Linkwitz Multiband Comp Question...
-
How's it going all?
Racking my brains here... I've been trying for a few hours to make a dynamic eq/compressor with a linkwitz filter network the same as I did before with a de-esser but I can seem to get the ranges right. I'm probably over complicating the setup using the same one that I used for the de-esser before and maybe don't need so many?
I'm trying to only affect around 250hz (the full range would be something like 160hz - 400hz with 250hz in the middle) similar to the way that is does in Fab Filter Pro MB (screenshot below)
Is this possible to do with the linkwitz filters?
I'll put a screenshot of how I had my de-esser setup in the back end as well....
-
@SteveRiggs Will be hard to get a bell curve from a HP+LP
-
Here lies the problem!
The curve doesn't have to be exact, just something that has a similar effect really just ducking the 250hz.
It will be static and wont need to change the frequency from 250hz. It's not looking good with these filters though so far. Wondering if theres another solution using something else...
-
@SteveRiggs Why don't you simply use another filter that has a bell curve like svf_eq or biquad?
-
Winner! That should do it! I had a feeling I was overcomplicating it and the answer was right there. Its been a long day Thanks man. I'll try that now...
-
@SteveRiggs I've tried biquads a couple of times before. You can see there is a phasing issue if you inspect in Plugin Doctor. Yeah there are bell shapes on the crossover points, but because of that phasing issue :D No flat frequency line possible
-
Cheers @orange. This seems to have done the job now using the svf_eq
I did notice that there was a 4db boost to the overall level of the plugin when turning the dsp module on which is weird, but a gain module above it has levelled that out. So far, so good!
-
@SteveRiggs The boost isn't weird at all, in fact, it should be 3dB. You have two nodes in parallel which logically double the signal... I said 3dB in an ideal case scenario, but some eq boost the signal a little bit by default
-
@ustk Makes sense! Cheers. I've gotta make this one as transparent as possible so it's all good to know.
-
@SteveRiggs Are you trying to make multiband compressor? Acording to the picture it doesn't seems like a compressor with crossover filter. It is a dynamic eq/filter.
-
@orange Kinda, but only for 250hz. The other bands wont be used
-
@SteveRiggs Ok so you don't need HP-LP crossover filters
-
You need to wrap the compressor into a chain and a
math.clear
node after it so that you can't hear the compressor but just the dynamic EQ. -
@Christoph-Hart Genius! Thanks. Like this?
-
@SteveRiggs Yeah that's how it should be done, you can now remove the gain compensation
-
And the next thing to do is to wrap the split container into another container that splits up the buffers so that the update rate of the compressor is faster. Try a fix32 or fix64 container (the 100% correct solution would be a frame container but the performance overhead does not justify the sound improvement for this use case)
https://docs.hise.audio/scriptnode/list/container/fix8_block.html
-
@Christoph-Hart Is there anything to do for this compressor behaviour? The gain reduction curve is unnatural?
https://forum.hise.audio/topic/3626/dynamics-compressor-scriptnode-question
-
Thanks. Just tried this but when I add the split container into a fix_64 or fix_32 container it breaks the connection between the compressor and the filters gain. It doesn't seem to show up anymore in the connections list to re-connect it.
Is there another way of connecting them back up once in the fix container?
This is the ancient build I'm using btw, incase that makes a difference...
-
It's cool, I worked it out. I moved the compressor outside of the chain first so that is shows back up in the connections list to re-connect
-
I'm wondering now if there is a way to have these on the front interface in floating tiles?