Emotiva RMC-1 processor

[richb]

I use DBPowerAmp for ripping. There are times where the checksum numbers do not agree and others fail and fall back to secure rip mode. So, my experiences is not that they are perfect if read. However, basically, it is not a problem.

PC's pretty much automatically select asynchronous transfer when using Kernel Streaming and WASAPI modes but some programs allow you to select Sync or Async.

There was a poster a while back called Marlene (I believe) and he was measuring frequency response, noise, and distortion with various cables. Eventually, he discovered that these results were largely based on a ground loop problem which was eliminated by running his laptop on battery. I found it noteworthy because, it was an example where environmental factors were reducing performance and the results differed with the cable selection.

- Rich

[Casey Leedom]

Thanks KeithL, most of that I knew from your previous messages and my own background in Computer and Network Systems — but please continue with your lengthly and incredibly educational messages!

So let me focus on the parts that I'm trying to tease out:

Isochronous versus Asynchronous Isochronous:
Thanks for the clarification on the terms.

As I understand it, Isochronous refers to the Source sending Digital Data in a continuous stream of smallish Packets, and Clocking them out on the USB Link in frequency which the Sink needs the Data. The Sink can either simply depend on the Data arrival for its Clock to send that to the DAC (which is thus subject to Jitter concerns), or buffer it up and reClock the Data for consumption by the DAC.

And Asynchronous Isochronous refers to the Source sending Digital Data in somewhat larger Packets as the Sink requests more Data. The Sink in this case must have a Clock in order to properly Clock the buffered Data to the DAC.

USB Power Noise:
As I understand this, a good DAC would have Galvanic Isolation to prevent this. If not, then it might be reasonable to consider a special "USB Regenerator" which presumably would have a much cleaner Power Signal for the DAC, but honestly, this just sounds like a badly designed DAC.

Errors:
Yes, there are very few higher level Data Formats (like PCM, DSD, etc.) which contain their own Error Correction. Most Application Data Formats depend on the underlaying Data Transport Technologies to assure Digital Data copied from Point A to Point B isn't corrupt. For instance, at a UNIX Shell prompt typing "cp A B", most of us don't worry whether the File B is a non-corrupt copy of the File A. We unthinkingly depend on a whole array of Error Detection/Correction Technologies which range from Disk Sector CRCs to Parity or ECC Memory. And of course the example you offered of the Red Book CD use of layers of Reed-Solomon Error Correction.

But in this case, USB is the underlaying Transport Technology and it does offer Error Detection via a CRC on the Packets. Error Correction occurs when the Sink fails to send an ACK for a Corrupt Packets and the Source must then resend that Packet.

The other error which could occur is that the Audio Digital Data simply doesn't arrive at the Sink soon enough to satisfy the DACs needs to drive its Digital-to-Analog circuitry. I.e. the DAC is sucking on a [temporarily] empty straw.

What I was wondering about is: are there any DACs which display these USB Transport issues? On a UNIX system for instance, I can type "ifconfig {ethN}" or "ethtool -S {ethN}" and I can typically see listed various RX and TX Error Counters.

And again, I know that these are typically not an issue. But I see articles like the one I referenced where people are slaving away over bits of the system which should be trivially correct. These are fundamentally solved problems. Otherwise we wouldn't trust Computer/Network Systems with our Bank and Medical Records. And yet, we still see people spending large sums of money on Audiophile-grade USB/Ethernet Cables without any objective reason (like observing error counts, etc.). There really are only a few things which can go wrong upstream of the DAC and it would be nice to have some way on knowing quantitatively that things are working/not working. Anything to shoot down the proponents of Green Felt Tip Markers ...

Casey

[KeithL]

ATMOS - ABOUT CHANNELS AND OBJECTS

I still seem to be seeing an awful lot of confusion about speakers, and channels, and objects.

There also seems to be significant confusion with the labeling on discs - which doesn't help matters.

The three numbers (for example 7.1.4) refer to the number of SPEAKERS you have.

So 7.1.4 means:

- 7 regular speakers (fronts plus surrounds)

- 1 subwoofer

- 4 height speakers

This has NOTHING WHATSOEVER to do with how many OBJECTS are used in the Atmos sound track itself.

You can have one object assigned to three speakers, or three objects assigned to one speaker; the decoder looks at the xyz coordinate of each object and assigns it to the appropriate speakers.

If you're used to virtualization this all makes sense - if not, then it can be sort of confusing.

Your Atmos system has two types of "speaker things" - bed channels and speakers.

Bed channels are the standard surround channels we're used to - like Front Left and Right Surround (in a home system each bed channel is usually going to be one speaker - in a theater it may be "all the speakers in the front left corner").

In addition to bed channels, each individual speaker you have is individually addressable, and counts as its own channel.

(I most home systems, your non-height speakers are BOTH bed channels and channels, and your height speakers are channels - but not bed channels.)

The sound track on an Atmos DISC has two different sorts of "audio things" on it.

One is the bed channels - which is the exact equivalent of what you had before with something like TrueHD.

The other are separate things called sound objects; each object has an xyz coordinate and a size.

Audio in each bed channel is always played through the bed speaker that goes with that channel (so something in the Left Front bed channel plays through your Left Front speaker).

