You should try running powertop on it. It will scrape sysfs and look for things that seem misconfigured, and suggest changes to fix them. On one of my machines it enabled some peripheral power saving mode that made a pretty dramatic saving!
(I also heard that it sometimes suggests power saving modes that are usually switched off for a good reason, like apparently you really don't want some USB controllers going into certain sleep modes as they take seconds to come back).
It's a server motherboard, not a consumer motherboard, so it doesn't surprise me that power management is not great. AMD Rome is also not exactly known for its low idle power.
Looking at powertop, the various qemu virtual machines have high events/sec compared to host processes. There are also a couple of timers - tick_nohz_handler and hrtimer_wakeup - which seem to generate hundreds of events per second.
I wonder if there's a business model in leasing cars from Mexico to Americans and swapping around once a year to get around the problem of having Mexican plates. Then you can get Chinese cars into America.
This made me laugh out loud thank you. They will clamp down on this so fast though. If its something that will help the average joe it will never get done but something like this threatens leadership so it will get shut down in record time.
If theres any Chinese entrepreneurs that have a line into their EV companies reading this, use some of that China speed and get on this now. You might as well squeeze out a little profit before they clamp it down! :D
I suspect solar + batteries will be dominant (but should not be the exclusive source) in places which get relatively good sun even in the winter. California and Hawaii, for example, will do pretty well with a mix that is heavy on those technologies. But if you're in North Dakota or Alaska and want to electrify your grid, you might not have a good time in December and January.
There's arguments to be made about having more transmission so you can move electricity from one place to another, but that's also expensive and difficult to build and comes with downsides like vulnerability to natural disasters and attack along a much longer path. Or, as in California, the transmission is its own latent source of disaster that can immolate the state.
I'm not an expert, but I suspect there will be a few driving use cases. Electrifying places far from the equator means heat pumps and/or district heating and providing for their power demand in the dark. In some places wind can do that, but not everywhere.
There are also loads that want very large, high-availability power and/or process heat. Reactors would pair well with things like metal refining or electrolysis to get the hydrogen for ammonia production.
At the end of the day, there's never one source of energy which is a silver bullet for everything and the best approach is probably a diverse mix of supply.
Electrifying heat far away from the equator is challenging. My area is largely heated by natural gas furnaces. During a cycle called a polar vortex, the utility company sent out a call to people to lower their thermostats because they were struggling to keep the lines pressurized with how much demand there was.
The wind chill would drop below -40 most nights, sometimes significantly lower. Wind power won't help much because they need to shut down at the worst times- either too windy or icy.
As much as I love heat pumps, having thousands upon thousands of homes switching to resistive heating because the pumps can't keep up in the evening is going to get ugly.
District heating won't save you; the metro doesn't actually burn enough stuff to heat the cities and a significant part of the population is in semi or very rural areas that wouldn't benefit anyway.
Edit: that same metro just put together a fund to renovate a few blocks of an underserved area. It's in the millions of dollars. I can't imagine the cost of converting the entire area to district heating; it would surely eclipse the entire government budget. This is the sort of thing that will only happen if you have the kind of fiat power of an imaginary wand.
Copenhagen and much of its suburbs installed district heating, mostly after the oil crises in the 1970s.
I found this article very interesting, although it didn't tell me how many kilometres of district heating pipe were laid each decade: https://www.mdpi.com/1996-1073/15/24/9281
The city I was referring to has a population density half that of Copenhagen. Much of its energy is also supplied by power plants out in the suburbs rather than being closely sited, so it would likely need to start with the least economical places. The main city also hasn't recovered terribly well from the COVID lockdowns and boom of remote work keeping a lot of computers away.
It wouldn't be impossible to do, but they would need someone else to help fund it.
My point is that Copenhagen is easy. Both "EU defined" metros in Spain are 2x and 3x than that. So Copenhagen is not an example of "it is posible everywhere".
I mean, for example, in Madrid a neighbourhood has no electricity because is has been taken over by drug gangs. But at the same time it has electricity to grow marihuana. I guess that is a problem Copenhagen does now have.
You don't need to burn stuff for district heating.
Giant heat pumps can source heat from seas, rivers, underground reservoirs etc and the latter can even store energy seasonally that might otherwise be curtailed or wasted by e.g. data centers.
There's a lot of unusual and flaky design in the boring plumbing parts of these things that makes them prone to self destruction. For example, using the MCU to make discrete buck regulators that power the MCU is a o.O choice that I think has no reasonable justification. There are perfectly wonderful multi output integrated chips with external synchronization that cost almost no money, take up less space, are more performant, and make it impossible for software to immolate the hardware upon which it runs.
Gifted embedded software, but the electronics is simply not that robust. The company could really use an electrical engineer to complement its software engineer.
The QMX works well enough that I'm not going to critique the choice of voltage regulation or other plumbing bits. The boring parts work, even if they weren't designed by a professional EE. It uses low power on receive and is reliable, which are my primary concerns. I don't expect to get the same design as a team of Yaesu engineers would use on a $5k transceiver. If it's not hurting RF performance, then it is fairly low on the wish list.
Overall, the kit is still a killer value and buildable by someone who can follow directions and is decent at soldering. This is in stark contrast to many other kits which require a lot of sleuthing to figure out issues because the instructions aren't clear, or they changed some components and didn't update their documentation. And these other kits often don't have any better design for their power management sides either.
There have been a few board revisions to address some of the bigger issues that have come up with early board spins, and I like that mods have been made public for most of the early builders to keep their kits working.
The most common problem seem to be blown transistors from transmitting into a mismatched load, and that is mostly the operator's fault and common in many radios. I like that Hans put some SWR protection in the QMX to keep the finals from blowing as often as they might.
