Old 03-05-2018, 01:37 PM   #121
ChristopherT
Human being with feelings
 
Join Date: Apr 2017
Location: South
Posts: 573
Default

mrlimbic
- will we be able to put more speakers into this?
- is it possible to have a speaker circle instead of square room?
ChristopherT is offline   Reply With Quote
Old 03-05-2018, 01:46 PM   #122
mrlimbic
Human being with feelings
 
mrlimbic's Avatar
 
Join Date: Nov 2009
Location: UK
Posts: 594
Default

Quote:
Originally Posted by ChristopherT View Post
mrlimbic
- will we be able to put more speakers into this?
- is it possible to have a speaker circle instead of square room?
This one is pretty much geared towards film sound & theatres are rectangle with speakers on the walls. The square UI is also necessary so you can hard pan to just one or between only two adjacent speakers without bleed to other speakers by slamming the puck against the walls.

I could make a different type of panner later with an arbitrary number of moveable speakers for more art installation type scenarios. Are there none like that already?
__________________
Convert NLE XMLs to RPP.
http://vordio.net
mrlimbic is offline   Reply With Quote
Old 03-05-2018, 03:33 PM   #123
ChristopherT
Human being with feelings
 
Join Date: Apr 2017
Location: South
Posts: 573
Default

Yes there are others - but the panning movement has major problems.
Sigh...
ChristopherT is offline   Reply With Quote
Old 03-05-2018, 03:52 PM   #124
mrlimbic
Human being with feelings
 
mrlimbic's Avatar
 
Join Date: Nov 2009
Location: UK
Posts: 594
Default

Quote:
Originally Posted by ChristopherT View Post
Yes there are others - but the panning movement has major problems.
Sigh...
OK. If you specify the behaviour you want in enough detail, I may be able do something after this one is ready in a month or so. Write it all down. Try to describe what would be the ideal. Also handy to know is what to avoid, i.e. what annoys you about others, so same mistakes aren't made.
__________________
Convert NLE XMLs to RPP.
http://vordio.net
mrlimbic is offline   Reply With Quote
Old 03-05-2018, 03:55 PM   #125
Ice
Human being with feelings
 
Join Date: Aug 2014
Posts: 659
Default

Another good feature of the PT panner is clicking (when using a mouse that is) anywhere in the box allows you to grab the puck no matter where it is. In other words you don't have to "catch it", which can be uber tricky if its already automated and you're chasing it around. (currently the case in ReaSurround - very hard to catch)

Ideally this means it doesn't jump immediately to where you clicked - so a relative position.

PT does have a modifier key to allow jumping to the clicked location but personally I never use that.

And the other benefit is keeping your eyes on the movie instead of the GUI.

This might be tricky when more than one puck is in the same box. Spanner's iPad app grabs the closest puck to the touch/click.
Ice is offline   Reply With Quote
Old 03-05-2018, 04:01 PM   #126
mrlimbic
Human being with feelings
 
mrlimbic's Avatar
 
Join Date: Nov 2009
Location: UK
Posts: 594
Default

Quote:
Originally Posted by Ice View Post
Another good feature of the PT panner is clicking (when using a mouse that is) anywhere in the box allows you to grab the puck no matter where it is. In other words you don't have to "catch it", which can be uber tricky if its already automated and you're chasing it around. (currently the case in ReaSurround - very hard to catch)

Ideally this means it doesn't jump immediately to where you clicked - so a relative position.

PT does have a modifier key to allow jumping to the clicked location but personally I never use that.

And the other benefit is keeping your eyes on the movie instead of the GUI.

This might be tricky when more than one puck is in the same box. Spanner's iPad app grabs the closest puck to the touch/click.
Mine at the moment moves the puck to the cursor if you continue to hold the mouse down when you click somewhere. A short click doesn't trigger it. With multiple pucks it uses the last puck that was touched. But you could have an option for closest too.

Not sure what you mean by doesn't jump immediately/relative position? Do you want it to start moving towards the cursor at some specified speed/acceleration instead?
__________________
Convert NLE XMLs to RPP.
http://vordio.net
mrlimbic is offline   Reply With Quote
Old 03-05-2018, 04:08 PM   #127
Ice
Human being with feelings
 
Join Date: Aug 2014
Posts: 659
Default

Quote:
Originally Posted by mrlimbic View Post

Not sure what you mean by doesn't jump immediately/relative position? Do you want it to start moving towards the cursor at some specified speed/acceleration instead?
Personally I like the way PT handles it now. A click just takes over automation while held, but the puck doesn't move toward the clicked position. I think of it more the other way around, as if the click made the mouse jump to the puck's position. Does that make sense?
Ice is offline   Reply With Quote
Old 03-05-2018, 05:42 PM   #128
ChristopherT
Human being with feelings
 
Join Date: Apr 2017
Location: South
Posts: 573
Default

Mrlimbic - I posted this youtube video earlier on in this thread.
This is what I hope Reaper can achieve as far as surround panning goes.

https://www.youtube.com/watch?v=3K6cLKSKBe4

At around 4 mins on the video - is selection of 8 speaker panning movement.
ChristopherT is offline   Reply With Quote
Old 03-05-2018, 06:59 PM   #129
Ice
Human being with feelings
 
Join Date: Aug 2014
Posts: 659
Default

Quote:
Originally Posted by ReaDave View Post
Typical workflow using Ambisonics to do your panning would be quite straight forward.

