|
|
Post by Gary Cook on Jun 6, 2018 18:40:42 GMT -5
The movie sound mixer/engineer acquaintance also spent some time explaining to me the differences between cinema Atmos and HT Atmos. There are 2 types of of Dolby Rendering and Mastering Unit (RMU). The Dolby RMU for commercial cinemas and the Dolby RMU for home theatre which is designed for Blu-ray and digital streaming (his HT RMU is a Dell). The 2 RMU's are not interchangeable. I found this quite revealing in that it means to me that we (HT enthusiasts) will not be getting cinema Atmos 23.1.12 or whatever on disc or streamed. The fact is the rendering software for HT consumption is different to the software for cinemas and the rendering hardware (computers/processors) are completely different. As a result no matter, what processor we have in our HT and how many amplifier channels and speakers we have, we are not getting the same Atmos as we experience in a commercial cinema. For sure we can access processors that extrapolate/matrix additional channels, but that's not cinema Atmos. Plus with Dolby's latest decisions regarding 3rd party up mixing even that option is becoming more limited. Cheers Gary |
|
|
|
Post by Bonzo on Jun 6, 2018 20:06:09 GMT -5
Too bad we didn't have nearly ample time (1 hour during EST drive time home?) too copy and paste our very legitimate discussions about the latest on Atmos from the other thread to this one. Now we will be forced to rehash everything. Nice. 🙄
|
|
|
|
Post by mgbpuff on Jun 6, 2018 20:19:36 GMT -5
from FilmMixer on the AVS Atmos thread - "Well, I discussed this in an earlier post.
Snap to speaker is a mixing tool.
If you used “snap to speaker” on an object (and it is a PER OBJECT parameter) a sound panning over head in a circle it would jump around in a system with only 4 overhead locations. That’s not what is being discussed. It was really designed for a theatrical setting where you have access to a large number of speakers (I’ve mixed in rooms with 52 available speakers).
You can play back a home Atmos mix on anywhere from a 2.x.2 to 24.x.10 output system. If you want to capture that output as 5.1.2 or 7.1.4 or 9.1.6 you simply set the rendering engine to the appropriate desired format and capture it (either by doing a linear capture and recording of said output or with other tools.) The overheads have assumed locations within the codec and you can then take those predefined positions and use those numerical value to set the panners for the appropriate objects. The bed channels are already defined positions within the codec. As I postulated earlier you might do this for a number of reasons (workflow, bandwidth, artifact reduction, etc...)
That is one way to do it. I believe there are others either currently available or coming soon in the authoring and mastering software.
You’re confusing a mixing tool (snap to speaker) with a way to output and author an Atmos master.
That’s not a criticism. It is a distinction that wouldn’t be clear to most enthusiasts as a deep understanding of the production workflow isn’t really that important to enjoying and understanding how the home versions works (except to illustrate the differences...)
IMO it’s fairly irrelevant how it is done (rendering to channels...
Here’s where we stand today :
We have seen a handful of titles released that are labeled 7.1.4.
And it’s been confirmed that is what is indeed on said discs.
I haven’t heard of this becoming the new norm for optical discs from most studios."