Objects are assigned to one or more speakers (the decoder decides which speakers each object should play through to end up in the proper xyz coordinates).

THERE IS NO 1:1 CORRELATION BETWEEN OBJECTS AND SPEAKERS.

If you have a 7.1.4 system then you have seven bed channels - which will end up playing the seven bed channels on the disc.

However, you could have 426 different objects.... one over here, one over there, one up there, and one that starts here and zooms back and forth like Tinkerbell after too many cups of coffee.

The object limitation has nothing at all to do with ow many speakers you have.

In fact, any Atmos disc can have any number of objects at all.

However, the number of objects that can be in use SIMULTANEOUSLY is limited by how many objects your decoder can keep track of.

The decoder used in the cinema version of Atmos is more powerful than most home processors, and so can keep track of more objects (and theaters tend to be a lot bigger and have more speakers).

Therefore, the Atmos version used on discs intended for home use is limited to a smaller number of objects (the two standards are slightly different).

(As a result, when a cinema Atmos track is converted for home use, if there are too many objects, some may have to be combined together as part of the remixing process.)

There is no specific reason why a disc would be labeled as "7.1.4" when it must comply with the "Atmos for the home standard", which can be played on a home Atmos system utilizing any number of speakers.

I suspect that, in response to all the confusion, some disc makers have simply come to label certain discs as "optimized to play on a home system with a 7.1.4 speaker configuration".

And what they would really be saying is that, since they've used height objects, and perhaps even some height objects that move from front to back, that disc will BENEFIT from being played on a 7.1.4 system.

In other words, they're essentially saying: "This disc was optimized to be played on a 7.1.4 system and will sound best when played on such a system".

However, any Atmos disc can be played on any Atmos system.

And, as I mentioned in another post, especially if starting with a movie that wasn't mastered with Atmos in mind, it's reasonable to assume that certain mastering engineers might choose to "pin objects to the height speakers".

(Each Atmos object can be positioned, sized, and moved dynamically.... which is very flexible.... but is also a lot of work. In your modern war movie, you might choose to assign each bullet whizzing around overhead as an object.

However, if you're re-mixing an old movie, and you have two tracks of "overhead bullet noises", it's probably easier just to assign one "full size top left" and the other "full size top right" and call it a day.

In order to have each bullet fly in a different direction, you would have to separate them into individual objects, and then assign each its own path... which would be a lot of extra work.)

As I understand it, the way the compatibility works is that all of those Atmos objects are combined into the bed channels, which are what plays on a standard TrueHD system.

That way you don't miss anything if you play the disc in TrueHD.

Then, based on the instructions contained in the metadata, the Atmos decoder "pulls the individual objects out of the mix" and applies the object localization to them.

May 21, 2018 21:19:53 GMT -5 lrobertson said:

24.1.10 is more accurate if you were to assign Atmos a channel layout with at least the real Atmos. 7.1.4 is 7.1.4. Atmos is 24.1.10 all the way to 2.1. So am I really not going to see any new movies in the future released in Atmos? That’s what everyone is making it sound like. Would a director be able to push for real Atmos?

[cwt]

May 24, 2018 8:54:56 GMT -5 mgbpuff said:

AT least Emotiva has shown the back of the RMC-1 for at least two years now with 16+ balanced outputs to line up with their claims. The company site photo of the NAD you're talking about only shows the standard 7.1 in balanced outputs. They're even behind Emotiva. (plus, they're European)!

Yes; that's there achilles heel ; they are hobbled by a design that dates before atmos and so you get anachronistic analog video switching and 4 miniature xlr's for the 4 ceiling speakers crammed in there 1st time ive ever seen such small connectors ; certainly quite hard to spot amongst the redundancy

[KeithL]

To answer your short question first... I don't recall ever seeing a DAC that reported packet statistics or detectable errors (although I wouldn't be surprised if a few do exist).

Exactly correct about asynchronous... with one slight qualification. Virtually ALL non asynch USB connections rely on the sink device locking onto and regenerating the source clock. While a buffer seems like a viable alternative, it turns out not to be. The problem is that the source may not only have lots of jitter - but may also run at a significantly different speed that what it should. Therefore, if you use a buffer, you run the risk of its overflowing or running dry. Therefore, in order to use a buffer, it would have to be a very large buffer, and it would take quite a long time to fill up partway (remember that, in order to have extra data, you have to wait for some of it to accumulate in the buffer... at regular speed.) The standard solution is to lock onto the incoming packet boundaries using something like a phase-locked loop, which work OK, but not great.Also, since USB is packet-based, it is inherently jumpy in time (it's not even fair to say it has jitter when the data is deliberately broken into discrete packets which much be reassembled at the other end.) Asynch USB interfaces use a buffer.... but, in that case, the sink gets to request more data only when it needs it, making it much easier to keep a small buffer partly full but not have it overflow.

