Any hard drive can fail at any time with or without warning. Worrying too much about individual drive families’ reliability isn’t worth it if you’re dealing with few drives. Worry instead about backups and recovery plans in case it does happen.

Bigger drives have significantly lower power usage per TB, and cost per TB is lowest around 12-16TB. Bigger drives also lets you fit more storage in a given box. Drives 12TB and up are all currently helium filled which run significantly cooler.

Two preferred options in the data hoarder communities are shucking (external drives are cheaper than internal, so remove the case) and buying refurb or grey market drives from vendors like Server Supply or Water Panther. In both cases, the savings are usually big enough that you can simply buy an extra drive to make up for any loss of warranty.

Under US$15/TB is typically a ‘good’ price.

For media serving and deep storage, HDDs are still fine and cheap. For general file storage, consider SSDs to improve IOPS.

I don’t remember if they fully closed the loopholes, but there are inputs that programs cannot catch unless you actually replace the OS.

Here in NZ they do a factory reset on your calculator at the start of every exam.

Exercising eminent domain can mean a long and expensive legal and media process. I’m not sure about Texas (or the rest of the US, for that matter), but many projects in the first world do everything possible to avoid using it.

Boeing doesn’t make many of the parts in the aircraft, especially things like pressurization controllers. Those come from contractors like Honeywell.

What they do is design the systems around the parts, including selecting the desired level of redundancy, and commission the custom parts needed.

The 737 is still mostly a 1960s design built mostly to 1960s rules. There have been plenty of improvements but that’s not the same as a clean sheet design built to be entirely automatic even when stuff breaks.

Even 95% is on the low side. Most residential-grade PV grid-tie inverters are listed as something like 97.5%. Higher voltage versions tend to do better.

Yeah, filters essentially store power during one part of the cycle and release it during another. Net power lost is fairly minimal, though not zero. DC needs filtering too: all those switchmode power supplies are very choppy.

I’m not sure there are any power grids past the tens-of-megawatt range that aren’t just a 2/3/4 terminal HVDC link.

Railway DC supplies usually just have fat rectifiers and transformers from the AC mains to supply fault current/clearing and stability.

Ships are where I would expect to start seeing them arrive, or aircraft.

Almost all land-based standalone DC networks (again, not few-terminal HVDC links) are heavily battery backed and run at battery voltage - that’s not practical once you leave one property.

I’m sure there are some pretty detailed reports and simulations, though. A reduction in cost of multi-kV converters and DC circuit breakers is essential.

PV inverters often have around 1-2% losses. This is not very significant. You also need to convert the voltage anyway because PV output voltage varies with light level.

Buck/boost converters work by converting the DC current to (messy) AC, then back to DC. If you want an isolating converter (necessary for most applications for safety reasons) that converter needs to handle the full power. If it’s non isolating, then it’s proportional to the voltage step.

Frequency provides a somewhat convenient method for all parties to know whether the grid is over- or under- supplied on a sub-second basis. Operating solely on voltage is more prone to oscillation and requires compensation for voltage drop, plus the information is typically lost at buck/boost sites. A DC grid would likely require much more robust and faster real-time comms.

The AC grid relies on significant (>10x overcurrent) short-term (<5s) overload capability. Inrush and motor starting requires small/short overloads (though still significant). Faults are detected and cleared primarily through the excess current drawn. Fuses/breakers in series will all see the same current from the same fault, but we want only the device closest to the fault to operate to minimise disruption. That’s achieved (called discrimination, coordination, or selectivity) by having each device take progressively more time to trip on a fault of a given size, and progressively higher fault current so that the devices upstream still rapidly detect a fault.

RCDs/GFCIs don’t coordinate well because there isn’t enough room between the smallest fault required to be detected and the maximum disconnection time to fit increasingly less sensitive devices.

Generators are perfectly able to provide this extra fault current through short term temperature rise and inertia. Inverters cannot provide 5-fold overcurrent without being significantly oversized. We even install synchronous condensers (a generator without any actual energy source) in areas far from actual generators to provide local inertia.

AC arcs inherently self-extinguish in most cases. DC arcs do not.

This means that breakers and expulsion type fuses have to be significantly, significantly larger and more expensive. It also means more protection is needed against arcs caused by poor connection, cable clashes, and insulation damage.

Solid state breakers alleviate this somewhat, but it’s going to take 20+ years to improve cost, size, and power loss to acceptable levels.

I expect that any ‘next generation’ system is likely to demand a step increase in safety, not merely matching the existing performance. I suspect that’s going to require a 100% coverage fibre comms network parallel to the power conductors, and in accessible areas possibly fully screened cable and isolated supply.

EVs and PV arrays get away with DC networks because they’re willing to shut down the whole system in the event of a fault. You don’t want a whole neighborhood to go dark because your neighbour’s cat gnawed on a laptop charger.

That’s not really practical. For a lot of on-call roles, you need the experience of working on the equipment daily to know how to fix it.

Even if you alternated e.g. a week of on-call then a week of normal shifts, you end up reducing the amount of time you have that is completely non-work.

Firemen are absolutely not doing nothing until called out. There’s training, maintenance, cleaning, post-incident paperwork, and probably other duties.

There should be no need for tuning, tweaking, or optimizing on functionality this basic.

If you ask the processor, it will spit out a graph like this telling you what threads/cores share resources, all the way up to (on large or server platforms) some RAM or PCIe slots being closer to certain groups of cores.

I didn’t realise it was possible to hate every side of an argument this strongly.

Essentially no processors follow a standard. There are some that have become a de facto standard and had both backwards compatibility and clones produced like x86. But it is certainly not an open standard, and many lawsuits have been filed to limit the ability of other companies to produce compatible replacement chips.

RISC-V is an attempt to make an open instruction set that any manufacturer can make a compatible chip for, and any software developer can code for.

Kinda yes, kinda no. There have certainly been times, particularly after 9/11 and various crises, when demand dropped significantly.

There’s also airliners that just haven’t sold well. A340NG, A380, 747-8, 767-400, the MD-11, until recently the Cseries/A220. The A330neo has also not sold particularly well and you could probably get a slot within a year easily.