Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Eh, it is my opinion that we are a long way from even having a net-positive prototype reactor (if ever).

Even if we get one operational, I suspect the construction cost alone would be an order of magnitude more than a traditional nuclear reactor. Then, how much of a net output will you actually be capable of generating, probably a fraction of the energy it consumes which is something but by the time you factor in construction costs as well as maintenance/upkeep costs you're probably going to have a fantastically expensive power source.

Then consider that if we had a viable fusion reactor TODAY, and only needed 10,000 of them to replace the 62,000 something power plants in the world it would likely take many decades to build them all. Then of course, energy consumption increases over time.

Between 1990 and 2008 world energy consumption increased 39%. Since 2008 we've added considerably more electronics to our lives just in 'the first world' nations so I suspect that increase has been even greater in the past decade. Factor in electric vehicles being added and...

So even if we do get fusion working in the near future, it's not going to be anywhere near a miracle.



The consumption probably grew due to Asia and Africa using more electricity, but in the western world?

In 2008 I had a stationary computer taking 350W-400W, a bunch of 100W lightbulbs in my house, a TV, and an old fridge. Now I have 60W laptop that took over my TV as well, and most of the lightbulbs are the power efficient ones taking 10-20W, 100W ones are not even being sold in Europe (a few exceptions aside).

(Just checked - in Europe, electricity consumption was on the rise to around 2008, and then remained static and even fell a bit)


>The consumption probably grew due to Asia and Africa using more electricity, but in the western world?

Fair, looking at more recent data it looks like between 2008 and 2017 the US has bounced back and forth about 100 TWh in total use.

BUT if you look at things like Bitcoin... one estimate ( https://arstechnica.com/tech-policy/2018/05/new-study-quanti... ) put the Bitcoin network alone using 30-70 TWh a year.

If we look at https://yearbook.enerdata.net/ we see more recent data and yeah the biggest culprit is Asia with 1-3% increases each year. I imagine a good chunk of this is manufacturing for 'first world' countries though which one could argue should be assigned to the countries being exported to not the country doing the manufacturing but that's going to be way too complicated to estimate. I work in international freight and I've definitely seen freight grow in the past decade, especially Asia (if we exclude the past few months with Section 301 and China) and if EVs start to catch on in the next 5-10 years you'll likely see a large jump in actual electrical demand as well.

The middle east is consistently growing, I'm guessing a lot of that isn't 'third world' areas though and more likely cities like Dubai.

Something to keep in mind though is global warming, if budgets allow I imagine we're going to see cost of cooling increase in the coming years.


https://en.wikipedia.org/wiki/Energy_in_the_United_States#Co...

In 2016 the total US consumption of electricity was 4,137.1 terawatt-hours (TWh).


"bounced back and forth about 100 TWh" as in on year 4000, one year 3900, one year 4100.


Oh, OK, I misunderstood.


I figured :)


I think we can expect electricity consumption to go back up quite soon. Contributing factors are electrification of transportation, transition to electric heating, ever growing software bloat, increasingly compute-intensive ideas like deep learning and blockchains requiring more and more powerful data centres... and let's not forget that software hasn't eaten the world yet.

Some of it will be a demand shift from fossil fuels, which is a good thing, but it still means we need more electricity.


Construction cost of ITER is very high, but these new designs use new superconductors and are ten times smaller for the same output (500MW). The JET reactor in the U.K. is about the same size, and was built in four years for half a billion dollars. That's without the facilities to actually extract electricity, but it's also a one-of-a-kind experimental device and you get economies of scale by producing lots of them in factories (which we could do, since it's not that big).

So a billion dollars a gigawatt seems about the right order of magnitude, and that's comparable to power sources we use today.


So that goes from being 400x worse than a PWR to only being 40x worse (metric: power density of the primary reactor vessel). It's still not going to be competitive.




Consider applying for YC's Winter 2026 batch! Applications are open till Nov 10

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: