Scriptnode modchain issue.

[lalalandsynth]

So I am recreating the ensemble section of the Logan String Melody Mk1.

This used to work (pre-develop) but after develop something happened.
First I had to change the stk delays to the JDSP, which is fine.
But it seems that my main modchain does not modulate anything.
See pic below.

I made a test modulator and that seems to modulate both gain and delays.
I tried connecting one of the peak-core outputs of the main modchain to a Gain module as well and it does not do anything.

If I open in an earlier version of HISE it works fine.

See pic Below.

Main Scriptnode

Main Modchain.

[lalalandsynth]

@lalalandsynth SOLVED:
It was still "sending" to the previous delays ...somehow .
when I removed those connections , its fine.

[Christoph Hart]

@lalalandsynth Just from looking at the patch I think you're overusing the send / receive nodes.

I would suggest to use the "normal" signal flow as much as possible and only resort to the send / receive nodes when you actually need the routing (for feedback loops et al).

If you need three mod signals that are summed up to a single mod signal try to use a split node inside the mod chain and a peak node after the split node.

[ustk]

@lalalandsynth @Christoph-Hart Sometimes when deleting modulating nodes, the connection persists in the modulated one and prevent for it to work normally until you remove the connection manually

[lalalandsynth]

@christoph-hart Its been a while since I made this patch so maybe I should revisit it.
At the time I felt I needed the Send and returns to make this work.

Its essentially 2 LFO´s 0.6hz and 6 hz blended together.

Then I need 3 versions of those LFO´s" blended together in a precise manner" and phase shift applied to each one or more precisely 120 deg and 240 to 2 of them..

Then I need control of the amount of each blended/shifted version as well as a global amount into each delay.

Not sure why I did it this way but cannot see another way to achieve this at a glance but will revisit it now that I have it working.

[lalalandsynth]

@christoph-hart I was using Math.Mul (although I cannot remember why )to achieve the phase shift , is there a more precise way to achieve phase shift now?

EDIT ! I just realised , I must have used delays for the offset but they were STK so they are gone !Makes sense why I dont have offset anymore.

[ustk]

@lalalandsynth said in Scriptnode modchain issue.:

is there a more precise way to achieve phase shift now?

A fixed delay

[lalalandsynth]

@ustk Of course . No idea why I used mul ?

[lalalandsynth]

Man would it be helpful to have a scope or something to look at multiple modulators at the same time and compare, probably useful for a lot of things.
Unless there is and I am not aware ?

Have to go to photoshop to see that my phase offset is gone.

[Matt_SF]

@lalalandsynth if you mul by -1 you get the signal in opposite phase... ?

[lalalandsynth]

@matt_sf If I remember correctly yes.

Like i said , its been a long time , cant really remember how I achieved the phase shift , let alone a precise one.

EDIT ! I just realised , I must have used delays for the offset but they were STK so they are gone !Makes sense why I dont have offset anymore.

[Christoph Hart]

@lalalandsynth The oscillators have a phase parameter now, so maybe you don't need the delay.

I would use 6 oscillators with the correct phase offsets.

[ustk]

@christoph-hart The phase delay would be cool for this but with a parameter for the phase percentage on top of the existing frequency.

[lalalandsynth]

@christoph-hart OK, will try that .

Is it possible that the fixed delays do not work , I have to use the JDelay for the phase offset , fix delay does nothing.?

For some reason the ensemble sounds much better in a previous HISE version.

[lalalandsynth]

So in my previous version I had 3 stereo delays - total 6 delays.
One stereo delay had one of the delays modulated in the inverted direction , I did this by inverting the max/min of the core peak.

I cannot seem to be able to achieve this in the new develop.
I can change the max/min on the destination (delay time knob) and that does something to the sound but I cannot see the modulation indicator of the knob going in the inverse direction?

If I invert it at the core.peak , it does nothing ?

This is how its set in the older project and worked fine.

This is the effect on the modulators - inverted modulation.
<

EDIT: Figured it out - make sticky ( arrow icon next to button ) invert range.

[ustk]

@lalalandsynth Indeed, it's broken somehow...

[lalalandsynth]

@ustk Figured it out - make sticky ( arrow icon next to button ) invert range.

[Matt_SF]

@lalalandsynth Use the range editor and edit the min/max values by SHIFT-click on them :

@lalalandsynth said in Scriptnode modchain issue.:

invert range

Edit : you can set the 'start' value greater than the 'end' value right away, they will be displayed as min & max but the curve will be inverted.

[Matt_SF]

@ustk since the new_layout & last develop version, if you don't use the range editor to edit the values, you have to define them for both the source AND the target. Otherwise that doesn't work.
In old develop branch, editing just the source was sufficient.

[lalalandsynth]

@matt_sf Why do they come up like this by default ?
Shouldnt they be linear ?