
I cannot find the bidule on needful things........

EoH

[@ndy]

Originally posted by Eye of Horus
BTW under which name did you upload them ?
I cannot find the bidule on needful things........

EoH

Hello EoH,
I guess you wouldn`t find the *.zip file on needfulthings cause daphy is on holidays. So this site isn`t uptodate also unless the last provider killed our webspace.

You should find this file on my FTP- Server (ftp://daphy.mine.nu) in the incomming folder. The original Post from ursamtl is dated 23.05. and I didn`t moved any files ;)

BTW: I played with creating myself a website to have a second mirror of the FTP- Server which is really slow :(

I`ll post the url this evening when I`m finished

CU
@ndy

[ursamtl]

Originally posted by Eye of Horus
BTW under which name did you upload them ?
I cannot find the bidule on needful things........

EoH

I uploaded a zip file called MatrixMixer Emulator.zip to the incoming folder of daphy's ftp server.

[ursamtl]

Originally posted by Eye of Horus
But I did mention that :D !!
The 6 mono 32 bits files can be used in Surcode to create an audio CD. There is no need to convert them to 16 bits first ! Surcode itself handles the conversion back to 16 bits.

EoH

Ok, sorry I didn't see that. Does Surcode do any dithering or does it simply truncate the extra bits? The reading I've done on dithering says that simple truncation introduces digital artifacts and distortion.

If this is the case, then it would be better to convert back to 16 bites with dithering before loading the 6 files into Surcode.

I'll see what I can find out.

[ursamtl]

Originally posted by ursamtl
I uploaded a zip file called MatrixMixer Emulator.zip to the incoming folder of daphy's ftp server.

Since the ftp server doesn't seem to be active this morning, I've also thrown together a little web page on Yahoo. You can download the bidule there.

http://www.geocities.com/ursamtl/audio.html

[ursamtl]

By the way, for those of you looking for impulses to experiment with for your bidules, www.noisevault.com has just put up a new complete set of 44.1k, 32-bit impulses recorded from a Lexicon 960L! It's 38MB download, but it seems worth it.

These should fit in well with the discussion of 32-bit files we're having!

[kempfand]

@ursamtl: Just had a quick look at your bidule and MatrixMixer group. 2 questions regarding Kelly's Bass Manager which you use inside the MatrixMixer Group:

1) The manual says: This plug-in is to be used for monitoring only!
You need to turn off the bass manager before exporting your mix otherwise all the low-end will be removed from L, C, R, Ls, and Rs files, which is not what you want. The most common problem seems to be that people don't realize that the LFE and SUB are two different things Wouldn't that mean that your current bidule sould only be used for listening, but not for Audio File Recording ?


2) The manual says (and I can confirm this by a quick test I made): output assignment is L, R, C, SUB, Ls, Rs! I.e. Bass Manager's Pin-Out #4 is the SUB ... But: Your Matrix mixer assumes Pin-Out #4 as Surround Right (SUB on #6).

Not sure, but I think something is not correctly linked.

Kind regards,
Andreas

[@ndy]

Originally posted by ursamtl
Since the ftp server doesn't seem to be active this morning, I've also thrown together a little web page on Yahoo. You can download the bidule there.

http://www.geocities.com/ursamtl/audio.html

Finally I get along with a new mirror for my FTP- Server (ftp://daphy.mine.nu) :cool:

The new website (http://www.freewebtown.com/andiz/) is still under construction (panta rei (http://www.archaeonia.com/philosophy/presocratics/heraclitus.htm) ; for the greeks between us) but you have full access to every file you`ll find at my ftp- server.

Use my FTP- Server (ftp://daphy.mine.nu) to upload your files :D


Don`t forget: needfulthings (http://www.nedfulthings.webhop.org) is still alive ;)

[ursamtl]

Originally posted by kempfand
@ursamtl: Just had a quick look at your bidule and MatrixMixer group. 2 questions regarding Kelly's Bass Manager which you use inside the MatrixMixer Group:

1) The manual says: Wouldn't that mean that your current bidule sould only be used for listening, but not for Audio File Recording ?


kempfand, it depends what you want. I read the manual as well, and while I understand what the author is getting at, I want the low-end removed from the L, C, R, Ls, and Rs and sent through the SUB output. You'll notice I didn't connect anything to the LFE input because I want Bass Manager to trim the bass under the chosen crossover frequency from each channel, sum it and route it through the SUB output. In effect, I'm using this to generate an LFE channel. Obviously, people who don't want this should remove the bass manager altogether.

2) The manual says (and I can confirm this by a quick test I made): I.e. Bass Manager's Pin-Out #4 is the SUB ... But: Your Matrix mixer assumes Pin-Out #4 as Surround Right (SUB on #6).

Not sure, but I think something is not correctly linked.


Thanks for pointing this out. I did my bidule using the Bass Manager beta 0.3 that I downloaded from Needful Things as part of the zipped VST plugins. I see now that the latest version switches to the ITU standard of L, R, C, SUB, Ls, Rs. I did update to SIR_1005 from the old SIR_086 that was in the zip file but I didn't realize that Kelly had updated their VST. Just goes to show that it pays to keep an eye out for updates. :)

Thanks,
Ursa

[Tantulus]

I did download the Lexicon impules and did a quick test by adding SIRs to the SAD5.1 bidule of the one called ambience. Sounded great so far.

BTW what are the SFK files for?

Still planning to use the matrixemulator after I finish a two disc set I've already started. I plan to experiment with the Lexicon impulses in your bidule.

Regards,
Scott

[kempfand]

kempfand, it depends what you want. I read the manual as well, and while I understand what the author is getting at, I want the low-end removed from the L, C, R, Ls, and Rs and sent through the SUB output. You'll notice I didn't connect anything to the LFE input because I want Bass Manager to trim the bass under the chosen crossover frequency from each channel, sum it and route it through the SUB output. In effect, I'm using this to generate an LFE channel. Obviously, people who don't want this should remove the bass manager altogether. Thanks for the info. A small write-up such as this really helps, and I now see what you try to achieve :)

Will do some extensive tests coming long weekend here, with various types of music.

Kind regards,

Andreas

[ursamtl]

I've uploaded version 2 of MatrixMixerEmulator to the FTP server's incoming folder and also posted it at http://www.geocities.com/ursamtl.

This updates the entire signal path to L, R, C, LFE, Ls, Rs as per ITU standards. It's designed to work with Kelly Industries Bass Manager v0.5 and SIR 1.005.

I've also replaced the dummy SIR with a native Plogue delay line set to 8960 samples as per SIR 1.005 specs. I haven't tested it, but I imagine this will be less CPU and/or memory intensive than creating an extra instance of SIR just to compensate for the delay.

Thanks for the feedback.

Ursa

[Eye of Horus]

Originally posted by ursamtl
I've uploaded version 2 of MatrixMixerEmulator to the FTP server's incoming folder and also posted it at http://www.geocities.com/ursamtl.

This updates the entire signal path to L, R, C, LFE, Ls, Rs as per ITU standards. It's designed to work with Kelly Industries Bass Manager v0.5 and SIR 1.005.

I've also replaced the dummy SIR with a native Plogue delay line set to 8960 samples as per SIR 1.005 specs. I haven't tested it, but I imagine this will be less CPU and/or memory intensive than creating an extra instance of SIR just to compensate for the delay.

Thanks for the feedback.

Ursa

I did some testing today and I'm afraid I don't have good news.
The rears sounded terrible ! There was a definite delay, so bad that it sounded like echo !!
I will recheck all the settings, but I'm sure I got them right this morning.

In one of your previous mails here, (the long one) you made quite some comments about Ambisonics and that you want to achieve more separation. That's what we're working on too. SAD51inBidule gives at this moment better results than Matrixemulator !
Of course all IMHO, because I only tested with two tracks.

I will do some more tests ASAP to see if I can get rid of the echo in the rears. Perhaps it's just an overseen setting in your bidule or in the new Plogue version, but for now I was not impressed......

I take for granted that you tested it thoroughly, so maybe you can point me to what can be wrong ?

kind regards,

EoH

[ursamtl]

Originally posted by Eye of Horus
I did some testing today and I'm afraid I don't have good news.
The rears sounded terrible ! There was a definite delay, so bad that it sounded like echo !!
I will recheck all the settings, but I'm sure I got them right this morning.

It may be a problem with your levels or your SIR settings. The rear level should be low enough so that the sound appears to come from the front but if you shut off the rear, the whole soundfield collapses to more or less the original stereo field.

Also, make sure you have the center level high enough because what appears in the front left and right channels is basically what is unique to each channel. The center needs to be high enough to re-construct the original soundstage.

This also reveals the one downside with my approach in that one really needs to be able to monitor the mix in surround and not simply monitor everything to the two Microsoft Mapper channels as in your original guide. It's not necessary to have a proper 5.1 computer setup. I`m just using a 4.0 setup and mixing the center and LFE channels into the front. So far, the tests I've done when burning DTS CDs and playing them on my true 5.1 system have shown that I`m getting close with the monitoring.

As for ambisonics, I was playing around with a 3rd-order setup last night and while it still was terrible in terms of separation, it does some interesting things to the music, bringing out instruments that were buried in the stereo mix. I`m thinking of trying a combination of the MatrixMixer Emulator with the ambisonics.

By the way, have you tried the real MatrixMixer DirectX filter from SourceForge? I'm very impressed with its sound. If we could just fit it into a bidule! It provides exactly the kind of big, expansive sound I'm looking for in an upmix.

