[sdiy] AI code generation

[Eric Frampton]

Hey! Thanks.
Some short answers follow. Look for a personal email with the longer answers and attachments, I’m not going to spam the group with all that.

Testing is 3 layers of static/structural proofreading, then running on emulated Z80 machine code internally, then sending binaries out to me along with Swift test beds for the emulator to run on my end.

LR = Lionel Richie. I’ve been keyboard tech’ing for him and his band for 11 years.

I heard that Tony K. (of Encore and Sequential fame) is working on a new CPU board and possibly a replacement for the MIDI/Sequencer board for those machines that don’t have one already, so I’m standing by to see what kind of awesomeness he whips up.

e

> On Jul 27, 2026, at 6:40 AM, cheater cheater via Synth-diy <synth-diy at synth-diy.org> wrote:
> 
> Hi Eric,
> Tremendous description, thank you so much.
> 
>> We figured out a 4-layer protocol for us to check its work as we finished each new build
> 
> What are the 4 layers? Can you tell us something about them?
> 
>> if there’s any new code to be built, we’d produce a “FINDINGS” document with what we’d learned and what we’d expect the code to look like, then start a new BUILD thread to do the actual coding
> 
> Could you please show some such FINDINGS docs? And could you show a
> BUILD thread? Even if they're boring, they would provide insight into
> how such things work internally. Feel free to spam attachments and/or
> send them off-list.
> 
> Who are LR? :)
> 
> Have you considered upgrading the CPU? It seems like a pico would be
> insanely more effective at quickly parsing things compared to a
> memorymoog; a full fledged pi could even hook up to an external
> monitor so you don't have to squint over a stupid tiny LCD when
> browsing patches and menus; a combination pico and pi could work where
> the pico does the realtime stuff, and the pi communicates with it for
> the gui and video output and e.g. sd card storage for patches.
>