Galvanic isolation is the best way to avoid power and ground plane noise. However, not all DACs offer it, and, to be quite blunt, most people don't need it. From my experience, ground noise tends to be a problem with USB powered DACs, especially when you connect them to an AC powered stereo system, but I've never seen it to be a problem with any DAC that plugs into the wall (it certainly could be - it just usually isn't). The other thing is that noise is generally obvious. You literally hear little chortling noises, vaguely related to what's displayed on your computer screen, in the quiet spots between songs... or you hear it in your speakers when your air conditioner switches on and off. I see far too many vendors suggesting that "even though you don't hear any noise, it's there - and it's secretly degrading the sound of your equipment". I suppose it's possible, but I've never seen that to be the case. HOWEVER, don't confuse "USB Regenerators" with solutions for USB power and ground noise. Although there are devices that do both, the main purpose of a "USB regenerator" is to regenerate the data packets - which is exactly what a USB hub does. So, in other words, a USB regenerator is a fancy USB hub. Some of them ALSO feature various sorts of power filtering and power isolation.

Although most storage formats, like files and CDs, and USB packets, have error correction, the basic PCM data stream itself does not (you can change a bit in a packet and nothing will know).

In theory the DAC could keep track of things like bad packets, but I don't recall ever seeing a DAC that monitors or reports on any of this.

I've always thought it would be a cool idea to put a little green "perfect data" LED on a CD transport that would stay lit as long as no unrecoverable errors occurred.... but I've never seen one.

I once owned an expensive USB-to-S/PDIF converter that offered a "bit perfect data test".... but it involved actually playing a special test file, on your playback software, which the device would then verify had arrived without alteration.

I believe a few DACs may also do so.... but it's a run-once diagnostic process, intended to confirm that your player and O/S aren't deliberately altering the data, rather than a continuous monitoring for errors.

I should also mention ASRCs (asynchronous sample rate converter) - which many modern DACs include.

An ASRC was designed to convert data from one sample rate to another.

It does so by "locking onto" the original data, and a new output clock, then resampling the data to match the new clock.

The internal mechanism includes a remarkably powerful dedicated DSP.

In terribly oversimplified terms: "it locks onto the incoming data, and creates new data, synched to the new output clock, that is equivalent to the original data".

However, since it uses the digital equivalent of a phase-locked loop to lock onto the incoming data, it is able to "ignore" any jitter that may be present.

In functional terms, it takes the incoming data, removes any jitter that may be present, and regenerates it, referenced to a new local clock.

(The ASRC works on PCM data, so it works on all inputs, not just USB.)

While technically the ASRC doesn't remove ALL jitter, it does reduce it by a minimum of 30-40 dB, and a lot more over some frequency ranges, which is darn close.

To keep the Green Felt Tip Pen Cult at bay, just point to the result of a good CD rip....... which proves that all of the bits are perfect (remembering that jitter doesn't exist for the contents of a file). 

May 24, 2018 11:48:13 GMT -5 Casey Leedom said:

Thanks KeithL , most of that I knew from your previous messages and my own background in Computer and Network Systems — but please continue with your lengthly and incredibly educational messages!

So let me focus on the parts that I'm trying to tease out:

Isochronous versus Asynchronous Isochronous:
Thanks for the clarification on the terms.

As I understand it, Isochronous refers to the Source sending Digital Data in a continuous stream of smallish Packets, and Clocking them out on the USB Link in frequency which the Sink needs the Data. The Sink can either simply depend on the Data arrival for its Clock to send that to the DAC (which is thus subject to Jitter concerns), or buffer it up and reClock the Data for consumption by the DAC.

And Asynchronous Isochronous refers to the Source sending Digital Data in somewhat larger Packets as the Sink requests more Data. The Sink in this case must have a Clock in order to properly Clock the buffered Data to the DAC.

USB Power Noise:
As I understand this, a good DAC would have Galvanic Isolation to prevent this. If not, then it might be reasonable to consider a special "USB Regenerator" which presumably would have a much cleaner Power Signal for the DAC, but honestly, this just sounds like a badly designed DAC.

Errors:
Yes, there are very few higher level Data Formats (like PCM, DSD, etc.) which contain their own Error Correction. Most Application Data Formats depend on the underlaying Data Transport Technologies to assure Digital Data copied from Point A to Point B isn't corrupt. For instance, at a UNIX Shell prompt typing "cp A B", most of us don't worry whether the File B is a non-corrupt copy of the File A. We unthinkingly depend on a whole array of Error Detection/Correction Technologies which range from Disk Sector CRCs to Parity or ECC Memory. And of course the example you offered of the Red Book CD use of layers of Reed-Solomon Error Correction.

But in this case, USB is the underlaying Transport Technology and it does offer Error Detection via a CRC on the Packets. Error Correction occurs when the Sink fails to send an ACK for a Corrupt Packets and the Source must then resend that Packet.

The other error which could occur is that the Audio Digital Data simply doesn't arrive at the Sink soon enough to satisfy the DACs needs to drive its Digital-to-Analog circuitry. I.e. the DAC is sucking on a [temporarily] empty straw.

What I was wondering about is: are there any DACs which display these USB Transport issues? On a UNIX system for instance, I can type "ifconfig {ethN}" or "ethtool -S {ethN}" and I can typically see listed various RX and TX Error Counters.

And again, I know that these are typically not an issue. But I see articles like the one I referenced where people are slaving away over bits of the system which should be trivially correct. These are fundamentally solved problems. Otherwise we wouldn't trust Computer/Network Systems with our Bank and Medical Records. And yet, we still see people spending large sums of money on Audiophile-grade USB/Ethernet Cables without any objective reason (like observing error counts, etc.). There really are only a few things which can go wrong upstream of the DAC and it would be nice to have some way on knowing quantitatively that things are working/not working. Anything to shoot down the proponents of Green Felt Tip Markers ...

