Black Sphere Studios News Today : Breaking News, Live Updates & Top Stories | Vimarsana

Metaprogramming, or the ability to inspect, modify and generate code at compile-time (as opposed to reflection, which is runtime introspection of code), has slowly been gaining momentum. Programmers are finally admitting that, after accidentally inventing turing complete template systems, maybe we should just have proper first-class support for generating code. Rust has macros, Zig has built-in compile time expressions, Nim lets you rewrite the AST however you please, and dependent types have been cropping up all over the place.

You can't have a 4-slot GPU. You just can't. We have finally left sanity behind, with nvidia's 4000 series cards yielding a “clown car” of absurd GPU designs, as GamersNexus put it. These cards are so huge they need “GPU Support Sticks”, which are an actual real thing now. The fact that we insist on relegating the GPU to interfacing with the system while hanging off of a single, increasingly absurd PCIe 6.

I am currently porting an experimental language to Windows. This experimental language was built in C++ with LLVM, and relies heavily on GCC extensions like VLAs and Compound Statement Expressions, which basically made it impossible to build with MSVC (although I have a truly horrifying idea I may attempt later). Luckily, you can now build things on Windows with Clang, which solves a lot of problems. However, clang-cl simply compiles the code - it still uses the Microsoft C++ headers and links to the Microsoft C++ runtime.

What exactly happens when you write Foo* foo = new Foo();? A lot is packed into this one statement, so lets try to break it down. First, this example is allocating new memory on the heap, but in order to understand everything that's going on, we're going to have to explain what it means to declare a variable on the stack. If you already have a good understanding of how the stack works, and how functions do cleanup before returning, feel free to skip to the new statement.