Should I use this driver for TV surround sound?

mirko2

I'm a bit upset about the original Audigy 2 ZX drivers because they don't support decoding ProLogic stereo sources (upmix to 5.1). They only have those CMSS/2 upmix options which seem to be a very poor own imitation of ProLogic. There is a better ProLogic-to-5.1 upmix decoder in Creative's player, the NEO DTS thingy, but being in that player, it's quite useless: I want to watch television with 5.1 sound. So I need a ProLogic-to-5.1 decoder directly in the sound card driver and not in a specific player. Would this kx project be a solution? Does it have a ProLogic decoder in the driver and, if so, is the decoder good and configurable?

JohnMcClane

kX drivers will simply upmix your tv sound using surrounder, but you'll not get true surround sound. I'm not sure if ffdshow is able to decode prologic streams, but you can give it a try. http://www.free-codecs.com/

mirko2

> kX drivers will simply upmix your tv sound using surrounder, but you'll not get true surround sound. I'm not sure if ffdshow is able to decode prologic streams, but you can give it a try

I've in the meantime taken the courage to uninstall the Creative drivers and install kx. So far, so good: unlike the original driver, there is now an EQ before the center and bass from the center (lowpass) is routed to the bigger front speakers. Sounds good. What do you mean with not getting true surround? Is a stereo ProLogic decoder supposed to do more than the "surrounder"? As to the surrounder: I've noticed a surrounder+ and a prologica and other interesting-sounding things but do not yet understand how to use them. One barrier for experiments is the message "not enough iTRAM" error message (on that matter: if one wants to adjust treble and bass only, is the 10 band EQ the only option or is there a less-complex object that saves ram?).

Maddogg6

Well, you seem to be off to a great start, just by playing with the dsp.

I believe the iTram is the RAM attached directly to the DSP hardware. xTram being the buffer ram on PC allocated for some plugins/modules. And yes, these will limit what you can and cant do.

To make use of more effects you should look into using the 'ProFX' plugins that directly replace Prolog, Epilog and FXBuss modules.
It takes some getting used to and learning what the modules do, but its well worth it.

Experiment by right clicking in the DSP window and clearing the dsp - (no worries, re-initialize dsp, or reloading the defailt.kx file settings will get it all back)

and add the modules that spark your interest. Buts its best first to study the help file to uderstand the difference between FXBuss 0/1 - 31 lines and the outputs etc.. That was the most confusing to me, and took some time to learn and memorize the things important to me (which may not be important to you.

SAVE SETTINGS - for every siyuation..

Like I have one for watching videos, one for music production, one for MSN Messenger... and it did take some to to learn to do, but the end result was WAY worth it. (I use the default for playing the occasional game)

KX kinda forces you into learning about the 10Kx DSP - because of the flexibility it provides.

KX is an experimenters dream, cuz you wont blow anything up if you keep the speaker amplifier at a lower level.

ps. i interchange the terms 'plug-ins' and 'modules' meaning the same thing.

mirko2

Thanks for the tips!

> Buts its best first to study the help file to uderstand the difference between FXBuss 0/1 - 31 lines and the outputs etc..

About this FXBus:

1. why does the manual mention 16 buses but the router displays 32 buses (0-63)? For example, in the default setup, the E routing of the voice Wav 0/1 is connected to bus 63. Why, what is bus 63? In the DSP window, only 0-31 are visible. Is 63 the equivalent of "not used"?

2. If the Wave 0/1 stream is selected in the router, the right side shows the routings A-H for this stream. Is it implicit that A is left and B is right and C is left and D is right and so on?

3. There are volume sliders for all routings of the Wave 0/1 stream (except for routing a/b, but that's explained in the manual with internal use for panning). But I don't get any sliders for the routings of stream Wave 4/5, 6/7 etc. Why is that?

Maddogg6

SBLive (10K1 chip) has 32 mono busses, Audigy (10K2) has 64.
FXBuss 2 gives access to 32-63

The busses can be used like effect sends on a physial mixer (mixing console/desk ) for the various sources. Except, you also choose the output with it also. (on a mixer theres dedicated outputs, in KX you assign what 'pin' the signal is sent to -

for instance:
Wave0/1 is set up to send to the FXBuss 0/1 pins (2 monos, making it stereo) and 13/14 is the stereo send to the FXMix which allows you to add reverb and chorus to that.