[specise_8472]

By the way, have you tried the real MatrixMixer DirectX filter from SourceForge? I'm very impressed with its sound. If we could just fit it into a bidule! It provides exactly the kind of big, expansive sound I'm looking for in an upmix. [/B]

My Ambiophonics methods are based upon this. I have kicking around somewhere in my backups, my VST version of MM. I originally was going the MM method, but found the needed paremeters for Ambiophonics. So went that way.
Don't forget that there are two versions of Ambiophonics. The version I use that takes WXYZ and computes speaker feeds.
Or the Stereo Dipole method. Which is why I created the X-talk VST. Which I am overhauling to produce binaural and transaural mixes and conversions.

Try my Allinone VST for rear channel seperation. It is 3rd order and computes the rears independant of each other. What I have done to really enhance the seperation all round is to use my x-talk impulses.

Just use three instances of SIR.
Left and Right Fronts using one of the front impulses.
Left and Right Surround using one of the Rear Impulses.
Center using the Zero impulse.

Really spreads out the sound image. And by feeding all signals through SIR's there is no delay.

At the moment, among other things, I am working on a better V3 ambiophonic mix using some enhanced methods to extract better channel seperation.

[ursamtl]

Originally posted by specise_8472
My Ambiophonics methods are based upon this. I have kicking around somewhere in my backups, my VST version of MM.

Hey, if you find it, please upload it! That would be very cool. By the way, MM is now part of the new AC3Filter release candidate. I noticed last night when playing with it that adding a 20ms delay to the left rear and 30ms to the right (in the filter's own delay pane) really gave me that big spread I hear in commercial 5.1 releases. The delay isn't really big enough to produce discreet echoes, but it does work well at giving the sound a sense of space. The 10ms difference between channels is based on the "TISDU" method used by traders of live mono recordings who use the delay to simulate stereo.

Try my Allinone VST for rear channel seperation. It is 3rd order and computes the rears independant of each other. What I have done to really enhance the seperation all round is to use my x-talk impulses.

Just use three instances of SIR.
Left and Right Fronts using one of the front impulses.
Left and Right Surround using one of the Rear Impulses.
Center using the Zero impulse.

Really spreads out the sound image. And by feeding all signals through SIR's there is no delay.

Actually I tried the AllinOne VST again last night and I'm afraid I had the same problem with it as with the ambisonics methods (no separation, rears sound like copies of the fronts), but I forgot about the impulses. I had read about them in the thread a few weeks ago and downloaded them but I forgot about them last night. I look forward to trying them with the Allinone.

By the way, instead of loading a SIR with no impulse, why not try what I did in v2 of my MM Emulator? Add a Plogue Delay Line and connect a Plogue Constant to its right pin entering the SIR delay value (8960 for v1.005, or 16384 for SIR 0.86). This does the same thing without having to load in an entire instance of SIR.

At the moment, among other things, I am working on a better V3 ambiophonic mix using some enhanced methods to extract better channel seperation.

Sounds cool. Getting the separation into it makes the sound come alive and gives the sense of realistic space.

[Eye of Horus]

Originally posted by ursamtl
Sounds cool. Getting the separation into it makes the sound come alive and gives the sense of realistic space.

The sense of realistic space ???????????

I have some old quad conversions here, where the level of separation is unbeatable. A guitar from FL, a singer from FR, drums from RL and another guitar from RR. What I miss is the band :D

Realistic space can only be achieved by Ambisonics and certainly not by separation. There is nothing realistic about separation at all, actually it will only be "realistic" when we have 5.1 ears !

IMHO separation is artificial !
Just simply because our ears are not build that way. We will hear from all directions at once, thus a left channel also influences what we hear with our right ear ! In other words, even if you can completely separate 5 sounds, there is still the fact that we will not hear it completely separated. So why this search for complete separation ?
Are we that much indoctrinated by Dolby ??

You mentioned the excellent results on DVD's. But...... are you comparing methods like Ambisonics with a music DVD or with a DVDA ?
Or were you talking about movie DVD's ?
I am just wondering because you will never get that separation from stereo to 5.1, whatever method ! You can come close, but not with the bidule you presented us ! I have tested it to it's limit this morning and it's not what I hoped it would be !
But I have to confess that I used a very difficult piece of music.
When I use DPL II, this happens : a guitar starts in the front right channel. The guitarplayer starts to sing. His voice is in the Center channel and the guitar remains in the right channel. The separation between C and R is perfect here with DPL.
Playing it in regular stereo you get the guitar from the right channel and the singer from both. And a very small bit of the guitar from the left, but you really have to play it vey loud to hear that.
With all methods I used on this piece of music, not one gave the same results. They just don't do a good job on this separation !
The only thing that came close was to use Gerzon LCR, and process the L=R from there and leave the Center as is.
All others made a mess of it ! Including yours.

For Ambisonics this piece is also difficult, but although no perfect separation, you still can sense where the singer comes from (Center) and where the guitar comes from (C+R). It's sounds more natural in Ambi because it's more fitted to our 2 ears. It makes a complete soundfield, which actually resembles what we hear with our 2 ears.

Also I would like to point out the zillion of music DVD's where the rears are nothing more than some weak reverb from the fronts. I even have this on a few DMP demo SACD's !
We can do better :-)

So there's still work to do !

kind regards,

EoH

[puzio]

Hello All

I tested ursamtl's method. The result is very bad. I think
E O H is right.

sorry for my english

[ursamtl]

Originally posted by Eye of Horus
The sense of realistic space ???????????

I have some old quad conversions here, where the level of separation is unbeatable. A guitar from FL, a singer from FR, drums from RL and another guitar from RR. What I miss is the band

Realistic space can only be achieved by Ambisonics and certainly not by separation. There is noth-ing realistic about separation at all, actually it will only be "realistic" when we have 5.1 ears !

Since everyone's reaction to sensory stimulation is subjective, there are no absolutes. What sounds "realistic" to one person sounds like crap to the next! Yes, in theory, Ambisonics looks to be a wonderful solution, but obviously some people don't agree or else this thread would be much shorter. Instead, it's full of alternative bidules, different approaches, changes, modifications, debates, etc.

All I'm saying is that I'm extremely disappointed with the results from Ambisonics so far. I came upon your guide quite by accident and read with enthusiasm its promise. I kept redoing the connections thinking surely I must be doing something wrong because the resulting 5.1 recording sounded small and boxy. I took Floyd's Comfortably Numb, Sarah Brightman's Captain Nemo, Marillion's Lavender, Zeppelin's Gallows Pole, Deep Purple's Lazy, Elton's Tiny Dancer, plus a couple of classical recordings, and tried over and over again to get some sense of the acoustic space you say Ambisonics gives. They all sounded like I remixed the tracks to mono and then routed the same mix to the rears. There was no sense of a realistic soundstage as if the real instruments were being played in front of me. The closest I achieved with Ambisonics was with the Surround configuration in Emigrator and only after tweaking the bpan controls (and obviously without the 36° rotation). This was on the classical guitar intro to Rush's A Farewell to Kings. It almost sounded as if the guitar were there right in front of me.

Once I took this same track and played it through the MatrixMixer filter, I got a much better sense of realism. I put together my bidule with a similar approach and with similar positive results. Whereas Comfortably Numb sounded really small and dull when rendered in Ambisonics, it became massive through my bidule. I dug out the remastered version of Deep Purple's Lazy and reveled with the sound and realism through my bidule. Blackmore's guitar sounded as if it were emanating from the amp in that Swiss hotel corridor so long ago, with gigantic Hammond groans coming from the right and sweeping through the room. In Ambisonics, this same song sounded like a 60's mono single mix (albeit with improved fidelity).


IMHO separation is artificial !
Just simply because our ears are not build that way. We will hear from all directions at once, thus a left channel also influences what we hear with our right ear ! In other words, even if you can completely separate 5 sounds, there is still the fact that we will not hear it completely separated. So why this search for complete separation ?
Are we that much indoctrinated by Dolby ??

I think you misunderstood what I meant by separation. I don't mean isolating instruments on one channel, either as a musical approach in the late 60s or as a sound effect in a modern movie soundtrack. I mean using all available channels to give a sense of the original sound of the instrument in an acoustic space, be it the original recording space or some other artificial space. You might complain that the rears of some music DVDs are nothing more than a weak reverb of the front, but if that's what it takes to make the sound seem to come from a stage in the front, then it's much more realistic than some artificial routing of sound to the back.

You mentioned the excellent results on DVD's. But...... are you comparing methods like Ambisonics with a music DVD or with a DVDA ?
Or were you talking about movie DVD's ?
I am just wondering because you will never get that separation from stereo to 5.1, whatever method ! You can come close, but not with the bidule you presented us ! I have tested it to it's limit this morning and it's not what I hoped it would be !

EoH, I never presented my bidule as the ultimate approach or even a "good" approach. If you check the first post I made to this thread, you'll see that I was extremely disappointed by the results of Ambisonics and I mentioned getting better results from a bidule I'd put together. Someone asked me to post it, so I cleaned it up a bit and did so. If you "hoped" it would be so good and if there is "still work to do" then obviously you see a deficiency with the Ambisonics approach. :)

Did it ever occur to you that perhaps you're expecting too much from Ambisonics? The theory seems good and probably sounds quite stunning if one is listening to material recorded with the proper miking, and correctly encoded. However, most music that was recorded for 2-channel stereo reproduction was not properly recorded for Ambisonic reproduction. Moreover, many modern recording studios record in rooms specially treated to add no coloration to the sound whatsoever. Therefore, any soundstage produced by its decoding method is going to be artificial. Perhaps Ambisonics does work wonders on a properly recorded sound that was recorded in an acoustic space capable of being reproduced, but even with proper recording, if an sound is recorded in a space that has little or no reflective capability--as is the case with many recording studios--there is no natural spatial information present in the recorded sound.