|
|
|
|
Post by Gary Cook on Jun 6, 2018 21:18:22 GMT -5
from FilmMixer on the AVS Atmos thread - "Here’s where we stand today : We have seen a handful of titles released that are labeled 7.1.4. And it’s been confirmed that is what is indeed on said discs. I haven’t heard of this becoming the new norm for optical discs from most studios." I would debate the "handful" number, more like "multiple handfuls" and growing daily. But what about discs that aren't labelled 7.1.4 but when we play them they are in fact 7.1.4? Does "new norm" refer to only discs labelled 7.1.4 or discs that play back 7.1.4 (even though they aren't labelled 7.1.4)? At the risk of being repetitive, the question I asked weeks ago, does anyone have a disk that actually plays back 9.1.6 (or more)? That's without up mixing, extrapolation or matrixing of course. Cheers Gary |
|
|
|
Post by musicfan on Jun 7, 2018 5:45:27 GMT -5
from FilmMixer on the AVS Atmos thread - "Here’s where we stand today : We have seen a handful of titles released that are labeled 7.1.4. And it’s been confirmed that is what is indeed on said discs. I haven’t heard of this becoming the new norm for optical discs from most studios." I would debate the "handful" number, more like "multiple handfuls" and growing daily. But what about discs that aren't labelled 7.1.4 but when we play them they are in fact 7.1.4? Does "new norm" refer to only discs labelled 7.1.4 or discs that play back 7.1.4 (even though they aren't labelled 7.1.4)? At the risk of being repetitive, the question I asked weeks ago, does anyone have a disk that actually plays back 9.1.6 (or more)? That's without up mixing, extrapolation or matrixing of course. Cheers Gary Like I said. I have OVER 100 atmos discs and only those that were distributed by Disney SAY 7.1.4 on them. And there is NO objective way to tell if it is >7.1.4. It’s a guessing game |
|
|
|
Post by mgbpuff on Jun 7, 2018 6:57:00 GMT -5
I would debate the "handful" number, more like "multiple handfuls" and growing daily. But what about discs that aren't labelled 7.1.4 but when we play them they are in fact 7.1.4? Does "new norm" refer to only discs labelled 7.1.4 or discs that play back 7.1.4 (even though they aren't labelled 7.1.4)? At the risk of being repetitive, the question I asked weeks ago, does anyone have a disk that actually plays back 9.1.6 (or more)? That's without up mixing, extrapolation or matrixing of course. Cheers Gary Like I said. I have OVER 100 atmos discs and only those that were distributed by Disney SAY 7.1.4 on them. And there is NO objective way to tell if it is >7.1.4. It’s a guessing game Those who have Atmos systems that can do 16 channels discrete can tell if it is >7.1.4, and some of them have said that the latest batch from Disney have absolutely NO sounds beyond 7.1.4. That would be those who own Trinnov, or one of the latest Dennon/Marantz 16 channel A/V units. |
|
|
|
Post by musicfan on Jun 7, 2018 7:28:19 GMT -5
Like I said. I have OVER 100 atmos discs and only those that were distributed by Disney SAY 7.1.4 on them. And there is NO objective way to tell if it is >7.1.4. It’s a guessing game Those who have Atmos systems that can do 16 channels discrete can tell if it is >7.1.4, and some of them have said that the latest batch from Disney have absolutely NO sounds beyond 7.1.4. That would be those who own Trinnov, or one of the latest Dennon/Marantz 16 channel A/V units. IM not talking about sticking your ear up to a speaker and if it doesnt have sound you can rule out 9.x.x that is unscientific...cause the first few atmos movies to come out (transformers was the first) it had VERY little x.x.4 sounds coming out....so unless the PROCESSOR shows you whats encoded...there is NO WAY TO ACTUALLY know..... I want OBJECTIVE evidence....not "what my ears tell me" |
|
|
|
Post by mgbpuff on Jun 7, 2018 8:30:40 GMT -5
Those who have Atmos systems that can do 16 channels discrete can tell if it is >7.1.4, and some of them have said that the latest batch from Disney have absolutely NO sounds beyond 7.1.4. That would be those who own Trinnov, or one of the latest Dennon/Marantz 16 channel A/V units. IM not talking about sticking your ear up to a speaker and if it doesnt have sound you can rule out 9.x.x that is unscientific...cause the first few atmos movies to come out (transformers was the first) it had VERY little x.x.4 sounds coming out....so unless the PROCESSOR shows you whats encoded...there is NO WAY TO ACTUALLY know..... I want OBJECTIVE evidence....not "what my ears tell me" So in a live performance with an orchestra that includes 5 trumpets, how would you objectively determine if one trumpeter was faking it? |
|
|
|
Post by thrillcat on Jun 7, 2018 8:34:32 GMT -5
