Comment by devbent | Hacker News Reader

Comment by devbent | Hacker News Reader

Parent Full thread

IMHO it isn't an AMD or Intel problem, it is a "every driver and piece of hardware on the machine has to cooperate with low power modes" problem.

Apple owns the stack and iterates slowly.

Most laptops change to some of their chips (wifi, graphics, etc) every year. But it can easily take a couple of years of experience working with a shop and writing drivers for it to fix every power usage bug.

For many years (!!) Microsoft actually kept the same wireless chipset in their Surface machines. Was it the latest and greatest? No, but it worked!

Then or course you have software. For years Firefox had a bug where if any open tab has GPU accelerated content, it would prevent the entire machine from going to sleep.

If Microsoft clamps down on that sort of behavior it'll break someone's software (complaints about laptops going to sleep fall into 1 of 2 camps, it happens too often or it doesn't happen enough !)

I think what we're discussing is "S0 sleep" vs S3 sleep, AKA "machine stays running and pretends to sleep while hopefully saving enough power to be okay" vs "the only thing getting power is RAM and the chip(s) that will wake the machine back up when you hit a button". So yes, largely an AMD/Intel problem; while other hardware parts are somewhat relevant, they're secondary. At least, that's how I think it goes; happy to be wrong. I'm sufficiently behind the curve that all I have is second-hand accounts, including people complaining that S0 sleep doesn't even work right on Windows.

Wow I never actually thought about what is happening at the hardware level during sleep. The thought of a trickle charge going to RAM to keep the data alive while everything else sleeps is really cool.

It can really be anywhere. Intel/AMD probably have done due diligence in ensuring that the SoC itself sleeps correctly, but once system integrators start glueing on external hardware like radios, external graphics, floppy drives ( :P ), things get messy fast. For example, a wireless driver which doesn't correctly power down the radios when the system sleeps will end up firing interrupts at the SoC and dragging it back from S3 constantly. Or maybe the SoC does actually power gate correctly but the external graphics doesn't and so you end up burning a non-negligible amount of power over there for no good reason. And so on and on and on.

Integrators here are really on the hook for validating their configuration and chasing down the appropriate vendor to fix it (be it Intel/AMD or, I dunno, Qualcomm).

Reply on news.ycombinator.com