So there's still work to do !
Yes indeed. Obviously none of us has a perfect solution. Let's try to enjoy this experimenting with sound. I'm sure you didn't mean it, but the tone of your message seemed to me as if you felt any criticism of Ambisonics was almost criminal. Again, I'm sure this isn't intentional but even with some smiles, grins and "IMHO"s thrown in, it comes across as such. We're all just having fun exploring sound and technology and we all have valid contributions to make.

Regards,
Ursa

[kempfand]

@ puzio: I agree with a BUT: It's not easy to come up with a method that works well on many pieces of music. I still appreciate that ursamtl shared his method, which allows us others to learn.

Ambisonics / AmbioPhonics is truely exceptional and hard to beat :) Just to repeat that Ambisonic's goal is not to bring the orchestra into your living room, but bring you into the concert hall, so no need to look or search for single instruments coming from defined speakers.

For non-Ambisonics, I recently posted LCR Upmix Bidules (http://forum.doom9.org/showthread.php?threadid=70678). Amongsts these is the Gerzon LCR, but the others are also worth checking :sly: as they produce some reasonable LCR from LR.

Regards,
Andreas

[ursamtl]

Originally posted by puzio
Hello All

I tested ursamtl's method. The result is very bad. I think
E O H is right.

sorry for my english

I'm sorry to hear that. Did you check your levels and monitor them?Could you describe what was wrong with the result? If the rears echoed too much, try backing off on the reverb level in SIR. If the front seemed too wide and indistinct, try increasing the center channel. If the LFE was either too loud or too soft, try adjusting accordingly. With all of these methods, I've found that each piece of music requires its own adjustments. None of these will give you "plug and play" satisfaction, even the almighty Ambisonics!;)

Ursa

[kempfand]

All I'm saying is that I'm extremely disappointed with the results from Ambisonics so far.
...
I kept redoing the connections thinking surely I must be doing something wrong because the resulting 5.1 recording sounded small and boxy. Just keen: What speakers do you use for listening the DTS-coded end product ? Did you take a DTS CD and listened in a store over different speakers. Doesn't have to be $$$ ones, but outcome for Ambisonics can be very different. Most speakers these days are designed to sound good with Dolby stuff ...

Regards,
Andreas

[ursamtl]

Originally posted by kempfand
Ambisonics / AmbioPhonics is truely exceptional and hard to beat :)
Unfortunately, my experience so far is that everything else beats them, even the kpex method that you guys seem to ridicule so much!
Just to repeat that Ambisonic's goal is not to bring the orchestra into your living room, but bring you into the concert hall, so no need to look or search for single instruments coming from defined speakers.
But you see I don't get this result from either Ambisonics / AmbioPhonics (to be fair, I haven't properly tested the latter). I don't want single instruments coming from defined speakers, but neither do I want everything mashed into a single sound with no definition. To me these approaches so far have sounded like a mono mix piped through 5.1 speakers. There is no sense of any concert hall known to man!


For non-Ambisonics, I recently posted LCR Upmix Bidules (http://forum.doom9.org/showthread.php?threadid=70678). Amongsts these is the Gerzon LCR, but the others are also worth checking :sly: as they produce some reasonable LCR from LR.
Yes, I heartily agree. These produced a really nice front sound.

[Eye of Horus]

Originally posted by ursamtl
[B]Unfortunately, my experience so far is that everything else beats them, even the kpex method that you guys seem to ridicule so much!

English is not my native language, perhaps that's why I don't understand you.
We rewrote his method in a bidule and we haven't had any negative comments, except for one. We wouldn't have done that if we didn't see the potential of Kpex' method !!!!
So I cannot see where you get that statement from...... or I must misunderstand you !

EoH

[kempfand]

@ ursamtl / All: But you see I don't get this result from either Ambisonics / AmbioPhonics (to be fair, I haven't properly tested the latter). I don't want single instruments coming from defined speakers, but neither do I want everything mashed into a single sound with no definition You might want to try some real (i.e. 3D) B-Files. Some short demos are available:
- Soundfield B-Format (http://www.soundfield.co.uk/downloads/b-format.php)
- Farina's OSER-recordings (public) (http://www.ramsete.com/Public/OSER-recordings/Reharsal-Room/B-format/)

You might have to resample to 32/44 or 16/44 (using CoolEdit or similar), but you then can use Gerzonic's B-Player (http://www.gerzonic.net/?prod=1&rub=2) and decode using Emigrator as Pentagon or Square.

This will give you an idea of what you are can expect from Ambisonics. Sure these samples have been properly recorded (Soundfield Mic), but you can get similar (good) results from Stereo (but 2D only).

Good luck,

Andreas

[ursamtl]

Originally posted by kempfand
Just keen: What speakers do you use for listening the DTS-coded end product ? Did you take a DTS CD and listened in a store over different speakers. Doesn't have to be $$$ ones, but outcome for Ambisonics can be very different. Most speakers these days are designed to sound good with Dolby stuff ...

Regards,
Andreas

My speaker system sounds quite good. It's a combination Infinity/Polk Audio system. While it's not state-of-the-art, it provides a nice balanced sound with crisp, transparent highs, a smooth midrange and tight, solid bottom end. My PC speaker system is also nicely balanced for music without the overwhelming super subwoofer sound that's so prevalent for gaming fans.

Your point is valid but only somewhat. After all, if my system were skewing my impressions of the different bidule results, it would do likewise to my impressions of movie soundtracks, DVD-As, etc. I don't get what you and a couple of others mean when you talk about the "Dolby stuff." Do you mean movie sound effects? If so, then that's not at all what I'm aiming for. I want something like Kevin Shirley's wondeful mix on the Led Zeppelin DVD last year. He mixed the surround to sound as if the listener were sitting about five rows back from the front center of a Led Zeppelin show. The rears mainly provide the ambience of the hall and the audience and serve to reinforce the front, which is presented as an accurate soundstage of the band from the vantage point I just mentioned. If this were done with the Ambisonics bidule as I've heard them so far, the entire band would be cramped into a thick middle that's repeated in the rears. It would sound something like the old Motown remasters that have been released in mono. They might sound punchier than a stereo mix, but only 1 speaker is necessary for such reproduction and 4.1 speakers are being completely wasted on such material.

I'll retry some Ambiophonics testing this evening to see if the results are better, but certainly Ambisonics by itself seems to be an utter disappointment.

Ursa

[ursamtl]

Originally posted by kempfand
@ ursamtl / All: You might want to try some real (i.e. 3D) B-Files. Some short demos are available:
- Soundfield B-Format (http://www.soundfield.co.uk/downloads/b-format.php)
- Farina's OSER-recordings (public) (http://www.ramsete.com/Public/OSER-recordings/Reharsal-Room/B-format/)

You might have to resample to 32/44 or 16/44 (using CoolEdit or similar), but you then can use Gerzonic's B-Player (http://www.gerzonic.net/?prod=1&rub=2) and decode using Emigrator as Pentagon or Square.

This will give you an idea of what you are can expect from Ambisonics. Sure these samples have been properly recorded (Soundfield Mic), but you can get similar (good) results from Stereo (but 2D only).

Good luck,

Andreas

Thanks for the links. I tried a couple of these. While they certainly are more impressive than the results from the ambisonics bidules, I don't see them lending a convincing sound to a 5.1 mix of music. The vantage point would be somewhat like sitting on stage in the middle of a performance instead of sitting in the audience of a concert hall. I still got no sense of an acoustic space.

I'll try some more tests tomorrow. I have some domestic stuff to take care of now.

[Eye of Horus]

Originally posted by ursamtl
My speaker system sounds quite good. It's a combination Infinity/Polk Audio system. While it's not state-of-the-art, it provides a nice balanced sound with crisp, transparent highs, a smooth midrange and tight, solid bottom end. My PC speaker system is also nicely balanced for music without the overwhelming super subwoofer sound that's so prevalent for gaming fans.

Your point is valid but only somewhat. After all, if my system were skewing my impressions of the different bidule results, it would do likewise to my impressions of movie soundtracks, DVD-As, etc. I don't get what you and a couple of others mean when you talk about the "Dolby stuff." Do you mean movie sound effects? If so, then that's not at all what I'm aiming for. I want something like Kevin Shirley's wondeful mix on the Led Zeppelin DVD last year. He mixed the surround to sound as if the listener were sitting about five rows back from the front center of a Led Zeppelin show. The rears mainly provide the ambience of the hall and the audience and serve to reinforce the front, which is presented as an accurate soundstage of the band from the vantage point I just mentioned. If this were done with the Ambisonics bidule as I've heard them so far, the entire band would be cramped into a thick middle that's repeated in the rears. It would sound something like the old Motown remasters that have been released in mono. They might sound punchier than a stereo mix, but only 1 speaker is necessary for such reproduction and 4.1 speakers are being completely wasted on such material.

I'll retry some Ambiophonics testing this evening to see if the results are better, but certainly Ambisonics by itself seems to be an utter disappointment.

Ursa

A few IMHO's wouldn't hurt !!

Ever thought of that it can be yourself ???

I have had friends listening here to the Ambi mix of War of the Worlds. 6 out of 7 : I didn't know this was available in surround.
The 7th reacted like you : I don't hear it !

