Meta will beam sunlight from space to power AI data centers, solar-collecting satellites will orbit 22,000 miles above Earth — firm reserves 1 Gigawatt of orbital solar energy and 100 Gigawatt-hours of long-duration storage
Posted by sr_local@reddit | hardware | View on Reddit | 203 comments
russia_delenda_est@reddit
Can we use metric at least in space related topics?
Homerlncognito@reddit
Asked Mars Climate Orbiter.
MiniHos@reddit
RIP
KizunaXxX@reddit
Meta? The same group that fumbles every product they push?
kearkan@reddit
Buying up oculus and then releasing the quest 2 was huge, like it could have been the popularity point for VR, but then instead of letting it launch as an entertainment product and pick up support they instantly pivoted to trying to push it on business customers which then flopped.
No-Improvement-8316@reddit
Buying up oculus and then releasing the Rift DK2, Rift (CV1), Go, Quest, Rift S, aaaand after that the Quest 2...
The fuck are you talking about, lol? Quests (and earlier Go and Rifts) were all about entertainment -- and still are. Meta pumped billions into first-party and third-party games and entertainment apps.
kearkan@reddit
I couldn't remember at what point they bought them.
They pumped even more into metaverse that completely flopped.
No-Improvement-8316@reddit
You have no idea what you're talking about. And at this point I think you should just stop.
Those 80–83.6 billion cited in media represent the cumulative operating loss of the Reality Labs division between 2020 and 2025. It is not spending on a single project, but rather a total loss balance that includes:
Industry estimates suggest that the direct investment in developing Horizon Worlds falls in the range of… $2–5 billion USD, including creator incentives for content on the platform.
KizunaXxX@reddit
This reeks of a bot :0 meta horizon worlds is doing great /s
IamGeoMan@reddit
Meta Glasses Gen 2 sold over 7 million units in 2025 per EssilorLuxottica, a triple increase of 2023 and 2024 combined.
Homerlncognito@reddit
For anyone wondering: Meta reserved capacity at a startup company Overview Energy. They're intending to launch satellites in geosynchronous orbits collecting solar energy. Then transferring it to Earth in near-infrared to "receivers", targeting exact locations. They're supposedly only using commercially available tech and demonstrated power transfer from an aircraft.
Sounds kinda cool at least on paper.
bitflag@reddit
I can't imagine setting solar panels into orbit is more cost effective than just spreading them everywhere in deserts.
Homerlncognito@reddit
This is trying to solve storage so the question is whether it's more effective than batteries.
AbhishMuk@reddit
Hmm yes is a spacecraft more cost effective than a lithium ion battery, I wonder.
Sorry, I'm may be a tech nerd but my background is energy. Unless Facebook is planning to put their servers in places where it's hard to connect to batteries (eg on top of a fucking Himalayan mountain), I wonder what exactly they're thinking of doing.
Most places where connecting/adding batteries is very expensive are also places where running internet/connectivity is also expensive.
Dismal_Guidance_2539@reddit
I read this company paper, they think if they deploy the solar farm on GEO, they can provide energy for 1/3 of earth surface that can be seen from that orbit.
That means they can beam the energy to place that on night times and then switch to another location. So they can use much less number of panel. Solar panel also work much better than on space than on earth.
If you factor that Starship can lower the launch cost to 100-200$/kg or lower. The economic math is feasible now.
tgwombat@reddit
What if you factor in the cost of maintenance? I have to imagine sending someone to space if something goes wrong is significantly more expensive than sending someone out to the desert in a pickup truck.
beenoc@reddit
Space doesn't have the environmental conditions that can cause things to go wrong in that same way. There's no humidity, no dust, no weather, etc. - it's extremely consistent. There are functional satellites that have been in space for decades without any kind of maintenance at all. The Voyagers have been up there for almost 50 years and the only reason they've lost some instruments is because the RTG is running out of power, nothing has mechanically failed. You just need to build for that (which we're pretty good at, as a species.)
Strazdas1@reddit
The radiation enviroment of space is a killer for solar panels though. Voyager is run by a nuclear battery and is as far away from the sun as we ever got.
beenoc@reddit
It is, that's why I specifically mentioned how it wouldn't last as Voyager but could still make it 20 years like LRO (which is in moon orbit, well outside of the magnetosphere and more exposed to radiation than these ones would be, and turns 19 this year and is still going.)
tgwombat@reddit
Do you trust the “move fast and break things” company to output satellites of the same quality as NASA in the 70’s?
Joezev98@reddit
SpaceX has the most reliable rocket ever. Yes, I'd absolutely trust them to put those satellites in orbit.
tgwombat@reddit
I wasn’t talking about the rockets. Please re-read my post.
