|
|
|
12-01-2017, 11:27 AM
|
#41
|
Human being with feelings
Join Date: Aug 2006
Location: Berlin
Posts: 11,817
|
I use that setup for Netflix mixes, which usually demand 2.0 and 5.1. Routing is easy in Reaper.
Panning? Well, we're not here for shits and giggles.
|
|
|
12-01-2017, 03:39 PM
|
#42
|
Human being with feelings
Join Date: May 2006
Location: Saskatoon, Canada
Posts: 2,110
|
Quote:
Originally Posted by serr
The last couple mixes I've worked on I've tried making both the stereo and 5.1 mix in the same session. (Mute either the 2.0 bus or the 5.1 bus to listen.)
I like not having the back and forth towards the end of the mix. (Pasting new work into the other session as it comes up.) The downside is some extra time spent routing and a few clumsy busing scenarios that seemed to take a little too long to dial up.
I've got to say at the end of the day I think the bigger PITA is working around the minimum 2 channel track count, not the lack of an integrated surround panner.
|
Delivering mono stems caught me out a while ago when I was delivering some 5.1 mixes. The broadcasters QC kept telling me my mixes were well below spec level. My playback bus said otherwise and my mono track meters said otherwise but because I was tapping the mono mixes off of 2 channel tracks with the audio coming to only 1 channel I wound up with the panning law lowering the level of the actual output files. I changed from 1 > 1 to 1 > 1-2 in my routing and the problem was solved.
Thankfully their QC engineer was a sharp one and a very kind person who raised the level using my reference tone and noticed my mixes were indeed on spec other than being delivered low. It saved me further emberassment. I have returned to checking the actual file levels in a wave editor to be sure. Funny how confident and complacent we can get when software seems to just work.
A mono routing would make that a more transparent situation I think.
|
|
|
12-02-2017, 04:57 AM
|
#43
|
Human being with feelings
Join Date: Aug 2006
Location: Berlin
Posts: 11,817
|
That's why I keep normalization separate from mixing. Makes it easier to do broadcast and Netflix versions in on fell swoop too.
What panners do you guys use ?
I just got my first surround reverb, Verberate. I'll still be tricking my way around with stereo reverbs in LRLsRs though.
Any recommendations there ?
|
|
|
12-08-2017, 10:13 PM
|
#44
|
Human being with feelings
Join Date: Dec 2016
Location: Los Angeles
Posts: 226
|
I feel like this issue is really important so I UP vote it here, sorry guys =3
|
|
|
12-09-2017, 08:58 PM
|
#45
|
Human being with feelings
Join Date: Mar 2010
Location: Adelaide, South Australia (originally from Geelong)
Posts: 5,598
|
Quote:
Originally Posted by airon
That's why I keep normalization separate from mixing. Makes it easier to do broadcast and Netflix versions in on fell swoop too.
What panners do you guys use ?
I just got my first surround reverb, Verberate. I'll still be tricking my way around with stereo reverbs in LRLsRs though.
Any recommendations there ?
|
I'm mixing almost everything in third order Ambisonics so I use the Ambisonic upmixers and panners from Blue Ripple Sound at present.
Regarding surround reverbs, I've created an Ambisonic template that enables ANY reverb you own to be used in 3D (width, height and depth). I could make a modified version of that which would work in 5.1 if you want. I can set it up to use the REAPER ATK for the encode/decode process (ATK is free).
|
|
|
12-09-2017, 09:04 PM
|
#46
|
Human being with feelings
Join Date: Mar 2010
Location: Adelaide, South Australia (originally from Geelong)
Posts: 5,598
|
One more thing Airon, you might want to take a look at mixing in Ambisonics too. It is easy to monitor and export in 5.1 from Ambisonics and I much prefer the result to mixing in 5.1 natively. It's also MUCH easier to export from Ambisonics to ANY other format (stereo, mono, 7.1, beds for Dolby Atmos for example) without needing to remix.