OK, as you probably know : I did a listening test. I took 10 stereo songs with all kind of styles of music. I converted these 10 songs with 10 different methods to surround. I made a poll online where people could fill in which version they liked the most. I didn't say before the poll what the used methods were.
End result : 70 % preferred the conversions with Ambisonics !

So obviously a. you're doing something wrong (BTW did you convert the stereo first to 32 bits ?) or b. you just don't (want to ?) hear it.

But keeping on saying how thrash Ambisonics is without adding a few "IMHO" gives me indeed the impression you're behaving like a troll. Eespecially since your own published bidule didn't give 3 people a good result. And believe me : if it was better than anyting already published, people sure would let it know here !

So come up with something better (your MM bidule was definitely worse). If not, than please stop all this whining and groaning !
Thanks.

EoH

[ursamtl]

Originally posted by Eye of Horus
A few IMHO's wouldn't hurt !!

Ever thought of that it can be yourself ???

I have had friends listening here to the Ambi mix of War of the Worlds. 6 out of 7 : I didn't know this was available in surround.
The 7th reacted like you : I don't hear it !

OK, as you probably know : I did a listening test. I took 10 stereo songs with all kind of styles of music. I converted these 10 songs with 10 different methods to surround. I made a poll online where people could fill in which version they liked the most. I didn't say before the poll what the used methods were.
End result : 70 % preferred the conversions with Ambisonics !

So obviously a. you're doing something wrong (BTW did you convert the stereo first to 32 bits ?) or b. you just don't (want to ?) hear it.

But keeping on saying how thrash Ambisonics is without adding a few "IMHO" gives me indeed the impression you're behaving like a troll. Eespecially since your own published bidule didn't give 3 people a good result. And believe me : if it was better than anyting already published, people sure would let it know here !

So come up with something better (your MM bidule was definitely worse). If not, than please stop all this whining and groaning !
Thanks.

EoH

EoH, I don't care if your first language isn't English, the tone of this message is extremely offensive. I have never attacked you personally, I've simply expressed disappointment with the method of 5.1 upmixing that you advocate. Your response is to attack me personally and characterize my comments as "whining and groaning." This is extremely childish.

I repeat what I said in my earlier post, if Ambisonics is so great, then why are there so many alternatives presented in this thread? Why did you yourself say there's lots of work to do? If Ambisonics is the great method you pretend, then surely your work is done. Why did you say you were working on ways to improve separation? Surely if Ambisonics were so wonderful, then there would be no need to "improve" anything!

Obviously, if my bidule is so bad, then so are the AC3Filter and Matrix Mixer on which it is based. Those seem to be quite popular. As well, my bidule's front generation is a greatly simplified version of the LCR Gerzon method that kempfand posted. His method is actually much better than mine in generating a solid soundstage (and far better than encoding traditional stereo signals with Ambisonics), but I didn't include it in my bidule because I wanted to honestly present it as what I said it was, an interpretation of the MatrixMixer DirectX filter. I repeat what I said in an earlier post: I never said it was the answer, just that I got better results for my listening taste with it than with your method. Someone asked me to post it, so I did. You seem to treat this whole thread as if it were some kind of competition and if someone comes along and criticizes Ambisonics, you pounce on them as if they were attacking you personally.

I don't think this is the first time I've seen you berate others in threads on the Doom9 forum. It's extremely unpleasant to read and it certainly adds nothing to the spirit of community present here. Please restrict yourself to a discussion of the methods and stop the name calling and personal attacks.

[Eye of Horus]

Originally posted by ursamtl
EoH, I don't care if your first language isn't English, the tone of this message is extremely offensive. I have never attacked you personally, I've simply expressed disappointment with the method of 5.1 upmixing that you advocate. Your response is to attack me personally and characterize my comments as "whining and groaning." This is extremely childish.

When you write in almost every post how bad the results are from Ambisonics and seldom add IMHO, yes, this is whining and groaning.
This has nothing to do with a personal attack, but simply with the fact that you are constantly saying how bad Ambisonics is for you in this originally Ambisonics thread. I consider your comments about Ambisonics offensive, because 7 out of 10 have another opinion, so you are insulting 7 out of 10 !


I repeat what I said in my earlier post, if Ambisonics is so great, then why are there so many alternatives presented in this thread? Why did you yourself say there's lots of work to do? If Ambisonics is the great method you pretend, then surely your work is done. Why did you say you were working on ways to improve separation? Surely if Ambisonics were so wonderful, then there would be no need to "improve" anything!


Talking about childish behaviour ! Stereo was so damn good, why search for anything better ??
All those remarks like "Ambisonics is so wonderful" in the above and other contexts, are suggestive and I think you meant it that way. So who is offensive here ?


Obviously, if my bidule is so bad, then so are the AC3Filter and Matrix Mixer on which it is based. Those seem to be quite popular. As well, my bidule's front generation is a greatly simplified version of the LCR Gerzon method that kempfand posted. His method is actually much better than mine in generating a solid soundstage (and far better than encoding traditional stereo signals with Ambisonics), but I didn't include it in my bidule because I wanted to honestly present it as what I said it was, an interpretation of the MatrixMixer DirectX filter. I repeat what I said in an earlier post: I never said it was the answer, just that I got better results for my listening taste with it than with your method. Someone asked me to post it, so I did. You seem to treat this whole thread as if it were some kind of competition and if someone comes along and criticizes Ambisonics, you pounce on them as if they were attacking you personally.

There is a difference between criticizing and putting down !
I can take good criticism without any problem. Otherwise Kempfand and me wouldn't have added so many extensions to the original bidule. I never criticized others bidules too. But...... when presenting your bidule with the attitude you spread out on Ambisonics, is IMHO not the way to discuss things here.
You can present your bidule without that constant nagging about Ambi !! Thanks !


I don't think this is the first time I've seen you berate others in threads on the Doom9 forum. It's extremely unpleasant to read and it certainly adds nothing to the spirit of community present here. Please restrict yourself to a discussion of the methods and stop the name calling and personal attacks.

You said something before about us ridiculizing Kpex methods ?
Now you make another suggestive remark.
Come with the facts !

The only one with his constant nagging about Ambisonics and who seems not to get the spirit of this forum is not me : it's you !
When I reacted to your remark about Kpex, you didn't come back with a "sorry". Not that I am waiting for that, but you set the tone ! Not me, I only react.

You stated clearly that your method gave you a better result. The first one you posted was completely wrong. The second should be fine. Well perhaps for your ears, but till now I have heard 3 negative experiences and not one positive one (except your own of course). When I say something about that you feel attacked ?
I would consider it a remark to stimulize you to come up with something working :)

Till now I sense more pretentions than inventions !
You put down Ambi, every time, you would come up with something better, but didn't.

On this forum we respect everyone's effort, but not by first putting down on the hard work of others first !

EoH

[ursamtl]

Originally posted by Eye of Horus

When you write in almost every post how bad the results are from Ambisonics and seldom add IMHO, yes, this is whining and groaning.

EoH, I've been involved with public forum messaging for well over a decade in many contexts, and most people I know consider using "IMHO" on a regular basis to be a sign of arrogance, as if those who use it really considers their opinion to be superior to opposing ones but want to make themselves appear more humble and avoid criticism for being arrogant. NOTE: I am not accusing you of this, just telling you why I avoid using such acronyms because, in my experience, they color the message. It is of course understood that if someone is expressing an idea on a forum and does not supply some sort of reference to another source or some sort of factual data as proof then he or she is obviously expressing his or her opinion, be it humble or otherwise. :)


This has nothing to do with a personal attack, but simply with the fact that you are constantly saying how bad Ambisonics is for you in this originally Ambisonics thread. I consider your comments about Ambisonics offensive, because 7 out of 10 have another opinion, so you are insulting 7 out of 10 !
This statement proves what I've been saying. If you consider criticism of the Ambisonics method to be "offensive" then you are taking it personally, as an attack on you as a person. I have never intended this! When I discovered your thread, I read it with a great deal of enthusiasm, as if I'd found some sort of holy grail. I wanted Ambisonics to work, but it didn't work for me. The more I read your comments about how superior it was, the more I tried redoing it to see if I'd overlooked anything. Eventually I came to the realization that it represents one approach to surround sound and not the ultimate solution. It is an approach or definition of surround that I personally find disappointing. I like surround sound to reveal something new and hidden in the music, some nuance that stereo couldn't deliver. To my ears, Ambisonics sounded like a collapsed stereo image across the fronts with the same image repeated in the rear speakers. Yes, there is a slight bit of separation, but it sounds to me like someone took a stereo signal and collapsed it to only about 20% of it's original width. I'm sure on a good sound system in a store, it sounds nice, but there's no magic. When I experimented with my bidule, the Gerzon LCR bidules, the AC3filter/MatrixMixer, and simply playing with the many stereo widener plugins, etc., out there, I discovered new sounds within my favorite music. Have you ever tried some of the QSound DirectX plugins? They take instruments that seemed buried in a mix and present them in a new light. Take a simple old Jimi Hendrix recording and run it through one of these and you get some stunning revelatory results. I believe it was you who said the Circle Surround II demo produced the best upmix surround you've yet heard (or perhaps kempfand, I don't have time to go searching through the forum to see who said it). These methods produce what I'm looking for, Ambisonics does not. If you take that statement as an attack on you or consider it offensive, then that's your problem, not mine. Do not make it mine by calling me a "troll" or dismiss my comments as "whining and groaning."


You said something before about us ridiculizing Kpex methods ?
Now you make another suggestive remark.
Come with the facts !

