Actually, I didn't. In fact, quite the opposite: I went out of my way to repeatedly clarify that it is not remixing (especially not what you're proposing with the centre channel content being spread to all three front speakers).

With PLII, there is no centre channel unless one is extracted by the processing. Since the extraction is already ocurring, it makes sense to have some user control over the extraction level.

This is very different from the remixing you're talking about, where no additional channels are created but the contents of the front three channels are literally redistributed. You can "consider" it anything you want. That will allow you to claim that the 990 has a feature (which it really doesn't) and then complain when that feature is not present for other modes.

If you actually admit what PLII is doing, then your complaint falls apart because you realize that there isn't any remixing going on to begin with. Again, you can "consider" it anything you want, but they are in fact different processes. Extracting a centre channel where there was none is different from redistributing discrete centre content to other channel and mixing it into the content that already exists in those channels. No, it's not "simply a case of software refactoring." Matrix extraction is different from downmixing. The above describes standard downmixing of three channels to two. But this is different from what you've been proposing, where you start off with three channels and end up with three channels, except with the contents redistributed. I don't understand what you mean by the above. Are you saying that imaging is shot when using a centre speaker for a listener in the sweet spot? And how does anchoring the dialogue to the centre channel end up collapsing the front soundstage?

The above describes standard downmixing of three channels to two. But this is different from what you've been proposing, where you start off with three channels and end up with three channels, except with the contents redistributed.

It is not different. Two channels of speakers is just a special case of three with the level of the third set to zero. Three input channels available for remixing is independent of whether some of those channels are derived from others, or not.

You appear to distinguish between special cases and the generic case, calling them different things. If we prefer, we can call what I propose a matrix transform of a vector function of the input channels. I prefer the term "remixing", to that actually.

I don't understand what you mean by the above. Are you saying that imaging is shot when using a centre speaker for a listener in the sweet spot? And how does anchoring the dialogue to the centre channel end up collapsing the front soundstage?

Yes. Add a center channel speaker when you have dialog slightly left or right of centre (which can be properly imaged with two front speakers for a listner in the sweet spot), and the dialog moves more toward the center.

This helps off-axis listeners, but hurts the listner in the sweet spot. Therefore, it should be tunable to the degree one has off-axis listeners (i.e. how far off-axis). I suppose a single center channel with a fixed level (relative to the others) suffices for most theatres, but I have never found one necessary when listening/watching alone. I do expect however, that a two channel downmix of discrete LCR sources might be worse than a stereo recording as far as placing dialog: when not downmixing 5.1 to stereo, I hear very little left and right channel dialog. With stereo recordings, I hear speakers (as in people speaking, not transducers) located quite precisely when seated at the sweet spot.

In the extreme, I've seen studies that suggest anything more than two front speakers is hogwash, and I have recordings with back to front and side to front and back effects that bear this out. But, they only work in just the right room, at the sweet spot, suggesting a lone viewer/listener with time and money for precise speakers and room treatment. Google for "phase accurate speakers" for an introduction to this kind of "voodoo" (as most discount it). The effect, however, while perhaps not as pronounced or important as proponents might suggest, is very real.

No, I'm distinguishing between matrix decoding (where a centre channel is extracted from other channels) vs downmixing (where the centre channel is combined with other channels). You can buy an inexpensive analogue mixer from pro-audio companies like Behringer to do the type of downmixing you're interested in. However, it won't do matrix extraction. They're two different things. I just tried this with a couple of sources that have dialogue moving across the front channels and the dialogue did not move towards the centre. One was a discrete 5.1 source (Monsters, Inc) and the other was a 2-channel source (Bad Day at Black Rock). I compared playback with and without a centre speaker. You may have a problem with your system.

No, I'm distinguishing between matrix decoding (where a centre channel is extracted from other channels) vs downmixing (where the centre channel is combined with other channels).

Then we have a difference in terminology. I consider downmixing the (usually linear) mixing of upstream data to be sent downstream, as in a pipeline process. The number of channels need not be reduced. Thus, a three channel to three channel downmix is possible.

What you say about extraction is true, but, because PLII extraction permits control over the distribution over the front three channels, I separated into a fixed extraction (which is likely a non-linear transformation) and a linear mixing transformation. It is the latter that I would like to see provided in general.

You may have a problem with your system.

Since I only have a 2.1 setup at the moment, I don't have a system to have a problem with. However, your observation is interesting, particularly with PLII extraction (I would expect a discrete mix to fare well).

Are you suggesting that a PLII extraction to two front speakers (no centre) is no different than a PLII extraction to three front speakers, including a fully extracted centre (i.e. what would be a "normal" extraction setting)? If so, how far apart are your two front speakers? My 520i sit 12 feet apart, and, on stereo test material, can image quite precisely between them, and to some small degree, outside them, at the sweet spot (not that this is surprising).

The "down" in downmixing refers to the fact that you're reducing the number of channels. The opposite of upmixing (e.g., from 5 channels to 7 channels). If you're going from 3 channels to 3 channnels, then that is not a downmix.

As for what PLII does in the front channels vs what the 'no centre' setting does, it's quite different:

PLII does matrix decoding and logic steering. First, correlated mono information is copied and sent to the centre output. Next, anti-phase cancellation signals are generated and sent to the three outputs. This surpresses L/R stereo info in the centre channel and surpresses centre info in the left & right channels. As part of the logic steering process, the gain of the three channels are ridden based on where sound is dominant in the front soundstage. This aids the cancellations signals in enhancing the sense of channel separation.

Compare that process to setting the speaker configuration to 'no centre', where discrete centre info is sent to the left & right speakers and the centre output is turned off. To be blunt: this is the electronic equivalent of a simple Y-splitter.

