News for hackers. Stuff that matters.
Re:Nim 1.0 by guggle on Tuesday September 24, 2019 @08:06AM (#21059103)
> so I'm not sure how big of an issue this is in reality.

I have audio programming in mind. Not that you can't do audio programming in GC-enabled languages, it's just that's it's quite frown upon in this circle (for good reasons). I'm sure there are workarounds though.

Re:Nim 1.0 by rayman22201 on Tuesday September 24, 2019 @01:54PM (#21062846)
Audio programming in Nim is completely possible and reasonable.

Here is an example of a super collider plugin written entirely in Nim with the GC turned off: https://forum.nim-lang.org/t/3625

The author seemed to find the experience very pleasant, and the performance was great.

Re:Nim 1.0 by etatoby on Monday September 30, 2019 @11:13AM (#21116189)
Not (yet) a Nim user, but another post mentioned that GC is thread-specific, so you can remove all traces of GC-types from your real-time thread(s) in order to avoid deallocations at the wrong times.

A related question is whether you can disable GC on a thread by thread basis.

Re:Nim 1.0 by PMunch on Tuesday September 24, 2019 @09:23AM (#21059840)
Audio programming? And why is it frowned upon? If it is because of the Java-like GC freezes that is something you can ensure will not happen in Nim by turning on manual control and only running it when its suitable.
Re:Nim 1.0 by monetus on Tuesday September 24, 2019 @01:00PM (#21062246)
Not the person you responded to, but yes. The audio must never be interrupted, everything else first.