[released] Masterpiece Generator - No AI, No Samples: 8-Track Algorithmic Groovebox

[mike_mgen_fun]

so, i've played with it for an hour, and got some thoughts about what's missing:

• ability to change the sample rate — it runs at 44100 but everything in my system runs at 48000.
• ability to make it MIDI clock slave. i'm ok with what it is but would like to run another instance when 8 tracks is not enough, and keep them in sync.
• MIDI mapping to connect MIDI controllers.
• since it's a synth-based groovebox — a super saw instrument.

Thanks a lot for the feedback and for taking the time to test it — really appreciate it!

1. I’ll add sample rate switching within the next week.
2. That’s an interesting idea — I honestly haven’t thought about running multiple instances like that, but I’ll look into it.
3. This would require adding MIDI input support — I’ll put it on the roadmap.
4. Supersaw — yes, I’ve been thinking about it as well, will definitely work on adding it.

Really glad to see so many great ideas — it’s very motivating and makes me want to keep pushing these features forward.

[mike_mgen_fun]

wait… have we just got a usable groovebox app on Linux?
there is serious lack of apps of this kind on Linux, so someone had to be first.

Linux is my home system — I work on it and relax on it.
So skipping Linux support when building something like this would feel like a bit of a betrayal

Really glad to hear it’s useful — that means a lot.

[chaocrator]

2. That’s an interesting idea — I honestly haven’t thought about running multiple instances like that, but I’ll look into it.

i just thought that launching another instance is definitely simpler than asking you about increasing the number of tracks
and indeed, another instance just launches and works, so the only problem (from the user's point of view) is keeping them in sync with each other.

as for me, 8 tracks would be enough if one of them was a full-blown drumkit (like on Roland hardware grooveboxes from 90s) but MG has very different design, so running another instance is pretty obvious solution.

[mike_mgen_fun]

New version 1.0.3 is available. What's new:
1. Selectable buffer sizes (128–1024 samples)
2. Selectable sample rates (44.1 / 48 kHz)

[chaocrator]

1.0.3 appears to behave better with 48 kHz on my system — at least i haven't got sporadic xruns yet.

[mike_mgen_fun]

v1.0.5 is now available

This update makes everyday use smoother:

- Better bass and lead sound (less harsh, more musical)
- More usable random sounds (fewer “broken” patches)
- Easier sequencer editing:
- LMB drag = move one step
- RMB drag = shift the whole row

[mike_mgen_fun]

v1.0.6 is now availavle:
- Sketch mode was removed; the workflow is now unified around Pattern Bank;
- A clear unsaved-changes model was introduced for bank patterns: edits are no longer auto-saved;
- A confirmation dialog for unsaved changes was added for critical actions (navigation, app exit) with Save / Cancel / Discard options;
- Save-on-close behavior was fixed: patterns are no longer silently overwritten.
- Copy/Paste in Pattern Bank was fixed: it now transfers the full pattern bundle (rhythm + instruments/state), not just step data;
- Redundant UI control actions were removed to keep a single source of truth for mode switching;
- A Song generation regression was fixed: distinct Break/Outro sections (breakdown and progressive stripping) were restored, including the fallback pipeline;
- Accent and slide expression for leads and basses was improved: accent is more audible, and slide is more stable and clearly heard in the mix.

[Uncle E]

Would it be possible to add the bass synths as options for the lead tracks? It seems like those would sound pretty good as lead synths.

btw, considering your low price, you could probably sell plugin licenses separately, if you ever choose to go that route.

[mike_mgen_fun]

Would it be possible to add the bass synths as options for the lead tracks? It seems like those would sound pretty good as lead synths.

btw, considering your low price, you could probably sell plugin licenses separately, if you ever choose to go that route.

Hmm, that's an interesting idea. I will definitely test it. thank you.
according to my first impressions, the project is unsuccessful and people don't need it. I will maintain and polish it, but I don't see any point in developing it globally at all, people aren't interested in this.

[Uncle E]

People prefer to buy plugins. You can take a look at my store, it’s almost all plugins: www.jrrshop.com