have datacenters get their power only from renewables and limit the amount of area they have to build them and watch renewable efficiency skyrocket as they either have to develop them or be without power.
Renewables blow past nuclear when it comes to cheap datacenter juice
Submitted 4 days ago by kalkulat@lemmy.world to technology@lemmy.world
https://www.theregister.com/2025/09/26/renewables_vs_smr_datacenter/
Comments
reksas@sopuli.xyz 4 days ago
justsomeguy@lemmy.world 3 days ago
Are you trying to trick tech companies into being useful? That’ll upset them.
HeyThisIsntTheYMCA@lemmy.world 3 days ago
Someone tell Silicon Valley: They should put datacenters on trains so no one knows where they are. gonna need HSR for it to work properly tho.
UnderpantsWeevil@lemmy.world 3 days ago
Making a data center profitable would be bad for the economy and the jobs
Tja@programming.dev 3 days ago
Renewable efficiency is close to the theoretical limit. Solar cell have a limit just over 33% and current models have efficiency of around 25%.
UnderpantsWeevil@lemmy.world 3 days ago
Renewable efficiency is close to the theoretical limit.
There’s still plenty of juice to squeeze in terms of cost to manufacturer, deploy, and maintain. This isn’t purely a question of cell efficiency.
reksas@sopuli.xyz 3 days ago
if the component is at its limit, then you can come up with ways to use that component more efficiently. Also reducing the size of the whole thing also increases efficiency singe you can stuff more of them in same area
SpaceCowboy@lemmy.ca 3 days ago
AI is another dot com style bubble. How about we all just be quiet about that so billionaires blow a lot of hype driven investment dollars on green energy?
Once the bubble bursts there will be a surplus of cheap green energy we can use for powering homes and EVs and such. Obviously there’s better ways to do this than scamming billionaires into a hype train, but global warming is a problem now and we can’t wait for our society to change to be able to address the problem in a rational way.
So… sure… AI is the future! We need to build a lot of wind and solar power so we can have AI! We don’t need this for woke global warming reasons, no no no. We need this for $$$$$$AAAAAAAAAAAAIIIIIIIIIIIIIIIIIII$$$$$$$$$$$$$$$ reasons! Increase shareholder value by making wind turbine and solar panels, you must do this because it’s illegal not to maximize shareholder value!!! Build wind and solar so you can someday fire all of your employees! For the shareholders!
Aatube@kbin.melroy.org 4 days ago
a commenter:
They claim to compare the cost of powering a 120MW data centre from a dedicated 470MW RR SMR compared to powering it from an 80MW gas turbine plus some unspecified number of wind, solar, and battery installations. For a study supposedly promoting wind, solar and battery technology, you would think they would tell us how many, what size, and what model of wind turbines they are modelling. But no, that's left to vague hand waving.
Aatube@kbin.melroy.org 4 days ago
on review this doesn't appear to be entirely true; they don't disclose how they get the "43.4%" but say they get 31.7% cheaper than nuclear if gas is 5%.
Forester@pawb.social 4 days ago
Get out of here with your facts, reason and logic it’s not allowed. Don’t you understand we have to turn off all the deadly evil nuclear plants and stop them from poisoning the Earth and killing so many more people than all other methods of power combined /s
FishFace@lemmy.world 4 days ago
Wind and solar ᵃⁿᵈ ᵍᵃˢ
Schlemmy@lemmy.ml 4 days ago
Of course. Renewable blows nuclear out of orbit when it comes to price. Nuclear plants take decades to build and are generally a lot more expensive than estimated.
humanspiral@lemmy.ca 4 days ago
The Vogtle scam’s end cost was $17/watt. $8B or $4/watt was just financing costs prior to eventual operation that Georgia Power got to charge its customers for its share, over the 20 years before it gave them power from the boondoggle.
Solar costs under $1/watt to deploy, and batteries in a container (can fit under solar) costs $1 per 10 watt-hours of storage. Both last over 30 years.
