Well, I know many Lisp folks had tried to revive Lisp Machine year after year, but I would say LambdaChip is not aiming that way to go.
The idea behind LambdaChip bases on a prominent trend: dynamic languages are trying to challenge C language in embedded software development. The campaign has MicroPython (and its famous fork CircuitPython), eLua, JerryScript (JS ES5), so on. LambdaChip is trying to join with its Functional Programming features. Another difference is that LambdaChip is not an interpreter, it contains an optimized compiler.
In the past, Functional Programming features are troublesome to optimize to be efficient. For example, the closures, TCO (proper tail call optimization), and Garbage Collector are not easy to optimize for a compact system. But nowadays, in modern compiler theory and practice, we can make it fast. LambdaChip is one of the efforts to show the power of Functional Programming implemented in an optimized compiler.
On the other side, LambdaChip is not only for Scheme programming language. There are reasons why LambdaChip supports Scheme as its first language. The Scheme on LambdaChip is compatible with R7RS-small, which is a minimalism standard Scheme. It's especially suitable for embedded systems. The Laco compiler of LambdaChip also implemented with Scheme. Because Laco uses CPS (Continuation Passing Style) as its IR (Intermediate Representation), and Scheme is suitable to implement it. The Laco compiler is flexible enough to support other Functional featured languages frontend. We planed to implement Lua as its second language, and we do have an unpublished Lua-5.2 frontend. There could be more languages on it in the future.
The expected future of LambdaChip is more like the .Net framework on the embedded system, rather than yet another Lisp Machine. You may reuse the modules written in different languages.
We hope LambdaChip can reduce the complexity of embedded software development. In the IoT era, there will be more developers hired to develop IoT products. We hope LambdaChip can help pure software engineers to pick up embedded development painlessly and work efficiently.