The only one with his constant nagging about Ambisonics and who seems not to get the spirit of this forum is not me : it's you !
When I reacted to your remark about Kpex, you didn't come back with a "sorry". Not that I am waiting for that, but you set the tone ! Not me, I only react.

Look, I don't have the time or desire to go searching back through every thread I've read over the last couple of months to quote examples of where I felt you were berating people. All I know is that I got that impression. If I'm wrong, I apologize, but that was how it seemed to me. If I remember correctly, you implied that kpex's SAD5.1 method was sort of a lowest possible benchmark, when you said that even it was better than mine. I seem to recall reading other comments where you implied it wasn't very good. Given my disappointing experiences with your Ambisonics method, can you not understand why I felt it unfair that you would berate someone else's method?

You stated clearly that your method gave you a better result. The first one you posted was completely wrong.

No, this statement is completely wrong! I already explained to you that my first bidule worked perfectly on the software I downloaded from needfulthings. The version of Kelly Industries BassManager in the needfulthings zipped file was 0.3. Check the Kelly Industries web site. It states very clearly that they changed the pinouts to the ITU standard in their v0.5. Still, they haven't changed their interface to match.

The second should be fine. Well perhaps for your ears, but till now I have heard 3 negative experiences and not one positive one (except your own of course). When I say something about that you feel attacked ? I would consider it a remark to stimulize you to come up with something working

Yes it did sound fine to my ears. I've also explained some possible reasons why there have been two other negative responses, the lack of adequate monitoring to achieve the levels required being the most probable. A professional recording studio monitors the mix of a recording while it is being mixed and using some sort of reproduction matching the mixing format. Look at section 0.8 your ambisonic guide. It suggests monitoring the front and rear lefts through the same left channel and the front and rear rights through the right channel. All that can give is a notion that the sound is present and at a certain loudness level. It cannot possibly give any sense of a surround mix. I have already said that the bidule I presented requires at least a 4.0 setup for monitoring to be effective. Every individual song requires such monitoring to bring out the best combination of factors. Perhaps a classical recording or a live performance recording could be set once and then left for the duration of the entire recording, but a modern studio album of popular music is full of many variations and nuances that require different approaches. For example, if the vocals were recorded with a lot of stereo reverb, this will dominate the rear dry level in my bidule and thus the wet level in SIR should be brought down. If the rears are without much reverb then the opposite is true. If the bass is tight and punchy already, then the LFE level needs to be brought down or it will overwhelm the final mix. If the original stereo mix was quite wide, the center level needs to be higher to avoid a washed out sound. If it was overly narrow, then bringing the front lefts and rights up with provide a nicer sense of a soundstage. But, if the fronts are too high, then a good vocal performance will lose its distinctiveness. I prefer this kind of approach because it allows me to tailor my mix to the music. Even if I put a separate gain on each channel in your Ambisonics bidule (which I've done), I don't have this kind of control to tailor the final mix. It just changes the balance of the overall sound the same way a balance control on a traditional stereo system works.

Till now I sense more pretentions than inventions !
You put down Ambi, every time, you would come up with something better, but didn't.
Surely if you read the detailed description of mixing levels I just wrote with any sense of objectivity, you must see that I didn't just throw my bidule together for fun, I did put some thought into it. You may not agree with the methodology or the results, but you cannot dismiss it as "pretention." That's an arrogant word that I consider offensive. No I don't consider my bidule an "invention" by any means and I do give you credit for presenting your application of Ambisonics as an approach to upmixing. I do not agree with you when you make a sweeping, absolute statment such as "Realistic space can only be achieved by Ambisonics and certainly not by separation." Unfortunately, I get the sense that just because I disagree with this, you feel I'm some sort of competitor or enemy of yours and so you dismiss me as a "acting like a troll" or "whining and groaning." EoH, if you feel offended that I criticized your Ambisonics method, I'M SORRY. Now, get over it!

[kempfand]

Thanks for the links. I tried a couple of these. While they certainly are more impressive than the results from the ambisonics bidules, I don't see them lending a convincing sound to a 5.1 mix of music. The vantage point would be somewhat like sitting on stage in the middle of a performance instead of sitting in the audience of a concert hall. I still got no sense of an acoustic space.
ursamtl: I have no problems that you are personally disappointed with Ambisonics/Ambiophonics. But for the rest of the readers, I would like to make a few (also subjective) comments from my side.

1) I think you (and others) are "victims" of what you (others) are used to in terms of 5.1 music. Note that this is not meant as a 'personal attack' or criticism. It's just how I see things.

I think the discussion in the previous few posts of the threat are not so much about "Ambisonic or Other Method". It's more about how the recording was done (mics used etc.) and how it was mixed before it came to the stores.

In detail: We are used to what we are used to, and I think the enormous (marketing) power of commercial recording companies and patent holders (e.g. Dolby & DTS just to name two) is often not considered enough as influencing factor on our behaviours. After all, if we are made to believe that what we buy is good, this means $$$ for the seller.

2) The same applies to how music is recorded and mixed before it ends on 5.1 DVD's or DTS CD's. Most 'modern' "Tonmeisters" (German expression for "mixing masters") have their style, which is often similar, i.e. music soundfield in the Fronts (incl. C), Ambience in the Surrounds. That's it. Full stop. In the worst case, there is but some echo and some audience applauding in the Surrounds. Horrible for my taste ...

More so: Professional listening tests (Verdie -> Listening Test (http://www.stud.tu-ilmenau.de/~proverdi/indexen.html)) raised the criticism, that Ambisonics gives too much room sound, and that isn't what the "Tonmeisters" want (even if it is correct!). It is also so that most people are so used to Stereo so hearing to much of the room won't sell (even if it is right !).

3) I agree with your point about "coloration" with Ambisonics, but this can be avoided by doing it right. Note that the most of the Ambisonics methods here (such as the guide at the beginning of this threat) are suited for the masses. Real Ambisonics uses signed filters (real-world impulse repsonses) to creat the B-format, as well as creating the decode to xyz speakers. See Conversion between UHJ and B-format (http://www.ramsete.com/aurora/conversion_between_uhj_and_b.htm#Audiomulch) for an outline on howto use signed filters for the B-format.

4) The vantage point would be somewhat like sitting on stage in the middle of a performance instead of sitting in the audience of a concert hall. That's how Ambisonics works. It brings you to the concert hall (restricted by how it was recorded and mixed, as outlined above). If you want to experience different positions in the audience, the only way to go is to use Impulse Responses (such as the ones from Noisevault) you mentioned and used.

Regards,
Andreas

[ursamtl]

Originally posted by kempfand
1) I think you (and others) are "victims" of what you (others) are used to in terms of 5.1 music. Note that this is not meant as a 'personal attack' or criticism. It's just how I see things.

No, I don't see this as a personal attack at all. It's a valid point, but.... what if we like being victims? In other words, mean what if we like the sound we are used to from 5.1 music? Give us some credit for being discerning listeners. When I first picked up my 5.1 system and started listening critically to the sound on the DVDs I rented or bought, I was astounded at the fidelity and clarity of reproduction. I'm not some newbie. I've been an audio fanatic since the mid-70s and have heard some pretty fine systems over the past 30 years! Properly mixed and presented 5.1 sound can be beautiful, providing a dimension and depth I've never heard otherwise. It's like the difference between looking out a small window at a lush garden and actually stepping out into the midst of it. I can't speak for anyone else, but if I'm a victim then I am a damned happy one!! :)

I think the discussion in the previous few posts of the threat are not so much about "Ambisonic or Other Method". It's more about how the recording was done (mics used etc.) and how it was mixed before it came to the stores.

This leads to the point I made earlier to EoH. Ambisonics might be an incredible approach given the right recordings but to take tracks that were not recorded properly for Ambisonics and attempt to encode and decode them as such doesn't provide the same effect. Call it Ambisonics for the masses if you like, but even following the instructions for what you call Real Ambisonics with signed filters, I found the result lacking life, or depth. It was certainly better than the original guide here, but the effect produced sounded similar to others I've heard where some work is done on phases at certain frequency. It didn't give me that "stepping into the garden" feeling I mentioned above.

Don't take my word for it, check out Ralph Glasgal's site (which you mentioned yourself when you presented your Ambiophonics bidule):

from Ambiophonics by Ralph Glasgal at http://www.ambiophonics.org/2ndEd_Preface.htm
There are two basic theoretical technologies that are prime candidates to replace stereo in elite high-end music reproduction, where mass marketing and complex technical concepts should not be (but of course are) major stumbling blocks. One is the wavefront reconstruction method often employing hundreds of microphones and speaker walls or, where recording is involved, Ambisonics. The Ambisonic wavefront reconstruction method generates the correct sound pressure and sound direction in a region that at least encompasses one listener's head. The other is the binaural technology method that more directly duplicates the live experience, independently, at each ear. Of course, both tehnologies aim to deliver to the entrance of your ear canal an accurate replica of the original sound field. The Ambisonic method does have the advantage that it can reproduce direct sound sources from any angle and so is quite well suited to non-concert events or movies. But since the Ambisonic wavefront reconstruction method requires a special microphone, a minimum of three (or better four recording) channels and a very complex decoder, is not as accurate as the binaural technology methods, and does nothing for the existing library of LPs and CDs it will not be considered further here.

I have yet to explore his approach to the point where I can form a valid opinion on its effectiveness for upmixing 2.0->5.1, but Glasgal obviously knows his stuff.