Joezev98@reddit
SpaceX is the planned launcher, not the builder for these satellites.
tgwombat@reddit
I understand that. I need you to understand that I never said anything about the launch, I was talking about the satellites. If you would like to continue this conversation I need you to re-read my posts and make an effort to understand what I am talking about.
Strazdas1@reddit
What about Overview Energy, the company making the sattelites, do you have a problem with?
Dismal_Guidance_2539@reddit
No. You don’t fix anything. Just launch new satellites in the constellation and let the broken one burn in the atmosphere.
total_zoidberg@reddit
Not from GSO, LEO burns up in the atmosphere in a few months~years depending on altitude, but geostationary is 36,000 km up (22,500 miles, ~100x higher than LEO, give or take). Having the orbit decay and burn the satellite doesn't take 100x times longer, it's exponential because there's less atmosphere the higher up you go.
Strazdas1@reddit
deorbit thrusters exist for a reason.
zero0n3@reddit
You’d likely have a deorbit specific thruster to do this? Nudge it down and accelerate that burn up. Would be like the decommissioning process
tgwombat@reddit
Okay, so what does the cost look like if you factor that in as opposed to basic terrestrial maintenance?
Rodot@reddit
What do you mean now? Starship isn't operational now. Do you mean to say "hypothetically in the future" rather than "now"?
ChocomelP@reddit
What if we put the data centers in space also?
Exist50@reddit
How do we cool them?
ChocomelP@reddit
That is already a how question and not an if question.
zero0n3@reddit
It’s actually logical.
The only way to remove heat in space is black body radiation. So storing energy as heat in space in theory could be easier to do if you build a system that can’t radiate the heat. Satellite collects solar energy (may not even need to be panels. Maybe heating up a material using mirrors).
Then when needed, point it down and convert stored energy to infrared that gets beamed down.
Exist50@reddit
What efficient mechanism do you propose that can convert heat to a usable IR laser?
Strazdas1@reddit
They are using laser diodes to convert solar panel elextricity into infrared. At least thats the plan for this project.
wingman_anytime@reddit
Musk Magic! (/s)
mduell@reddit
Where in the world, with good network connectivity, can you readily get 200-1000 MW grid connections without waiting half a decade or more?
AbhishMuk@reddit
Except near perhaps some hydro projects or China (or Chinese dams), I really don't think it's super easy.
But if you really want to build it to the point you're considering spacecrafts, you might as well consider starting a PV factory and striking a deal with Panasonic/CATL etc.
One of these two is a proven and mature tech that has millions of engineers and can scale up very well. The other one requires rocket science.
Bubbly-Owl8707@reddit
Perhaps they're doing both. I'm sure all of these companies are looking at any potentially viable solution for energy. It makes sense to.
yabucek@reddit
Mimd sharing the source that states how much hardware they're launching, on which vehicle, how much energy output they're expecting and for what lifespan?
Because from all that arrogance I'm assuming you've actually run some numbers and you're not just dismissing it because "new thing stoopid".
ICC-u@reddit
Heating up sand and recovering the energy is probably most cost effective, but this can also transmit the energy to where it's needed, seems niche, feel like this is absolutely a military project in disguise.
Strazdas1@reddit
In theory yes. In practice not a single sand/salt power sink exists.
ICC-u@reddit
https://www.youtube.com/watch?v=EbEVOa9IEgQ
Strazdas1@reddit
Thank you for the interesting video. This seems to be finished construction this year?
ICC-u@reddit
I think it's operational now, it was completed earlier in the year
nickleback_official@reddit
How does this solve storage? Don’t you lose power when it’s cloudy?
Dismal_Guidance_2539@reddit
Only if you use laser to transmit energy back to earth. If you use microwave, I think you will be fine.
Strazdas1@reddit
This project is planning to use laser.
Rodot@reddit
Why would you use microwaves as your example? The wavelength range that is famously known for being absorbed by water
Something_Else_2112@reddit
You've never tried to watch satellite TV during a rain storm have you?
Thercon_Jair@reddit
I think it's a great idea to redirect sunlight, that would have never hit the earth, to earth, where it is ultimately turned into heat energy, while we're having a crisis of global warming. /s
Exist50@reddit
All things considered, probably a drop in the pond vs greenhouse gases.
Rodot@reddit
Green house gas radiative forcing amplifies solar radiation. Adding more solar radiation makes the greenhouse effect worse
greiton@reddit
Space has better angles, longer exposure, and higher efficiency from a lack of atmosphere filtering light.
radiation in Geosync is brutal on panels though and will lead to significant issues and wear over time. I am mostly curious at how they addressing this, if they have discovered some good radiation mitigating materials or techniques, it could be profitable over a long term.