SMR’s can pretend lower capital costs per watt, when excluding design/prototype time, but trade much more expensive enriched (proliferation risk) fuel that is less efficient, needs breeder reactors to provide likely from Russia, and carries higher security costs per watt. SMRs are simply a new scam to defraud investors with because nuclear is worthless as energy, and only ever is for military applications.
UnderpantsWeevil@lemmy.world 3 days ago
SMR’s can pretend lower capital costs per watt, when excluding design/prototype time, but trade much more expensive enriched (proliferation risk) fuel that is less efficient
The primary appeal of SMRs is their portability. Pointless for a data center, but vital for a large vehicle like a cruise liner or a shipping frigate.
Replacing our fleet of bunker fuel powered ships would be enormously beneficial.
kameecoding@lemmy.world 4 days ago
That’s mostly because the west has become a bad place to build things, bike-shedding and a general loss of nuclear building expertise lost due to successful campaigning against nuclear by the fossil fuel industry.
We could be scaling up nuclear right now to help the goals for 2050 to be reached and then coast for a while as renewables pickup pace and fusion is finally cracked.
But no only thing people care about is immediate cost.
Yes renewables are cheaper per kw at the moment but they are also putting a lot of strain on the grid that’s not accounted for that’s expensive to upgrade, they are also not scaling up fast enough, which means there will be added cost to climate change.
Vs we could build nuclear reactors at a loss and bring on serious gigawatts of clean energy in a decade that would provide a stable baseline.
Schlemmy@lemmy.ml 1 day ago
The west, the east, the north, the south… Wherever you build your reactor it will overshoot its estimated budget and wil be overshadowed by renewables.
But yes, there are many variables and the answer always lies in differentiating.
HugeNerd@lemmy.ca 3 days ago
Which is why I always laugh when people say to replace a 15 year old fridge to “save” on electricity. Why? It’s as cheap as the wind, making and shipping a new fridge isn’t.
UnderpantsWeevil@lemmy.world 3 days ago
Which is why I always laugh when people say to replace a 15 year old fridge to “save” on electricity.
Really depends on how much your electricity costs relative to your efficiency gain on the new fridge.
But refrigerators are also largely a “solved” technology. I ended up getting a new one recently because my old refrigerator’s repair bill was going to be as much as a new unit.
Now, if units were more modular and easier/cheaper to repair? The math changes.
Prox@lemmy.world 3 days ago
“I’m going to spend $1500 so I can save $8/month.”
Darkenfolk@sh.itjust.works 3 days ago
… For quite a few years and it pays itself back in 15/16 years, after which it probably still works for another 5 to 10 years.
jaykrown@lemmy.world 2 days ago
Thankfully wind and solar are cheap and require a low up front investment, otherwise it couldn’t be. We need to continue to invest in battery technology, sodium batteries are the way forward.
eleitl@lemmy.zip 1 day ago
Except you can’t power 24/7/365 with renewable alone, so you still need gas turbine backup.
kalkulat@lemmy.world 1 day ago
A funny thing happened back in the middle 1800s. A man ran a 7-ton electric locomotive a mile and a half. The motor was powered by a storage device. In the late 1800s, people drove their cars around all day using a storage device. These storage devices became better and better, until they could power trucks and buses for hundreds of miles.
They are still getting better and better. Of course they can be depleted, and it’s good to havea backup methods to cover these cases and to keep the storage devices charged when there’s no sun or wind. Hydroelectric dams powered by water-storage are widely-used, and some flat places still burn fossil fuels to do that as well.
eleitl@lemmy.zip 1 day ago
You need a buffer with at least 60 TWh in case of Germany. There is no economic electrochemical energy storage system for that capacity.
acosmichippo@lemmy.world 4 days ago
problem is solar and wind are variable and not feasible everywhere. for places like australia solar is amazing. Winter in canada? not so much. So for a baseline you’d have to store a massive amount of energy in some way.
if you plan on batteries that requires lots of precious metals we will need elsewhere to aid in the transition to electric power.
MonkderVierte@lemmy.zip 4 days ago
problem is solar and wind are variable and not feasible everywhere.
