I am considering rearranging the audio system by adding the 5-channel LPF inserting after the oscillators.　This is a common configuration for high-end polyphonic analog synthesizer, but the problem is the load and memory consumption required for processing, and the allocation of resources that are currently at the limit is the key to realization.

The current configuration with 4 channels of LPFs controls 2 channels of stereo pairs separately.　　If the extra control system is switched to the delay system, the user interface will be organized, which is a good thing….

Before writing a new code, let’s take a look at the current status of the system.

First of all, there are two oscillator outputs, normal and transition, and strictly speaking, the correct approach would be to reserve LPF for 2×5=10 units for each of them, but that seems to be too much considering the consumption of RAM1.　In reality, we may have to adopt a method of switching inputs according to the address selected by WaveEdit, but this method may also consume new resources to the extent that the wiring becomes more complicated.

The LPF code should be reorganized into a subroutine for each corresponding CH, and the function of selecting the output of the oscillator should be inside here.

After the LPF is duplicated,

The extra parameters were made to correspond to the controller of the Delay system.

In addition, as a countermeasure against crashes that occur when an excessive value is input to DelayLevel, the maximum value is restricted.

As a result of the refurbishment, the sound quality became clearer by moving the LPF to the front of the CombFilter.　

This is a benefit of the extra dynamic range obtained by placing an LPF for each oscillator, and the unpleasant distortion that sometimes occurred in the process of operating the transition has been eliminated.