In the KX Router
If you click on the '0' (or any of the A-H sends) - a pop out will allow you to change the pin that signal will appear at.

*I think* its because the 'X Routing' will manage the record sends (in KX Mixer.. if you remove XRouting from the DSP - the KX Mixer looses the vol sliders for wav recording (master and wave 01/ 2/3, 4/5, 6/7) in the recording screen.

If you adjust Sends A/B in the KX router also, it could get really confusing, you need to do too many adjustments to make recording work. ??
I could be wrong, but thats my opinion/guess. XRouting is a bit counter intuitive (IMHO) and its name makes no implication to its purpose.

Also back to FX Buss, I look at it like this... The FX Buss plug-in/module - is NOT a plug in per se like a delay or reverb is - but represents, and gives interface to, the DSP's internal routings ability much like the Prolog and Epilog give access to the physical (and virtual - ASIO) ins and outs. To me, they should be called something more 'mixer like' as Ive never seen a 'prolog' on a phyical mixer (those are programming terms as far as I recall/seen) - Phyical & Virtual ins/outs - would have been more 'musician' friendly.

JohnMcClane

The surrounder plugin does not decode the prologic stream. It's a simple upmix.

Surrounder+ is almost identical to surrounder - surrounder only supports up to 5.1 speakers. Prologica is a more simple plugin, that will not allow to do bass redirection and has not as many speaker modes as the surrounder plugin.

mirko2

> The surrounder plugin does not decode the prologic stream. It's a simple upmix

What's the difference? Prologic is, I believe, a simple addition of a center signal and a rear signal (both mono) to L and R in the sound studio. The signals are added in a way that allows removal in a ProLogic decoder: the center signal is added to L and R, so getting it back is as simple as extracting the monaural part from L/R. The rear signal is added to L and R but with the phase shifted by 180 degree, ie. one channel gets the signal upside-down. As with the center signal, getting it back is as easy as a few additions and substractions. So in what way does the surrounder not execute these primitive steps and is "a simple upmix"?

On getting good suround sound: I understand that many ProLogic decoders have options such as speaker distance [m], speaker position [degree], speaker size [small|large], room size [m*m] and vertical speaker position [m]. Distance probably translates into the [ms] options of the surrounder element (Creative's driver has a good THX calibration console; surrounder needs something like that, too, or at least I have no clue what to enter in the ms fields). Room size probably translates into reverb (?) and speaker size into bass redirection (?). I wonder to what the position [degrees] and the vertical position [m]translate?

Totally unrelated: what element in the DSP window implements a stereo volume slider? I would like to balance the front channels after the prologica/surrounder (in a simple setup with ProFX elements). Or are there mono sliders only so that two of those have to be inserted into the L/R paths?

mirko2

I understand that part. But each router block (the A-H sends) is related to a signal with two channels. For example, Wave 0 and Wave 1 in case of the eight Wave 0/1 routings. So it must be an implicit, hard-coded fact that the first routing (A) routes Wave 0 and the second routing (B) routes Wave 1 and then (routing C) it starts with Wave 0 again? The router is extremly counter-intuive. Much more logical visual representations come to mind...

Maddogg6

Yes - you are correct

WAVE 0 - A,C,E,G
WAVE 1 - B,D,F,H

As I understand it.

And yes - better lables even could be less confusing (like WAVE0/1 - would make more sense (when relating to a physical mixer) if were called WAVE 1 L/R..... BUT it wouldnt make sense when working in surround. Thus Im sure its why the lables arent more specific, and you just have to remember - 0/1 is the 'reminder' of those busses being 2 channels.

Maddogg6

Some of those adjustments are in the 'TIME BALANCE' plugin I mentioned earlier - btw... TIME = DISTANCE FROM SPEAKER - and not playback speed (which is what I thought it was ay first.) I dont think vertical position is part of it tho.