Add the Ambisonic panner in the last FX slot on a given track and turn off the master output for that track.
Do the same for all other tracks you want to pan.
Bus them to a track with the Ambisonic decoder.
Feed the output of the decoder bus to your surround master (you can use these as stem tracks too).
This is an interesting concept. I remember trying this once before but I could never get it going right. I just tried again, and I got it sort of working, but in no instance can I ever get to any channel discreetly. I also can't find a 7.1 decoder setup.

If you get the chance it would be interesting to see a simple session set up correctly. Say 1 mono, and 1 stereo track, bussed and configured to route out of a 7.1 system. It can be without audio. I'd just like to see the routing. There are a bunch of plugins on that site and I can't tell which one goes where and pairs with what exactly. I watched some tutorial videos but didn't see the plugin setups.

I'm on a Mac so it'd be great to have the session be Mac and ask for the correct plugins.
Ice is offline   Reply With Quote
Old 03-05-2018, 07:29 PM   #130
ChristopherT
Human being with feelings
 
Join Date: Apr 2017
Location: South
Posts: 573
Default

+1 on a detailed explanation of the Ambisonic "how to" set up ReaDave.
A video would be even better
ChristopherT is offline   Reply With Quote
Old 03-05-2018, 10:07 PM   #131
RobinGShore
Human being with feelings
 
Join Date: May 2013
Location: New York
Posts: 371
Default

I've done a lot of sound work for VR and 360 videos over the past few years and ambisonics has become a de facto standard for that stuff. It's a really neat format and I like working with it, but it has some drawbacks that make it not that great for more traditional post work.
Quote:
Originally Posted by Ice View Post
I just tried again, and I got it sort of working, but in no instance can I ever get to any channel discreetly.
For me, this is the biggest downside of Ambisonics. Because it represents an encoded soundfield instead of a fixed channel layout, there is no way to get it to play sound discretely through a particular channel. The spatial image will always be created by combining sound from all the channels of the decoded output. Some decoders (namely Harpex) give you a lot more channel separation than others, but you'll never be able to get it completely discreet.
Quote:
I also can't find a 7.1 decoder setup.
I'm not sure there many good free options for a 7.1 decoder. As far as I know the Bruce Wiggins plugins that Dave linked to don't include 7.1. I think the ambisonics toolkit has a plugin that will decode to 7.0, but I haven't used that one very much. My favorite decoder by far is Harpex (https://harpex.net/). It does do 7.1 as well most other standard surround formats. It's a little pricey, but there's a 30 day free demo, so you may want to give it a try if you're interested. It comes the closest of anything I've found to being able to decode to discrete channels.
Quote:
If you get the chance it would be interesting to see a simple session set up correctly. Say 1 mono, and 1 stereo track, bussed and configured to route out of a 7.1 system. It can be without audio. I'd just like to see the routing.
Here's a link to a very simple Reaper project set up as you described:

https://drive.google.com/file/d/1xYK...ew?usp=sharing

No media, just tracks, plugins and routing. There is one track with a stereo panner on it, another with a mono panner, and both are feeding a buss with a 7.1 decoder. The panners are from the free Blue Ripple Core plugins (http://www.blueripplesound.com/products/o3a-core) and the decoder is the aforementioned Harpex ( https://harpex.net/).

A few important things to note about this project:

-The channel order for the 7.1 is L, R, C, Lfe, Lss, Rss, Lsr, Rsr. You may need to adjust the plugin pin connectors to get it to play out correctly on your system

-The sends to the decoder buss are multi-channel sends. The panners I'm using output in third-order ambisonics, which is a 16 channel format, so my sends have to be 16 channels.