I don't see how you can consider these two processes to be even remotely similar. Making the downmix function variable would require actual mixing capability, which is why I suggested you get an inexpensive analogue (pro-audio or DJ) mixer. Wait a second: you're only using two speakers? Rene, have you been guessing about some of the stuff you're claiming or have you actually tested those things out (as I did last night)? What extraction are you talking about? With two front speakers, PLII is not extracting any centre content. If you're using two speakers total (no surrounds), then PLII is not doing any processing whatsoever. Two channels, two speakers; what's there to process? You simply route each channel to its respective speaker.

Wait a second: you're only using two speakers? Rene, have you been guessing about some of the stuff you're claiming or have you actually tested those things out (as I did last night)?

I performed a mathematical analysis. On the BG dispersion patterns (and those of planar magnetic speakers), I refer you to the BG site (www.bgcorp.com). As I own a pair of Radia 520i as well as Carver ALS, I am familiar with them personally.

I note that the PLII extraction with variable centre channel level can be decomposed, mathematically, into a non-linerar transform followed by a linear 3x3 matrix multiply. I refer to the latter as downmix, because that's what it's typically used for. Use "remix" if you prefer. That remix can be used for extracted as well as discrete channels.

The bottom line is that you are saying that matrixed sound is decoded by Flcrs(X) and remixed discrete sound is handled by MY. No disagreement.

I'm noting that this can be transformed to Flcrs(X) = M*F'lcrs(X) and M*Y = M*Y, refactoring M. M, of course, is a linear transformation (i.e. remix) matrix, Flcrs() and F'lcrs() being a non-linear vector function over a two dimensional (stereo) vector argument, and Y being a discrete signal vector.

Given that M exists for discrete sources (to map channels to speakers) and can be derived for matrixed sources, I say, why not make the processing offered by M orthogonal to the encoding format?

What I'm not hearing from you (and what might very well be the case), is that there is no linear decomposition M*F'lcrs(X) = Flcrs(X) possible to achieve the degree of center channel extraction for PLIIx sources. I was under the impression that such a linear decomposition did, in deed, exist.

What extraction are you talking about? With two front speakers, PLII is not extracting any centre content. If you're using two speakers total (no surrounds), then PLII is not doing any processing whatsoever

On the 990 it most certainly is, because I can't tell it there is no centre speaker. I can either tell the 990 to downmix to stereo (which is idempotent for PLII, as you noted and it is not clear if the 990 will extract and remix or just leave the signal alone), OR tell the PLII processor to extract none of the centre.

If M*F'lcrs(X) = Flcrs(X) decomposition is possible for PLII, then there is no reason why the 990 can't just let me control M independent of source. It lets me do this for some values of M knowing how many speakers I have for discrete sources with one fixed value of M for each speaker configuration, and several values of M for PLII decoding, controlling centre channel extraction. If PLII-style control of M was available for all sources, then it would offer a workaround for the bug where I can't specify the lack of a centre channel speaker.

Sure, I can select stereo downmix from the 990, but it defeats this when it sees a PLII encoded source! So, there are two bugs consipring against me. If Outlaw had refactored M as above, I could use creative settings of M to get around the bugs.

I suppose, I can get around the problem until my BG 220i arrives by setting the centre channel extraction to none for PLII matrixed sources.

The answer is not an 'inexpensive' analog mixer (I have not found good mixers to be inexpensive -- the BG 220i is by far the cheaper and more correct solution). The answer is a centre channel speaker.

There is no reason you can't do that but it would require the 990 to have an actual mixer built-in (something different than the simple Y-splitter functionality it currently has). There is no interest (aside from you) in such capability because the goal of home theatre technology/processing has been to localize and stabalize centre content, not smear it across three speakers. It's the antithesis of what companies like Dolby and DTS want to do with a discrete centre channel. Again, that's a bug. On a normally operating procesor, PLII does not extract centre content when only two front speakers are configured. That would be the equivalent of discarding centre content (it's being extracted but not played back). Again, PLII does not do any downmixing. The Centre Width parameter can be adjusted to vary the amount of centre content extracted. But PLII has no capability of combining centre content with other channels. OK, I'll bite. When you do get a centre speaker, how will you suddenly have the capability to variably bleed discrete centre content into the L/R channels? Is the centre channel somehow going to change the processing capabilities of the 990 to add the remix function that you want?

This is an interesting thread to say the least. I'm wondering why Rene's 990 doesn't allow selecting NONE for the center channel. Is there anyone else who has this problem? I would think Outlaw would exchange the unit if it were a defect of that unit alone.

Unless it's a bug with the model, not Rene's specific unit?

OK, I'll bite. When you do get a centre speaker, how will you suddenly have the capability to variably bleed discrete centre content into the L/R channels? Is the centre channel somehow going to change the processing capabilities of the 990 to add the remix function that you want?

Obviously, I won't have that capability.

But, if remixing were implemented as I propose, it would have (a) provided a workaround for the no centre speaker bug; (b) allowed fine-grained control over the front soundstage which might be useful for some; (c) been easier to document since remixing would be orthogonal to source format. I'm of the opinion that this is just a software implementation issue, and not constrained by hardware. I'd like to know if that assumption is not correct.

A centre channel speaker would, of course, eliminate the need for the hack, but the ability to experiment with front channel remixing would be handy to see the tradeoff between sweet spot imaging and off-axis centre channel anchoring.

Yes, yes, an analog mixer could be used to do this. But, if the 990 has all the capabilities to provide such a mixer in the digital domain, why not expose them to hackers?


I'm also guessing that the inability to specify no center speaker is a software bug with the model. Does anyone know otherwise?