Offshore wind is constant.
brachiosaurus@mander.xyz 2 days ago
The problem is something else. Energy in winter canada shouldn’t be used to power factories, the industry should be moved south but you have invisible lines on a map preventing that from ever happening.
Schlemmy@lemmy.ml 4 days ago
Hydro energy to the rescue.
Aatube@kbin.melroy.org 4 days ago
the studied location is the UK
echodot@feddit.uk 4 days ago
The UK has one of the largest wind farms in the world, I think it actually is the largest in the world. One of the wind farms was built just off the coast of Scotland right next to trump’s golf course and I’m sure it was built mostly just to annoy him.
Solar however is a lot less reliable, just because it’s not particularly sunny here and also with it being so far north during the winter the nights are quite long.
The government says that the intention is to go 100% renewable but what they actually mean is as much as renewable as possible, plus nuclear cover the load. No one thinks you can 100% be on solar and wind.
kameecoding@lemmy.world 4 days ago
if you plan on batteries that requires lots of precious metals we will need elsewhere to aid in the transition to electric power.
Umm, what about sodium-ion that are now getting put into production?
sobchak@programming.dev 3 days ago
A lot of the companies and people responsible for having all these datacenters built are heavily invested in SMR. So they’ll probably be used anyways.
UnderpantsWeevil@lemmy.world 3 days ago
For a modern scaled up data center, there’s no real benefit to nuclear miniturization. That’s the sort of technology best employed on shipping frigates and space stations - places where portability is a priority.
You don’t need to pick up a date center the size of 70 football fields and send it anywhere.
SpaceCowboy@lemmy.ca 3 days ago
Shipping frigates? Sure, lets give the Houthis and Somali pirates the capability of building dirty bombs.
And if solar power is cheaper on Earth, think of how much more cheaper it is in space where there isn’t an atmosphere getting in the way.
Sometimes a tech is really cool, but there just isn’t any viable use case for it.
Aatube@kbin.melroy.org 4 days ago
Actual paper (not calling this a study since this appears to be non–peer-reviewed and only self-published): https://microgridai.centrefornetzero.org/ Be advised that this website relies on some Chromium-only trickery.
renewable microgrids [...] compared to nuclear small modular reactors
A 95% renewable microgrid with 5% gas backup - in line with the UK’s Clean Power 2030 target - was modelled at almost a third (31.7%) lower cost than [nuclear] in today’s prices.
dharmacurious@slrpnk.net 4 days ago
Without having read it myself, perhaps they mean 5% of total usage. So the gas generation is built to be able to handle 2/3rds of the power demand, in case of outage as a backup, but in normal operation will only contribute 5% of the energy demand. That way, in the event of a failure of the renewable energy source for whatever reason, or a failure in the batteries, the gas can kick in and keep the servers online while cutting disposal operations that represent 1/3 of the total.
humanspiral@lemmy.ca 3 days ago
Extreme incompetence in modeling. the github is complete crap. otoh their angelfire site does actually list some costs for the SMR.
They make the SMR side look absurdly cheap. $55/mwh power costs with 30c/watt capital costs is just absurdly low. Conservative SMR power estimates start at $180/mwh, and so actual microgrid costs would be over 80% lower.
More incompetence has their microgrid using off shore wind which is just stupid for HVDC requirement for small scale. Automatically too incompetent to trust their modeling. They don’t specify cost assumptions for any of the microgrid components.
Aatube@kbin.melroy.org 3 days ago
wdym the github is complete crap? it has everything you mentioned you wanted to look at. and wouldn't a too-low estimate for nuclear costs give extra validity to the claim that microgrids are much cheaper?
individual@toast.ooo 4 days ago
interesting, never heard this before
FreedomAdvocate@lemmy.net.au 4 days ago
The “analysis” was done by “Center for Net Zero”………definitely not biased at all…….lol
Also offshore wind is one of the most environmentally destructive methods of power generation.
kalkulat@lemmy.world 4 days ago
Offshore wind is one of the most environmentally destructive methods of power generation.
Interesting claim (as compared with coal mining and its fly-ash ponds, Canadian tar sands, hundreds of bankrupt and leaking well sites in New Mexico and the Gulf of America, rivers stripped by nuclear heat waste, etc). What exactly does most mean?