-Ambisonics uses spherical panners instead of the x-y coordinates we're all used to. The panners in this project may look like standard rectangular grids, but they're actually "unwrapped" spheres. The far left and right edges represent 180 degrees on the sphere (i.e directly behind the listener) and the grid lines marked left and right represent 90 degrees to the left and right of the listener, not the Front left and right speakers. Any vertical movement on the panning grids represents height, so for normal 7.1 panning you'll probably want to keep the pucks along the "equator". Basically you're limited to panning along the edge of the sphere. There's no way to pan in between front and back without also panning up or down.
__________________
www.silversound.us

Last edited by RobinGShore; 03-05-2018 at 10:34 PM.
RobinGShore is offline   Reply With Quote
Old 03-06-2018, 09:55 AM   #132
Ice
Human being with feelings
 
Join Date: Aug 2014
Posts: 659
Default

Quote:
Originally Posted by RobinGShore View Post
For me, this is the biggest downside of Ambisonics. Because it represents an encoded soundfield instead of a fixed channel layout, there is no way to get it to play sound discretely through a particular channel. (snip)

Here's a link to a very simple Reaper project set up as you described:

https://drive.google.com/file/d/1xYK...ew?usp=sharing

No media, just tracks, plugins and routing. There is one track with a stereo panner on it, another with a mono panner, and both are feeding a buss with a 7.1 decoder. The panners are from the free Blue Ripple Core plugins (http://www.blueripplesound.com/products/o3a-core) and the decoder is the aforementioned Harpex ( https://harpex.net/).
Thanks for this - I'll take a look in the coming days. I spent way top much time yesterday on diversions! I can't afford another day like yesterday.

I've seen Harpex but never pulled the trigger. I still have the free Surroundzone plugin from Soundfield (Pre Rode days) which will do 7.1 and tried that yesterday as a decoder but couldn't get to the discreet channels, as you've confirmed. I still use that to decode stuff I record with my AMBEO as its a decent enough decode, though not as good as the Harpex.

As you mention, if panning can't get to a channel discreetly then I'm afraid thats going to be a deal breaker for what I'm doing. Its a very clever concept, but for film, at least what I'm working on and who for, they'll never accept that. There was just way too much bleed into too many other channels. For VR I can certainly see it being viable so I'll keep exploring it, but at the moment I'm not doing any VR.

In any case thanks for the thorough session setup. I'm missing all 3 of those plugins as installs. I thought maybe I could replace the Harpex with Surround zone but with the decoder track being 16 channels, maybe not.
Ice is offline   Reply With Quote
Old 03-06-2018, 05:41 PM   #133
RobinGShore
Human being with feelings
 
Join Date: May 2013
Location: New York
Posts: 371
Default

Oh wow, I had totally forgotten about Surroundzone, that's actually the first ambisonics plugin I ever used. I would give the Harpex demo a shot when you have some time. It will give you much much better channel separation and can do some other neat decoding tricks if you use it on your Ambeo recordings (the shotgun mode is a really nice feature).
Quote:
In any case thanks for the thorough session setup. I'm missing all 3 of those plugins as installs.
Follow the links! They're all free to download and easy to install.
Quote:
I thought maybe I could replace the Harpex with Surround zone but with the decoder track being 16 channels, maybe not.
You can feed the first four channels of a 16 channel track into the Surround Zone plugin and it will work just fine. Higher order ambisonics signals are compatible with all the orders below them. The first four channels of a third order signal is actually a complete first order ambisonics signal. Harpex is actually just a first order decoder as well. I only used sixteen channel sends in the session for the sake of being a completionist, in case someone decided they wanted to test the routing with a third-order decoder.

The problem you will run into if you replace Harpex with Surround Zone in my session is that Surround Zone uses the older Furse-Malham channel ordering and normalization standard while the panners in my session use the newer ambiX standard ( If anyone wants to go down a real nerdy rabbit hole have a look at the wikipedia page on ambisonics exchange formats and the papers it cites: https://en.wikipedia.org/wiki/Ambiso...change_formats). You would need another plugin on the decoder track before Surround Zone to convert to the right channel ordering and normalization, otherwise the decoding will be way out of wack.


Anyway this is veering way off topic. Hopefully mrlimbic and Schwa release usable versions of their panners soon so everyone in this thread can have something new to chat about.
__________________
www.silversound.us
RobinGShore is offline   Reply With Quote
Old 03-07-2018, 10:47 AM   #134
introvert
Human being with feelings
 
Join Date: Dec 2011
Location: A cold place
Posts: 69
Default

Quote:
Originally Posted by RobinGShore View Post
Anyway this is veering way off topic. Hopefully mrlimbic and Schwa release usable versions of their panners soon so everyone in this thread can have something new to chat about.
Indeed!

I tried the setup with SurroundZone just as Ice did and had no luck due to the channel separation. I will try the Harpex demo today or tomorrow when I can find some time. But hopefully a permanent solution pops up from Schwa or mrlimbic so we don't have to jump through hoops for surround panning anymore.

I've been working around this for a couple of years now by just hard routing stuff. It works, but it is a lot more effort and it's anything but intuitive/creative. I'm very excited that this seems to finally be getting some momentum here.
introvert is offline   Reply With Quote
Old 03-07-2018, 02:53 PM   #135
plush2
Human being with feelings
 
Join Date: May 2006
Location: Saskatoon, Canada
Posts: 1,828
Default

Quote:
Originally Posted by introvert View Post
Indeed!