There are several volumes and mixers with adjustments.
Theres FXMix2 - which is a 4 St input mixer with 1 send. (like 8x4)
Theres St. Vol. and various sums (i never really use)
Stereo mix (with gain too)


AND.... MX6 - is a 6 stereo In - with 2 sends, a main and record bus.
The 'M' and 'R' are MAIN and RECORD bus enables - which are dis-abled when adding the module - so no sound until those are enabled.

(its like a 12x8 mixer - but hardwired with no pans - so more like a 6x4 mixer)

ThomasBarnes (? spelling ? - a user in these forums) has a nice tutorial on setting up ProFX - its aimed at using for Sonar - but the DSP setup technique is the same.

Experiment around - adding modules - dbl click on them to get the GUI. And look to see what it does. Not all mudules/plug-ins have a gui.

Hope that helps

JohnMcClane

Sorry I thought you know what I mean. kX drivers do not decode the stream. The surrounder plugIn and also the prologica plugin simply upmix the stereo source into multichannel sound according to your settings. If you select Rear=Front, then all speakers will have the same output (similiar to stereo surround from creative), if you choose Movie Mode you'll get a more cinematic effect, where the center speaker will be in foreground. But you'll not get positional sound, as the center and surround streams are not decoded.

Maddogg6

It sounds like your describing a old matrix (and crappy - Dolby Pro Logic) way of surround before discrete surround became to be - by simple adding and subtracting l and r to derive a 'surround' effect - then added a delay line (10 - 50ms or so)for the rear speakers so you can 'electrically' emulate more distance between your ears and the rear speakers sound relative to the front channels. Then add a lowpass filter for a subwoofer (before the decoding) and a mixer/band pass filter for the center (after decoding the front channels).

discrete (Dolby Digital) surround on the other hand has no such encodings and all channels are their own channels.

- if using proper software (as I have read - FFDSHOW (audio), a filter somewhat like a codec - that many players can utilize) can properly decode (seperate from a single multi-channel stream) dolby digital encoded material as the decoding is done BEFORE the audio enters the 10Kx DSP. You may need to make a DSP setup to do so - like directly connecting respective FXBus signals to the Epilogs 5.1 (L,R,C,RR,RL) outputs. (Note, thse signals on the xRouting are 'Compass based' - N - north, S-south E W. etc... )

Now, the surrounder/prolocica, as John mentioned - just takes normal stereo and makes it so all your 5.1 speakers will make some sound.(I had wondered about the technique used in those plugins, if the did do the add/subtract method to 'upmix')

Read this:
Surrounder Explained

Its my understanding modern discrete digital surround requires MORE than the 48Khz sample rate as locked in to the 10Kx DSP - thus decoding with a 10Kx DSP needs software, as the dsp wont be able to on its own. The CL drivers Im sure did the decoding.

Using FFDSHOW *should* decode BEFORE running any signal through the DSP Like CL drivers would have had to - and I was under the impression *WOULD* in fact give a discrete 5.1 surround decoding. As long as, of course, a 5.1 encoded signal is sent through the KX WAVE HQ (as I understand it, the only sound device capable of playing anything other than 48Khz streams) - or possibly the discrete channels are routed (with in the decoding software) to the appropriate output device (IE. 2/3 front, 4/5 rear, 6/7 LFE & center)

I need a 5.1 speaker system so I can better experiment/understand/explain... but thats how I understand it as Im no expert on the subject and I have much to learn my self Im sure.

I DO KNOW - 'surround' is a generic marketing term that has turned into a huge fiascko (IMHO) - meant to confuse consumers.

heres some good explanation of the different surrounds that have been out (and note some of the dates for some of the systems/techniques):
http://www.informit.com/articles/art...?p=341486&rl=1
and
http://www.surroundassociates.com/fqmain.html

Now it seems they just went and added matrix to discrete to get 6.1 and 7.1 - Im sure thats just giving plenty of consumers headaches too.

mirko2

Yes! I'm talking about getting a surround effect from analog stereo material (old videos, TV). Isn't that what Dolby Surround, Dolby ProLogic, CMSS and DTS NEO:6 are all about?! I'm not talking about discrete digital multichannel audio.

