eusti
|||
Nicht die Lösung, aber hier das Statement von Vladi, das dem Update beigefügt ist:
Multi-Mode Tips & Tricks
The M’s multimode has received considerable criticism over the years. I understand this, as
users often expect it to function similarly to digital-only synthesizers. However, the multimode
was originally intended as a highly flexible system, allowing users to configure the four
dedicated outputs for four independent instruments and process them via external effects. This
led to a complicated voice management structure with reservation of hardware voices to the
outputs, parts priority, and an inability to steal voices between parts.
Later additions to the synthesizer engine have further complicated matters. However, it is
possible to create a working multi-arrangement by following several building rules. Here I will
attempt to explain them.
Rule 1 – Parts Priority for Voice Reserve! The V.Total parameter of an arrangement sets the
number of hardware voices that M will mark as reserved and locked by the DSP for each part.
However, this is not a guaranteed reservation, as one might assume. Part priority (1 → 2 → 3 →
4) plays a crucial role here; priority always determines voice allocation. For example, if you
reserve six voices for part 1, four for part 2, and two for part 3 on an 8-voice M, the actual
distribution will be six voices for part 1, but only two for part 2 and zero for part 3. On a
16-voice M, reservations are honored as requested. Therefore, rule one is to carefully manage
your voices, considering hardware limitations and part priority.
Rule 2 — Mind the PLAYMODE parameter for the part’s sound. It is makes no sence to set more than
1 voices for Monophonic sounds, the Voice Stealing does not matter too.
Rule 3 – Only the First Part Can Use Modern Mode! This is an important restriction related to
the limited memory resources of the M’s DSP. Remaining parts can be in classic or MVA modes, but
only the first part can utilize extended RAM for the “modern” (i.e., Microwave 2) wavetable
layout.
Rule 4 – Consequently, No Transition-Enabled Parts Can Mix with Part One in Modern Mode! Because
transition replay for “classic” (i.e., PPG 2.3 / Microwave 1) mode uses the same extended RAM as
the wavetables for “modern” mode, loading a part with transitions enabled will immediately
corrupt the wavetable data for modern mode!
Rule 5 – There is Only One Arpeggiator in the M. Loading or creating an arrangement with
multiple parts containing arpeggiator-enabled sounds is pointless; only one (first) part’s
arpeggiator will function.
Rule 6 – Cross-Parts VCF Lock. Regardless of which part is selected, if a part's LOCKVCF
parameter is set to On, it will follow the VCF behavior of another part. This parameter allows
all locked parts to track the controls of the currently selected locked part. For example, if
you have three parts and parts 1 and 3 have LOCKVCF = On while part 2 has LOCKVCF = Off,
adjusting the VCF section’s controls on part 1 will cause part 3 to follow the same settings
(i.e., lock to part 1), and vice versa. However, this will not affect part 2. Consequently,
editing part 2's VCF section will not influence these locked parts.
Rule 7 – Be Mindful of MIDI Filters. The M has a flexible set of MIDI filters capable of
blocking/passing certain messages like mod wheel, pitch bend, program/bank changes, aftertouch,
channel volume, and panorama. However, there is no CC filtering! Therefore, if you have two
parts in an arrangement on the same MIDI channel, both will respond to all CCs received on that
channel. And, don’t forget about the global MIDI filters of the M, which applied first!
Rule 8 – Voices Routed to Different AUX Outputs Still Appear on Main/Headphone Output. The M
does not exclude voices routed to different AUX outputs from the main output. In other words,
the main out is a sum of all Aux outputs (which is also true from the M’s schematic
perspective).