I tried the setup with SurroundZone just as Ice did and had no luck due to the channel separation. I will try the Harpex demo today or tomorrow when I can find some time. But hopefully a permanent solution pops up from Schwa or mrlimbic so we don't have to jump through hoops for surround panning anymore.

I've been working around this for a couple of years now by just hard routing stuff. It works, but it is a lot more effort and it's anything but intuitive/creative. I'm very excited that this seems to finally be getting some momentum here.

The designed goal of ambisonics is different than that of film surround. Ambisonics is an attempt to capture/design and play back an entire sound scene (an uninterupted sphere of sound production around a listener). It will never, at any reasonable order of complexity and even with harpex offer a completely discrete signal to a single speaker in a film surround speaker array. That said, it can provide a really nice sense of envelopment when that is required. It also can provide nice smooth movement around the decoded speakers because it relies on an ideal encoded sound scene which is then decoded to the desired speakers.

Going forward I think a combination of distance based amplitude panning/object based and ambisonic/scene based mixing is likely to become the norm because of their respective strengths. Since the idea of a decode at playback is starting to take hold (atmos, VR of all kinds, games) I'm hoping a new panner would take into account this mix of approaches and make it easy to use them together. I'm probably not the voice of the industry on this as I'm not so much in the trenches at this moment, trying to churn out client and downstream pleasing mixes.
__________________
mymusic - http://music.darylpierce.com
mywork - http://production.darylpierce.com
plush2 is offline   Reply With Quote
Old 03-07-2018, 03:43 PM   #136
ChristopherT
Human being with feelings
 
Join Date: Apr 2017
Location: South
Posts: 573
Default

I'm with you plush2
The combination of both ambisonics and standard surround panning is where I am heading towards.
Lets hope Reaper gets the surround panning working properly so we can move forward.
ChristopherT is offline   Reply With Quote
Old 03-08-2018, 03:42 AM   #137
ReaDave
Human being with feelings
 
ReaDave's Avatar
 
Join Date: Mar 2010
Location: Adelaide, South Australia (originally from Geelong)
Posts: 5,495
Default

Quote:
Originally Posted by ChristopherT View Post
+1 on a detailed explanation of the Ambisonic "how to" set up ReaDave.
A video would be even better
Apologies for the delay replying here. I've had a very busy couple of weeks and with an interstate trip involved, I haven't had much time in my studio.

It looks like Robin has taken the requests on board and provided a project example already though (thanks for helping out Robin).

I personally find mixing in Ambisonics and decoding to discreet surround provides a better mix than discreet mixing precisely because it works with a sound field rather than isolated speakers.
In the end though, for many, the preference will come down to those paying the bills. Most clients (especially bigger commercial ones) are locked into a mindset which determines that preference and it may take some time to educate them to the benefits of higher order Ambisonic production.

I'm working solidly on my next album at the moment which is the biggest personal project I've undertaken to date. It is all being recorded and mixed in third order Ambisonics and I will be making it freely available in full Ambisonic format, 5.1 decode from the Ambisonic mix and regular stereo also decoded from the Ambi mix.
It is my hope that the results will demonstrate how much BETTER than discreet mixing this is. Sound localization in higher order Ambisonics is more convincing between physical speaker locations than the phantom images created by regular mixing methods (pan and level based) and the experience is much more enveloping.

Last edited by ReaDave; 03-08-2018 at 03:48 AM.
ReaDave is offline   Reply With Quote
Old 03-08-2018, 05:04 AM   #138
ChristopherT
Human being with feelings
 
Join Date: Apr 2017
Location: South
Posts: 573
Default

PMed you ReaDave
ChristopherT is offline   Reply With Quote
Old 03-08-2018, 05:15 AM   #139
ReaDave
Human being with feelings
 
ReaDave's Avatar
 
Join Date: Mar 2010
Location: Adelaide, South Australia (originally from Geelong)
Posts: 5,495
Default

Quote:
Originally Posted by ChristopherT View Post
PMed you ReaDave
Got it. Will reply shortly.
ReaDave is offline   Reply With Quote
Old 03-09-2018, 09:33 AM   #140
mrlimbic
Human being with feelings
 
mrlimbic's Avatar
 
Join Date: Nov 2009
Location: UK
Posts: 594
Default

Little bit more progress on my panner. Now experimenting with making the basic panner strip into a dual panner (unlinked at the moment).

Also experimenting with how to show output feedback. Not the right panning volume algorithm yet (it's just showing puck-speaker distance at the moment).