Vb_33@reddit
How is beaming sunlight at night gonna affect the local environment?
greiton@reddit
It isn't sunlight. It's far spectrum infra-red
HappyBengal@reddit
What about animals?
mutl8@reddit
Pumping heat isn't much better, I'd say it's the most problematic part with rising temperatures.
Exist50@reddit
Which you then lose by needing to convert to a different wavelength and beaming it through the atmosphere anyway.
alpacadaver@reddit
Regulation, cleaning, uptime.
CatsAndCapybaras@reddit
We are nowhere close to fully picking the low hanging fruit when it comes to solar. This would be like Henry Ford pitching flying cars to investors instead of an assembly line for autos.
Strazdas1@reddit
we have overabundance of solar power which is why we have negative energy prices at midday and horrible prices when sun goes down.
Strazdas1@reddit
When you account for the fact that in orbit they get more than 6 hours where they actually work.
waxwayne@reddit
These companies are interesting they hire some the smartest people in the world and then have them pick out shades of blue for an icon. We have a whole generation of geniuses wasted on bullshit. They could be curing cancer or fixing our economy instead they pumping stock value by doing stuff like this.
MrKvic_@reddit
Well its a little bit strange. Setting solar panel keeps the infrastructure locked onto the position where they were built, you need place with a lot of sunlight (the "desert") and batteries to store the energy for night (which is harder than it sounds AFAIK) if you need the 24/7 power.
With this solar in space you could theoretically "redirect" the energy wherever you need it - just rotate the space solar few degrees and there you go. You get the power 24/7 and the solar in space is more efficient than on earth.
I ts probably not cost effective and operating a solar in space must be expensive as fuck now, but it could have some usecase and could be viable in future - this is just exploring that possibility
NedixTV@reddit
so a temu dyson swarm /s
i was thinking this will make us see a light beam on the sky at night ?
AssCrackBanditHunter@reddit
Virginia is allegedly about to get a fusion reactor of which Google has reserved a gigawatt. AI is an insane waste of resources but thank God there's finally investment in energy infrastructure
Exist50@reddit
What fusion reactor? No one's made one that can even (net) produce energy yet.
Strazdas1@reddit
isnt the record net producing for something like 140 seconds?
Exist50@reddit
Not up to date on the latest, but it needs to be sustainable indefinitely and harvestable to use as an actual power plant. I'm not going to be so pessimistic as to claim that'll never happen, but we're clearly not there yet.
Strazdas1@reddit
Oh sure i agree we are not there, but net production is technically achievable.
ICC-u@reddit
If it's so simple and so effective then why aren't we using the same technology to divert energy AWAY from population centres and decrease global warming.
nickleback_official@reddit
Goodness… I thought we were on a tech subreddit lol. That makes no sense…. Think about it, I’ll wait.
ICC-u@reddit
Tell me why it makes no sense?
Instead of using solar panels in space to capture energy and beam it to earth, why not use solar shields in space to cut temps in overheated areas with high AC usage and cut energy usage instead.
Think about the planet as a global consumer of energy, it's more effective to cut usage than it is to beam in more energy from space.
nickleback_official@reddit
Well I gave you a chance haha. Global warming is caused by greenhouse gases trapping heat. Redirecting sunlight could reduce some heating but putting a giant sunshade over cities would have some very severe consequences for the people living under it. Just stop burning fossil fuels and let the atmosphere return to its preindustrial levels of CO2.
Strazdas1@reddit
It wont return. we are long past the runaway chain effect now. Even if we stopped today and emitted zero CO2 (impossible, human breathing alone is responsible for about 8%) the planet would kept getting warmer in a self fueling cycle. The only thing we can do now is mitigate effects.
ICC-u@reddit
If you follow most scientists we're past the point of just stopping now, we need to actively revert things
nickleback_official@reddit
Debateable but sure we can try scrubbing too. Solar reflectors are not part of the plan tho for myriad reasons.
ICC-u@reddit
Probably for the same reasons that satellite based energy storage/distribution isn't a serious consideration.
nickleback_official@reddit
Well clearly meta thinks it’s serious but we will see what happens. Two very different problems.
raptorlightning@reddit
"Let the atmosphere return to its preindustrial levels of CO2."
Buddy, I have some bad news for you.
nickleback_official@reddit
And that is? CO2 does not last in the atmo indefinitely. There’s a long lag but the best we can do is stop producing. Scrubbing is possible but not really feasible at this time. More feasible than a type 1 kardeshev Dyson sphere like OP suggested tho haha.
raptorlightning@reddit
Even if we stop now there's enough to cause massive effects on civilization, starvation, wars, etc. In a few thousand years it might go down some but we won't have much of a society left. We have to sequester at this point if we want to survive.
nickleback_official@reddit
Ok. If that’s the case then we can sequester. Space mirrors still aren’t a solution lol.