IM not talking about sticking your ear up to a speaker and if it doesnt have sound you can rule out 9.x.x that is unscientific...cause the first few atmos movies to come out (transformers was the first) it had VERY little x.x.4 sounds coming out....so unless the PROCESSOR shows you whats encoded...there is NO WAY TO ACTUALLY know..... I want OBJECTIVE evidence....not "what my ears tell me" So in a live performance with an orchestra that includes 5 trumpets, how would you objectively determine if one trumpeter was faking it? I believe it should be in the metadata section of this evening's printed program. Enjoy the show. |
|
|
|
Post by rbk123 on Jun 7, 2018 8:57:39 GMT -5
IM not talking about sticking your ear up to a speaker and if it doesnt have sound you can rule out 9.x.x that is unscientific...cause the first few atmos movies to come out (transformers was the first) it had VERY little x.x.4 sounds coming out....so unless the PROCESSOR shows you whats encoded...there is NO WAY TO ACTUALLY know..... I want OBJECTIVE evidence....not "what my ears tell me" So in a live performance with an orchestra that includes 5 trumpets, how would you objectively determine if one trumpeter was faking it? His point is the processor can mux up sound, ala Dolby IIx or the various spoof formats, and send it to the speakers. No one wants that, rather they want sound specifically intended/encoded for that speaker. If you hear sound from the speaker, there's no way to tell - by listening - if it's the former or the latter. |
|
|
|
Post by mgbpuff on Jun 7, 2018 9:13:17 GMT -5
So in a live performance with an orchestra that includes 5 trumpets, how would you objectively determine if one trumpeter was faking it? His point is the processor can mux up sound, ala Dolby IIx or the various spoof formats, and send it to the speakers. No one wants that, rather they want sound specifically intended/encoded for that speaker. If you hear sound from the speaker, there's no way to tell - by listening - if it's the former or the latter. Matrixing or discrete decoding is a known characteristic of a given A/V unit. For a discrete > 16 channel Atmos A/V unit, even if one can look through the Atmos file to identify if it contains positional metadata or not, the only way to acertain that a given speaker is participating in an immersive objective sound is to listen for it or hook a meter or scope up to it. |
|
|
|
Post by davidl81 on Jun 7, 2018 9:21:27 GMT -5
So in a live performance with an orchestra that includes 5 trumpets, how would you objectively determine if one trumpeter was faking it? His point is the processor can mux up sound, ala Dolby IIx or the various spoof formats, and send it to the speakers. No one wants that, rather they want sound specifically intended/encoded for that speaker. If you hear sound from the speaker, there's no way to tell - by listening - if it's the former or the latter. In a way isn't Atmos sorta of a matrix by design? The object audio is never designed to go to any specific speaker (ie this helicopter sound is on the top right in ceiling speaker) but the sound is coming from X,Y,Z coordinate. It is then up to your AVR to know where your speakers are actually placed in your specific setup (based on doing the calibration etc) and place the sound in the correct speakers that match the XYZ's given by the metadata. Therefore Atmos object audio is never designed to send sound to a specific speaker, just to a specific part of your room and your AVR through the processing should decide what speaker (or speakers) to place that audio from. |
|
|
|
Post by socketman on Jun 7, 2018 9:27:26 GMT -5
NOt a matrix. meta data tells it where to go, the OAR tells it which speakers to use.
|
|
|
|
Post by Bonzo on Jun 7, 2018 9:33:08 GMT -5
or one of the latest Dennon/Marantz 16 channel A/V units. The last time I checked out a Denon unit, they were still using front screen identification images (little square boxes) to show what channels were coming in and what channels were being played. I wonder if the new ones still have this, and if they would then show the owner what's going on? Just talking out loud. |
|
|
|
Post by davidl81 on Jun 7, 2018 9:36:02 GMT -5
NOt a matrix. meta data tells it where to go, the OAR tells it which speakers to use. Right, it is not a matrix, but the metadata should not call out a specific speaker to play the object sound from (ie DD 5.1 has each specific sound tided to one specific channel), where with Atmos the OAR takes the data and computes which speaker to play out of. It is just similar to a matrix that would take sounds and try to interpret where they should go then place the sound on those specific channels. Atmos is superior in the sense that the OAR does not have to interpret where the sound should come from (it is given that XYZ data from the metadata) it just has to place that sound based on the specific speaker layout of your room. In theory Atmos should make setting up your HT easier since now you would have some flexibility in speaker placement if your room limited you to having speakers in the precise perfect location. The OAR would recognize where your speakers are after calibration and then place the object audio in the right combination of channels to best reproduce the intended sound. |
|
|
|