__________________
Convert NLE XMLs to RPP.
http://vordio.net

Last edited by mrlimbic; 03-09-2018 at 09:39 AM.
mrlimbic is offline   Reply With Quote
Old 03-09-2018, 10:08 AM   #141
Ice
Human being with feelings
 
Join Date: Aug 2014
Posts: 659
Default

Quote:
Originally Posted by ReaDave View Post
Apologies for the delay replying here. I've had a very busy couple of weeks and with an interstate trip involved, I haven't had much time in my studio.

It looks like Robin has taken the requests on board and provided a project example already though (thanks for helping out Robin).

I personally find mixing in Ambisonics and decoding to discreet surround provides a better mix than discreet mixing precisely because it works with a sound field rather than isolated speakers.
In the end though, for many, the preference will come down to those paying the bills. Most clients (especially bigger commercial ones) are locked into a mindset which determines that preference and it may take some time to educate them to the benefits of higher order Ambisonic production.

I'm working solidly on my next album at the moment which is the biggest personal project I've undertaken to date. It is all being recorded and mixed in third order Ambisonics and I will be making it freely available in full Ambisonic format, 5.1 decode from the Ambisonic mix and regular stereo also decoded from the Ambi mix.
It is my hope that the results will demonstrate how much BETTER than discreet mixing this is. Sound localization in higher order Ambisonics is more convincing between physical speaker locations than the phantom images created by regular mixing methods (pan and level based) and the experience is much more enveloping.
I look forward to seeing how this plays out, and if Ambisonic panning has a place in film mixing. I haven't heard a film mixed in it, so have no basis whatsoever to comment on that.

The only thing I can add, is regarding channel separation. This application/result may differ if using an ambisonic panner over a traditional panner. On one project where I was premixing in a near-field room and final mixing in a theatre-sized stage, I decided to try to maintain a slight stereo image on stereo sources as they panned around the room. It sounded great on the near-fields, but in the bigger room, even just maintaining a slight stereo image compromised the movement around the room and I was disappointed with the result. It was too washy and not localized enough. When I collapsed the stereo to mono and panned, I got the desired more directional movement and therefore better separation of specific things on screen. In film sound effects particularly, things have to get out of the way of other things sometimes pretty drastically. After that experiment, when I see things bleeding into other channels it makes me nervous, especially if there's no way to make it any more discreet.

Maybe there is really something about ambisonic where that bleed really doesn't matter. Maybe it IS superior. I've only done small tests in a near-field environment. From what I've heard so far in here, I couldn't get enough localization.

I'd have to hear mixes of the same material done both ways to say for sure.
Ice is offline   Reply With Quote
Old 03-10-2018, 01:50 PM   #142
introvert
Human being with feelings
 
Join Date: Dec 2011
Location: A cold place
Posts: 69
Default

Quote:
Originally Posted by mrlimbic View Post
Little bit more progress on my panner. Now experimenting with making the basic panner strip into a dual panner (unlinked at the moment).
This is AWESOME progress. Please keep us posted on how things go. Very excited to check it out.
introvert is offline   Reply With Quote
Old 03-12-2018, 02:27 AM   #143
swiiscompos
Human being with feelings
 
swiiscompos's Avatar
 
Join Date: Mar 2011
Location: Lausanne (CH)
Posts: 1,109
Default

You might all want to have a look at the prerelease forum...
swiiscompos is offline   Reply With Quote
Old 03-12-2018, 06:25 AM   #144
RobinGShore
Human being with feelings
 
Join Date: May 2013
Location: New York
Posts: 371
Default

FYI for everyone following:

The latest pre-release version has a few changes to ReaSurround that address some of the things brought up here. There's still problems, but it looks like they are indeed trying incorporate the feedback from this thread. Head over to the pre-release forum to check it out, and most importantly give some feedback!
__________________
www.silversound.us
RobinGShore is offline   Reply With Quote
Old 03-12-2018, 10:28 AM   #145
introvert
Human being with feelings
 
Join Date: Dec 2011
Location: A cold place
Posts: 69
Default

Quote:
Originally Posted by RobinGShore View Post
FYI for everyone following:

The latest pre-release version has a few changes to ReaSurround that address some of the things brought up here. There's still problems, but it looks like they are indeed trying incorporate the feedback from this thread. Head over to the pre-release forum to check it out, and most importantly give some feedback!
Excellent, will go grab it and check it out. Thanks for the heads up!
introvert is offline   Reply With Quote
Old 03-12-2018, 11:53 AM   #146
airon
Human being with feelings
 
airon's Avatar
 
Join Date: Aug 2006
Location: Berlin
Posts: 10,405
Default

Quote:
Originally Posted by Ice View Post
I look forward to seeing how this plays out, and if Ambisonic panning has a place in film mixing. I haven't heard a film mixed in it, so have no basis whatsoever to comment on that.

