Whats your favorite LP filter with a good balance between cpu/sound quality?

Octaviu5

https://community.native-instruments.com/discussion/comment/39443#Comment_39443

Yes.

Im going to actually recreate it with reaktor and create a custom midi controller solely for it.

And i agree... the monologue is imho the greatest synth that couldnt be.

If they only had put a k35 or an svf in it.

But i guess the korg guys are very idiosincratic, and maybe they felt that people would be bored by it. Or something...

ANDREW221231

https://community.native-instruments.com/discussion/comment/39444#Comment_39444

I've heard it argued that it doesn't really need as spectacular of a filter as it has a lot of other interesting stuff going on on the tone generation side

it could also be down to cost, as the minilogue was introduced as a low cost polysynth and with a lot of other stuff packed in they had to save somewhere

gentleclockdivider

https://community.native-instruments.com/discussion/comment/39439#Comment_39439

Yud on't know what you're missing the core filters are awesome .

Here' s a simple comparison bewteen core setup filter and primary , I colour coded the elements .

In core the filters take linear input values , but you can use log values with a pitch to freq converter , I also clipped th input values but that's not a necessity

Woow ,the forum really downgrades the quality of the images

Octaviu5

https://community.native-instruments.com/discussion/comment/39537#Comment_39537

OMG thank you very much for taking the time.

I am going to be trying them this weekend.

Octaviu5

https://community.native-instruments.com/discussion/comment/39537#Comment_39537

Hey! i tried my best to clone your circuit.

But i think i failed.

test core filter.ens

Would you mind taking a look and giving me a hint on how to make it work?

Thanks in advance. i really appreciate your help.

ANDREW221231

https://community.native-instruments.com/discussion/comment/39578#Comment_39578

the problem is your parameter ranges, the inputs going into the pitch to frequency converter are at most 2, which is a frequency of 9.17 something

so multiplying by 130 before the clip will give you the fuller frequency range

test core filter.ens

ANDREW221231

also, a more general building tip... turn on wire debugging circled in red.... then you can place your mouse onto wires and see what values are propagating thru your structure

Octaviu5

https://community.native-instruments.com/discussion/comment/39585#Comment_39585

Amazing! thank you I really appreciate your help.

i had the idea of ripping the svf block out and reducing it to it's most essential parts, and just switching the core filter inside of it and give it the appropiate ranges.

This is amazing. The core filters are so cheap on the cpu. Im so thrilled.

Thanks again for your very precious help.

gentleclockdivider

https://community.native-instruments.com/discussion/comment/39590#Comment_39590

It's most essential parts are already there , these are called core macros , you just have to assgn control and the ecessary preparation .

Just right click in an empty core cell and choose a filter under audio .

Also , you really need to understand the difference between log en linear ranges , the p inputs of primary take log values , this is the range 1-127 this covers the whole frequency range and midi scale .

But as you have noticed this is not enough to cover the frequency of a filter or osc(for linear fm) that takes linear input values because 127 is just 127 HZ , so you need to convert them .

There are lots of methods ,you either set your knob range 1-127 and use the exponentiator to convert it to hertz values or you can use a knob range 0-1 (with decimal points for fine tuning ) and scale it afterwards .

FOr example , you want the travel range of a knob-slider to gradually increase it's effect ( meaning less changes in the beginning and more towards the end ) for that you can use range 0-1 and square or cube it , at halfway position 0.5 it will actually output 0.25 when squared , or 0.125 when cubed .etc

ALSo , for instant atttacks you wan't negative values for P range since zero in p range is not exactly zero etc...