In detail: We are used to what we are used to, and I think the enormous (marketing) power of commercial recording companies and patent holders (e.g. Dolby & DTS just to name two) is often not considered enough as influencing factor on our behaviours. After all, if we are made to believe that what we buy is good, this means $$$ for the seller.

2) The same applies to how music is recorded and mixed before it ends on 5.1 DVD's or DTS CD's. Most 'modern' "Tonmeisters" (German expression for "mixing masters") have their style, which is often similar, i.e. music soundfield in the Fronts (incl. C), Ambience in the Surrounds. That's it. Full stop. In the worst case, there is but some echo and some audience applauding in the Surrounds. Horrible for my taste ...

That last word says it all, this is a matter of taste. We all have a right to our individual taste. I personally am looking for the big, multi-dimensional pristine sound I hear on 5.1 mixes. I like the soundstage in front of me. Since I play guitar, I know what it's like to be on stage in the midst of a band in full flight. That has its appeal and its unique sound as well. I don't want that kind of surround sound on my home system when I sit down to appreciate a piece of music.

4) That's how Ambisonics works. It brings you to the concert hall (restricted by how it was recorded and mixed, as outlined above). If you want to experience different positions in the audience, the only way to go is to use Impulse Responses (such as the ones from Noisevault) you mentioned and used.

Actually, that brings me to another point that's only somewhat related. While I'm enthralled by the quality of reverb impulse responses give as well as the ability to cpature the sound of vintage amps, mics, etc., I have yet to discover convincing impulses for giving me that "being there" feeling. For example, at work I can listen to music through a nice set of headphones, but the plugins or impulses suggested when I search on the net for "out of head" sounds just don't seem to work. In theory, with the right impulses I should be able to sit there and have the music sound as if I had a PA sitting in front of me, but it hasn't worked so far. But, I digress...

Anyway, time to do some more testing.

Regards

[pelmen]

first off a big thanks to all the people here who have worked hard on this whole 2.0 to 5.1 issue. i've been playing around with several of the posted bidules and have settled on the SAD5.1 for now as being the better of the bunch for my tastes.

1)
i'm certainly no expert when it comes to audio though but i was wondering if it was possible somehow to integrate a way of making use of dolby surround (and i think pro logic) if it exists in a 2.0 stream? from my understanding both dolby surround and prologic were multichannel signals mixed into a 2.0 format and re-expanded by the amp at the playback stage. i have a dolby surround test CD which i use to test the bidules (mostly using the channel identifier to make sure i ended up with all 6 channels going to the correct speakers :) and by far the result after processing isn't as good as the original. so i don't know if its even possible or not to detect if any such surround encoding is already existing in the 2.0 WAV file and once identified be able to make use of that knowledge to get an even cleaner 5.1 seperation?

2)
also it seems the main use for people in this thread is for turning music into 5.1. do people feel the same bidules are just as good for tv/movie audio or would a bidule layout specific for non-music sources provide a better end result?

3)
my setup is a computer with sound audigy and cambridge soundworks 5.1 amp/speakers and i connect my dvd players audio to the same amp. with the audigy card it plays 2.0 sources to all 5.1 channels already (but i'd like to put things onto DVDR so i can do other work with my computer). does anyone know how the bidule approach compares to the audigy? or even making use of the audigy hardware somehow if it is better to do hardware processing into 6 channel wav files?

4)
ideally what is the end goal people here are aiming to achieve? complete sound seperation with for example voices staying in the centre channel instead of "leakage" into all channels.

its a great thread so far, the only problems i've ever come across have been related to 2/4GB filesize limitations with various apps. i'm certainly happy with the current results when compared to boring old stereo :) but i hope it'll be possible to improve and get results closer to proper 5.1 source results :)

thx

[Eye of Horus]

Originally posted by ursamtl
EoH, I've been involved with public forum messaging for well over a decade in many contexts, and most people I know consider using "IMHO" on a regular basis to be a sign of arrogance, as if those who use it really considers their opinion to be superior to opposing ones but want to make themselves appear more humble and avoid criticism for being arrogant. NOTE: I am not accusing you of this, just telling you why I avoid using such acronyms because, in my experience, they color the message. It is of course understood that if someone is expressing an idea on a forum and does not supply some sort of reference to another source or some sort of factual data as proof then he or she is obviously expressing his or her opinion, be it humble or otherwise. :)

Although I can understand your point here, you should be aware that this is an international forum ! Every now and then making clear that it is your personal opinion, doesn't harm, but would make it much more clear to users who don't have English as native language.
TO ME it sounded as if you condemn the whole Ambisonics thing and when the majority of users is very satisfied with the results, I feel like you're nagging. A bit more IMHO would lead ME to the conclusion : hey, for this guy Ambi doesn't seem the solution !, instead of : hey, this guy is saying Ambi is thrash !
These are only nuances, which make it easier for non English speakers. I hope you understand that !


This statement proves what I've been saying. If you consider criticism of the Ambisonics method to be "offensive" then you are taking it personally, as an attack on you as a person. I have never intended this! When I discovered your thread, I read it with a great deal of enthusiasm, as if I'd found some sort of holy grail. I wanted Ambisonics to work, but it didn't work for me. The more I read your comments about how superior it was, the more I tried redoing it to see if I'd overlooked anything.


You must take into consideration that : a. it's more than a year ago ! and b. we were the first to use AND publish this way of using Bidule and Ambisonics ! Did you ever try the very first method of Ambi with Cooledit and the Aurora plugins ? The thread is still on this forum. Perhaps you can get better results with that approach....


Eventually I came to the realization that it represents one approach to surround sound and not the ultimate solution.


Again : things develop ! The first method lead to the Bidule method and that method lead to ALL the others ! It's the ease of Bidule that you can experiment so easy with all kinds of filters and layouts. More and more I get the impression you reacted to the first method only, without taking the time ago it was made into your opinions. For instance : you could have said the same about 78 rpm mono recordings :-) It's not very realistic to have a go at Ambi only, based on old material. We improve where we can. (actually our latest development, which is almost ready for publishing will be more for your taste !)


It is an approach or definition of surround that I personally find disappointing. I like surround sound to reveal something new and hidden in the music, some nuance that stereo couldn't deliver. To my ears, Ambisonics sounded like a collapsed stereo image across the fronts with the same image repeated in the rear speakers. Yes, there is a slight bit of separation, but it sounds to me like someone took a stereo signal and collapsed it to only about 20% of it's original width. I'm sure on a good sound system in a store, it sounds nice, but there's no magic. When I experimented with my bidule, the Gerzon LCR bidules, the AC3filter/MatrixMixer, and simply playing with the many stereo widener plugins, etc., out there, I discovered new sounds within my favorite music. Have you ever tried some of the QSound DirectX plugins? They take instruments that seemed buried in a mix and present them in a new light.


I have the same experience with the Ambi mix of War of the Worlds !!


Take a simple old Jimi Hendrix recording and run it through one of these and you get some stunning revelatory results. I believe it was you who said the Circle Surround II demo produced the best upmix surround you've yet heard (or perhaps kempfand, I don't have time to go searching through the forum to see who said it). These methods produce what I'm looking for, Ambisonics does not. If you take that statement as an attack on you or consider it offensive, then that's your problem, not mine. Do not make it mine by calling me a "troll" or dismiss my comments as "whining and groaning."


Understood, but now you explain. You didn't in one of your first posts ! I said this before : a simple : "Ambisonics doesn't produce what I'm looking for....." would have been enough.
Hey, even we have records where we have the same opinion !! That's why we are still developing new methods. I said it many times before and also in this thread : use the method that gives YOU the best results. Listen and do not just base your opinions on theory !
I never said it was the ultimate method !



Look, I don't have the time or desire to go searching back through every thread I've read over the last couple of months to quote examples of where I felt you were berating people. All I know is that I got that impression. If I'm wrong, I apologize, but that was how it seemed to me.


It's always dangerous to suggest based on things you think you remember without quoting the facts. Thus you can get a very strange opinion about people and I don't want to be a subject like this. thanks !


If I remember correctly, you implied that kpex's SAD5.1 method was sort of a lowest possible benchmark, when you said that even it was better than mine. I seem to recall reading other comments where you implied it wasn't very good.


The only comment we ever had on Kpex method was the fact that using 16 bits material (his method didn't work by then with 32 bits) made the music sound dull and that we could (and did) improve it with adding some reverb to the rears. We even made a Bidule to be able to work in 32 bits and : excellent results on especially electronic music ! So again : I don't know where you get that impression from that we condemn his method ! Again : don't be too fast with making YOUR own conclusions about people. Thanks !


Given my disappointing experiences with your Ambisonics method, can you not understand why I felt it unfair that you would berate someone else's method?


Again : where do you get this idea from ? And there is a difference in the tone of the messages...... Whatever I write here or in other threads and how hard I do my best : I'm still not English or American. Things I put in English can sound to you strange, while in my own language it sounds normal. We even had this discussion with Kpex, when I called the added reverb a major improvement, which is a quite normal expression in non English spoken countries, but at the same time I guess it sounded to Kpex as if we invented the wheel :-)
For you I only suggest you ask me how I meant something, instead of take it for granted that your interpretation of what I wrote is exactly how I meant it .It's easier for you to adjust your English than it is for me !


No, this statement is completely wrong! I already explained to you that my first bidule worked perfectly on the software I downloaded from needfulthings. The version of Kelly Industries BassManager in the needfulthings zipped file was 0.3. Check the Kelly Industries web site. It states very clearly that they changed the pinouts to the ITU standard in their v0.5. Still, they haven't changed their interface to match.


OK!