The only thing I can add, is regarding channel separation. This application/result may differ if using an ambisonic panner over a traditional panner. On one project where I was premixing in a near-field room and final mixing in a theatre-sized stage, I decided to try to maintain a slight stereo image on stereo sources as they panned around the room. It sounded great on the near-fields, but in the bigger room, even just maintaining a slight stereo image compromised the movement around the room and I was disappointed with the result. It was too washy and not localized enough. When I collapsed the stereo to mono and panned, I got the desired more directional movement and therefore better separation of specific things on screen. In film sound effects particularly, things have to get out of the way of other things sometimes pretty drastically. After that experiment, when I see things bleeding into other channels it makes me nervous, especially if there's no way to make it any more discreet.

Maybe there is really something about ambisonic where that bleed really doesn't matter. Maybe it IS superior. I've only done small tests in a near-field environment. From what I've heard so far in here, I couldn't get enough localization.

I'd have to hear mixes of the same material done both ways to say for sure.
The ambisonics approach might be pretty neat for backgrounds and some special effects, but from what I've gathered in the past years, Atmos wouldn't be able to use it all that well. Isn't the 9.1 bed (i.e. the non-object mix) without access to height speakers ?


Nice to see progress on a new panner(more choice) and ReaSurround. Have to check that out.

That reminds me. I've created a Lemur layout for the VST version of Stige's Surround Panner MK2. You can check it out here: https://stash.reaper.fm/v/33093/Reap...180310_pan.zip (Screenshot)

The Lemur layout also includes a page for a send mixer for the last selected track.
__________________
Dialogue/FX Editor & Re-Recording Mixer
(Video)Using Latch Preview
"My ego comes pre-shrunk" - Randy Thom
airon is offline   Reply With Quote
Old 03-14-2018, 08:26 AM   #147
ReaDave
Human being with feelings
 
ReaDave's Avatar
 
Join Date: Mar 2010
Location: Adelaide, South Australia (originally from Geelong)
Posts: 5,495
Default

Quote:
Originally Posted by airon View Post
The ambisonics approach might be pretty neat for backgrounds and some special effects, but from what I've gathered in the past years, Atmos wouldn't be able to use it all that well. Isn't the 9.1 bed (i.e. the non-object mix) without access to height speakers ?
From my understanding of Atmos, the beds are typically 7.1.2 format. The .2 are the two height speakers. The Blue Ripple Sound upmixers and decoders both support Atmos beds in 7.1.2 format.

Dolby themselves have the bed definition here...
https://www.dolby.com/us/en/guide/do...-2-setups.html
ReaDave is offline   Reply With Quote
Old 03-14-2018, 09:18 AM   #148
plush2
Human being with feelings
 
Join Date: May 2006
Location: Saskatoon, Canada
Posts: 1,828
Default

Quote:
Originally Posted by airon View Post
The ambisonics approach might be pretty neat for backgrounds and some special effects, but from what I've gathered in the past years, Atmos wouldn't be able to use it all that well. Isn't the 9.1 bed (i.e. the non-object mix) without access to height speakers ?
7.1.2 has access to height by virtue of the .2 speakers above. It doesn't have access to the dynamically calculated audio objects as that is quite proprietary and derived from encoded metedata.
__________________
mymusic - http://music.darylpierce.com
mywork - http://production.darylpierce.com
plush2 is offline   Reply With Quote
Old 03-20-2018, 08:12 PM   #149
introvert
Human being with feelings
 
Join Date: Dec 2011
Location: A cold place
Posts: 69
Default

I'm excited about the progress, but I've also been out of touch for several days now so I've lost track of things. Is there more progress coming for ReaSurround? Is schwa looking for more feedback?
introvert is offline   Reply With Quote
Old 03-21-2018, 03:22 AM   #150
The Byre
Human being with feelings
 
Join Date: Dec 2011
Posts: 680
Default

Quote:
Originally Posted by ReaDave View Post
From my understanding of Atmos, the beds are typically 7.1.2 format.
No! Atmos packages up to 128 audio tracks — a 9.1 channel-based bed and up to 118 virtual audio positions. Your decoder then assigns these to whatever you have in the room, up to 64 outputs.

DTS:X works differently and is more sophisticated, in that it gives each sound a virtual position and can ascribe an unlimited number of such positions and movements.

In the average movie theatre, you do not hear any difference (IMO) between DTS:X and Atmos and certainly not in the average living room (or even control room) with just two or four ceiling speakers. That means that you can simulate a full 3D sound with just 12 discrete channels.

But before we get all excited - most TV households do not even have 5.1, or if they do, it is via some pretty dreadful cheap speakers. Pretty close to 100% of your output for TV and BR is going to be the usual mono/stereo/5.1 delivery package. Atmos and DTS:X may come to households in meaningful numbers one day, but I fear that will be mostly with cheap soundbars at $1,000 or less, rather than with properly laid out speaker arrays! Most living rooms are far too small anyway, to have speakers in the ceiling without creating hot-spots.

