Hi list,
what is the current state of the art in binaural externals? I see in the archives that there are several around, each with advantages/disadvantages.
I would need it to:
windows 32b
record the output of an already existing patch with x-y
coordinates/trajectories into binaural audio
sample rate fixed at 44.1KHz
there are around 500 samples with an individual player, so each will
need its own enconder. They don't play necessarily all together, though.
Best,
Joao
Personally, I always use [vstplugin~] + "StereoEncoder" / "BinauralDecoder" from the IEM plugin suite, but I'm of course biased :-)
On 01.06.2021 15:12, João Pais wrote:
Hi list,
what is the current state of the art in binaural externals? I see in the archives that there are several around, each with advantages/disadvantages.
I would need it to:
windows 32b
record the output of an already existing patch with x-y
coordinates/trajectories into binaural audio
sample rate fixed at 44.1KHz
there are around 500 samples with an individual player, so each will
need its own enconder. They don't play necessarily all together, though.
Best,
Joao
Pd-list@lists.iem.at mailing list UNSUBSCRIBE and account-management -> https://lists.puredata.info/listinfo/pd-list
I'm having a look, but I'm apparently not getting something. When I try [vstplugin~] + "StereoEncoder", it only outputs a signal to the left-ish, changing the position doesn't affect the output )0th order). Or do you put a [vstplugin~] + "BinauralDecoder" afterwards?
Personally, I always use [vstplugin~] + "StereoEncoder" / "BinauralDecoder" from the IEM plugin suite, but I'm of course biased :-)
On 01.06.2021 15:12, João Pais wrote:
Hi list,
what is the current state of the art in binaural externals? I see in the archives that there are several around, each with advantages/disadvantages.
I would need it to:
windows 32b
record the output of an already existing patch with x-y
coordinates/trajectories into binaural audio
sample rate fixed at 44.1KHz
there are around 500 samples with an individual player, so each
will need its own enconder. They don't play necessarily all together, though.
Best,
Joao
Am 5. Juni 2021 01:06:08 MESZ schrieb "João Pais" jmmmpais@gmail.com:
I'm having a look, but I'm apparently not getting something. When I try
[vstplugin~] + "StereoEncoder", it only outputs a signal to the left-ish, changing the position doesn't affect the output )0th order).
this sounds very much as if you were listening to the W-channel.
Or do you put a [vstplugin~] + "BinauralDecoder" afterwards?
if you want to listen to the spatialisationvia headphones, then: yes.
.mfg.jdr.afgg. IOhannes
strangely, now Pd (win32) always crashes when opening BinauralDecoder and dsp on - independently if the other vstplugin is being used. That wasn't the case yesterday, I'm loading it exactly in the same way, with the same configuration.
I'm having a look, but I'm apparently not getting something. When I try
[vstplugin~] + "StereoEncoder", it only outputs a signal to the left-ish, changing the position doesn't affect the output )0th order).
this sounds very much as if you were listening to the W-channel.
Or do you put a [vstplugin~] + "BinauralDecoder" afterwards?
if you want to listen to the spatialisationvia headphones, then: yes.
Hi,
now Pd (win32) always crashes when opening BinauralDecoder
and dsp on
This is a known bug with BinauralDecoder: it crashes in the process method if you provide less than 4 input channels (= 1st order). I *think* they have fixed this already. Maybe you are using an older version?
Generally, I have the suspicion that you don't use the correct number of channels. Assuming you want to use 3rd order, the StereoEncoder needs 2 input and 16 output channels, i.e. [vstplugin~ 2 16]; conversely the BinauralDecoder needs 16 input and 2 output channels, i.e. [vstplugin~ 16 2].
Christof
Am 5. Juni 2021, 12:17, um 12:17, "João Pais" jmmmpais@gmail.com schrieb:
strangely, now Pd (win32) always crashes when opening BinauralDecoder and dsp on - independently if the other vstplugin is being used. That wasn't the case yesterday, I'm loading it exactly in the same way, with
the same configuration.
I'm having a look, but I'm apparently not getting something. When I
try
[vstplugin~] + "StereoEncoder", it only outputs a signal to the left-ish, changing the position doesn't affect the output )0th
order).
this sounds very much as if you were listening to the W-channel.
Or do you put a [vstplugin~] + "BinauralDecoder" afterwards?
if you want to listen to the spatialisationvia headphones, then: yes.