@John:
Really? I was under the impression matrix-decoding is too easy not to be implemented (see below). And I don't see any other way to get the cinematic effect you are mentioning. Well, maybe a bandpass that routes frequencies associated with human speech to the center. Are you absolutely, 100% sure that the surrounder module does not implement matrix decoding?

--- quote from doom site

Dolby Pro Logic (1 or 2 or whatever) is a decoding algorithm. [...The] DPL decoder is about interpolating more than two channels out of a 2ch track. (like in a 2.0 --> 5.1 process). The DPL/DPL2 implements an adaptive 2x5 matrix:

[Lt Rt] * MTX = [L R C SL SR LFE]'

The left channel output, L, would be equal in this case to : Lt * MTX(1,1) + Rt * MTX(2,1)
(and so on for R, C, SL,..)

This matrix is adaptive and that means that the coefficients change from time to time, according to the input data. The dpl algorithm doesn't "care" if the source is a dolby surround material or a plain stereo track.

Let's think of this very primitive process with a static dpl-like matrix. Given a 5.1 source L, R, C, SL, SR, LFE we would create a 2ch downmixed track with channels Lt & Rt:

Lt=L+C+(SL+SR)
Rt=R+C-(SL+SR)

The (static) decoding process would look like:

Lr=Lt;
Rr=Rt;
Cr=(Lr+Lt)/2; [=L+C+(SL+SR)+R+C-(SL+SR)=L+R+C]
Sr=(Lr-Lt)/2; [=L+C+(SL+SR)-R-C+(SL+SR)=(L-R)/2+(SL+SR)]

creating Lt & Rt is principly what Dolby Surround Downmix is about, and creating Lr, Rr, Cr, Sr is principly what Dolby Pro Logic is about. Aas mentioned before, the (adaptive) DPL algorithm can reproduce 4/5/6 channels out of ANY 2ch source, but in case the source was encoded as Dolby Surround, the decoded track would be very close to the original. As you can now understand, Dolby Surround "encoding" is a very simple process of simply weighted summings of 5.0/5.1 channels into two channels.

As for DPL2 & Dolby Surround 2, in Dolby Surround (1), we would sum the surround channels into one mono channel, and treat them just like in the example as "(SL+SR)", while in Dolby Surround 2 we would "help" DPL2 seperating them back by giving different weights to each channel. For instance:

Lt=L+C+SL+0.5*Sr
Rt=R+C-SR-0.5*Sl

That's it.

Note: all above equations are only for demonstration! For the real downmix coefficients download BS_Downmix.dll from my webpage.

---

(I don't know why the matrix has to be adaptive but to offer a guess, it changes if it detects speech?)

JohnMcClane

Yes I'm sure - you can take a look at the source code if you don't believe me. The surrounder does no decoding in any form - it simply copies your front output to center and rear.

mirko2

I could look at the source code all day long without getting any smarter What about the prologica plug-in. The name suggests...

Maddogg6

With out looking at (or even understanding it) the prologica plugin - its name implys it will do the matrix decoding expected in a 'Dolby Pro Logic' manner.

And, there is one point I'd like to address about John saying sourounder only copies - there IS, at the very least some delay and EQ happening when 5.1 speakers and surround mode 'ON' is selected. Thus, makes me wonder if a matrix has been implemented in it (in a newer version than what John has??). Which IS adjustable. (Rear speaker width/distance (delay). It creates a pleasent effect with 4 speakers - which is all I have.

Even looking at source code wouldnt mean anything to me as Im no programmer.
But it may JUST add delay and a comb filter possibly.

BUT, to answer your original question.
The Prologica module has NO adjustments. BUT, the time balance plug-in *could* be used with it to give (at least some) of the adjustments (speaker distances) you mentioned. I dont beleive 'placement/locations' vertical nor horizontal are implemented - it 'sounds' (from reading the thread) like its hard coded into the plugin, and has been tweaked in different versions. (I cant check cuz I didnt install it as its not compiled for KX V3638h, that I have to use)
look here:
I made a TimeBalance plugin

I say, install KX - give it ago - many people are very happy with the results - if your JUST looking to have 5.1 speakers be used when playing stereo material. It will, with plenty of adjustments and ability to make your own prologic decoder if you have C++ (M$ Visual C++ 6.0 is best as I understand) experience and understand microcode. All the source code is available and the SDK is included (it was with V3538h anyway) - the time balance plug in, on the otherhand, was developed after - thus I dont have the source or the compiled code for my version of KX. And can only regurgitate whats in the thread I linked to.