HOWEVER

Movie houses are slowly digging their own graves by providing a pretty sub-standard experience. Far too often, projectors are under-sized, the audio is mushy and unclear and the LFE is mostly a pretty feeble affair. On the other side, home audio setups are improving and TV sets are getting larger, better and brighter and much sharper as 4K becomes the norm.

This is particularly noticeable in Germany, where the average person only goes to the movies 1.5 times a year. In the UK, where TV sets are much smaller and almost nobody has a decent sound system, they go to the movies 2.5 times a year. The average main TV in Germany is over 50", whereas the average main TV set in the UK is under 40".

All over the World, when living standards permit, there is a clear trend towards home entertainment, with key movies and other material on BR or as a download and away from going to movie houses. There is now as well, a clear move by young people away from Netflix/Amazon-Prime and to just watch/use free-to-air, augmented by Blu-Ray disks or downloads for the latest big hits.
The Byre is offline   Reply With Quote
Old 03-21-2018, 04:17 AM   #151
schwa
Administrator
 
schwa's Avatar
 
Join Date: Mar 2007
Location: NY
Posts: 10,307
Default

Quote:
Originally Posted by introvert View Post
I'm excited about the progress, but I've also been out of touch for several days now so I've lost track of things. Is there more progress coming for ReaSurround? Is schwa looking for more feedback?
We're about to release 5.78 without any of the reasurround changes, because there are other stable improvements that don't need to wait on the reasurround changes.

After 5.78 is released, Many of the improvements discussed in this thread will be back in 5.79pre1, and we can continue the discussion then!
schwa is offline   Reply With Quote
Old 03-21-2018, 01:23 PM   #152
ChristopherT
Human being with feelings
 
Join Date: Apr 2017
Location: South
Posts: 573
Default

great news
ChristopherT is offline   Reply With Quote
Old 03-21-2018, 05:43 PM   #153
Reaktor:[Dave]
Human being with feelings
 
Reaktor:[Dave]'s Avatar
 
Join Date: Jun 2010
Location: Berlin
Posts: 458
Default

@Schwa, will we get ReaSurround natively built into tracks replacing the stereo pan knob on the MCU on multi-channel tracks? The GUI mockup by The Byre is the way to go, I think.

Quote:
Originally Posted by The Byre View Post
Here are some suggestions for 5.1, 7.1 and 3D audio GUIs



In the 3D version, a height fader has been added which moves the audio from the front/side speakers to the ceiling speakers.

[...]
Reaktor:[Dave] is offline   Reply With Quote
Old 03-21-2018, 07:49 PM   #154
introvert
Human being with feelings
 
Join Date: Dec 2011
Location: A cold place
Posts: 69
Default

Quote:
Originally Posted by schwa View Post
After 5.78 is released, Many of the improvements discussed in this thread will be back in 5.79pre1, and we can continue the discussion then!
You are awesome sir. Thank you!!
introvert is offline   Reply With Quote
Old 03-22-2018, 02:10 AM   #155
airon
Human being with feelings
 
airon's Avatar
 
Join Date: Aug 2006
Location: Berlin
Posts: 10,405
Default

Quote:
Originally Posted by The Byre View Post
No! Atmos packages up to 128 audio tracks — a 9.1 channel-based bed and up to 118 virtual audio positions. Your decoder then assigns these to whatever you have in the room, up to 64 outputs.
.....
There's got to be more space between those 118 positions for transitions, but really, 118 start(and end I assume) positions only ? What's that, 3x3x2 bits of coordinates ? :P

Well, I'm ok with 5.1 for now, seeing as that's the big max Netflix/BD is likely asking from me. Most of my mixes are geared towards stereo anyway, as I don't mix for theaters, and mostly dialogue.

Certainly looking forward to giving an improved ReaSurround a spin on test material.
__________________
Dialogue/FX Editor & Re-Recording Mixer
(Video)Using Latch Preview
"My ego comes pre-shrunk" - Randy Thom
airon is offline   Reply With Quote
Old 03-22-2018, 07:13 AM   #156
The Byre
Human being with feelings
 
Join Date: Dec 2011
Posts: 680
Default

Quote:
Originally Posted by Reaktor:[Dave] View Post
@Schwa, will we get ReaSurround natively built into tracks replacing the stereo pan knob on the MCU on multi-channel tracks? The GUI mockup by The Byre is the way to go, I think.
Thanks for that 'Honourable Mention'!