Yes it did sound fine to my ears. I've also explained some possible reasons why there have been two other negative responses, the lack of adequate monitoring to achieve the levels required being the most probable. A professional recording studio monitors the mix of a recording while it is being mixed and using some sort of reproduction matching the mixing format. Look at section 0.8 your ambisonic guide. It suggests monitoring the front and rear lefts through the same left channel and the front and rear rights through the right channel. All that can give is a notion that the sound is present and at a certain loudness level. It cannot possibly give any sense of a surround mix. I have already said that the bidule I presented requires at least a 4.0 setup for monitoring to be effective. Every individual song requires such monitoring to bring out the best combination of factors. Perhaps a classical recording or a live performance recording could be set once and then left for the duration of the entire recording, but a modern studio album of popular music is full of many variations and nuances that require different approaches. For example, if the vocals were recorded with a lot of stereo reverb, this will dominate the rear dry level in my bidule and thus the wet level in SIR should be brought down. If the rears are without much reverb then the opposite is true. If the bass is tight and punchy already, then the LFE level needs to be brought down or it will overwhelm the final mix. If the original stereo mix was quite wide, the center level needs to be higher to avoid a washed out sound. If it was overly narrow, then bringing the front lefts and rights up with provide a nicer sense of a soundstage. But, if the fronts are too high, then a good vocal performance will lose its distinctiveness. I prefer this kind of approach because it allows me to tailor my mix to the music. Even if I put a separate gain on each channel in your Ambisonics bidule (which I've done), I don't have this kind of control to tailor the final mix. It just changes the balance of the overall sound the same way a balance control on a traditional stereo system works.


Fully understood, but we tried to make a method that can be used by everyone. Newbies and pro's. You're more into the pro approach, while we try to make a method where you only have to input stereo and output surround. No adjustments in the bidule to make !!
Of course : gimme a few thousand dollars and I will buy the C3 soundanalyser, buy Wavelab 5 and Nuendo and a few other goodies and I can make a perfect mix with the output from C3.
But that's not our goal ! As soon as you have to taylor anything, we think our bidule is not good enough !!! And that's a very different apporach. Don't you agree ?
And : we will try to keep closer to the original stereo. I think that what we want is : when you play a surround CD and put your receiver into stereo mode, it should sound the same as the original stereo. In other words : We try to change the mix, but not the sound and when possible in an "convert only" bidule. If the stereo is narrow, we will leave it narrow ! That's the intention !
On a sidenote : every receiver will do DPL. Why bother at all ?
Why trying to re-invent the wheel ?
a. because it's just so much fun to experiment
b. because we're not always satisfied with DPL


Surely if you read the detailed description of mixing levels I just wrote with any sense of objectivity, you must see that I didn't just throw my bidule together for fun, I did put some thought into it. You may not agree with the methodology or the results, but you cannot dismiss it as "pretention." That's an arrogant word that I consider offensive. No I don't consider my bidule an "invention" by any means and I do give you credit for presenting your application of Ambisonics as an approach to upmixing. I do not agree with you when you make a sweeping, absolute statment such as "Realistic space can only be achieved by Ambisonics and certainly not by separation."


OK, so we disagree on that. No big deal ! Glad we're not all the same !! But I tried to nuance it, with the statements about hearing with both ears, in one of my previous mails.


Unfortunately, I get the sense that just because I disagree with this, you feel I'm some sort of competitor or enemy of yours and so you dismiss me as a "acting like a troll" or "whining and groaning." EoH, if you feel offended that I criticized your Ambisonics method, I'M SORRY. Now, get over it!

Feeling offended is too strong ! Irritated about how you put it in words in the first place : yes.
But these exchange of posts should clear things up and for me it did ! Did it for you ?

kind regards,

EoH

[Eye of Horus]

Originally posted by pelmen
first off a big thanks to all the people here who have worked hard on this whole 2.0 to 5.1 issue. i've been playing around with several of the posted bidules and have settled on the SAD5.1 for now as being the better of the bunch for my tastes.

1)
i'm certainly no expert when it comes to audio though but i was wondering if it was possible somehow to integrate a way of making use of dolby surround (and i think pro logic) if it exists in a 2.0 stream? from my understanding both dolby surround and prologic were multichannel signals mixed into a 2.0 format and re-expanded by the amp at the playback stage. i have a dolby surround test CD which i use to test the bidules (mostly using the channel identifier to make sure i ended up with all 6 channels going to the correct speakers :) and by far the result after processing isn't as good as the original. so i don't know if its even possible or not to detect if any such surround encoding is already existing in the 2.0 WAV file and once identified be able to make use of that knowledge to get an even cleaner 5.1 seperation?


I am not 100% sure, but wouldn't Softencode give that information when you load in a WAV with integrated Dolby ? Worth to try !



2)
also it seems the main use for people in this thread is for turning music into 5.1. do people feel the same bidules are just as good for tv/movie audio or would a bidule layout specific for non-music sources provide a better end result?


It's just a matter of trying ! Only you can decide what sounds the best for what you want. I can imagine you want more on the LFE when you're dealing with non-music. Just experiment and try to find what you like most. There is no general rule, unfortunately !


3)
my setup is a computer with sound audigy and cambridge soundworks 5.1 amp/speakers and i connect my dvd players audio to the same amp. with the audigy card it plays 2.0 sources to all 5.1 channels already (but i'd like to put things onto DVDR so i can do other work with my computer). does anyone know how the bidule approach compares to the audigy? or even making use of the audigy hardware somehow if it is better to do hardware processing into 6 channel wav files?


See the above answer ! Just try it and compare the results and use what you like most !
I found out that a simple listening test says more than the opinion or theory of others :-)


4)
ideally what is the end goal people here are aiming to achieve? complete sound seperation with for example voices staying in the centre channel instead of "leakage" into all channels.


Simple : the end goal is to get the best surround possible from stereo. How ? That is so difficult to answer, because the original stereo more and more decides which bidule you should use to get the best result. There are recordings where we can make a perfect (almost !) separation 5.1 from with indeed a voice only in the center speaker. But there are also recording that will sound better with Ambisonics or with Ambiophonics. It's too difficult to give a general answer to which bidule to use.
A lot of people are very enthusiastic about the SAD51inBidule we made. I like it very much on electronical music, but not on music with singers ! It's also a matter of taste and personal preferences.



its a great thread so far, the only problems i've ever come across have been related to 2/4GB filesize limitations with various apps. i'm certainly happy with the current results when compared to boring old stereo :) but i hope it'll be possible to improve and get results closer to proper 5.1 source results :)

thx

LOL !!! Define proper !!! LOL !!!

kind regards,

EoH

[pelmen]

EoH,

by "proper 5.1" i basically meant what a professionally produced 5.1 track can sound like. like what you'd expect on a good special effects movie with things audibly moving clearly throughout the room. so while atm we may be able to take a 2.0 copy of say the battle sequence from Master and Commander it might sound "big" when we turn it into 5.1 but it is still a long way from the original 5.1 soundtrack.

i guess there's no such thing as a "surround sound benchmarking" system or something. something that could analyse a 5.1 track and show you a visual "spread" of the soundstage...maybe something like a colourspace graph. it could look at how much each channel has in common at any given instant and graph it over time. so when you put a single mono track into all 6 channels it would show you a dot in the middle of the room because each channel is identical. something along those lines anyway i was hoping might exist, it might be a way of helping to benchmark bidules and fine tune them? or am i just fantasising? would also help people like myself who dont have a very good surround setup nor a good "ear". Also being able to use an app to test a file i'd probably trust more anyway since the audigy spits everything i play out to 5.1 channels anyway so its hard to tell if i'm hearing the results of the upmixing or the audigy's own processing.

so far i've been using a channel seperation dolby surround track for my testing. it is a voice speaking each channel in turn (left, centre, right, surround) followed by a few seconds of (pink? white?) noise. then i load up each of the 6 mono wavs at the end and "measure" the height of the noise sections for each channel mentioned by the audio. from that i can see how much cross channel leaking there is easily. since i can't subjectively trust my hearing for small differences between techniques i tend to use this one track as my benchmark to compare a new bidule layout to what i've used already and go from there. its probably the best testing method i'm going to be able to use.

does anyone know if there are prologic channel seperation tests available free online in a two channel format? my dolby surround test cd rips to a 2 channel wav and that wav clearly plays in each designated speaker as the original cd does. from memory prologic should be the same, just it gives more channels but i dont have a prologic test cd but something like that i'd have thought would be available free online to help developers.

i've been having a lot of fun playing around and exploring this process anyway since i recently got a dvd burner. i originally stumbled across this thread, it was not even something i'd considered possible but its certainly given me some inspiration to put a bit of extra effort into my vhs to dvd conversion so i end up with something that will last a long time to come. can't wait to see what new tricks people come up with! jsut wish i understood it all :)

thx

btw.. SMILE!!! its the weekend!!! :)

[ursamtl]

Originally posted by kempfand
ursamtl: I have no problems that you are personally disappointed with Ambisonics/Ambiophonics. But for the rest of the readers, I would like to make a few (also subjective) comments from my side.


Andreas,

Just to follow up on this, let me say that I retried your Ambiophonics dipole last night and was quite impressed with the results. First of all, for some reason the wave files in the www.app.demon.nl download would not load into SIR. However, the version on Needfulthings did work.

While the separation was still lacking and the rears were disappointing, the biggest thing I noticed about the front sound was a significant enhancement to the bass. There was much more clarity, definition, and punchiness without it sounding over-EQed. On some tracks it was actually quite stunning. In addition, the overall mix seemed to possess the kind of clarity and richness I normally associate with 5.1 mix.