Maddogg6

BTW - the prologica module has a stereo in and 5 outputs.

Heres the micro code for prologica.

I beleive it to be doing 'Dolby Pro Logic' decoding - but the micro code is mostly greek to me. But I know its much more than just routing signals.

PS - micro code - 'MACS' = Multiply and carry Saturation (Saturation is like Assmblers 'Overflow' - sorta i think its a DSP equivelent of the overflow anyway)

The iTRAM is the delay - which you CAN adjust in the micro code. (its using 20ms - which I think is the DPL spec. - but many decoders would allow adjusting that anyway)

UPDATE:
I did more testing with speakers in surround mode(s)
Common conditions in all tests...

5.1 ALtec Lansing ANALOG system (bought cheapest for this test) - will return tomorrow. hehe

DVD - 'Monster in Law ( I know, not greatest surround movie - but was pronounced enough for me to hear the different channels)

Win XP SP 2 -
K-LITE codec pack installed (provided AC3 filter as I recall - in addition to all the other codecs)
Win DVD 7 Installed (wont work with KX - no audio - but provides DVD codecs)
AC3 filter - set as 3/2 SW 5.1 Channels - direct sound device


format below is:
CONDITIONS
TEST
RESULTS

C: DSP CLEARED - WMP10 w/ AC3 filter - direct sound device.
T:DVD - 5.1 Dolby Digital Surround Encoded
R:5.1 source works great - set speaker mode in WMP10 options/dvd/advanced (match the reported stream detected for best results) Audio TAB.

C:DSP - default (re-init DSP) then change speaker mode to 5.1 and 'ON'
T: Need for speed Underground II - only game I have loaded
R:No discrete surround.

C:DSP - using 1 src, prologica and KXLT2
T1:DVD - 2channel - surround turned off (AC3 dilter settings in WMP10)
T2:NFSUGII
R1&2:Prologica - sure seems like a Matric (DPL) decoder with delayed rears (20ms according to mivrocode).

C:DSP - default (re-init DSP) then change speaker mode to 5.1 and 'ON'
T1:DVD - 2channel - surround turned off (AC3 dilter settings in WMP10)
T2:NFSUGII
R1&2: Surrounder - does similar as prologica - with adjustable delay for rears and even an adjustble 'rear width' via what sounds like a comb filter.

So I am pretty confident to say the SB0350 (Audigy 2 ZS PLAT) with KX V3538H will decode matrix and discrete surround.

mirko2

Creative's mediasource player has a stereo to 6.1 upmix called DTS NEO 6. The manual says it's mutually exclusive with their CMSS upmix, so I assume it's a DSP program and not a software solution in the player. Would it be possible to dump this program out of the DSP and try it out with kx?

Maddogg6

I understood the media source player would actually decode DTS and is not just upmix. - But ONLY if played in their mediasource player - which as I understand is the *only* way to play DVD-A discs, DVD-A discs being the only thing I've seen thats DTS encoded personally.

I still think (quite sure) that theres significant PC code that would accompany the micro code for the DTS playback/decoding.

DVD-A - Ive read is looking to avoid pirating in a big way... so much so, I think it will ultimately backfire on them - I could be wrong but the DTS site reaks of 'EXPENSIVE TO LICENSE'.

The DTS site is not very informative as to how it works - a quick look didnt find any technical specs even. - My guess its all about DRM and anti-piracy more than its about improved end user experience.

http://www.dts.com/

for anyone who care to look.

mirko2

> I understood the media source player would actually decode DTS and is not just upmix. - But ONLY if played in their mediasource player - which as I understand is the *only* way to play DVD-A discs, DVD-A discs being the only thing I've