Post by mgbpuff on Jun 7, 2018 9:50:05 GMT -5
or one of the latest Dennon/Marantz 16 channel A/V units. The last time I checked out a Denon unit, they were still using front screen identification images (little square boxes) to show what channels were coming in and what channels were being played. I wonder if the new ones still have this, and if they would then show the owner what's going on? Just talking out loud. I have a Marantz AV7702 and those identification images only indicate what channels are enabled for the format being used. They do not indicate if sound is actually being delivered to each location. |
|
|
|
Post by Bonzo on Jun 7, 2018 10:03:54 GMT -5
The last time I checked out a Denon unit, they were still using front screen identification images (little square boxes) to show what channels were coming in and what channels were being played. I wonder if the new ones still have this, and if they would then show the owner what's going on? Just talking out loud. I have a Marantz AV7702 and those identification images only indicate what channels are enabled for the format being used. They do not indicate if sound is actually being delivered to each location. I believe Denons readout system is slightly different, or at least it used to be, and that's why I only mentioned Denon. Mine definitely shows whats coming in on the left, and what is going out after processing on the right. The incoming information on the left has nothing to do with what format is being used, it's simply showing what the unit is sensing is coming in. In this way a unit could detect if it's even receiving 7.1.4 or 9.1.6 etc. If the boxes on the left only show 7.1.2 for example, then that's all its getting from the disc player. The boxes on the right really mean nothing for the point of this discussion. |
|
|
|
Post by socketman on Jun 7, 2018 10:07:54 GMT -5
NOt a matrix. meta data tells it where to go, the OAR tells it which speakers to use. Right, it is not a matrix, but the metadata should not call out a specific speaker to play the object sound from (ie DD 5.1 has each specific sound tided to one specific channel), where with Atmos the OAR takes the data and computes which speaker to play out of. It is just similar to a matrix that would take sounds and try to interpret where they should go then place the sound on those specific channels. Atmos is superior in the sense that the OAR does not have to interpret where the sound should come from (it is given that XYZ data from the metadata) it just has to place that sound based on the specific speaker layout of your room. In theory Atmos should make setting up your HT easier since now you would have some flexibility in speaker placement if your room limited you to having speakers in the precise perfect location. The OAR would recognize where your speakers are after calibration and then place the object audio in the right combination of channels to best reproduce the intended sound. Strangely , dolby still has very definite speaker postitions and the direction they should face. I am talking real atmos not some sound bar/bouncy shite . |
|
|
|
Post by musicfan on Jun 7, 2018 10:31:09 GMT -5
IM not talking about sticking your ear up to a speaker and if it doesnt have sound you can rule out 9.x.x that is unscientific...cause the first few atmos movies to come out (transformers was the first) it had VERY little x.x.4 sounds coming out....so unless the PROCESSOR shows you whats encoded...there is NO WAY TO ACTUALLY know..... I want OBJECTIVE evidence....not "what my ears tell me" So in a live performance with an orchestra that includes 5 trumpets, how would you objectively determine if one trumpeter was faking it? one of the worst analogies I have read in some time. you are talking about real life individuals...vs what is ENCODED on a disc. if a speaker is not playing any sound (because you stuck you big hairy ear up to it) does NOT mean the disc was not authored to use that speaker...it just means at that given time nothing is being played out of it.....I have a PLETHORA of 7.1 DD discs that have little to just about nothing coming out of the rear speakers (or even the LFE channel)....should I be skeptical of those mixes as well...because...welp...its what my ears tell me.... NO!!! i go by what my processor is reading out. but to go back to your anaology....if each musician is Micd up....I can go to the control board to see OBJECTIVELY if a sound is coming from his mic or not...if i dont do that its just a guessing game |
|
KeithL
Administrator 
Posts: 10,273
|
Post by KeithL on Jun 7, 2018 11:04:08 GMT -5
I don't find this strange at all.
When you define an object, you basically give it an X,Y,Z coordinate and a size. Because of this you can position an object wherever you want to... and you can move it around dynamically. This is done using what looks very much like a 3D drawing program - and the object appears like "a 3D red beach ball". You can put it wherever you want to, dial in a size, set a motion path, and all such cool things.
