Hacker News Reader

Comment by pansa2 | Hacker News Reader

> I find that the dumbest code is the best code.

This is the idea behind Golang, isn’t it? That everything should be written out explicitly and not hidden behind abstractions. Some people love that, others hate it.

View on HN

I'd love to try Golang, but as a game developer, the gc makes it a no go due to perf. I begrudgingly accept C# due to unity. GC is a different topic altogether, but has also always been a pain. You end writing code to avoid allocations, which at that point you ask "Why am I not just writing C++?" I also discovered GC makes bad programmers worse by allowing them not to care about ownership, enabling them to develop systems in an adhoc way. Some programmers need a segfault screaming at them to make them realise they are writing terrible code.

While I’m not suggesting using Go for game development, don’t most games include code written in a language with garbage collection? If not C#, then a scripting language like Lua?

And everyone wishes they hadn't :P Even modders that use Lua have to write code that avoids allocations, caching and reusing all the objects they can. I can't stress enough how much time is spent on optimising code to avoid GC pauses. The main goal of my C# coding style is avoiding unnecessary allocations at all costs. I'm not exaggerating.

Not to mention acceleration is an issue with the go/c ffi

Weirdly enough I actually have a 3D project that I started in C# that I moved over to Go so I have a little experience with it. So far it's been much less painful on the GC front. Basically the GC pauses are around 0.5 ms even on very large heaps [1], so that changes the conversation from "everything must must be manually managed / pooled in a language designed to offer zero help with this in order not to drop frames" to "you'll make frame rate as long as you can leave some performance on the table and don't go crazy on hot paths" which was a lot easier to live with.

It's also much easier to avoid allocations to begin with than C# - pointers can be used as interface types without boxing (and taking interior pointers to fields / array elements is allowed), and it's possible to allocate regular Go objects / arrays in unmanaged memory and feed it into any API.

Obviously it's not really designed for it and the ecosystem isn't there but using it hasn't been completely terrible so far.

[1] https://blog.golang.org/ismmkeynote

Some games have servers, too. And many of those are request-response (not "hold socket open") servers - which are ideally suited for GC.

There is always a middle ground as there are other native languages. In regards to C++ you can limit yourself to convenient feature set you are comfortable with.

C++ can be a real bitch and commit one to a mental institution or it can be very helpful if one does not try to play PhD.

I write servers and some other stuff in C++ and find it incredibly easy to use. I just do not do any esoteric things.

Reply on news.ycombinator.com