Pd-list@lists.iem.at mailing list UNSUBSCRIBE and account-management -> https://lists.puredata.info/listinfo/pd-list
now Pd (win32) always crashes when opening BinauralDecoder
and dsp on
This is a known bug with BinauralDecoder: it crashes in the process method if you provide less than 4 input channels (= 1st order). I *think* they have fixed this already. Maybe you are using an older version?
no, I just downloaded them for the first time. using 4 channels works.
Generally, I have the suspicion that you don't use the correct number of channels. Assuming you want to use 3rd order, the StereoEncoder needs 2 input and 16 output channels, i.e. [vstplugin~ 2 16]; conversely the BinauralDecoder needs 16 input and 2 output channels, i.e. [vstplugin~ 16 2].
ok, since I don't know anything about this topic, I didn't knew what was happening. 1st order with 4 channels seems to already provide stereo spacialization (I just want to turn a stereo signal into stereo binaural).
I couldn't also find any documentation on how to control the vst parameters in the website - although there is info on osc control, maybe I'll try that later.
1st order with 4 channels seems to already provide stereo spacialization
Sure, but higher ambisonics orders give more accuracy.
(I just want to turn a stereo signal into stereo binaural).
I guess you also want to spatialize the signal in some way? There is no benefit in just "converting" a stereo track to binaural IMO, the stereo image just gets worse...
I couldn't also find any documentation on how to control the vst parameters in the website
Just like any other VST plugin. Have a look at the "pd parameter" section in vstplugin~-help.pd
Christof
On 05.06.2021 23:18, João Pais wrote:
now Pd (win32) always crashes when opening BinauralDecoder
and dsp on
This is a known bug with BinauralDecoder: it crashes in the process method if you provide less than 4 input channels (= 1st order). I *think* they have fixed this already. Maybe you are using an older version?
no, I just downloaded them for the first time. using 4 channels works.
Generally, I have the suspicion that you don't use the correct number of channels. Assuming you want to use 3rd order, the StereoEncoder needs 2 input and 16 output channels, i.e. [vstplugin~ 2 16]; conversely the BinauralDecoder needs 16 input and 2 output channels, i.e. [vstplugin~ 16 2].
ok, since I don't know anything about this topic, I didn't knew what was happening. 1st order with 4 channels seems to already provide stereo spacialization (I just want to turn a stereo signal into stereo binaural).
I couldn't also find any documentation on how to control the vst parameters in the website - although there is info on osc control, maybe I'll try that later.
1st order with 4 channels seems to already provide stereo spacialization
Sure, but higher ambisonics orders give more accuracy.
I'll try out. there are at least 6 sources to spatialize, which means 6x enconding+decoding plugins adding to the patch. I'm assuming these aren't the heaviest processing, but there are already up to 1000 tabplay~ playing sometimes.
(I just want to turn a stereo signal into stereo binaural).
I guess you also want to spatialize the signal in some way? There is no benefit in just "converting" a stereo track to binaural IMO, the stereo image just gets worse...
you're right, a spacialization will be added.
I couldn't also find any documentation on how to control the vst parameters in the website
Just like any other VST plugin. Have a look at the "pd parameter" section in vstplugin~-help.pd
ah, now I found out that the values are scaled between 0 and 1. This makes sense now.
Since I have you here (without abusing), do you know the most adequate formula to convert a xy or xyz (cartesian??) coordinate system to the stereoencoder coordinates (which I assume are polar?)?
On 06.06.2021 12:24, João Pais wrote:
1st order with 4 channels seems to already provide stereo spacialization
Sure, but higher ambisonics orders give more accuracy.
I'll try out. there are at least 6 sources to spatialize, which means 6x enconding+decoding plugins adding to the patch. I'm assuming these aren't the heaviest processing, but there are already up to 1000 tabplay~ playing sometimes.
Just to make sure: you only need *one* BinauralDecoder instance where you would send all your encoded signals.
ah, now I found out that the values are scaled between 0 and 1. This makes sense now.
Yes, VST parameters are always normalized between 0 and 1. Just watch out for the actual range/mapping. You can easily check it by setting the parameter with [param_set( and watching the result in the GUI.
Since I have you here (without abusing), do you know the most adequate formula to convert a xy or xyz (cartesian??) coordinate system to the stereoencoder coordinates (which I assume are polar?)?
https://en.wikipedia.org/wiki/Spherical_coordinate_system#Cartesian_coordina...
Christof
Pd-list@lists.iem.at mailing list UNSUBSCRIBE and account-management -> https://lists.puredata.info/listinfo/pd-list
-
Christof Ressi -
IOhannes m zmölnig -
João Pais