You just convolve the two impulses with each other. That will give you a new filter with a length that is the sum of the lengths of the two original filters. The new filter likely has very small amplitude at the ends, so you can probably re-window and cut the length down a bit without losing anything of value.
I just checked, yes it's different. The GUI doesn't use the mute functionality, it just sets the volume to -99dB instead to mute. I don't remember why it's like this (haven't touched that part for quite a while), but it's definitely not good. I'll change it to use the set/get_mute commands.
Thanks for the explanation. If I understand this correctly, convolving impluse a with impulse b would lead to the same result as convolving b with a, right? Sorry for these "explain to me like I'm five" type of questions. I suck at math, my greatest talent is putting square pegs into round holes.
Thanks for the feedback. Not a huge deal for me as I don't use the GUI for volume control but figured it might be something worth changing,
Michael
Websocket is a very common standard. You can use any language as long as there is a websocket client library available for it (which probably means nearly all languages). You also need libraries for reading and writing json and yaml (also probably nearly all languages).
Sounds good...i use java daily so prefer that...json and yaml should be fine..
The graphic display of filters... I'm on a bit old build so might have changed, but - I don't get the display... the vertical lines - what are these - there are 8 of them between 10Hz and 1k... ?? Pointing to the curve to find 100 Hz and do but don't hit a line... below is a low shelf, 150 Hz Q 0.5. There must be something fishy going on here ;-D
//
//
Attachments
The level meters take the values from the playback thread, just before the samples are send to the output device. They should definitely be identical, there may be a bug here.
Could you show the pipeline tab of the gui?
Does it behave differently if you flip the order of the sources for channel 1 (it's first 1, then 0 now, try to put them the other way).
Got the files and looked through the config. The confusion is that the mixer is placed at the beginning of the pipeline. That makes the two channels identical to begin with, but then the pipeline contains quite different filters for the two channels. For example the -6dB Gain filter is only applied to channel 1.
Just move the mixer to the end!
Just move the mixer to the end!
It's not that the channels are not levelled - its that the differences between them vary, like between stereo channels... as mono, they should have a constant level relation - they don't here - thats whats so strange. I still wonder where in the pipeline the out level meters measure?
This is for development of a speaker of which I only have one individual for now. So ch1 is tweeter and ch0 is bass - 2-way speaker. I dont want L in tweeter and R for bass.... hence the differences in filters.... but I had hoped for mono into the pipe-line by mixing incoming channels x-wise...
Mixer placing in PL... how to think about that? Rules?
//
This is for development of a speaker of which I only have one individual for now. So ch1 is tweeter and ch0 is bass - 2-way speaker. I dont want L in tweeter and R for bass.... hence the differences in filters.... but I had hoped for mono into the pipe-line by mixing incoming channels x-wise...
Mixer placing in PL... how to think about that? Rules?
//
The level meters measure input level just after the capture device, before the data gets sent to the pipeline, and output level after the pipeline just before the data gets sent to the output device.
If you start with mono as two identical channels, and then apply very different filters to the two channels, then I would expect the levels to behave exactly as you describe. Depending on the input, they will have stronger and weaker passages at different times.
Try on simple test. Just remove all filters and keep just the mixer in the pipeline. Now the output levels for the two channels should be identical.
Where in the pipeline to place the mixer, well that depends on what you want to accomplish.
If you start with mono as two identical channels, and then apply very different filters to the two channels, then I would expect the levels to behave exactly as you describe. Depending on the input, they will have stronger and weaker passages at different times.
Try on simple test. Just remove all filters and keep just the mixer in the pipeline. Now the output levels for the two channels should be identical.
Where in the pipeline to place the mixer, well that depends on what you want to accomplish.