However, the ramifications of this are very different depending on the situation.
If I'm CREATING an Atmos sound track, it's very cool to be able to do this. I can take each thing that has its own sound track, make it into an object, put it wherever I want, and move it wherever I want. I don't have to worry about where the speakers are; the Atmos system will sort that all out for me. However, the important thing is that I can only do this if I have the sound of that airplane, all by itself, in its own track, which I can then assign to an object.
If I'm converting an old movie, or if the sound for the movie I'm working with wasn't recorded with this in mind, I may not have that separate track.
I may simply have "a bunch of stuff panned to the back right" and "the sound of the plane, mixed in with all that other stuff, now panned to the front left". If so, then the sound I would LIKE to use for my "object: plane #3" is mixed in with a whole bunch of other stuff. Therefore, I don't have the separate track I need, of just the sound of that airplane, essentially in monaural in a single track, or its own set of tracks, in order to "create" that object.
If that's the case, the easiest thing for me to do is to simply take what used to be "a top left back track" and make it directly into "a single object located immovably at the top back left". And, if I do it that way with all the original "tracks" - simply turn each into an object that never moves - all the original pans will work as expected.
You are basically looking at a subset of the possible features. Just because Atmos objects CAN move around doesn't mean that they have to. It sounds like what you're describing is simply "a piece of software that allows you to easily produce perfectly valid Atmos files that simply don't use all the available Atmos features".
If you have a mix that only contains two "height channels".... "stuff up and to the left" and "stuff up and to the right"..... It would be a lot of work to convert that into a bunch of separate objects that move independently. It would be much simpler to just create two objects, send each of your original tracks to one of them, and leave them in one location.
(And, if I have a single object, located "up and to the left", there's no reason for it to be playing from more than the single speaker or speakers that serve that xyz location.)
It would be nice if the upmixer can "fluff it up a bit" - but your guess is as good as mine as to how useful that will be in practice.
However, in any case, you're not talking about any limitation in Atmos itself... you're just talking about a specific limitation in a specific piece of software that was used to create that particular disc.
I believe the problem here is that a lot of people still seem to be confused about the difference between "Atmos objects" and "height channels". (And, to be quite honest, according to the way Dolby has consistently explained what the three numbers in the designation mean, those discs are in fact labelled wrong.) According to Dolby's information, the first number describes your "regular channels", the second digit describes your subwoofers, and the third digit describes HOW MANY HEIGHT CHANNELS/SPEAKERS YOU HAVE. Note that THE NUMBER OF ATMOST OBJECTS IN THE MIX IS IN NO WAY RELATED TO THE NUMBER OF HEIGHT CHANNELS YOU HAVE. You could have fifty objects zooming around between two height speakers; and you could have two objects zooming around between fifty speakers; they are NOT INTRINSICALLY RELATED. From the latest information I have (which is usually quite good), many recent discs are being encoded with A LIMITED NUMBER OF ATMOS OBJECTS. This would reduce the processing requirements, which might make life easier for some hardware. (I'm inclined to think of it as "Atmos lite".)
This means that, at any given time, only four simultaneous Atmos objects may be "in play". So, you CANNOT have twenty drones, each flying in a different pattern, flying around over your head at the same time.... you can only have four.
HOWEVER, each of those objects may still be mapped anywhere on or between any of the speakers you have, including six height channels, if you have that many.
Hi Keith, generally we agree on almost everything but in this case not so much. Around 2 years ago (very quietly it seems) Dolby released Atmos software that facilitates the pinning of objects to channels. I don't know if this was the intention, but the effect has been a lot of (most) movies being released since with the Atmos objects pinned to 7.1.4. Not only is this stated on the case but many people, including me, have confirmed this to actually be the case. Without extrapolation/matrixing there is no output for the 2 (out of 6) ceiling speakers and no output from the front wides (2 of the 9). I have confirmed this as best I can with an acquaintance of mine who works as a sound engineer/mixer, plus by the well respected FilmMixer on AVS. DTS-X to my knowledge has always been 7.1.4 and the Dolby software simply aligns Atmos with DTS-X, which I believe was the intent by Dolby and/or the request by the movie studios. Yep. I'm still in Camp B  Cheers Gary |
|