|
|
|
12-10-2017, 03:12 AM
|
#47
|
Human being with feelings
Join Date: Jan 2011
Location: Porto, Portugal
Posts: 89
|
Quote:
Originally Posted by the_drumreaper
I need a sweetspot mix for my 5.1 wav (or sometimes 6 mono) to a 4 channel ambisonic layout that best represents the original 5.1 mapping. I cant get it right in Reasurround and all the plugin mashups Im experimenting with. Getting close, but ends up a mess. I have a 5.1 test file with a person saying "left, center, right, right surround, left surround, (boom for LFE)". Anybody got a template for what I need? I want an ambisonic mix to do as best as possible to preserve it.
|
I am not sure if i am getting this right but ReaSurround is not encoding to A or B Format so you cannot just do it directly there. You must use a dedicated ambisonics encoder.
|
|
|
12-10-2017, 03:30 AM
|
#48
|
Human being with feelings
Join Date: Jan 2011
Location: Porto, Portugal
Posts: 89
|
Quote:
Originally Posted by ReaDave
One more thing Airon, you might want to take a look at mixing in Ambisonics too. It is easy to monitor and export in 5.1 from Ambisonics and I much prefer the result to mixing in 5.1 natively. It's also MUCH easier to export from Ambisonics to ANY other format (stereo, mono, 7.1, beds for Dolby Atmos for example) without needing to remix.
|
Totally agree with you here Dave.
|
|
|
12-10-2017, 08:08 AM
|
#49
|
Human being with feelings
Join Date: Aug 2006
Location: Berlin
Posts: 11,817
|
Quote:
Originally Posted by ReaDave
One more thing Airon, you might want to take a look at mixing in Ambisonics too. It is easy to monitor and export in 5.1 from Ambisonics and I much prefer the result to mixing in 5.1 natively. It's also MUCH easier to export from Ambisonics to ANY other format (stereo, mono, 7.1, beds for Dolby Atmos for example) without needing to remix.
|
That's a good point.
Basic plugins are free. The upmixer isn't cheap, but I'll give the free plugins a try after the current project when I can experiment a little. If it can save me a rebalance pass for the 5.1 version(I primarily mix for the LTRT stereo version first), it may be worth it.
|
|
|
12-10-2017, 08:43 PM
|
#50
|
Human being with feelings
Join Date: May 2006
Location: Saskatoon, Canada
Posts: 2,110
|
Quote:
Originally Posted by airon
That's a good point.
Basic plugins are free. The upmixer isn't cheap, but I'll give the free plugins a try after the current project when I can experiment a little. If it can save me a rebalance pass for the 5.1 version(I primarily mix for the LTRT stereo version first), it may be worth it.
|
I love to hear of others using HOA and such in post mixing. The main caveats to mixing with ambisonics for conventional surround are - There is no elegant way to create a discrete center channel. 7th order or higher gets close but you will always have some bleed in the adjacent speakers.
- Same issue for discrete LFE.
Neither problem is insurmountable. I would even go so far as to say that dialog locked exclusively to the center channel is becoming somewhat passe anyway. If and when you need this then a simple bypass routing mixed in with the decoded ambisonic signal (decoded to 5.1, 7.1 etc.) does the trick.
|
|
|
12-11-2017, 05:45 AM
|
#51
|
Human being with feelings
Join Date: Mar 2010
Location: Adelaide, South Australia (originally from Geelong)
Posts: 5,598
|
Quote:
Originally Posted by plush2
I love to hear of others using HOA and such in post mixing. The main caveats to mixing with ambisonics for conventional surround are - There is no elegant way to create a discrete center channel. 7th order or higher gets close but you will always have some bleed in the adjacent speakers.
- Same issue for discrete LFE.
Neither problem is insurmountable. I would even go so far as to say that dialog locked exclusively to the center channel is becoming somewhat passe anyway. If and when you need this then a simple bypass routing mixed in with the decoded ambisonic signal (decoded to 5.1, 7.1 etc.) does the trick.
|
I totally agree on all points here. Third order mixes decoded to 5.1 still use all the speakers to localize the sound source positions by creating a sound field rather than simple panning. I find that to be much more convincing than regular discreet mixes even when the discreet mix uses bleed intentionally. The sweet spot for the Ambisonic mixes is bigger even decided to 5.1 too.
