Because of the relatively poor state of multithreading in Nim and the reliance on external libraries like Arraymancer for heavy numerical workloads (also the performance issues with boxed values due to `ref object` everywhere), I started writing a language from scratch, with built-in support for concurrency via parallel blocks (without macros) and a C backend, similar to Nim.
GC is optional and the stdlib will work with or without the GC.
Example:
int glob_value = 0
float glob_value_2 = 0.0
parallel:
glob_value = some_heavy_task()
glob_value_2 = some_other_heavy_task()
Also there are parallel fors, like so:
parallel for x = 1 to 5:
print "x = %d" | x
parallel for y = 10 to 20:
print "y = %d" | y
sleep 0.1
print "Nested parallel for loop completed."
Compiler implemented in Go, originally with Participle, recursive-descent approach. All examples in the examples directory compile.
Parallel/threads is some whole other can of worms, of course. It is unfortunate that the stdlib is weak, both here and for numerics, and for other things, and that people are as dependency allergic as in C culture.
Anyway, "easier to optimize" is often subjective, and I don't mean to discourage you.