The road to native performance

A common question is how fast Norscode really is, and how fast it can become. The short answer: the language is moving upward, step by step, and the goal is native performance — without letting go of readability and safety.

From runtime to machine code

Today a finished program runs as a linked bundle via the runtime. It is compiled and statically typed, and well ahead of interpreted languages — but between the program and the machine there is still a layer. The next step is AOT: compiling straight to native machine code for Linux, macOS and Windows. Then the intermediate layer disappears, and you are left with an entirely ordinary executable file.

What AOT gives

Removing the interpretation layer is the single biggest leap a language can make in performance. Computation-heavy code that previously went through the runtime is now run directly by the machine. For the vast majority of services, that means performance stops being something you think about.

Honest about the ceiling

Norscode has a garbage collector that cleans up memory for you. That is a large part of why the language is safe and easy to read — and at the same time the reason the very last layer of performance belongs to languages with manual memory control, like Rust and C. With AOT, Norscode ends up in the company of fast, compiled languages like Go and C#: more than enough for almost everything, but not quite on the metal.

Being built now

The toolchain already has the building blocks for this — code generation for x86-64 and ARM64, and writing native executables for the three major platforms. It is not finished, and we do not promise numbers we have not measured. But the direction is set, and it is set without sacrificing what Norscode is for.

Back to the overview