Even stereo and mono decodes from HOA sound more consistent and convincing IMHO. I'd find it very hard to go back to 5.1 and stereo mixing now.
Last edited by Dannii; 12-11-2017 at 05:52 AM.
|
|
|
12-11-2017, 05:50 AM
|
#52
|
Human being with feelings
Join Date: Mar 2010
Location: Adelaide, South Australia (originally from Geelong)
Posts: 5,598
|
Quote:
Originally Posted by airon
That's a good point.
Basic plugins are free. The upmixer isn't cheap, but I'll give the free plugins a try after the current project when I can experiment a little. If it can save me a rebalance pass for the 5.1 version(I primarily mix for the LTRT stereo version first), it may be worth it.
|
If you send Richard from Blue Ripple Sound an email, he will hook you up with fully functional, time limited demo versions of the paid packages. I ended up purchasing the upmixers and the decoders. Combined with the free core plugins, these have all my bases covered. You might find some of the other packages worth checking out too though.
I'll have to update the post I made with my Ambisonic reverb templates. I have done some tweaks lately that improve the imaging considerably. These templates allow you to use any FX plugin you have in true B format mode (width, height and depth in and out). They're only first order at this stage but I find reverbs, delays and modulation FX fit very convincingly into higher order mixes even though they are first order. All your primary sources are still HOA and these give you the bulk of your directional cues.
|
|
|
12-11-2017, 01:16 PM
|
#53
|
Human being with feelings
Join Date: Aug 2006
Location: Berlin
Posts: 11,817
|
My chain might look like this :
O3A --> 5.1 Decoder --> two sends
1) 5.1 Mix
2) LTRT Encoder, with a rebalanced center channel. The stereo decoder is useless to me I'm afraid. But I'd love to be wrong.
Working with three reverbs per fx send is a big downer. So far that's not practical, since I do not have the time to do three reverb setups all the time.
We'll see how this pans out.
|
|
|
12-11-2017, 02:35 PM
|
#54
|
Human being with feelings
Join Date: May 2006
Location: Saskatoon, Canada
Posts: 2,110
|
You shouldn't need 3 verbs. You can do 2 verbs with the ATK BtoA and AtoB plugins on either side. That will get you a perfectly good first order verb.
|
|
|
12-11-2017, 09:44 PM
|
#55
|
Human being with feelings
Join Date: Mar 2010
Location: Adelaide, South Australia (originally from Geelong)
Posts: 5,598
|
Quote:
Originally Posted by airon
My chain might look like this :
O3A --> 5.1 Decoder --> two sends
1) 5.1 Mix
2) LTRT Encoder, with a rebalanced center channel. The stereo decoder is useless to me I'm afraid. But I'd love to be wrong.
Working with three reverbs per fx send is a big downer. So far that's not practical, since I do not have the time to do three reverb setups all the time.
We'll see how this pans out.
|
IMHO, you'll get a much better result by using something like this....
———O3A———
——/———\——
—5.1———2.0—
In other words, run the O3A mix into two sends, one to the 5.1 decoder and one to the stereo decoder simultaneously.
Regarding the reverbs, I've already created track templates for the Valhalla reverbs with all parameters already linked. You only need to adjust the first instance of it and the others follow.
Workflow is insert track from template, create sends to that track, adjust the first instance, adjust level... Done.
These templates retain all the directional information of the sends.
|
|
|
12-12-2017, 02:57 AM
|
#56
|
Human being with feelings
Join Date: Aug 2006
Location: Berlin
Posts: 11,817
|
@Readave
The reverbs are worth a look at the very least. Thank you Dave. I remember the effort you put in to those, and Valhalla plugins are at least civil in their use of CPU. The only other reverbs I'd try this with are Exponential Audio ones. Stuff like Aether is big on CPU use, and Acon Digital can't handle automation transitions without giant CPU hits(still usable though).
The 2.0 mix is not an option. LTRT is a delivery requirement for streaming 5.1 projects. It's my primary mixing target, since almost everyone uses two speakers. Like mobile phones or tablets :P .