DeHub94@reddit
You want to roof entire cities with solar panels and beam the energy somewhere else?
Strazdas1@reddit
Caves of steel vibes intensify.
ICC-u@reddit
No, but imagine you could cut the number of sunlight hours or intensity in some areas to increase vegetation, improve farming, reduce poverty etc. Or cut temperatures in areas with high AC usage to save energy. Cutting energy usage on cooling must be more efficient than capturing and transmitting heat from space.
Lee1138@reddit
Instead of just using the energy directly in favor of the carbon fuel sources we use now?
ryanvsrobots@reddit
That’s how you get dystopian underground cities where access to sunlight is a luxury limited to the wealthy who live on the surface
alvenestthol@reddit
If Archimedes theorized a mirror heat ray that can burn ships, why didn't he also theorize a mirror anti-heat ray that gives you shade?
Oh right, that's an umbrella. You need to cover something to divert energy away from it.
ICC-u@reddit
Archimedes didn't theorise/build a mirror shield because they didn't need one back then, but we do now.
fafatzy@reddit
Because that doesn’t make anyone money
Atomic-Avocado@reddit
Huh
Sh1rvallah@reddit
Isn't that going to start increasing energy/heat absorbed by the planet? And also seems potentially dangerous if they get hacked.
Strazdas1@reddit
Yes but the amount is not significant enough to worry about.
Homerlncognito@reddit
Increased power absorption isn't a concern unless this would be on a pretty ridiculous scale. The Earth is absorbing around 122 PW of solar energy. 1GW is therefore 1/122 000 000.
The security is a much bigger concern though. You can absolutely cook entire neighborhoods with it once it's deployed in sufficient capacity.
Sh1rvallah@reddit
And this is phase 1 from a single company.
How much does the increase in energy absorption compare to CO2 accumulation from NG power?
AssCrackBanditHunter@reddit
Absorbing more heat really doesn't matter. Earth can radiate off all accumulated heat no problem. The issue is when you start smothering the planet with a huge CO2 blanket
greiton@reddit
the amount they could do in the next century is negligible. the planet is really really really big, and the amount of solar panels we have are really really really small in comparison.
when you look at the numbers it is mind-blowing just how much energy the sun provides earth everyday.
AbhishMuk@reddit
Well it's a private company, so if they do get hacked, they'll plead oopsie dasies.
Sorry, you said something about global warming? The corporates weren't paying attention.
Sh1rvallah@reddit
Yeah it's just frustrating that they'd spin this that it's good because solar without presenting research or evidence that it's actually a good idea.
TRKlausss@reddit
Dumb question: even if they send it with a laser, what is the beam spread/cone on those? Anything having the potential to generate anything will fry anything passing above the receiver while in operation…
Also: how inefficient is this? Are they doing conversion photon -> electricity -> photon -> electricity or are they somehow converting the photons from the sun directly to infrared to beam back down?
michaelsoft__binbows@reddit
I thought it was just gonna be some higher orbit satellites that have big mirrors they could point. For something like that I could see some cool military applications... night time spec ops? light up the whole site on short notice. that would be so wild
mgalexray@reddit
It’s one of those scam startups designed to siphon away VC money. Dumb idea that has no merit to it. Eevblog did a good analysis on technical feasibility (it sucks)
SemanticTriangle@reddit
If it works, it is a direct increase to the effective cross section of insolation of the planet when light is captured which otherwise would not have interacted with the atmosphere, ergo, more energy into the earth atmosphere and direct warming.
If light which would have otherwise hit the surface is collected, then that's a local shading effect, reducing insolation there.
Setting all that aside, building in space is much, much more expensive than just building more solar and storage on Earth, so the reasonable conclusion is that this is all just smoke and mirrors to bamboozle shareholders.
Dismal_Guidance_2539@reddit
If building in space is much more expensive than solar and storage on Earth, the no discussion of space base solar in the first place. Right now, we still far from the point battery storage can provide days of electricity need to totally replace fossil generation.
And if starship can lower the launch cost by an order of magnitude, we can bypass many obstacles of space equipment manufacturing.
I think space solar and solar + storage will be the future.
panick21@reddit
We are back the the solar space fantasy, funny how things come around and remain being dumb ideas.
DivideFluffy1279@reddit
Gonna be great for global warming
TophxSmash@reddit
This sounds so fake. Even on a small scale wireless charging is bad and scaling it up to space to earth???
ar-dll@reddit
Oh great. Even more orbital debris.
jhenryscott@reddit
Before you ask, yes, we still have people going bankrupt from an ambulance ride. Thank you for asking.
Dismal_Guidance_2539@reddit
So we shouldn’t develop 24/7 solar power to solve renewable energy around the world just because of health care problem in the US alone.
mediandude@reddit
Yes, we shouldn't, because that would disrupt natural environment.
