Switching mod sources in core , filter becomes dull

gentleclockdivider

Yep , figuerd that out soon after .

Thanks for the help .

gentleclockdivider

Doube post

gentleclockdivider

And once agian I am stuggling hard here .

IN the screenshot you see the mod sources (red rectangle ) works perfectly .

The bipolar knob ( negative -1, +1)defines the destination , negative value modulates osc 2 pitch(yellow rectangle) , positive value modulates PW (green ) , it also functions as a variable attenuator .

Middle postion(zero )=no mod

Sort of same issue as before , switching between negative (pitch mod ) and positive ( pw ) , the knob affects the base pitch of the osc ,.

Adding a merger , no luck

gentleclockdivider

Here's the essence of the issue , I made a small ensemble out of it without all the other modulations

bIssue i the same , turn mod knob to the left for pitch mode , turn to the right for PW mod , but base pitch also changes

portion.ens

gentleclockdivider

Anyone ?

I know why the issue occurs , the value that is stuck is because whatever value was last send by the mod source and kept in the latch write output and send to the multply memory , and since the modulation knob

( SEL bus) is mutliplying that stuck value, pitch changes occur when it shouldn't

A merge would solve it but I have no idea where to place it ,since I can't be put but it after the bracnhed mods

Seqsual

Not tested but it could work.

portion2.ens

gentleclockdivider

@seqsual , that does it , but do yo have the solution when using the router switch instead instead of the clipping method ?

I would like to know how to get rd of the stuck value

I managed to make improvements with adding an latch , but for some reason it's still outputtung a value of 1

Edit found the solution , the added memory latch did it , and changing the comparison to smaller/equal

Altough your method is much more streamlined

gentleclockdivider

Cried victory too soon , now some left over values from thepulse width mod is there

Still wanting the switch/latch solution

bolabo

Seqsual's clipping method is probably easier to achieve your dual function control, the routing method is going to give you issues zero-ing the modulation when the router switches the clock to each of the 2 destinations, leaving the previous value stuck where it was.

Seqsual

It's the same idea as before - if you are comparing against 0, router will not give you 0 resulting path on both outputs - this can be solved by latching 0 for upper path (which would never give 0) when router's lower output is "active":

portion3.ens

gentleclockdivider

https://community.native-instruments.com/discussion/comment/39695#Comment_39695

THanks , but this structure gives PITCH mod but also alters the PW ( no mod ) when modulation knob is counterclockswise;it should not alter the pw;

Seems that clipping method is the only working one , altough I find that hard to believe

colB

Maybe this?

You might want to stick a dup filter in an appropriate place depending on how FMMOD and PPWW are used downstream

gentleclockdivider

@ colb ,the latch zerolt doesn't seem to be necessary at all

Perhaps I was to fixated on using the SR.C at the routing inlet instead of the Modsource itself

colB

https://community.native-instruments.com/discussion/comment/39800#Comment_39800

I suppose that's because the last 'sel' value that gets used will often be close to zero. No guarantees though (e.g. very fast wide knob movements, or some external modulation), so be careful with that ;)