@Plush2
I've got ATK installed and never used it so far. I should give that a whirl too.
Most of my work is high efficiency. I hope I can find a way to fit this stuff in to my workflow, if it can get me improvements in sound quality.
Last edited by airon; 12-12-2017 at 03:02 AM.
|
|
|
12-12-2017, 04:32 PM
|
#57
|
Human being with feelings
Join Date: May 2006
Location: Saskatoon, Canada
Posts: 2,110
|
Quote:
Originally Posted by airon
@Plush2
I've got ATK installed and never used it so far. I should give that a whirl too.
Most of my work is high efficiency. I hope I can find a way to fit this stuff in to my workflow, if it can get me improvements in sound quality.
|
Check out FocusPressPushZoom. I find it extremely useful for visualizing my soundfield manipulation.
Last edited by plush2; 12-12-2017 at 04:51 PM.
|
|
|
12-17-2017, 05:05 AM
|
#58
|
Human being with feelings
Join Date: Nov 2015
Location: Cologne
Posts: 1,636
|
I believe we'll have that in v6. As soon as Justin reads that he probably will experience a very emotional moment and the good will and his mercy will be with us. :P
|
|
|
12-18-2017, 11:12 AM
|
#59
|
Human being with feelings
Join Date: Dec 2016
Location: Los Angeles
Posts: 226
|
Quote:
Originally Posted by Gass n Klang
I believe we'll have that in v6. As soon as Justin reads that he probably will experience a very emotional moment and the good will and his mercy will be with us. :P
|
Let's hope so!
It would be a huge addition to ARA implementation.
|
|
|
12-21-2017, 11:35 AM
|
#60
|
Human being with feelings
Join Date: Mar 2010
Location: Adelaide, South Australia (originally from Geelong)
Posts: 5,598
|
Quote:
Originally Posted by plush2
You shouldn't need 3 verbs. You can do 2 verbs with the ATK BtoA and AtoB plugins on either side. That will get you a perfectly good first order verb.
|
I'm not sure that would give you a complete sound field though. From my understanding, A format is just the equivalent of four signals from a tetrahedral mic capsule array. By using two stereo reverbs on such a signal, you wouldn't get interaction between some of the signals.
The method I use is to decode the three m/s signals (left/right, up/down, front/back) into three stereo signals, feed them into three true stereo reverbs (or any other such effect) and then re-encode them back to B format using m/s encoding on the three reverb outputs. This method completely retains the original 3D sound field integrity. If you bypass the reverbs, the input to output is totally transparent.
Quote:
Originally Posted by airon
@Readave
The reverbs are worth a look at the very least. Thank you Dave. I remember the effort you put in to those, and Valhalla plugins are at least civil in their use of CPU. The only other reverbs I'd try this with are Exponential Audio ones. Stuff like Aether is big on CPU use, and Acon Digital can't handle automation transitions without giant CPU hits(still usable though).
The 2.0 mix is not an option. LTRT is a delivery requirement for streaming 5.1 projects. It's my primary mixing target, since almost everyone uses two speakers. Like mobile phones or tablets :P .
|
Fair point on the LTRT. Makes sense. You'd still be able to your two buss monitor workflow using what you mentioned above, ie Ambi to 5.1 to LTRT using your current 5.1 fold down solution. That would still sound more cohesive than mixing in traditional 5.1 format.
If you haven't seen the update yet, I uploaded the fine tuned versions of my B format FX track templates and have added a couple more (Waves TrueVerb and VoS Nasty DLA MK2).
You can actually just use any of the templates I've uploaded and substitute your own FX. The time consuming part is linking each parameter between the three FX instances. You only need to do that once for each effect and save it as a template though.
Here's the link to my templates...
https://forum.cockos.com/showthread.php?t=185676
|
|
|
12-21-2017, 03:34 PM
|
#61
|
Human being with feelings
Join Date: May 2006
Location: Saskatoon, Canada
Posts: 2,110
|
Quote:
Originally Posted by ReaDave
I'm not sure that would give you a complete sound field though. From my understanding, A format is just the equivalent of four signals from a tetrahedral mic capsule array. By using two stereo reverbs on such a signal, you wouldn't get interaction between some of the signals.
