Hacker News Reader

Comment by iforgotpassword | Hacker News Reader

> who don't bat an eye on the perspective of having a home server, sucking up electricity 24/7.

Maybe they need it? A router isn't a home server. But even if, why do you keep your power efficient router running 24/7? You could turn it off when you're not home to save even more energy.

> a full-fledged PC (probably consuming like x100 the power

Maybe do some research before claims like these. He states this pc idles at 48W. Please show me a router capable of handling 25gbit/s that consumes 0,5W at idle.

> how would you build a "something" that monitors the network for packets sent to powered-off machines, then somehow caches the request, powers the destination machine On, and finally lets the request continue to its target?

That's what "wake on LAN" is for.

View on HN

> But even if, why do you keep your power efficient router running 24/7? You could turn it off when you're not home to save even more energy.

You jest, but I turn off my router for the night and each time I leave home for more than a day. Not just router to be exact, everything that's connected to a power strip goes down, as I turn off all the power strips. (Not OP.)

Not a fan of the "turning the power strips off" method. Power supply failures are the number one reason why expensive electronics turn to bricks and subjecting them to the large inrush current that happens on the primary side every time you do the mechanical switch thing is a great way to significantly accelerate that process. Penny wise, pound foolish kind of thing.

That's the primary method of turning embedded devices on and off though. :) During development it's done around the clock and noone bats an eye. Rule of thumb: if a device doesn't have a power button, it's fine to turn off using the power strip.

I'd assume because development process doesn't need to deal with repercussions of accelerated power supply degradation

What do you mean? Things that don’t have an in-built switch, are meant to be fine hot plugging and as such, controlling via a power strip is fine.

Why not just go all in and shut off at breakers?

Breakers are not made for very frequent on/off cycles. They won't last as long as a light switch for instance. I've never seen a light switch wear out, but I have seen breakers wear out (at which point they trip very easily, and eventually seem to just permanently fail open).

Fridge.

Then don’t flip the fridge breaker.

Plenty places don't have a separate breaker for that.

I have physical switches on or near every outlet. Anything that's not actively being used gets turned off. With the exception of my oven and hob, because those are nearly impossible to reach. I use roughly 2kWh/day. This approach probably doesn't make much sense if you're using an order of magnitude more power. But it makes sense for me.

Crazy. My wife and I used 619 kWh last month

We were at 5MWh last month, so it’s all relative. Don’t feel bad.

That is impressive! Did you do that on a 200A 120/240V service?

What is your load? Ac?

It’s 240V/400A split-phase service, PNW, around $0.11/kWh. Air conditioning is a big part of it when the temperatures are high. Also charging multiple bEV, a lot of 24x7 loads: half-dozen fridges, a couple large freezers, wine cellar, about 1kW of IT+AV equipment, etc.

Not the person you were replying to, but my June statement lists 1,800 kWh. I'm in a two-bedroom apartment in Texas. The primary consumer of power by far is AC, which is kept at 68 degrees in the summer (partly for temperature, partly for humidity). My effective rate after all the fees was $0.11/kWh.

Reply on news.ycombinator.com