If you want 24/7 solar then you should go to space.
Dismal_Guidance_2539@reddit
Everything we do will disrupt natural environment. So you happy with the gas turbine run 24/7 instead.
jhenryscott@reddit
You know what I think we should do? Ban private jets, ban leisure commercial air travel, Dismantle every military on earth, make plastic illegal outside of medical devices, line up the 10 richest people every year and walk them off the Golden Gate Bridge, seize the entirety of the NYSE, hold it as a global wealth fund with publicly appointed members serving as the replacement board members of every member institution who serve for life with a comfortable salary but cannot earn any additional income from any source, use what’s left of private profits to build plain electric cars with 50 mile range and ban any other non emergency vehicle, set a “one home per household” law, make all medical care free at the point of service, let all power, medical infrastructure, and every form of insurance be run as a publicly owned non profit and operate as simple and straightforward as your local water and sewer systems, build a worldwide public transportation system, mandate the Chinese language and abolish every prison on earth.
Dismal_Guidance_2539@reddit
Yeah so instead of solving problem one by one. You choose to stop fixing anything and just complain on Internet.
jhenryscott@reddit
Lmao. I’m literally at my job where I develop and build housing for people exiting homelessness. You don’t know what the hell you’re talking about.
Strazdas1@reddit
so you are not actually helping the homeless, because housing isnt what they are lacking. Homelessness is the side effect of untreated mental illness and drug use in vast majority of cases.
Dismal_Guidance_2539@reddit
Ok, you are expert at building house, so you are clearly not an expert at energy generation. This is an energy problem so I pretty sure you also don’t know what the hell you are talking about.
Strazdas1@reddit
What is this shizoposting i dont even. Did you intentionally try to come up with every insane idea you could?
mediandude@reddit
Those already exist, they are called e-bikes and e-pedelecs. With 80km range, some even longer.
Dismal_Guidance_2539@reddit
Or you should go out and live. Most what you just type will never happen in real life so why bother typing all of that ??
jhenryscott@reddit
Since we’re all wishcasting 🤷🏼♂️
mediandude@reddit
Strawman.
We don't need as many data centers.
And we can use PV panels on the ground, for 1500-3500 hours per year, with energy storage devices.
Dismal_Guidance_2539@reddit
Even most distinct expert frequently fail to predict the future. I don’t think any one here can predict that we need data centers or not.
We can use PV on the ground but battery storage still far from the point we can totally replace fossil fuel with renewable. Try a new approach is never a bad thing, especially when we already spend hundreds of billions on developing solar pv and battery. Just a billion dollars commitment can’t change the landscape at all.
mediandude@reddit
Either fully use renewables or no permit, it is that simple.
Strazdas1@reddit
there is nothing wrong with disrupting natural enviroment. You do this every time you take a breath.
Innocent-bystandr@reddit
Can you explain how that pertains to this article?
andrerav@reddit
Everything about this is stupid. Distilled and purified end-to-end stupid. Quite the achievement.
Dismal_Guidance_2539@reddit
I think Chinese and Japanese governments don’t agree with you.
https://asia.nikkei.com/business/science/japan-eyes-world-first-transmission-of-space-based-solar-power-to-earth
Exist50@reddit
Any "by 2050" article is nothing more than someone's fantasy.
Dismal_Guidance_2539@reddit
Every new tech come from someone’s fantasy
Exist50@reddit
You need actual a workable hypothesis and the RnD to get there. These ideas have neither.
Strazdas1@reddit
The company in OP (not meta, Overview Energy) has demonstrated a working prototype using airplane. So its a bit more than workable hypothesis here.
Dismal_Guidance_2539@reddit
I think there a reason both China and Japan support these projects. It because they provide a workable hypothesis.
The RnD money is other problem but Meta already burn $80 billions on Metaverse so I think they have couple of billions to spend on space solar.
Exist50@reddit
With actual money?
Dismal_Guidance_2539@reddit
Space solar actually better for military applications than civilian one so I think there is actual money.
andrerav@reddit
Nice appeal to authority there, but this idea is still stupid.
JaggedMetalOs@reddit
To capture 1GW with solar panels in space (no atmospheric loss) you still need something like 3km² of collectors, and that's assuming 100% efficient energy beaming (which it won't be).
This smells like bullshit designed to deflect from all the fossil fuel gas turbine powered datacenters they are building.
Seanspeed@reddit
Technology has to start somewhere. The first sort of solar cells back in the late 1880's had like a 1% efficiency rate.
These companies know full well that they are going to be energy-limited with their AI datacenter buildout plans going forward and are obviously looking at alternative sources of energy.