The method I use is to decode the three m/s signals (left/right, up/down, front/back) into three stereo signals, feed them into three true stereo reverbs (or any other such effect) and then re-encode them back to B format using m/s encoding on the three reverb outputs. This method completely retains the original 3D sound field integrity. If you bypass the reverbs, the input to output is totally transparent.]
|
A format is exactly equivalent to B format of the same order. Much like mid-side and stereo, one can convert back and forth with absolutely no loss. Your method of 3 stereo verbs on a 6 speaker axial decode will achieve a good result. At the same time 4 channels of reverb on the 4 channels of A-format will achieve a good result as well.
|
|
|
12-21-2017, 10:28 PM
|
#62
|
Human being with feelings
Join Date: Mar 2010
Location: Adelaide, South Australia (originally from Geelong)
Posts: 5,598
|
Quote:
Originally Posted by plush2
A format is exactly equivalent to B format of the same order. Much like mid-side and stereo, one can convert back and forth with absolutely no loss. Your method of 3 stereo verbs on a 6 speaker axial decode will achieve a good result. At the same time 4 channels of reverb on the 4 channels of A-format will achieve a good result as well.
|
I might have to experiment with that. My main aim is to get a true stereo reverb image on each axis (f/b, l/r, u/d) rather than multiple mono reverbs combined. Even better still would be to have a six in, six out reverb where everything interacts cohesively but that would require a purpose designed Ambisonic reverb VST.
|
|
|
12-22-2017, 08:25 PM
|
#63
|
Human being with feelings
Join Date: May 2006
Location: Saskatoon, Canada
Posts: 2,110
|
Quote:
Originally Posted by ReaDave
I might have to experiment with that. My main aim is to get a true stereo reverb image on each axis (f/b, l/r, u/d) rather than multiple mono reverbs combined. Even better still would be to have a six in, six out reverb where everything interacts cohesively but that would require a purpose designed Ambisonic reverb VST.
|
Something like this or this?
I know that second one is expensive but the first is free.
|
|
|
12-23-2017, 05:13 AM
|
#64
|
Human being with feelings
Join Date: Mar 2010
Location: Adelaide, South Australia (originally from Geelong)
Posts: 5,598
|
Quote:
Originally Posted by plush2
Something like this or this?
I know that second one is expensive but the first is free.
|
Bruce's stuff is pretty good, especially his decoders. The reverb is not in the same league as Valhalla though.
The Blue Ripple Sound stuff is top notch and their upmixers and decoders are now the foundation of my Ambisonic workflow. They are relatively expensive but well worth the money.
To be honest through, the reverb is the only one of their products I'm not really thrilled with. The main problem is that only the early reflections have true directional i/o. The reverb itself does not even though it is third order in and out.
The one plugin in the reverb VST pack that I really DO love though is the Shoebox. This one is fantastic for automating movement in 3D space on single sound sources. At this point though, I do not have the funds spare to purchase that pack and the Australian dollar to British pound exchange rate is terrible.
|
|
|
12-23-2017, 10:53 AM
|
#65
|
Human being with feelings
Join Date: May 2006
Location: Saskatoon, Canada
Posts: 2,110
|
I have a few more comment but I think I should make them in your reverb preset thread.
|
|
|
12-23-2017, 03:30 PM
|
#66
|
Human being with feelings
Join Date: Mar 2010
Location: Adelaide, South Australia (originally from Geelong)
Posts: 5,598
|
Quote:
Originally Posted by plush2
I have a few more comment but I think I should make them in your reverb preset thread.
|
Good call.
Apologies Airon for the tangent.
|
|
|
12-23-2017, 04:43 PM
|
#67
|
Human being with feelings
Join Date: Aug 2006
Location: Berlin
Posts: 11,817
|
Oh I'm fascinated.
It doesn't let Cockos off the hook of course .
|
|
|
06-19-2018, 10:40 AM
|
#69
|
Human being with feelings
Join Date: Nov 2014
Posts: 798
|
It's impossible to render out a LCR / LRC file.
