JNEXT
Skip to content

5.4 Sound

JNEXT emulates everything the Next can make noise with, and none of it needs setting up:

  • Beeper — the original single-bit speaker.
  • AY / TurboSound — the AY sound chip that gave the 128K its music, up to three of them (TurboSound) on the Next, with stereo panning.
  • DAC — 8-bit sample playback (Specdrum, Soundrive and Covox compatible).

Which of these a program can actually use is up to the program and the machine it was written for. Stereo separation and channel balance are likewise set by the software, not by you — they are properties of the emulated hardware.

Settings > Preferences > Audio can trim or boost the master, beeper, each TurboSound AY chip, and the DAC family independently from -24 dB to +24 dB. For a single run, use --audio-gain-db DB for the master and --audio-gain-beeper-db DB, --audio-gain-ay0-db DB, --audio-gain-ay1-db DB, --audio-gain-ay2-db DB, or --audio-gain-dac-db DB for one subsystem. Preferences changes apply live; all six gains affect playback and WAV/video recordings without changing the emulated machine's audio state. Complete mute remains available through Settings > Preferences > Startup > Start muted or --silent; muting also skips sound synthesis entirely, which speeds up runs that do not need audio. Tape loading still works while muted.

If the sound stutters or clicks

JNEXT paces itself against your sound card, so audio should stay clean indefinitely. Glitches almost always mean the host cannot keep up.

Check the two frame-rate figures in the status bar:

FPS: 50.0 emu / 50.0 shown

emu is how fast the machine is being simulated; shown is how many of those frames reached the screen. If emu sits below about 50, your host is behind, and the audio will suffer with it. Things that help, roughly in order:

  1. Close whatever else is busy on the machine.
  2. Turn off the CRT filter and drop the window scale.
  3. Reduce the emulated CPU speed (Machine > CPU Speed) if the software tolerates it — 28 MHz is many times the work of 3.5 MHz.
  4. Use --silent if you do not need sound at all.

shown being lower than emu on its own is not a fault: it is expected when you run faster than 1x, and during brief audio catch-up.