Quote:
Originally Posted by airon View Post
There's got to be more space between those 118 positions for transitions, but really, 118 start(and end I assume) positions only ? What's that, 3x3x2 bits of coordinates ? :P
Like stereo (just two positions) or 5.1 (just five positions) the various positions just get quieter or louder. Rocket science - it am not! Your encoder tells the audio stream to give this particular noise one of the 118 virtual positions and a relative volume for that position. The decoder then assigns that noise to whatever is in the theatre or living room. That is why I say that in reality, 12 (or maybe a few more for theatre release) discrete channels is enough for all practical purposes. 7 surround, 1 LFE and 4 height channels.

These 12 discrete channels should be enough for the mastering house/engineer to create the encoded Atmos or DTS:X digital signal for a realistic experience.

Right now, in stereo, you have the ability to place a sound anywhere between two speakers. With 7.1, you have the ability to place the sound anywhere in front or behind the listener. With 3D sound (Atmos or DTS:X) you have the ability to place the sound anywhere in a hemisphere (i.e. above, but not below them!) around the speaker, thereby simulating the real world of sound.

Next step - a smell-channel (it has been tried a few times!!!) Dolby Sniffmos? DTS:Olfacto? ReaSmell? ProTools - Elastofart?
The Byre is offline   Reply With Quote
Old 03-22-2018, 07:37 PM   #157
plush2
Human being with feelings
 
Join Date: May 2006
Location: Saskatoon, Canada
Posts: 1,828
Default

Quote:
Originally Posted by Reynaud View Post

It's 118 mono or stereo Objects, specified by positional Metadata, not 118 positions.
Yes, these are objects that are dynamically moved around the array (bed + however many ceiling speakers you have) using Dolby special sauce (they won't tell anyone so probably ambisonics or dbap). It is a mash up of the scene based and object based paradigms I was mentioning in an earlier post.
__________________
mymusic - http://music.darylpierce.com
mywork - http://production.darylpierce.com
plush2 is offline   Reply With Quote
Old 03-23-2018, 12:51 AM   #158
airon
Human being with feelings
 
airon's Avatar
 
Join Date: Aug 2006
Location: Berlin
Posts: 10,405
Default

Quote:
Originally Posted by Reynaud View Post
Atmos doesn’t use Ambisonics nor DBAP, since an Object can be allocated to all speakers at the same time, where in actual use more speakers are assigned to Objects simultaneously than DBAP.

Instead, the amplitude weighting is based on a custom algorithm unique to Dolby that uses more variables in its estimation.

DTS:X also uses a custom algorithm in the rendering of Objects, instead of using a variant of DBAP or Ambisonics, since again, more variables are used in its estimation, and Objects may be assigned to a more flexible array of speakers than other systems.
There's a size parameter. I'm still studying the Dolby Atmos suite manual (PDF). Avid has made their own Atmos panner too.



Anyone know where the manual for that thing is ?

In a video I found something very interesting. A preset selection for speaker isolation. That would make things a little quicker for ReaSurround. How to do that though...



Something we can definitely use for ReaSurround already is Leap Motion. I foudn this via this article on Pro-Tools Expert. It's MIDI CC data . Check it out:

__________________
Dialogue/FX Editor & Re-Recording Mixer
(Video)Using Latch Preview
"My ego comes pre-shrunk" - Randy Thom

Last edited by airon; 03-23-2018 at 01:02 AM.
airon is offline   Reply With Quote
Old 03-23-2018, 06:50 AM   #159
jm duchenne
Human being with feelings
 
jm duchenne's Avatar
 
Join Date: Feb 2006
Location: France
Posts: 734
Default

Quote:
Originally Posted by Reynaud View Post
Atmos doesn’t use Ambisonics nor DBAP, since an Object can be allocated to all speakers at the same time, where in actual use more speakers are assigned to Objects simultaneously than DBAP.
Instead, the amplitude weighting is based on a custom algorithm unique to Dolby that uses more variables in its estimation.
I don't know of course what technique is used in Atmos, but there is no constrain in DBAP to adress only one speaker, it is just a parameter.

INA-GRM Spaces which is 100% DBAP has a Spread parameter that does this, and the same kind for the DBAP option in the Flux Spat Revolution.
jm duchenne is offline   Reply With Quote
Old 03-23-2018, 07:07 AM   #160
mrlimbic
Human being with feelings
 
mrlimbic's Avatar
 
Join Date: Nov 2009
Location: UK
Posts: 594
Default

Quote:
Originally Posted by Reaktor:[Dave] View Post
@Schwa, will we get ReaSurround natively built into tracks replacing the stereo pan knob on the MCU on multi-channel tracks? The GUI mockup by The Byre is the way to go, I think.
If you do build an icon into tracks for surround, please make the actual panner to be used a configurable option. Different panners have pros and cons depending the situation.
__________________
Convert NLE XMLs to RPP.
http://vordio.net
mrlimbic is offline   Reply With Quote
Reply

Thread Tools
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT -7. The time now is 10:55 AM.


Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2019, vBulletin Solutions Inc.