Casey

[KeithL]

I also use dBPowerAmp........

Honestly, as I recall, out of about 500 CDs, I had three "hard errors".

Two turned out to be just plain damaged discs, which were scratched badly enough that the error correction was unable to correct for them.

One worked after being well cleaned; the other I replaced.

One turned out to be an error in the disc master; I bought another copy - which had the same data error in the same location.

CDs are replicated optically, in a process much like how vinyl albums are pressed, and so physical defects in the data surface on the master are possible.

I've also seen many discs where, out of a dozen songs, some of them match checksums with twenty or thirty others, while five or seven only match half that number.

Since there were still multiple matches, I assume that there are two different versions of the physical master, and some of the songs are not identical between them.

Note that Accurate RIP doesn't look for disc errors per-se.... it compares a checksum on your rip to the checksum other users got when ripping the same album....

This means that, if the actual master was damaged or imperfect, it's quite possible that other folks got discs made from the same master, with identical flaws.

The two levels of hardware error correction on a CD drive are internal... most drives don't actually report how many recoverable and unrecoverable errors they received.

May 24, 2018 11:39:26 GMT -5 richb said:

I use DBPowerAmp for ripping. There are times where the checksum numbers do not agree and others fail and fall back to secure rip mode. So, my experiences is not that they are perfect if read. However, basically, it is not a problem.

PC's pretty much automatically select asynchronous transfer when using Kernel Streaming and WASAPI modes but some programs allow you to select Sync or Async.

There was a poster a while back called Marlene (I believe) and he was measuring frequency response, noise, and distortion with various cables. Eventually, he discovered that these results were largely based on a ground loop problem which was eliminated by running his laptop on battery. I found it noteworthy because, it was an example where environmental factors were reducing performance and the results differed with the cable selection.

- Rich

[mgbpuff]

May 24, 2018 13:24:16 GMT -5 cwt said:

May 24, 2018 8:54:56 GMT -5 mgbpuff said:

AT least Emotiva has shown the back of the RMC-1 for at least two years now with 16+ balanced outputs to line up with their claims. The company site photo of the NAD you're talking about only shows the standard 7.1 in balanced outputs. They're even behind Emotiva. (plus, they're European)!

Yes; that's there achilles heel ; they are hobbled by a design that dates before atmos and so you get anachronistic analog video switching and 4 miniature xlr's for the 4 ceiling speakers crammed in there 1st time ive ever seen such small connectors ; certainly quite hard to spot amongst the redundancy

I see the mini balanced connectors now. But I don't see the same number of RCA pre-outputs. Wierd! Why mix two different types of balanced outputs?

[Bonzo]

May 24, 2018 12:02:52 GMT -5 KeithL said:

ATMOS - ABOUT CHANNELS AND OBJECTS

I still seem to be seeing an awful lot of confusion about speakers, and channels, and objects.

There also seems to be significant confusion with the labeling on discs - which doesn't help matters.

The three numbers (for example 7.1.4) refer to the number of SPEAKERS you have.

So 7.1.4 means:

- 7 regular speakers (fronts plus surrounds)

- 1 subwoofer

- 4 height speakers

This has NOTHING WHATSOEVER to do with how many OBJECTS are used in the Atmos sound track itself.

You can have one object assigned to three speakers, or three objects assigned to one speaker; the decoder looks at the xyz coordinate of each object and assigns it to the appropriate speakers.

If you're used to virtualization this all makes sense - if not, then it can be sort of confusing.

Your Atmos system has two types of "speaker things" - bed channels and speakers.

Bed channels are the standard surround channels we're used to - like Front Left and Right Surround (in a home system each bed channel is usually going to be one speaker - in a theater it may be "all the speakers in the front left corner").

In addition to bed channels, each individual speaker you have is individually addressable, and counts as its own channel.

(I most home systems, your non-height speakers are BOTH bed channels and channels, and your height speakers are channels - but not bed channels.)

The sound track on an Atmos DISC has two different sorts of "audio things" on it.

One is the bed channels - which is the exact equivalent of what you had before with something like TrueHD.

The other are separate things called sound objects; each object has an xyz coordinate and a size.

Audio in each bed channel is always played through the bed speaker that goes with that channel (so something in the Left Front bed channel plays through your Left Front speaker).

Objects are assigned to one or more speakers (the decoder decides which speakers each object should play through to end up in the proper xyz coordinates).

THERE IS NO 1:1 CORRELATION BETWEEN OBJECTS AND SPEAKERS.

If you have a 7.1.4 system then you have seven bed channels - which will end up playing the seven bed channels on the disc.

However, you could have 426 different objects.... one over here, one over there, one up there, and one that starts here and zooms back and forth like Tinkerbell after too many cups of coffee.

The object limitation has nothing at all to do with ow many speakers you have.

In fact, any Atmos disc can have any number of objects at all.

However, the number of objects that can be in use SIMULTANEOUSLY is limited by how many objects your decoder can keep track of.

The decoder used in the cinema version of Atmos is more powerful than most home processors, and so can keep track of more objects (and theaters tend to be a lot bigger and have more speakers).