FistingEnthusiast@lemmynsfw.com 4 days ago
Damn,
You’re really stupid.
Please don’t breed.
FreedomAdvocate@lemmy.net.au 4 days ago
I know I’m right when your only reply is insults.
Aatube@kbin.melroy.org 4 days ago
"net zero" just means you don't elect for the far more environmentally destructive method of burning fossil fuels
also, it seems the jury's still out on offshore turbines' environmental impact. some say it creates artificial reefs while some say its tons of noise disrupt marine life
FreedomAdvocate@lemmy.net.au 4 days ago
Net zero organisations have shown a clear agenda against nuclear, which is ironic considering it’s the cleanest and most reliable power generation method, as well as taking up the smallest footprint with the least environmental disruption.
Offshore turbines require insane amounts of concrete, steel, oil, and non renewable non recyclable materials not just to make, but to maintain. There’s also no doubt about them altering the ecosystems around them, and not for the better. They also aren’t even a viable option in most countries.
acosmichippo@lemmy.world 4 days ago
generally agree but tech firms are looking into small modular nuclear reactors sized for datacenters.
FreedomAdvocate@lemmy.net.au 4 days ago
I’ve heard this as well, but in doing this they would either make a significantly smaller and cheaper one that isn’t outputting 5x the power required, or they’d do a deal with the local councils/government to provide lower for them as well.
This “study” is comparing the cost of 80 units of power generation for “renewables” to over 400 units for nuclear. Is just yet another dishonest agenda driven “study” for the anti-nuclear groups.
SparrowHawk@feddit.it 2 days ago
huh, so AI WILL solve climate change, lol
Zeoic@lemmy.world 4 days ago
Its wonderful how they just drop the “20% is gas” part from that headline. Yes, burning gas is cheap, but it is also aweful for the environment and shouldn’t be getting considered at all… 20% of a fuck ton of power is still a shitload of power. I think that’s how those units work anyway.
corsicanguppy@lemmy.ca 4 days ago
Fracking methane should be excluded. It’s 80 times worse for the environment than even CO2.
humanspiral@lemmy.ca 3 days ago
They undersell the benefits of renewables significantly overall. This is for UK which they come out with slightly lower costs for omitting solar. They also say 5 years to build a 120mw microgrid. 1 post driller, 1 crane for support posts, with 2 workers guiding post insertion and cleaning up, 1 “wall of panels” crane lifter, with 3 workers aligning connecting panels on the ground, and then connecting wall to posts can get 40kw/hour=320kw/day. Complete in little over a year. But, in solar, 9 crews can really make a baby in 1 month.
Microgrids don’t need permits, and utilities will give them an import connection.
FreedomAdvocate@lemmy.net.au 4 days ago
It’s pathetic propaganda. You know what’s even cheaper? Coal! Or just going 100% gas! So if it’s really about cost then the answer is zero renewables.
SolacefromSilence@fedia.io 4 days ago
Coal is more expensive, it's not the 90's anymore.
humanspiral@lemmy.ca 3 days ago
NG newbuild in UK is not cheaper than wind. Unless NG is extremely cheap due to no one using it anymore, and there are abundant supply options.
davidagain@lemmy.world 3 days ago
Onshore wind is the cheapest electricity in the UK, by far. It’s not even close.
Aatube@kbin.melroy.org 4 days ago
where does it say 20%
Zeoic@lemmy.world 4 days ago
That was an extrapolation from where they said renewables would cover 80% in the article. I can only assume the mentioned gas would be the other 20%
wordmark@mas.to 4 days ago
@Aatube @kalkulat @Zeoic It's a good thing that the #sun never shines here and the #wind never blows... In 1810, #Chile expelled the #Spanish #colonial rulers, only to then allow a Spanish company to exploit its own solar and wind energy for profit, Apparently, the highways are also controlled by a foreign Spanish investor who is making a hefty profit from the local population (#neocolonialism 🙈👎 also with #water and #nestle and #energy possibly also in your country