> Renewables can't meet the base energy needs.

That assertion is not something everyone agrees with. And baseload is hardly ever qualified with even a ballpark estimate in GW or GWH of capacity needed. So, it's a fairly hollow and meaningless term.

And the reality is that for every 100GW added to grids world wide, about 80% or more is renewable. Nuclear is only small portion of the remaining capacity. And SMRs are a rounding error on that. Most of the rest is gas based generation.

Besides, data centers are a great example of something that can easily scale up and down its energy consumption based on price signals, user demand, etc. So, it's actually ideal to pair with fluctuating supply and demand from renewables. Using e.g. spot instances makes it easy for data centers to scale down their demand if energy is scarce and expensive. Other things they could do is throttle CPUs/GPUs based on energy pricing or encourage people to time shift non critical jobs to when energy is plentiful.

SMRs won't have fixed anything until there are lots of them. Whether you believe this will happen or not, it won't be happening very soon. Realistically, SMRs will remain a niche solution for decades to come; even if they do work at reasonable cost levels.

▲

torginus 3 days ago | parent [-]

> And baseload is hardly ever qualified with even a ballpark estimate in GW or GWH of capacity needed.

If we close all the steel mills and ship off manufacturing to China, then yes, we won't have baseload, and we can be happy that we saved the planet using solar!

> Renewables can't meet the base energy needs. That assertion is not something everyone agrees with. And baseload is hardly ever qualified with even a ballpark estimate in GW or GWH of capacity needed. So, it's a fairly hollow and meaningless term.

> And the reality is that for every 100GW added to grids world wide, about 80% or more is renewable.

Do you have solar at home? Because I do, I have 10kW of panels on my roof. I just checked my stats and in December I approximately made about 15% o peak capacity. And even that isn't the whole picture, as there were chunks days where I basically made nothing and even the batteries couldn't pull me through it. And I have no idea how you're calculating this 100GW. If you count adding 2000 500W panels as adding 1MW, then even on the Caribbean your calculation is going to be incredibly generous.

> Nuclear is only small portion of the remaining capacity

As for nuclear, it was made way too expensive because the economy and money became fake, divorced from real value, and pearl-clutchers and concern trolls made it too expensive. But even in the 70s-80s when things were actually built in Europe, it was clear that Gen IV (of which SMRs are an example) was the future of nuclear, its just nobody bothered to build it because it was easier to ship off manufacturing into the 3rd world.

>Besides, data centers are a great example of something that can easily scale up and down its energy consumption.

Yeah when you buy millions of dollars of HW, the 'we'll need to run it at 15% capacity in December and during night, not at all' sounds like a sound return on investment. Way to cheerlead to get another industry shipped off from the continent.

> SMRs won't have fixed anything until there are lots of them.

SMRs are not small, they are scalable, and can be made in similar capacity to existing coal and gas plants. Once they reach EOL, they'd be a perfect slot-in zero emissions replacement. But since nuclear is the devil's work, I guess we'll get to keep burning gas for another half a century.

▲

cycomanic 3 days ago | parent | next [-]

That fact that people can't even agree on what SMRs even are tells you everything. You advocate that they are essentially the same thing as a regular power plant, others say they are small boxes that can be put at any neighborhood corner.

But everything because they are going to scale magically. Nobody ever explained how that scalability is supposed to happen. A large portion of any nuclear reactor is still a steam turbine, we have built lots of these without seeing prices fall exponentially (like solar) why should adding a couple (really a miniscule) number of SMRs suddenly fundamentally change the way their price scaling?

	▲	torginus 3 days ago \| parent [-]
		I might be wrong but I don't think anyone's seriously proposing on building these truly tiny reactors - I think every one of these is being built is powerplant sized. SMRs identifying trait is that these reactors are prefabricated and transported onsite and do not have to be built in pace.

▲

lukeschlather 3 days ago | parent | prev | next [-]

> If we close all the steel mills and ship off manufacturing to China, then yes, we won't have baseload, and we can be happy that we saved the planet using solar!

86% of generation added in China in 2024 is renewable.

	▲	torginus 3 days ago \| parent [-]
		If you want reliable all-year round-the-clock solar, you have to over build by 20x, not to mention the batteries. (Wind has different but analogous issues) That's not to hate on solar - I think it's great, and I personally have solar at home, but its not a substitute for everything

▲

seec 8 hours ago | parent | prev [-]

I'm glad I'm not alone thinking this.

Much of the renewable price/cost calculation is plain wrong because we have offloaded the cost to other countries (who gladly still use fossil fuels).

On top of that there is the problem of maintaining a real economy inside the country, to benefit people actually living in said country. I believe most of HN don't care because they get paid to make stuff that is sold globally and they profit from that. They can't understand that not everybody can do that and you need to create value inside of the country so that they can share/benefit from that value creation.

I also have access to solar generation statistics in the middle of France and they make no sense without the generous subsidies and electricity generation purchase from the provider. But plenty are profiting from that so of couse they are happy about it.