Not saying this is somehow a super promising route, but it's hardly surprising that they might be looking into it, even if just as an expensive experiment.
HuntKey2603@reddit
There's a single digit chance that everyone involved actually thinks this is a serious proposal.
There's like, almost literally no upside to this. Especially with the boon of solar being that's dirt cheap.
"Guys, guys... so, what if we made it super expensive?"
Strazdas1@reddit
Overview Energy, the company doing this, thinks this is serious. They demonstrated the technology working from a plane simulating a satellite.
JaggedMetalOs@reddit
They should be paying to build out kms of PV+storage on Earth right now instead of announcing obviously impractical solutions while continuing to build fossil fueled datacenters.
Seanspeed@reddit
These companies are literally exploring a bunch of avenues, including the development of private nuclear reactors.
JaggedMetalOs@reddit
They are "exploring a bunch of avenues" while they build extremely polluting gas turbine and diesel generators to power their datacenters, they should be building clean energy now not just stringing the world along with some pinky promise that some magical breakthrough technology to give them unlimited clean energy for their gigawatt AI datacenters is just round the corner.
itsjust_khris@reddit
The plan is both, solar on the ground and this solution. This solution isn't being proposed as a replacement of ground solar and battery systems, instead it seems to be designed to extend the amount of hours ground based solar panels can produce energy, reducing the energy storage needs.
Sensitive_Ear_1984@reddit
I'm fairly certain this Will be the hardware equivalent of vaporware.
itsjust_khris@reddit
Probably honestly, it's a big investment and the tech isn't even proven.
VenditatioDelendaEst@reddit
I would say that the promise of this technology is appropriately matched to the sophistication of the critics it is meant to appease.
Naaahhh@reddit
I was wondering where all the outer space solar panel experts were
Sh1rvallah@reddit
Wouldn't this cause global warming too though by sending energy into the planet that otherwise would have just radiated on by into space
nickleback_official@reddit
174 Petawatts of sunlight hit the earth continuously. Our measly gigawatts have no effect, it’s just a blip in the noise.
Sh1rvallah@reddit
Sounds like what people say about burning fossil fuels.
Flonkadonk@reddit
Fossil fuels heat up the planet through indirect means that are (unfortunately) vastly more effective. A 1GW solar array in space can at best beam down 1GW (theoretically assuming magical perfect efficiency) to Earth. 1GW from fossil fuels have much more than 1GW of a heating effect on Earths surface.
I still doubt the viability of this in the first place. But the heating effect from this is not as bad as one may think. That said, if this is theoretically scaled up into crazy numbers (think tera watts), that would be a real problem.
nickleback_official@reddit
174 Petawatts of sunlight hit the earth continuously. Our measly gigawatts have no effect, it’s just a blip in the noise.
szczuroarturo@reddit
I see that the world is going cyberpunk in more ways than one.
hasanahmad@reddit
if meta is doing this , there will be a mishap the they will burn people alive using this
Strazdas1@reddit
Overview Energy is doing this. Meta is just buying capacity.
VenditatioDelendaEst@reddit
No.
Focusing light on a tiny spot from space is difficult and you still have to cool the receiver. There is no reason for this to use more power density than you could leisurely walk away from.
dowhatisaynotwhatido@reddit
southpark_were_sorry.gif
dragenn@reddit
Assaulted Instantly...
truehd24@reddit
Meta pioneering SaaS (Sunlight as a Service)
TotalWarspammer@reddit
LOL. 😃
pnutnz@reddit
ahhh, you know sunlight is already beamed from space right...
vagrantprodigy07@reddit
There is no way this is cheaper and more reliable than putting panels on the roof of those data centers. We have solutions to the energy problems facing us, there are just political and corporate interests desperate to keep us from moving to those solutions. Instead, they want to distract us with bullshit vaporware like space power.
Dismal_Guidance_2539@reddit
You forgot the battery need for night hours. That most of the cost.
And no, not only the US billionaire wants space base solar power. Both Chinese and Japanese government sponsors these projects.
https://asia.nikkei.com/business/science/japan-eyes-world-first-transmission-of-space-based-solar-power-to-earth
0gopog0@reddit
While there are very useful applications for the technology, for use in powering AI datacenters is stupid. What the technology and methods let you do is deliver power to difficult to access locations. For instance delivering power to lunar bases/equipment, any technology already faces the launch to space, while the lunar surface also has a 14 day night and additional landing deltaV needed. Here, it makes more sense. Similarly, delivering power to extremely weight constrained or similar condition earth locations (high latitude nights) also offers some prospects
AI datacenters aren't any of those conditions. They are very large permanent installations built in easy to access locations at middling latitudes. Offloading that to high orbit satellites (that have batteries themselves) that have significant costs associated with them is never going to balance out.
