The GP's comment wasn't accusing there of being a nuclear waste problem (there isn't). They were talking about how nuclear waste can be burned in a breeder reactor, producing orders of magnitude more than the burning of a couple tenths of a percent of the natural uranium in a conventional reactor does.

Despite the press hype about thorium (which is way more popular among the media and nerds on the internet than with actual nuclear engineers), nuclear power is already basically unlimited, even without breeder reactors (which are very much viable tech, and much more mature than thorium). Only with an incredibly weak definition is it in any meaningfully way "limited" - if you limit yourself to currently quantified reserves, at current fuel prices, with production mining tech, you have a bit over two centuries worth at current burn rates. But this is obviously nonsense. Uranium production tech isn't going to advance in *two centuries*? Nobody is going to explore for more in *two centuries*? And as for "at current prices" - fuel is only a very small percentage of the cost of fission power, so who cares if prices rise? And rising prices or advancing production tech doesn't just put linearly more of a resource onto a market, they put exponentially more onto the market. As an example with uranium: seawater uranium could power the world's current (overwhelmingly non-breeder) reactor fleet for 13000 years, and current lab-scale tech is projected to be nearly as cheap as conventional uranium production at scale.

Also, if you switch to breeder reactors, you don't just extend the amount of fuel you have by two orders of magnitude - the cost of the raw mined uranium also becomes two orders of magnitude less relevant than its already very small percentage of the cost of fission power generation, because you need so much less per kWh.

As for any thoraboos in the comments section: thorium fuel is more complex and expensive to fabricate (fundamentally - thorium dioxide has a higher melting point and is much harder to sinter), it's more complex to reprocess (it's more difficult to dissolve), its waste is much more hazardous over human timescales, the claimed resistance to nuclear proliferation is bunk, the tech readiness level is low and the costs are very high, and it's unclear it'll ever be economically competitive - most in the nuclear industry are highly dubious (due to what's needed to actually burn it vs. uranium). Hence the lack of investment. And I say this with the acknowledgement that nuclear power is already a very expensive form of electricity generation.

Some understanding of chip making helps. There is a reason people stop at 6 layers or so. Yes, you can stack more. No, it does not make sense to do so.

What Samsung does is stack dies on top of each other. What these people here do is a die with more layers. But this is very likely just a stunt that is non-viable for any real use.

Nowhere. This is a meaningless stunt.

It actually is interconnect first, thermals second and then things like clock distribution and power. There always was more space for logic, but that logic could then not be used.

It is _really_ difficult to do more layers. If you have simple interconnect (FLASH or cache chips), you can stack chips, but more layers on a single chip comes with a ton of problems.

This thing here is probably just a stunt and will not be ready for production, ever.

The problem is that it is simply not needed for most folks. The workarounds will likely keep IPv4 alive for at least another 20 years.

Yep. I have a paid-for static IPv4, but I just get a free IPv6 block from my ISP. Not using it though.

Real world testing gives a wide variety of range reduction in cold weather, depending on the make and model of EV. Some are really good at maintaining range, some are lousy at it.
In any case, preheating the cabin and battery cuts that substantially, and you generally don't need to keep warming the battery while driving as the regular discharge and charging from regenerative braking keeps the battery at operating temperature to limit range loss.
It's a contribution, but it isn't something like 30% is what he's getting at. More likely ~5%.

So basically "bullshit...bullshit...firing people...bullshit....bullshit"? That would fit, agreed.

You really need to get over yourself. Rational arguments are arguments others can follow and verify. Your self-aggrandizements do not count as such...

Indeed, it does. And good men continue to do nothing or far too little.

Still, I have got a toast to make.

Also probably for yet another company not making smart hiring decisions...

Can anybody decode what they mean?

Well, you are just digging yourself deeper. Ok, lets go one step further in your self-disgracement: 1. There are no synapses in AI 2. Reverse PN is not a standard operating mode for transistors and only applies to bipolar transistors in the first place. 3. Computers are using MOS-FETs. 4. Computers are optimized for digital, entirely deterministic computations, nobody wants avalances in there. If they happen, the computer crashes.

But I bet all of that flies right over your head, because you are obviously completely clueless.