02-25-2019, 08:43 AM | #1 |
Human being with feelings
Join Date: Jan 2017
Posts: 43
|
mic surround techniques to Ambisonics
Hello,
I am trying to make my head around how to convert/decode stereo, multichannel surround techniques to Ambisonics, (pretty much the way round of most Ambisonics decoders). I would like to adapt my old recordings to my new spherical projects. Let's say I have a recording made with a stereo technique for the front and a stereo technique for the back. I know the distance between the mics: how can make that into ambisonics stream (AmbiX)? So far I found some plugins that could be in the right direction but they don't seem to do what I need: Waves B360 Ambisonics Encoder - looks like more a mix-to-ambisonics tool Harpex-X - It seems the only technique it can manage is DoubleMS (which I never used). SPARTA Array2SH - to me is not very clear what it does. Thanks. |
02-25-2019, 02:42 PM | #2 |
Human being with feelings
Join Date: Mar 2010
Location: Adelaide, South Australia (originally from Geelong)
Posts: 5,598
|
What are you using for decoding and what Ambisonic order are you using?
Pretty much all my Ambisonic mixes are third order and I'm using Blue Ripple Sound for the majority of my needs as well as some of the very good encoding and upmixing tools in the IEM Plugin Suite. The IEM plugins are free and can work right up to seventh order. There's a stereo encoder and a multi encoder included in the suite. The Blue Ripple Core plugins are free and include a stereo upmixer (encoder) which is also very good and works in third order. For front/back stereo pairs, I'd use two stereo upmixers (encoders) and feed the front pair to the left and right front at 45 degrees either side of the front center position. Do the same for the back stereo pair at 45 degrees left and right of the back center position. This is assuming a square mic setup. Adjust the angles to match if your mics are in a rectangle. There's a sticky in this section of the forum with links to various Ambisonic resources and the Blue Ripple Sound and IEM plugins are listed there. Here's a direct link to the IEM plugin descriptions.... https://plugins.iem.at/docs/plugindescriptions/ And here's a link to the sticky with other links..... https://forum.cockos.com/showthread.php?t=210108
__________________
|
02-25-2019, 02:50 PM | #3 |
Human being with feelings
Join Date: Mar 2010
Location: Adelaide, South Australia (originally from Geelong)
Posts: 5,598
|
Also, Daniel Rudrich from IEM is a member of this forum. This is his profile here....
https://forum.cockos.com/member.php?u=119131
__________________
|
02-26-2019, 05:55 AM | #4 |
Human being with feelings
Join Date: Jul 2017
Posts: 24
|
Hi there,
in general what you can do is simply encode your microphone signals with their directions during recording, as ReaDave has already proposed. There are quite a lot of tools available which are capable of this, as it's basic Ambisonic panning: - IEM MultiEncoder or StereoEncoder - ambix_encoder - O3A panner I personally would recommend using tools which support higher order Ambisonics, at least third order, as anything below is not very good in my opinion. Unfortunately, Waves B360 Encoder only supports first order. Double-MS is basically 2D Ambisonics. You can see first order Ambisonics as a tripple-MS, as you have one Omni-signal (mid) and three figure of eights (sides). Regarding the array2sh by SPARTA: this plug-in is brilliant, however it's not what you are looking for. It encodes spherical (and cylindrical) microphone arrays into Ambisonics. Those arrays should be as coincident as possible, we are talking about a radius of only few centimeters. I suppose your front/back stereo pairs had a bigger distance. tl;dr; treat your microphone signals as separate sources, and encode them with their directions into higher order Ambisonics. best Daniel |
02-27-2019, 01:50 AM | #5 |
Human being with feelings
Join Date: Jan 2017
Posts: 43
|
Thanks guys for your answers, I will play with the mentioned plugins to see if I can find the right set-up for my specific technique, which of course won't have the Z signal as it's only orizontal surround.
All my projects are in 3OA so if I could implement all my old recordings would be fabulous. The following step will be how to create a fake but convincing Z plan out of those, but hey, a step at a time. :-) Many thanks. |
02-27-2019, 04:48 AM | #6 | |
Human being with feelings
Join Date: Mar 2010
Location: Adelaide, South Australia (originally from Geelong)
Posts: 5,598
|
Quote:
It has had some very positive reviews though..... https://nugenaudio.com/haloupmix/
__________________
|
|
02-27-2019, 03:09 PM | #7 | |
Human being with feelings
Join Date: Jan 2017
Posts: 43
|
Quote:
https://www.blueripplesound.com/prod...a-manipulators |
|
02-27-2019, 06:03 PM | #8 | |
Human being with feelings
Join Date: Mar 2010
Location: Adelaide, South Australia (originally from Geelong)
Posts: 5,598
|
Quote:
The only thing I'm not a fan of regarding the Blue Ripple paid plugins is the licensing system. It needs an Internet connection to refresh the licenses every few weeks. I'm curious to know if the Halo stereo upmixer actually derives some form of discreet information for the height axis based on the content of the input. I'm hoping it does. That would be preferable to me compared to just adding ambience.
__________________
|
|
02-28-2019, 03:20 AM | #9 |
Human being with feelings
Join Date: Jan 2017
Posts: 43
|
I've seen the O3A in action with a mono signal and I was quite impressed. I don't know about the Halo but if is a matter of adding some ambience, I think is better to get an IR and use a convolution reverb, like the Ambi Verb HD by Noisemakers, which I have and I use quite extensively.
|
02-28-2019, 01:35 PM | #10 |
Human being with feelings
Join Date: Mar 2010
Location: Adelaide, South Australia (originally from Geelong)
Posts: 5,598
|
One of the major claims of the Halo upmixer is its downmix compatibility. To do that, it has to extract all it upmixes from what is already contained in the recording. This is my preferred approach for material I want to upmix but do not have in multitrack format for remixing any longer.
The approach of using 3D reverb to create the extra dimensions is good for adding mono or stereo sources to an Ambisonic mix. There's some good 3D convo reverbs floating around these days too. I'll have to check out Ambi Verb HD. If you are interested in using regular stereo FX in Ambisonic configuration, I have created some track templates for this here.... https://forum.cockos.com/showthread.php?t=185676
__________________
|
03-01-2019, 01:33 PM | #11 |
Human being with feelings
Join Date: Jan 2017
Posts: 43
|
So I had a go at the proposed plugins and I have to thank you all for the tips.
I was already using O3A panner, the only annoyance it has, is that it treats the two stereo inputs as an independent source to move, which is great for creativity. I ended up using IEM multiencoder applied to a parent track with excellent results. I would like to be more specific and ask if there is a more precise tool, where I can insert the distance between the microphones and what polar pattern I am using so that every technique will sound different with different depth of field? Last edited by lamacchiacosta; 03-03-2019 at 12:44 PM. |
03-02-2019, 04:33 AM | #12 | |
Human being with feelings
Join Date: Jul 2017
Posts: 24
|
Quote:
So when you encode the microphone signals, they will automatically sound different when you use different polar patterns, different distances between the microphones, or different angles between the microphones. When encoding you simply simulate how it would sound if you'd play back the mic signals over stereo loudspeakers, placed at the directions where you encode the signals. I doubt there's a (good sounding) tool to get more spatial information out of the two mics. With adding reflections, or full reverb, you can of course add some information from different directions. |
|
03-03-2019, 12:43 PM | #13 |
Human being with feelings
Join Date: Jan 2017
Posts: 43
|
Make sense.
Thank you. |
Thread Tools | |
Display Modes | |
|
|