Comment by sharpneli | Hacker News Reader

Comment by sharpneli | Hacker News Reader

Parent Full thread

One must also remember to differentiate the deployment targets. Not every piece of software is deployed in your own machine.

Let's take games as an example. You cannot say to users "Just buy more powerful compute/phone". And even if they would you may have literally millions of machines, due to having millions of users. In which case saving 1 dollar worth of machine per user is savings of millions which is worth a lot of developer time. Naturally you as a dev won't really be paying this, so the incentive is not directly there. But the incentive comes from people just not buying a game that runs like crap.

In these cases GC is nice and dandy at start but then you run into the inevitable "Why does my game stutter?", after which object pools are introduced and even then some annoying "leaks" may remain. If the amount of time had been used from the get go to actually think about memory management the whole issue could have been avoided.

Yes, but GC and object re-use aren't mutually exclusive.

Sometimes people argue as if they are, which is understandable because the overlap between semi-FP programming styles and GCd languages is fairly strong, and the FP world pushes immutable state as almost a matter of moral principle. If you tried to do functional programming with purely manual memory management you'd see high overheads too simply because there'd be so many small allocations and copies. In practice though, people tend to think harder about in-place mutation when they don't have a GC to clean up after them.

They aren't mutually exclusive, but if you're using a tracing garbage collector this millennium you're probably using a generational copying collector, and if you're using a generational copying collector you have a write barrier, and the write barrier means that allocating and initializing a new object is less expensive than overwriting all the fields of an existing object, probably by a factor of several. Also, if your reused object is only referenced from the nursery, and it references other objects in the nursery, those objects ought to be collected on a minor collection, but they won't be; instead they'll be promoted, possibly through several generations, until they reach the generation where your reused object is, which will finally allow them to die. Balanced against this you have less frequent minor collections — none at all during times when you're allocating no new objects.

So, with a tracing GC, introducing an object pool will usually make your program slower, not faster. (This is not true with reference counting.)

Reply on news.ycombinator.com