Dismal_Guidance_2539@reddit
Sure but the billionaire are pouring money into a new useful tech so why bother complaining about that ??
0gopog0@reddit
Because it's not very useful by a longshot at the scale and the configuration they are pursuing. Money is being poured into developing bad usecases. Flashy (and arguably investor bait) technologies like these have less practical and usefuls returns, and the development of science and technology shouldn't be dictated by a billionaire's opinion of "what is cool".
Dismal_Guidance_2539@reddit
So it is the billionaire’s money and we can’t decide what they spend it on. So why bother ?
And are you sure that this not useful?? I don’t think any one of us here is space solar expert. Many people used to think star-link and even reusable rocket is bad use cases.
0gopog0@reddit
There is a deeper discussion here around society, policy, politics, and a bunch of other things that isn't wholy suited to this subreddit so I'll hold off on that front.
Positive. The company plans to use low-intensity near-infrared light emitted from GEO satellite to utilize solar farms in nighttime hours. Right off the bat, we're immediately limited to high density solar installations, eliminating regions without one present or unsuitable for solar panels. So any place where solar is not a good choice to deploy a large farm is already out. The power transmission modality also means that clouds will significantly impact the efficiency of transmission, further limiting regions it can by deployed, nevermind usual atmospheric losses. Actually, not to understate the problem, but the modality being near-infrared vs radio/microwave doesn't bode well for transmission losses.
But can you do it? Of course you can. There's tons of things you can do. Another orbital body (moon or mars for instance) you'd honestly be silly not to do something like that because the costs compare to things already launched and up there, in additiona to the costs of bringing it safely down to the surface. But cost in context of earth is another matter, and it has to be weighed against other techniques. For instance, why not develop an agrivoltaics system that can readily integrate into already watched and monitored land with minimal work, and even improve the output of the field? Why not develop a drop in system for shading resovoirs in place of shade balls? Why not push the developments of silicon carbide or other high temperature semi-conductors so they can be run with less intensive cooling systems or even energy reclimation systems? Or even just cheaper and more energy denser batteries? Why not optimize and further develop SMRs in all respects from their actual design to the ease of deployment?
Frankly, because those aren't flashy or showy. Incrimental improvements and upgrades go overlooked of seemingly transformatal ones that actually fall behind. At best they misguidedly believe it and bought into what was being sold rather than pursuing cheaper options, at worst its greenwashing investory bait as they fire up another gas turbine to run their datacenters.
There's a loaded statement if I ever saw one that loops back around to the "society, policy, politics, and a bunch of other things" problems with a smattering of technologies available, as welll as it's fair share of truth and falsehoods. Also one heck of a fallacy to try and make a point.
Dismal_Guidance_2539@reddit
You write a 1000 words long comment but I think your hate for billionaire makes it mostly useless.
First, all the obstacles you said about power transmission is spot on. But Meta only invest in one space based solar company. There are many other working on different approaches. So without expert level knowledge about this sector, I think everything we discuss here just doesn’t worth that much.
Second, all of things you list is already under development of different institutions and companies as far as I know. Do you realize we can build different things at the same time ?? Even Meta commit to spend billions on SMRs nearly one years before this announcement. They are even working on recommissioned an old nuclear plant. They invest in few approaches just like everyone else.
So, turn out these billionaires not only invest in the flashy or showy tech to misguidedly people, you just need to stay out of your circle of news to learn that.
0gopog0@reddit
You know, I deliberately left my opinion for billionaires out of it; what makes you draw a conclusion from that? Science doesn't change depending on who/waht group is funding it. Doesn't matter where on the political or wealth spectrum the source of funding comes from. As implemented
Nah. The approaches that are feasible are dispersed near-infrared and radio waves. The windows of energy that are suitable for deployment both, tolerated by people, and spectral absorbtion/reflectance through the atmospheric dramtically limit what is possble. Start putting dangerous emitters - in a directed energy weapon sort of way - and good god the cost of the project would jump just from the safety legislation. Energy transmission from space is not a new idea, nor is it something that is cutting edge from a "what is possible" standpoint.
So, question here. Are you an engineer? Have you worked in an aerospace company designing satellites? Do you have an understanding of spectral interactions with the atmosphere? Have you ever performed energy analysis for different energy states aboard spacecraft? No? Great. Because if you keep appealing to authority and how us little plebians can never understand, I'm going to start waving my credentials and background around.
Stupid investments are stupid investments regardless of what else you are funding. They don't magically start making sense when you fund others, only thing that would be better is diverting the funding to other projects.
No shit. The article talks about it. But they should do more for projects which are sensible and not a joke.
Since you clearly don't have the background for this stuff, don't bother replying cause I won't be waiting for it.
phrstbrn@reddit
Governments invest in speculative tech all the time, that doesn't give this project any additional weight. A lot of that technology doesn't end up working out.
