The 5G revolution also comes with major battery pain in marginal areas. In Melbourne itâ(TM)s pretty easy to get an asymmetric link and have the phone (any type) burn through power fruitlessly retransmitting.

Iâ(TM)ve had to carry a battery bank around with me as Iâ(TM)ve been visiting more marginal areas lately.

Places like the zoo are also good 5G battery drains if youâ(TM)re using the cellular network (sending pictures to someone, for example).

Itâ(TM)s fantastic if youâ(TM)re sitting in front of a mobile tower or a small cell though. So pretty good in a few areas.

The 100% renewable energy storage problem is quite vast though, and the scale of the challenges involved are often downplayed quite strongly.

I've noted many of the numbers quoted for the energy storage required assume 2-3 days, or maybe a week. You can definitely assume some renewable capacity will still be online during a reduction in generation (maybe a large bushfire reduces your solar output significantly), so you don't need enough storage to cover all power consumption... but you still need a lot. And you definitely need to plan for longer periods of time, in the several week range, for emergencies, because a country running out of power in the middle of winter is Bad(tm).

The volume of storage required is immense for somewhere like Germany (500 TWh/year consumption, so 1.3-1.6 TWh/day average, depending on season, and you'd need storage for at least half of that). The volume of ammonia (or hydrogen or compressed air) required would be immense. Building dams for water-based storage sounds relatively easy but, ignoring the ecological impacts and actually finding suitable sites, you'd also need to build and maintain enough hydroelectric power plants to turn that storage back into energy (assuming existing hydro plants are still operating).

Batteries are cool as supplemental and for stabilising the grid, but not practical as primary storage. That battery in South Australia can store 194 MWh and supply 150 MW (i.e. it's empty in 78 minutes at full power). Australia consumes 190 TWh of electricity per year, so 520 GWh/day (naive averaging). To supply half of Australia's energy requirements (with naive averages 260 GWh/day at approx 11 GW) via batteries for one day would require another 1,760 batteries of similar scale (since each can only supply 150 MW of power for a bit over an hour); at the original price for the initial 100 MW battery (90 million AUD), that'd be 240 billion AUD. For 21 days of half power it'd cost 5 trillion AUD. And you probably don't want to fully discharge or charge the batteries, to reduce the replacement time, so might need 20% more than that -- and this assumes only half of the daily energy requirements. And ignores the mining capacity of raw materials.

You also need to be able to replenish that energy store in a reasonable period of time, to deal with fluctuating circumstances, so you need to build notably higher total energy generation than your average demand. If you had a 100% efficient storage mechanism, you'd probably need at least 10-20% over-capacity. If your storage efficiency is down at 55%, you'd need a lot more. If you have large amounts of solar, you'd have to have enough capacity to generate that surplus during the available daylight.

It's a problem worth focusing on addressing... but man, it's quite hard. Nuclear build times are long, but they may need to be included in more net zero by 2050 plans -- to reduce the total volume of energy storage required as economies grow.

The summary clearly states that it needs to be grievous and irremediable and be a state of irreversible decline along with enduring physical or psychological suffering. If they can recognise their mental illness and can enter a process of recovery then it isnâ(TM)t irreversible and isnâ(TM)t eligible with those conditions. The article just appears to be clickbait by suggesting something which is impossible (someone curably addicted to a drug being eligible) is somehow going to be normalised. Unless the legislation is more nuanced. It seems unlikely, but Iâ(TM)m not Canadian so I donâ(TM)t have a good feel for the political climate there.

Not quite. PlayStation does take a cut of all games sold and itâ(TM)s likely in a similar ballpark, same as Microsoft, even without being the store itself. And you can bet a car manufacturer makes a pretty penny on all of the genuine accessories - the fuel analogy is more akin to the electricity required to charge a phone. It may be worth limiting the amount for a public good reason, though given market forces have established current prices work we probably shouldnâ(TM)t expect much of a price reduction as consumers - but the businesses will benefit. We will definitely see an increase in the number of subscriptions that require shaving a yak in the alpines as a prerequisite for cancellation though. Itâ(TM)d be good if any law requiring open payment systems also enforces requirements for cancellation and fraud protection.

The majority of Tasmanians agree, given the effort we're putting into it, so it is essentially a fact; the birds aren't at risk, but we need nature's little garbage trucks.

For interest, Android does this too now with Google's RCS chat system. When you swap to an iPhone, as I recently did, you have to go and deregister your phone number from RCS or all messages from Android devices (which support RCS) are lost.

So it's a bit of ecosystem lock-in from both sides, as a side effect of providing a helpful service whilst you stay engaged.

nm, turns out that, like everything else in the world, it's not that simple. The density is only one factor, but the "10 nm" and "7 nm" are just marketing names. This video uses an electron microscope to inspect both processes, which was cool.

I was under the impression that the nanometre classification was irrelevant, and Intel's 10nm process and TSMC's 7nm process produce the similar transistor densities (approximately 100 MTr/mm2 for Intel and 91 MTr/mm2 for TSMC); is there some other advantage to TSMC's approach (other than yield, which appears to be improving)?.

AMD's problem is they won't be able to supply enough chips to OEMs and distributors due to competing with everyone else for resources (hence all their current CPUs, GPUs, and both videogame consoles being out of stock); Intel has an edge there, given the in-house manufacturing.

AMD spun off Global Foundries; TSMC is unrelated.

Kurzgesagt made a video on this topic a year ago which I found interesting:

Building a Marsbase is a Horrible Idea: Let’s do it!

It seems that living on mars wouldn't be particularly fun, but building the technology required to do so would be worthwhile.

It looks like a neat little unit; a few seconds into the video they show a potted device with a sigfox module on one side, another photo shows a GNSS module on the other (with a patch antenna on top, similar to this), and a Saft LS14250 LiSOCl2 battery (nominally 4.3 Wh). A GNSS position fix consumes up to ~1 mWh, and a Sigfox transmission should consume less than that, so they should have more than enough energy budget to last the three years comfortably.

Interestingly if they'd used LoRaWAN instead they may not have needed the GNSS module. Sigfox's geolocation via radio signal strength has a precision of 1 - 10 km, whereas LoRaWAN can use multilateration (time difference of arrival) and get down to ~200 metres. That should be more than enough precision to track a rhino's movement towards a dangerous area, which means the device could last longer without using a GNSS module. LoRa competes with Sigfox, so someone else would have to do it, but it could be an interesting approach to animal tracking.

I dual boot Windows and Linux at work, mostly using Linux. On rare occasion when I have to reboot into Windows it seems to take around 20 min to process whichever update it tried to apply last time I shut it down, wasting time. I'm not a fan of that little feature and the inability to control when updates occur.

Android was released 10 years ago. This anti-trust battle has been progressing for 8 years. I suspect your assumption is highly flawed.

Seems oddly common, probably due to legacy issues. Our entire surface mount assembly line and x-ray inspection machine run Windows 7 (except maybe the ovens). We are not fans, but there's not much you can do about it.

As that article is 4 years old now, I'd be interested to read an updated view on how systemd has changed since then (the single PID-1 thing appears to be at least partially outdated, if it's comprised of ~60 programs now). It may not have changed much, but 4 years is a fairly long time.