What was missing of course was the sense of width and depth to the rear. First of all, even if theoretically the speakers are supposed to be spaced at +-10°, this just collapsed the width of the soundstage too much for me. Returning my speakers to +-25° (the way my PC system is setup this is the maximum front width I have to work with), gave me a better sense of width. Still, on some music, it seemed too dry. For example, I have a beautifully remastered CD of Elton John's Madman Across the Water. Running "Tiny Dancer" through this bidule resulted in the vocals sounding incredibly dry, virtually devoid of any ambience. While one might argue that this is a more realistic recording of his actual voice, the effect was not as pleasant. I tried inserting a stereo expander VST and this improved the image significantly. There is a sweet spot at approximately 120-130% width. Anything over that improves the sense of ambience but starts to negatively affect that glorious bass response.

There remains the problem of the rears. I don't want to just slap some reverb on it as that can get tedious. But the Ambisonics rears in the original bidule just don't work for me. In fact, I find that they detract from the enhanced fronts.

I have a couple of ideas for both the front and rear that I'll try to work on this weekend. Nonetheless, this is truly an encouraging result and definitely a step in the right direction.

Ursa

[Eye of Horus]

Originally posted by pelmen
EoH,

by "proper 5.1" i basically meant what a professionally produced 5.1 track can sound like. like what you'd expect on a good special effects movie with things audibly moving clearly throughout the room. so while atm we may be able to take a 2.0 copy of say the battle sequence from Master and Commander it might sound "big" when we turn it into 5.1 but it is still a long way from the original 5.1 soundtrack.


Starting with 6 mono wavs recorded separately would be ideal. But we're using stereo and need to extract as much as possible from only 2 channels !
I don't think we can get the same results, but with music we can get close !


i guess there's no such thing as a "surround sound benchmarking" system or something. something that could analyse a 5.1 track and show you a visual "spread" of the soundstage...maybe something like a colourspace graph. it could look at how much each channel has in common at any given instant and graph it over time. so when you put a single mono track into all 6 channels it would show you a dot in the middle of the room because each channel is identical. something along those lines anyway i was hoping might exist, it might be a way of helping to benchmark bidules and fine tune them? or am i just fantasising? would also help people like myself who dont have a very good surround setup nor a good "ear". Also being able to use an app to test a file i'd probably trust more anyway since the audigy spits everything i play out to 5.1 channels anyway so its hard to tell if i'm hearing the results of the upmixing or the audigy's own processing.


I do think there is some software that can analyse surround sounds, but I am not sure. I found that the C3 analyser is the most promising, but I could only find a Chinese (or Japanese ?) site, and..... no english translation there. But seeing the screenshots for stereo, it makes me feel this could be the ideal analyse software !
OTOH what would you exactly do with such software ?
I mean : as you say with a single mono the spot is in the middle, but this doesn't represent the ideal situation for the listener.
Music can be loud sometimes in a song from one speaker only and than with the same benchmarking you're thinking of, that spot won't be in the middle. And that doesn't indicate the mix is wrong then !!


so far i've been using a channel seperation dolby surround track for my testing. it is a voice speaking each channel in turn (left, centre, right, surround) followed by a few seconds of (pink? white?) noise. then i load up each of the 6 mono wavs at the end and "measure" the height of the noise sections for each channel mentioned by the audio. from that i can see how much cross channel leaking there is easily. since i can't subjectively trust my hearing for small differences between techniques i tend to use this one track as my benchmark to compare a new bidule layout to what i've used already and go from there. its probably the best testing method i'm going to be able to use.


Aha ! Yes , this way makes sense !
Can you upload that track to alt.binaries.sounds ?
It would be very interesting to use as test sound for the new bidules we're developing (Kempfand and me).
They are already very nice and give more separation than the things I tried till now, but such a track would be very helpful to finetune even further !


does anyone know if there are prologic channel seperation tests available free online in a two channel format? my dolby surround test cd rips to a 2 channel wav and that wav clearly plays in each designated speaker as the original cd does. from memory prologic should be the same, just it gives more channels but i dont have a prologic test cd but something like that i'd have thought would be available free online to help developers.


I never heard of it, but perhaps some googling can help ?
I will do a search online too and let you know if I could find anything.....


i've been having a lot of fun playing around and exploring this process anyway since i recently got a dvd burner. i originally stumbled across this thread, it was not even something i'd considered possible but its certainly given me some inspiration to put a bit of extra effort into my vhs to dvd conversion so i end up with something that will last a long time to come. can't wait to see what new tricks people come up with! jsut wish i understood it all :)


I wish that too !! :-)
I am not the writer of the bidules made ! I am more the guy with the (sometimes very weird) ideas behind bidules and I try to translate every technobabble into a useable guide. Some of the techniques used to get the results are too difficult for me to understand. The outcome however is for me the only thing important !!



thx

btw.. SMILE!!! its the weekend!!! :)

ALWAYS smiling :-)

kind regards,

EoH

[ursamtl]

Originally posted by pelmen
my setup is a computer with sound audigy and cambridge soundworks 5.1 amp/speakers and i connect my dvd players audio to the same amp. with the audigy card it plays 2.0 sources to all 5.1 channels already (but i'd like to put things onto DVDR so i can do other work with my computer). does anyone know how the bidule approach compares to the audigy? or even making use of the audigy hardware somehow if it is better to do hardware processing into 6 channel wav files?


Pelmen,

If you haven't done so already, run (don't walk) all the way to www.kxproject.com and see if your Audigy model works with the kx drivers. These are free ASIO-compatible drivers for the Creative Labs cards based on either of two EMU DSPs (most internal Soundblaster Live's, Audigy's and Audigy 2's) and they can turn your card into a much more powerful tool for music. Their DSP applet is very similar to Plogue Bidule in that you can customize routing and connections graphically to suit your needs. With your system, you could easily create bidules with multichannel ASIO outputs for realtime monitoring of your surround mixes. In Plogue, if you're sample rate is set to 44.1k, just insert a kX ASIO output device (15 channels), or if at 48k, a kX ASIO Duplex input/output device (15 channels each way). Then check your routings in the kX DSP to make sure you're routing the 5.1 channels properly. After that in Plogue, you can connect each channel in real time while playing back and experimenting with different approaches, settings, etc.

[Eye of Horus]

Originally posted by ursamtl
Andreas,

Just to follow up on this, let me say that I retried your Ambiophonics dipole last night and was quite impressed with the results. First of all, for some reason the wave files in the www.app.demon.nl download would not load into SIR. However, the version on Needfulthings did work.

Hey, the link to my homepage is nice, but realize people clicking on that link won't get what they want :D :D :D

I will have another look into these wavs, because they should work in SIR, because we used them too.....

I also found this method giving a good sound, but my objection :
a. I cannot move my speakers and
b. I am not in the position to buy 2 extra ones :-)

(I'm sure you will like our new ones...... but they need some more work on them before we publish them.)

Just curious : do you convert your stereo to 32 bits first ?

kind regards,

EoH

[ursamtl]

Originally posted by Eye of Horus
Hey, the link to my homepage is nice, but realize people clicking on that link won't get what they want :D :D :D

Oops, sorry. I posted just the text without using the http:// button on the forum interface because I didn't want to repost a link to a file that might potentially be corrupted. I didn't notice that the forum software automatically converted it to a link.

I also found this method giving a good sound, but my objection :
a. I cannot move my speakers and
b. I am not in the position to buy 2 extra ones :-)

Well, did you try the dipole without moving your speakers? I found the results were better with my speakers at +-25°. At +-10° there was very little sense of width to the sound at all.

(I'm sure you will like our new ones...... but they need some more work on them before we publish them.)

Just curious : do you convert your stereo to 32 bits first ?
I look forward to the new bidules. As for the 32 bit conversion, yes, I have done this a couple of time for comparison sake and it does provide a marginally cleaner sound especially in high-frequency details such as cymbals, etc., but I didn't notice any appreciable difference in terms of the Ambisonic effect. I'm lucky in that, for one thing, I can monitor my mixes in real time on my PC so I don't always burn DTS CDs. In addition, my DVD player reads CDRWs so when I do write burn DTS CDs, they're temporary tests to see how the sound is transferring to my 5.1 system. Once I settle upon an upmix method that gives me the results I seek, I'll do some proper full CDs and start with 32-bit files. I still haven't found a satisfactory answer concerning the whole question of dithering. It's generally agreed that going from 32-bit to 16-bit with no dithering reduces sound quality, but so far the Surcode documentation hasn't mentioned whether it downsamples to 16-bit and if it does so, does it use dithering. All it says it that the program accepts 16- or 32-bit files for input. The CD redbook spec is 16-bit, but it's not clear that a DVD player recognizing a CD but playing a DD or DTS wave file does so only if the file is 16-bit.

By the way, yes, your earlier post did clear things up between us. I'm at work and don't have time for a detailed reply, but yes, there are surely cultural nuances to our communication that can cause misunderstandings. For example, I once worked in an office with some Swiss and German personnel who were working on a contract for the Canadian govt. here. They would get into discussions that seemed to me like they were ready to haul out weapons and kill each other, yet walk out of the office at the end of the day smiling at each other! Here I was the poor Canadian fellow shocked into thinking I was about to witness murder! These folks told me this was simply their European temperament and nothing I should be concerned about. Perhaps we Canucks are just too polite! Plus, having worked with them for awhile I found I was adopting some of their attitude.

Have a great weekend!
Ursa