Japan specifically, is a resource poor island nation with almost no energy reserves to speak of. Investing in high-risk, speculative energy products, practically speaking, is an issue of economic security for Japan.
Dismal_Guidance_2539@reddit
Yes, I agree with you. But I think if starship can lower the launch cost to $100-$200/kg or lower zone, a lot of new space tech can now be feasible.
So, let watch out for this idea.
vagrantprodigy07@reddit
I haven't forgotten it. Battery tech is moving forward at a crazy pace, and they are cheaper than ever.
Where in my comment did I say that only US interests want us to avoid the obvious solutions?
Dismal_Guidance_2539@reddit
Sure battery is getting cheaper everyday.
But I don’t think we have data center with full renewable energy plus battery storage. Not to mention, city scale energy storage that can provide days of electricity to totally replace fossil generator.
Looking for a new solution is not a bad thing.
And I think space base solar also have much less impact on environment. Mine billions ton of materials for future battery need definitely worse for environment than just launch them to space with methane and liquid oxygen in Starship.
Gr00m3d@reddit
This how we get heat rays
Loose_Skill6641@reddit
yes let's heat up the earth more with solar death rays, Genius!!
PhyrexianSpaghetti@reddit
No they won't.
It's a bullshit tesla-tier nonsensical claim to counter the "hurr durr we're gonna put the data centers in space" claim by the competition
jdoorn14@reddit
What could go wrong for the 99.9% of the population that aren’t megalomaniacal billionaires? Guessing no one has seen the Simpsons episode where Monty Burns builds a giant array to block the sun, forcing Springfield residents to buy more electricity from him at exorbitant rates?
Seanspeed@reddit
Are you thinking that these satellites are going to 'use up' all the sunlight or something? What are you so concerned about here?
This is just blind, thoughtless cynicism.
Sh1rvallah@reddit
Have you considered that the energy they will be collecting would not have reached the earth otherwise? This will lead to a direct increase in inbound energy from the sun. Which sounds terrifying when scaled up by short-sighted corporations.
Seanspeed@reddit
Like 99.99999999% of the sun's energy that reaches Earth is not 'captured' and utilized for direct energy usage. The idea that this is some big concern is well beyond laughable.
Sh1rvallah@reddit
I'm not talking about the sun's energy that reaches the earth. Some, if not all, of the energy collected by these satellites would never reach the earth. Thus, redirecting that energy at the earth is a net increase to energy absorbed from the sun.
IANVS@reddit
MOON TARP
bhop_monsterjam@reddit
no they won't
thanks for coming to my ted talk
JackhorseBowman@reddit
We're doing sci-fi shit so that you can put nipples on that bird.
amtom61@reddit
So basically use laser beams to transfer energy from outer space to earth surface....Geee...I wonder what happens if someone could move around the laser and let's say target stuff......
future_lard@reddit
Just what the planet needs. More heat
circa86@reddit
This is absolutely the kind of stupid shit Meta would invest in. Makes sense.
yabucek@reddit
Reddit experts when company uses conventional energy generation: grr bad company, taking our resources.
Reddit experts when company invests in R&D of nee energy sources: grr bad company, never gonna work, just do it the conventional way
subwoofage@reddit
Fried chicken, anyone?
ConejoSarten@reddit
This sounds just as stupid as data centers in space
wornoutseed@reddit
So we can expect Death Star like weaponry when they get hacked.
VenditatioDelendaEst@reddit
Elsewhere
This is not a good way to use solar energy in space. If you abandon the dream of scaling to replace all terrestrial power generation, you can do a lot without long-duration storage or conversion losses.
Steerable mirrors with no electricity generation. Sell light to offset streetlighting costs in jurisdictions with overpriced electricity.
Put the load in space, a la Musk et. al.
Military applications for jamming IR/EO seekers, if the reaction time is fast enough.
TryHardEggplant@reddit
Anyone remember the Microwave power plants in SimCity 2000? The satellite misses the dish and lights the surrounding area on fire?
Chipay@reddit
Can't wait for the day I have to watch an advertisement in order to get sunlight.
Noisii@reddit
Aaaaand can this be used for military applications..? ~ the investors probably
kyleleblanc@reddit
I’ll take Things That Will Never Happen for $1000 Alex.
AutoModerator@reddit
Hello sr_local! Please double check that this submission is original reporting and is not an unverified rumor or repost that does not rise to the standards of /r/hardware. If this link is reporting on the work of another site/source or is an unverified rumor, please delete this submission. If this warning is in error, please report this comment and we will remove it.
I am a bot, and this action was performed automatically. Please contact the moderators of this subreddit if you have any questions or concerns.