Therefore, the Atmos version used on discs intended for home use is limited to a smaller number of objects (the two standards are slightly different).

(As a result, when a cinema Atmos track is converted for home use, if there are too many objects, some may have to be combined together as part of the remixing process.)

There is no specific reason why a disc would be labeled as "7.1.4" when it must comply with the "Atmos for the home standard", which can be played on a home Atmos system utilizing any number of speakers.

I suspect that, in response to all the confusion, some disc makers have simply come to label certain discs as "optimized to play on a home system with a 7.1.4 speaker configuration".

And what they would really be saying is that, since they've used height objects, and perhaps even some height objects that move from front to back, that disc will BENEFIT from being played on a 7.1.4 system.

In other words, they're essentially saying: "This disc was optimized to be played on a 7.1.4 system and will sound best when played on such a system".

However, any Atmos disc can be played on any Atmos system.

And, as I mentioned in another post, especially if starting with a movie that wasn't mastered with Atmos in mind, it's reasonable to assume that certain mastering engineers might choose to "pin objects to the height speakers".

(Each Atmos object can be positioned, sized, and moved dynamically.... which is very flexible.... but is also a lot of work. In your modern war movie, you might choose to assign each bullet whizzing around overhead as an object.

However, if you're re-mixing an old movie, and you have two tracks of "overhead bullet noises", it's probably easier just to assign one "full size top left" and the other "full size top right" and call it a day.

In order to have each bullet fly in a different direction, you would have to separate them into individual objects, and then assign each its own path... which would be a lot of extra work.)

As I understand it, the way the compatibility works is that all of those Atmos objects are combined into the bed channels, which are what plays on a standard TrueHD system.

That way you don't miss anything if you play the disc in TrueHD.

Then, based on the instructions contained in the metadata, the Atmos decoder "pulls the individual objects out of the mix" and applies the object localization to them.

May 21, 2018 21:19:53 GMT -5 lrobertson said:

24.1.10 is more accurate if you were to assign Atmos a channel layout with at least the real Atmos. 7.1.4 is 7.1.4. Atmos is 24.1.10 all the way to 2.1. So am I really not going to see any new movies in the future released in Atmos? That’s what everyone is making it sound like. Would a director be able to push for real Atmos?

Thanks Keith. I think this is the best overview for dummies I've read on this subject.

Now onto some stupid Bonzo questions.

So if you have say, more than 7 bed speakers, say 9 for example, you would need a processor that recognized that. And you would need a disc labled 9.1.4 for it to be descrete, not derived. Correct.

So its totally impossible for the .4 at the end to be bed channels? There is no way an Atmos system could be 7.1.4 with 7 surround beds and 4 height beds? There just seems to be all sorts of confusion being made by discs getting labeled this way.

And one last stupid question, wouldn't this all work best if you could tell the processor exactly where your speakers were? Put them where you want and then feed the processor xyz coordinates for each speaker? Is that impossible? Or does it just require a schiit load more processing?

Sorry for the dumb questions. If I could ever get around to hooking my Marantz up I'd probably answer all my own questions.

Cheers.

[Gary Cook]

My understanding is that the bed channels are Dolby True HD which means 7.1. That anything more than 7.1 is an object and resides in the meta data that sits over the Dolby True HD track. For 7.1.4 movies (the vast majority available) the meta data contains the X, Y, Z coordinates of the .4. That's it, if your receiver/processor and set up is capable of more than 7.1.4 then that is matrixed using an algorithm (not unlike PL11z), which varies between manufacturers. In simple terms on a 7.1.4 disc there are no X,Y,Z coordinates for more than 7.1.4.

The above is based on my experiences with 6 x receivers/processors so far (Yamaha, Denon, Marantz and Onkyo), with set ups ranging from 5.1.2 to 11.4.6 and over 20 movies.

Cheers
Gary

[cwt]

May 24, 2018 14:48:38 GMT -5 mgbpuff said:

May 24, 2018 13:24:16 GMT -5 cwt said:

Yes; that's there achilles heel ; they are hobbled by a design that dates before atmos and so you get anachronistic analog video switching and 4 miniature xlr's for the 4 ceiling speakers crammed in there 1st time ive ever seen such small connectors ; certainly quite hard to spot amongst the redundancy

I see the mini balanced connectors now. But I don't see the same number of RCA pre-outputs. Wierd! Why mix two different types of balanced outputs?

Can only guess its to do with the topology of the original 7.1 circuitry and what would fit on the crowded back panel ? Yes the pre outs ; Too bad if your power amps were single ended ; unless there were some obscure mini xlr balanced to unbalanced cables ? maybe supplied or possibly they want to sell you one of their 4ch power amps with the same proprietary type connectors ? guessing .

[liv2teach]

Alright, I don't expect a parade or any thanks, but I'm sure I've insured the release of the RMC very soon. I broke down and bought a 7.1 processor off Audiogon. With the way my timing always works out, they zig when I zag, I'm sure now that I've taken that step...the RMC should be available any day now!

You're welcome!

[sveb]

May 25, 2018 1:22:42 GMT -5 cwt said:

May 24, 2018 14:48:38 GMT -5 mgbpuff said:

I see the mini balanced connectors now. But I don't see the same number of RCA pre-outputs. Wierd! Why mix two different types of balanced outputs?

