PCE LFO is broken (or not)

[freq-mod]

I don't know if it's a DefleMask that's incorrect or Furnace, but still: LFO_Ex.zip - wav recording of how it sounds on DM and dmf module in question. Difference is enormous,

[freq-mod]

PCE_FM_N.zip - more of that, even more severe

[freq-mod]

PCE_LFO_test.zip there are some differences with .HES output being played on Mednafen and original module file played in Furnace... which is odd, considering both use the same core

[Fragmare]

Awaiting equipment for proper testing on real hardware

[freq-mod]

PCE_LFO_test.zip real harware recording of an above , courtesy of @laoo

[laoo]

I've recorded it from classic Japanese PC-Engine

[tildearrow]

Conclusion: (thanks @laoo and @Fragmare for hardware tests!)

1. Channel 2 is NOT being muted, even when LFO is on. Only Mednafen is correct.
2. Direction is up. MAME and Mednafen are correct here.
3. There is no delay when starting LFO. Whether this is because the LFO is already running since the ROM does not start at power on is unknown. Ootake and MAME correct here.

The only thing left to test is the LFO noise one

[tildearrow]

Fix applied. Now it matches recording.