I have to render out a 4ch file and then remove the 4th channel in soundforge.
|
|
|
06-19-2018, 07:08 PM
|
#70
|
Human being with feelings
Join Date: May 2006
Location: Saskatoon, Canada
Posts: 2,110
|
Quote:
Originally Posted by Travesty
It's impossible to render out a LCR / LRC file.
I have to render out a 4ch file and then remove the 4th channel in soundforge.
|
When you go to render just type 3 for number of channels instead of the listed stereo, 4, 6, 8 etc. options.
|
|
|
06-20-2018, 03:54 AM
|
#71
|
Human being with feelings
Join Date: Aug 2006
Location: Berlin
Posts: 11,817
|
Quote:
Originally Posted by Gass n Klang
|
Pretty efficient.
|
|
|
06-20-2018, 08:43 AM
|
#72
|
Human being with feelings
Join Date: Nov 2015
Location: Cologne
Posts: 1,636
|
Quote:
Originally Posted by airon
Pretty efficient.
|
definitly. But that's at least what I need. I don't get that kind of tool realized with reasurround.
|
|
|
06-21-2018, 04:31 AM
|
#73
|
Human being with feelings
Join Date: Nov 2014
Posts: 798
|
Quote:
Originally Posted by plush2
When you go to render just type 3 for number of channels instead of the listed stereo, 4, 6, 8 etc. options.
|
Wow, didn't know that. Thanks
|
|
|
06-21-2018, 07:18 AM
|
#74
|
Human being with feelings
Join Date: May 2006
Location: Saskatoon, Canada
Posts: 2,110
|
Quote:
Originally Posted by Travesty
Wow, didn't know that. Thanks
|
I'm glad I was able to share that the problem had an easy solution. I've never needed to deliver LRC as a mix before but I knew about the arbitrary channel counts from other odd mix deliverables.
|
|
|
02-14-2019, 07:02 AM
|
#75
|
Human being with feelings
Join Date: Nov 2016
Posts: 79
|
I know this is an old post, but I'm gonna bump this one. I work with a lot of multichannel stuff, and having the option of choosing uneven track channels would be awesome.
+ a better reasurround panner would be nice
|
|
|
06-04-2019, 10:27 AM
|
#76
|
Human being with feelings
Join Date: Feb 2015
Location: Matosinhos, Portugal
Posts: 23
|
+1 on this
|
|
|
07-29-2021, 06:00 PM
|
#77
|
Human being with feelings
Join Date: Mar 2021
Posts: 18
|
+1 on this. I need to use LCR and 5.0 tracks and mix them quickly in a way that makes sense.
|
|
|
07-30-2021, 02:33 AM
|
#78
|
Human being with feelings
Join Date: Aug 2006
Location: Berlin
Posts: 11,817
|
We do have that native surround panner now.
I've built a quick LCR panner with it and uploaded it as a track template.
https://stash.reaper.fm/42525/4-ch%2...RTrackTemplate
It's a 4-channel track for which the panner only outputs to three.
You might need to play a bit with the Left and Right speaker angles, but it should suffice for mono sources.
|
|
|
01-03-2022, 09:27 PM
|
#79
|
Human being with feelings
Join Date: Dec 2010
Posts: 23
|
Quote:
Originally Posted by crackinkid
I know this is an old post, but I'm gonna bump this one. I work with a lot of multichannel stuff, and having the option of choosing uneven track channels would be awesome.
+ a better reasurround panner would be nice
|
Add my vote to odd numbered channels in tracks, or at least to be able to select a range of inputs to record for odd numbered channels from hardware sources. For example, if I have a 20 channel audio interface and I want to record 5, 7 or 19 channels only in one track.
|
|
|
02-10-2022, 05:23 AM
|
#80
|
Human being with feelings
Join Date: Feb 2022
Posts: 65
|
Other than having odd number of tracks, is there a way to decouple the number of input tracks with the number of output tracks?
|
|
|
Thread Tools |
|
Display Modes |
Linear Mode
|
Posting Rules
|
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts
HTML code is Off
|
|
|
All times are GMT -7. The time now is 07:45 AM.
|