Can only guess its to do with the topology of the original 7.1 circuitry and what would fit on the crowded back panel ? Yes the pre outs ; Too bad if your power amps were single ended ; unless there were some obscure mini xlr balanced to unbalanced cables ? maybe supplied or possibly they want to sell you one of their 4ch power amps with the same proprietary type connectors ? guessing .

They have adapters included.

[mahuzz13]

May 24, 2018 22:38:44 GMT -5 Gary Cook said:

My understanding is that the bed channels are Dolby True HD which means 7.1. That anything more than 7.1 is an object and resides in the meta data that sits over the Dolby True HD track. For 7.1.4 movies (the vast majority available) the meta data contains the X, Y, Z coordinates of the .4. That's it, if your receiver/processor and set up is capable of more than 7.1.4 then that is matrixed using an algorithm (not unlike PL11z), which varies between manufacturers. In simple terms on a 7.1.4 disc there are no X,Y,Z coordinates for more than 7.1.4.

The above is based on my experiences with 6 x receivers/processors so far (Yamaha, Denon, Marantz and Onkyo), with set ups ranging from 5.1.2 to 11.4.6 and over 20 movies.

Cheers
Gary

Thanks Gary for that explanation
My question is if you have a set up that is 9.x.6 will the RMC-1 do that processing to all the bed channels including wides and the 6 atmos no matter what the disc says on the back as long as the content is there to process?

[lrobertson]

I feel like Keith was suggesting that movies not really intended to be Atmos will have less than ideal object sound engineering. I’ve never been to an Atmos theater as there aren’t any around where I live but has anyone experienced one of these so called 7.1.4 labeled movies in an Atmos cinema and know that the theater made use of Atmos as intended? If the cinema version had a perfect Atmos and the uhd has a undermined version then I’d still be a little worried going forward. If the studio is just implementing a poor Atmos in general then they just don’t have respect for the need to make a piece of art over creating another box office movie that will sell well regardless of sound. Like he said it requires time and dedication to engineer the object sounds as needed and frankly looking at the movies like Star Wars and marvels that are geared towards young people the demand for works of art are probably less and corners might be cut by the studios and/or directors. I’m going to side with Keith here and expect the movies that are worth buying for me are naturally going to be the movies that take the time for real Atmos. The ones that might pin their objects to cut corners are the ones I’d probably rent anyways and would never considering watching it twice. It is disappointing but to expect perfect sound mixes on all types of movies doesn’t sound practical either. As far as moving past 7.1.4 with a 7 bed channel layout it still being misrepresented by some as not intended. The bed channels are meant to be in speaker arrays. It’s not a drawback of expanding. It’s based off of psychoacoustic principals to create diffuse soundstages. The more speakers playing the same material across an array the better. As long as there are no pinned objects then expanding past 7.1.4 is the job of the sound engineer to implement the right object code and to maintain the correct type of diffuse material in the bed channels. A bad 24.1.10 movie Atmos implementation would be nothing more than a bad sound engineer or obvious corner cutting for costs. Until we see a Saving Private Ryan type of movie released with pinned objects in a 7.1.4 format we should really drop the issue of no demand for 9.1.6+ because you’d be placing more importance for the kids movies over the masterpieces for all people. No point in acting like there is no point for the 24.1.10 systems till we all see it for the movies that count for people who would put that kind of money towards those systems. Something tells me marvel and Disney isn’t their target media.

[AudioHTIT]

May 24, 2018 22:38:44 GMT -5 Gary Cook said:

My understanding is that the bed channels are Dolby True HD which means 7.1. That anything more than 7.1 is an object and resides in the meta data that sits over the Dolby True HD track. For 7.1.4 movies (the vast majority available) the meta data contains the X, Y, Z coordinates of the .4. That's it, if your receiver/processor and set up is capable of more than 7.1.4 then that is matrixed using an algorithm (not unlike PL11z), which varies between manufacturers. In simple terms on a 7.1.4 disc there are no X,Y,Z coordinates for more than 7.1.4.

The above is based on my experiences with 6 x receivers/processors so far (Yamaha, Denon, Marantz and Onkyo), with set ups ranging from 5.1.2 to 11.4.6 and over 20 movies.

Cheers
Gary

So at least with Atmos we’ll get more 7.1 disks!

[lrobertson]

That’s what a lot here would like I feel. I remember the how many channels RMC-1 should be thread and 7 was very common with a lot on the forum as if the XMC-1 wasn’t going to fulfill the 7 channel audio bliss. Now it seems people are still pushing to limit things to what they see practical in implementing themselves. Bed layers in Atmos are not 7.1 channel discrete content. 7.1.4 uhd’s might be different but bed layers in Atmos are 24 speakers laid out in arrays. At this point suggesting otherwise is borderline trolling unless there is still genuine confusion. Because it’s backwards compatible to sample down to those who chose 7 channels or less doesn’t make its original intended 24 lower channels any less intentional in the mixing process.

