Hello All,
Just some thoughts on the topic:
- The architecture of Lattice PAC is very purpose-specific, tailored for control-loop, monitoring and power management tasks. It is in no way a general-purpose analog platform.
- Triad VCA is a better approach on the technical side. However, it has its own drawbacks:
1. It is explicitly stated that the platform is patented, so using it easily results in vendor lock-in with a single source.
2. They seem to use "usual" fabs behind the set and don't seem to be interested in the "custom IC in single quantities for realistic end-user price" business (as free software can be recompiled by the user, free Si shall be re-tape-out-able by the end user)
3. The concerns mentioned by David also hold (hidden circuitry in pads, ...)
- If a IC project exists that is analog and open-source and useful simultaneously: I don't think so. Analog designs are inseparable from the PDK and technology they were made for: a design made for an university or research process lacks the manufacturing capability that would be needed to be used for real-life volumes, while a design tainted by the PDK of a commercial process would be born secret (if everyone knows that R=Rsq*(L/W), and Rsq is secret, then any R,W,L triplets for that Rsq would also be secret).
Regards,
Ferenc