Jump to content
IGNORED

VST plugins to impulse responses (suitable for HQPlayer convolution engine)


bogi

Recommended Posts

Are you interested how to create impulse responses for convolution engine from any VST plugin?

It is a way how to do DSP in HQPlayer with DSD files.

After some experiments from the neighbor thread "Tilt Control" I think I now know how to do it.

 

Today I tried to create IRs from Panagement Binaural Distance Panner (the free edition).

It is stereo to binaural plugin for listening on headphones. You can play with the plugin GUI for example in Foobar2000 or JRiver to get idea what it does.

I used Foobar2000 to adjust settings:

 

2089533894_obrzok.png.5f418a2986cfca1b72ce1c3fadc92e65.png

 

You can click on the black area representing music scene to set the center place the music should come from (the circle around a point).

Then it is possible to set width of music scene. I lowered it at least a bit to be more adequate to increased distance from listener.

I disabled Reverb because the recordings already present some 'space' or 'room' and I didn't want to change it.

 

Here is my result, which works in HQPlayer, with DSD recordings too: Panagement2.zip

 

Let me know if the VST to IR conversion topic is interesting for you.

i7 11850H + RTX A2000 Win11 HQPlayer ► Topping HS02 ► 2x iFi iSilencer ► SMSL D300 ► DIY headamp DHA1 ► HiFiMan HE-500
Link to comment

@Miska, I found your 4 years old post and now I probably understood what you meant.

So my current understanding is: (assume the IRs to be created are intended to be used with stereo content):

1. I will play test tone with few seconds delay at the end at first on left channel, but the result will be stereo. It is because there is some intended crosstalk the filter does, which is needed for the effect to work. Actually that intended and wished type of crosstalk is called crossfeed. So I need to create IRs from both output channels for input left channel.

2. Then I need to do the same for the right channel.

3. Now we have 4 IR files and it comes to the Matrix pipeline you pointed to in the "Tilt Control" thread. Left side response from the left side source channel has to be mixed with left side response of the right side source channel. Right side response from the right side source channel has to be mixed with right side response of the left side source channel.

Something like this:

 

1237287134_obrzok.png.dfa8766409f12ca8ec8fd84fab9f9dfe.png

 

Is such a procedure sufficient for all types of crossfeed or binaural type of plugins? Does any category of plugins/filters exist where such a procedure would not be sufficient?

i7 11850H + RTX A2000 Win11 HQPlayer ► Topping HS02 ► 2x iFi iSilencer ► SMSL D300 ► DIY headamp DHA1 ► HiFiMan HE-500
Link to comment
On 10/5/2020 at 8:26 PM, bogi said:

@Miska, I found your 4 years old post and now I probably understood what you meant.

So my current understanding is: (assume the IRs to be created are intended to be used with stereo content):

1. I will play test tone with few seconds delay at the end at first on left channel, but the result will be stereo. It is because there is some intended crosstalk the filter does, which is needed for the effect to work. Actually that intended and wished type of crosstalk is called crossfeed. So I need to create IRs from both output channels for input left channel.

2. Then I need to do the same for the right channel.

3. Now we have 4 IR files and it comes to the Matrix pipeline you pointed to in the "Tilt Control" thread. Left side response from the left side source channel has to be mixed with left side response of the right side source channel. Right side response from the right side source channel has to be mixed with right side response of the left side source channel.

Something like this:

 

1237287134_obrzok.png.dfa8766409f12ca8ec8fd84fab9f9dfe.png

 

Is such a procedure sufficient for all types of crossfeed or binaural type of plugins? Does any category of plugins/filters exist where such a procedure would not be sufficient?

 

Hard to say exhaustively to "all", but for most yes. There are some DSP plugins that are not described by transfer function, such as dynamic range compressors. Also other non-linearities like distortion generators cannot be done this way. Things that depend on amplitude/phase response (including delays like reverbs) can be processed this way.

 

Signalyst - Developer of HQPlayer

Pulse & Fidelity - Software Defined Amplifiers

Link to comment

Thanks, Miska! It is fun for me to learn something new.

There is an update according to the previous 2 posts. Setup it according to the picture below, including the Gain setting. I could add gain in Voxengo Deconvolver, but HQPlayer may include better algorithm for volume control. The Gain setting is saved with the matrix configuration when you save the preset. Feel free to adapt it down a dB if you would get clipping. Panagement2_stereo2binaural.zip 

 

image.thumb.png.78b8ae23d1e83f6a61301ab0ff736b5c.png

 

Please take this only as an example what can be reached by the described method with free tools only.

The 60 seconds long test tone sine sweep was generated in Voxengo Deconvolver. The Panagement2 VST plugin was applied to the test tone using Foobar2000 Converter. For deconvolution of the response sweep Voxengo Deconvolver was used. For small operations like adding silence and splitting stereo impulse responses to mono I used SoX. To check test tone and the response I used Audacity. To check the generated impulse responses I used REW.

i7 11850H + RTX A2000 Win11 HQPlayer ► Topping HS02 ► 2x iFi iSilencer ► SMSL D300 ► DIY headamp DHA1 ► HiFiMan HE-500
Link to comment

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now



×
×
  • Create New...