[Gary Cook]

May 25, 2018 11:07:11 GMT -5 lrobertson said:

That’s what a lot here would like I feel. I remember the how many channels RMC-1 should be thread and 7 was very common with a lot on the forum as if the XMC-1 wasn’t going to fulfill the 7 channel audio bliss. Now it seems people are still pushing to limit things to what they see practical in implementing themselves. Bed layers in Atmos are not 7.1 channel discrete content. 7.1.4 uhd’s might be different but bed layers in Atmos are 24 speakers laid out in arrays. At this point suggesting otherwise is borderline trolling unless there is still genuine confusion. Because it’s backwards compatible to sample down to those who chose 7 channels or less doesn’t make its original intended 24 lower channels any less intentional in the mixing process.

From a sample base of 1, there's not a lot of Atmos equiped movie theatres in Australia, what you posted is right but wrong at the same time. It had;
4 speakers at the front, 2 left and 2 right, that's 2 bed channels
2 speakers behind the screen centre, that another bed channel
8 speakers with 4 along each side, that's another 2 bed channels
4 speakers across the rear, 2 each side, that's another 2 bed channels

That's 7 bed channels, 18 speakers, in 7 arrays with each speaker in the array passing the same signal. I could stick my head next to each speaker in the array and they sounded the same, same effect, same time, same volume.

There were also;
4 height speakers at the front, 1 left and 1 right and one pair above. I'd guess that's 2 or maybe 3 object channels.
12 ceiling speakers, which appeared to be grouped in pairs, but too high for me to be sure by listening. Based on that guess 6 object channels.
4 subwoofers at the front, 2 at the side and 2 at the rear, they sounded the same, same effect, same time, different volume. Hard to tell if they were grouped in 1 or 2 arrays.

My maths adds that to;
40 speakers arranged in 17 (maybe 18) arrays
7 bed channels
1 (maybe 2) LFE channels
9 (maybe 10) object channels

The movie was Wonder Woman and some time later played on my friends system with a disc labelled 7.1.4 it had 7 bed channels, 1 LFE channel and 4 object channels using a Yamaha CX-A5100 processor. The cinema mix was different to the HT mix, noticeably more effects from the sides and slightly more from the ceiling, about the same from the rear. Obviously it's an action movie so it benefits greatly from directional sound effects.

Cheers
Gary

[lrobertson]

I thought it was discussed that speaker limits will also limit the amount of objects able to be panned between speakers because speakers can only effectively pan so many before losing there effect so objects need to be spaced out between more speakers for more objects... yeah the 7.1.4 label thing does have me stumped as trinnov users said they lost wide support... So how the cinema compares to the ht version is very interesting to me. Like if it is really 7.1.4 is it still object based between all lower speakers like normal Atmos beds or are all individually pinned object speakers? And did they use a similar format for even your 40 speaker theater?... But that being said it really does sound like Atmos for home theater’s standards as Keith put it is here to stay. I’m not ready to jump on the assumption that we will see new 7.1.4 limits automatically placed on all media. I could see this as Atmos’s attempt to not lose business to DTS:X. For those movies that aren’t deemed necessary for the Atmos dedicated masterpiece which costs more and is more complicated to author. Like say if the target audience is more for kids in order to be more competitive with DTS:X they release a format that allows for 3d audio but much easier to author when in reality the studio or director wouldn’t care to go past that. I understand there is streaming formats talk too. I’d just like myself and everyone to speculate less until we actually see all the movies that should be a masterpiece Atmos come to us degraded. We probably just need to assume that if a movie comes to us in 7.1.4 Atmos then it could of just as easily been in DTS:X or 7 channel just because whoever made it that way didn’t see the point in dedicating time to making it sound as good. Marvel and Disney will sell tickets and media no matter the audio format. A movie trying to recreate reality as much as possible as it’s selling point whether it’s a war flick or not will probably be done to utilize the whole speaker layout and Atmos ability.

[Casey Leedom]

By the way, I assume that if one explicitly selects a Lossless 71 Dolby Soundtrack on a 5.1 system, that most Processors will simply direct the Rears to the Side Surrounds, right? I just picked up the new Matrix UHD Remaster and the AVS Forum Review say that the 7.1 Dolby Lossless Soundtrack is way better than the default Lossy 5.1 Soundtrack and all I have is a 5.1 system.

** NOTE: The default audio track is the lossy Dolby Digital 5.1 mix. Selection of the Dolby Atmos/TrueHD 7.1 channel track has to be done manually.**

Of course, I'm not sure what would happen if I tried to play that Dolby Atmos/TrueHD 7.1 channel track on my old DMC-1 ...

Casey

[KeithL]

You've got it almost right - but you're looking at it sort of sideways.

There are seven bed CHANNELS; that number is defined by the standard and the system itself; it is NOT specifically tied to how many speakers you have.
Bed channels are separate discrete channels that are stored in the mix.
However, each bed channel is assigned to an array of speakers (which could be one or multiple speakers).

Let's try a simple example......
And let's just look at the front of the room......

Let's assume we have a very large but more or less square room.
Across the front of the room we have NINE speakers... numbered, from left to right, 1 thru 9.
(1)    (2)    (3)    (4)    (5)    (6)    (7)    (8)    (9)

We might:
- assign speakers 1 - 3 to the left front bed channel
- assign speakers 4 -6 to the front center bed channel
- assign speakers 7 - 9 to the right front bed channel

And, when I do my mix:
- the track I mix to the left front bed channel will play through all three speakers in the left front bed array
- the track I mix to the center front bed channel will play through all three speakers in the center front bed array
- the track I mix to the right front bed channel will play through all three speakers in the right front bed array

HOWEVER, in addition to being part of the bed arrays, each of those speakers can also be addressed individually by the renderer at the same time.

