Digitone II Mods / bug fixes / LFO hold (superseded)
LFO Hold
The first fix for LFO TRIG MODE ONE and HALF. It is kept here as a record: what it did,
how it was tested, and why it was replaced. It is no longer offered as a download.
Superseded on 2026-10-09 by the LFO ONE/HALF fix.
Two things came to light after this fix was published:
- The MIDI tracks’ LFOs run through a second copy of the LFO code, which this fix did
not touch. On a MIDI track, ONE and HALF still jumped when they stopped.
- A second community report: with the start phase (SPH) moved, ONE and HALF end too early,
or in HALF too late. The firmware stops them at fixed points of the cycle, not one cycle (or
half) after the trig.
Source
The LFO ONE/HALF fix does what this one did, on both copies, and fixes the length too.
The two write the same bytes, so they were never meant to be combined.
What it did
- Fixed
- LFO TRIG MODE ONE and HALF jumping to the centre when they stop, instead of holding the value they reached
- ONE
- After one cycle the LFO stayed at the value its waveform ended on. A square on the filter ended closed and stayed closed
- HALF
- After half a cycle the LFO stayed at the waveform’s middle value. A square stayed open. A triangle stayed at its peak
- Why
- The firmware stops these modes by writing a fixed value per waveform into the LFO’s output, the centre for nearly all of them. The fix removed those two writes (audio tracks)
- Changes
- 2 in-place edits of 4 B in section 3, nothing appended
- Tested
- Flashed 2026-10-08 against factory 1.11 on the same patch (a square LFO on the filter). With the fix, ONE and HALF held. Without the fix, they jumped. TRIG and FREE did not change. The click at the edges of a square came from the factory firmware: it was the same with and without the fix
- Code
src/dnfw/mods/lfohold.py and site/js/mods/lfohold.js in the repository, kept as published
Reported by the community on official 1.10 as well.
Source