Hacker News Reader

Comment by miohtama | Hacker News Reader

Memory leaks are a feature of a software, not hardware. I read the article and while the Apple and Adobe software can be blamed, there is nothing in this post that it would pin this on the new hardware and the author does not seem to be technical enough to understand this.

View on HN

There’s nothing saying that the ARM macOS port isn’t to blame. Porting operating systems is hard.

Totally agree. I read the article as, when I am on my M1 I often experience memory pressure related issues. Not, because of my M1 I often experience memory pressure related issues.

It's very likely future software updates can help address these issues.

Honestly, the machines haven't had wide deployment in the field for all that long for Applications as huge as Photoshop. It takes time to discover, fix, and redeploy things like Photoshop. And it's not like Photoshop never leaked memory on an x86.

> There’s nothing saying that the ARM macOS port isn’t to blame.

There is. People have been running ARM macOS for about a year and I haven’t seen this complaint before.

The big thing is that M1 chips have RAM in the same package, which allows for performance improvement if the software takes advantage of the new architecture. It turns out that the system they came up with offers good performance but doesn't tolerate memory leaks very well.

It is made worse by the fact M1 MacBooks are limited to 16 GB of RAM, non upgradable of course.

So yes, it is a software problem, and it will get better, but the M1 is particularly sensitive to it.

> It turns out that the system they came up with offers good performance but doesn't tolerate memory leaks very well.

Can you be more specific here? It seems natural that a system with limited memory doesn't tolerate memory leaks very well.

It is just a wild guess based on the article, but the unified architecture is mentioned. So maybe each subsystem (system RAM and GPU in particular) all have direct access and therefore more of a tendency to fight for resources and maybe also complex and therefore more error prone locking patterns over shared memory.

Why doesn’t it tolerate memory leaks very well? The fact that the system remains responsive and able to close the application indicates that it does tolerate it well. I’d say macOS in general tolerates memory leaks much better than Linux does, which if you don’t have a swap disk and you run out of memory the entire system just stops functioning.

It's obvious you haven't read the article and are reacting to the headline. In fact, the author highlights that the issue is probably with the system software (macOS):

That’s extremely similar to the MacBook Pro and Monterey issues I’ve read this week—except I’m still running Big Sur ... It’s possible that macOS isn’t managing this unified memory structure properly, and will continue to allocate RAM beyond what is available without freeing up RAM that is no longer needed. This is commonly referred to as a “memory leak.” Performance gradually deteriorates until you need to either wait for the RAM to clear, force-quit the app, or restart the machine ... it’s entirely possible that Apple fixes it in a future version of macOS without ever addressing the issue publicly. That’s what happened with the excessive SSD usage earlier this year, though that Apple says that was a “data reporting error” and not an actual problem. The memory bug is an actual issue that needs to be fixed as soon as possible.

It's a fair criticism that the headline is misleading, even if the text is accurate.

It might be in Rosetta 2, or in the M1 builds of some apps.

Still not hardware

Apple hardware and software are closely linked, it's not like you can use Rosetta2 on any other hardware of your choice.

Rosetta 2 is an essential feature of M1 Macs due to their hardware, that is not is not a feature of Intel Macs.

Reply on news.ycombinator.com