So, if I assign a "small size object located at ground level just left of center front" it may play from only speaker (4).
And, if I make that object bigger, now it may play from speakers (3), (4), and (5).

And, at the same time, a DIFFERENT object might be playing from speakers (2)+(3).
And, a few seconds later, that object may move slowly to the right, going from (2)+(3) to (3) to (3)+(4) to (4) to (4)+(5).
And then it might grow in size until it's playing from (4)+(5)+(6)+(7)+(8).
However, it's still one object.

In a full theater setup, an array of speakers will be assigned to each bed channel (so, since there are seven bed channels there will be seven bed arrays).
In addition, each speaker may individually be addressed by the object renderer.
(It is possible that some speakers may not be assigned to any bed array, but may be used to render objects anyway, which would include all those height speakers.)

So, every Atmos mix has a set of pre-mixed "bed channels" - which are routed to the bed arrays (which, in a home system correspond to your "regular surround speakers").
And, in addition to that, it has a bunch of objects; and each object can be assigned to as many speakers as the renderer deems necessary.
Visually, you can visualize the two systems as being overlaid on top of each other, but independent from each other.

As you can see, the number of bed arrays much match the number of bed channels on the disc.
However, ANY number of objects can be mapped to ANY number of speakers.
(The only limitation there is how many objects can be described in the meta-data and how many your processor can keep track of.)

In a home system, the bed channels will normally correspond to "your regular main and surround speakers".
And, as you might expect, if you only have a 5.1 system, the surround and rear surround bed channels will be combined.

The theatrical system suppots a lot more simultaneous objects than the home system.
So, when a disc is re-mastered for home play, objects may be "aggregated" (combined).
As a result, for example, "the swarm of bees swirling around each other" in the theater may become "the group of bees moving as a group" on the home version.

I've heard that the latest theatrical mastering software may support an additional two "height beds" - but the home system does not.

Bear in mind that "bed channels" are simply channels in the mix which are selected to play from "a particular bed array" - rather than to be positioned specifically by the renderer.
So, in simplest terms, the left front bed channel plays from the left front corner of the room, while individual objects play from whatever speaker or speakers the renderer chooses to send them to.

In a "7.1.4 system" the .4 at the end specifies FOUR HEIGHT SPEAKERS.... it does not specify any particular channels in the mix.
Since a normal home disc doesn't contain any height beds, those four speakers are individual speakers, intended to be used by the renderer to render objects.
(I'm quiet convinced they're simply labelling those discs in that confusing fashion in an attempt to suggest that "this disc will sound best when played on a 7.1.4 system" or something like that.)

In order for this to make total sense, you have to adjust the way you think about channels and sounds.
For example, a single airplane, flying in circles over your head, is a SINGLE OBJECT.... and that object moves around between the four overhead speakers.
From the processor's point of view, that sound is being sent to four speakers, via four amplifier channels.
From Atmos point of view, it is simply one object whose position is changing.
(And, being a rendered object, it doesn't exist in the bed channels.)

Note.... the WAY in which objects and bed channels are stored in a theatrical mix is somewhat different then how they're stored in the home mix.
In the home mix, the rendered objects are merged into the bed channels, enabling them to be played on TrueHD systems... and the Atmos renderer "picks them out and renders them".
However, that needn't concern us as end users of the system.
To us, they "act" just as I described above.

Your last question isn't stupid at all.
According to the original Atmos literature, the system was intended to allow you to do just that....
Place your speakers wherever was convenient, tell the system the xyz coordinates of all your speakers, and let the system figure out how to map everything.
However, that possibility has not been realized, presumably due to a combination of requiring far too much processing power, and the realization that some sort of direction really is necessary.
(After all, you really can't put the speakers wherever you want... just try to make "enveloping surround sound" with seven speakers - all located in the left front corner of the room.)
It really does make more sense to require people to put their main speakers in the traditional locations - which happen to coincide with the intended bed channel locations anyway.

May 24, 2018 19:28:01 GMT -5 Bonzo said:

May 24, 2018 12:02:52 GMT -5 KeithL said:

.....

Thanks Keith. I think this is the best overview for dummies I've read on this subject.

Now onto some stupid Bonzo questions.

So if you have say, more than 7 bed speakers, say 9 for example, you would need a processor that recognized that. And you would need a disc labled 9.1.4 for it to be descrete, not derived. Correct.

So its totally impossible for the .4 at the end to be bed channels? There is no way an Atmos system could be 7.1.4 with 7 surround beds and 4 height beds? There just seems to be all sorts of confusion being made by discs getting labeled this way.

And one last stupid question, wouldn't this all work best if you could tell the processor exactly where your speakers were? Put them where you want and then feed the processor xyz coordinates for each speaker? Is that impossible? Or does it just require a schiit load more processing?

Sorry for the dumb questions. If I could ever get around to hooking my Marantz up I'd probably answer all my own questions.

Cheers.