I use codex daily and a moderately complex plan + a few files is more than enough context to make it blatantly ignore prompts like "stop and ask before adding new layers or data structures"
> Samuel B. Morris, the general manager and chief engineer of Los Angeles’s Department of Water and Power, traveled all the way to Geneva in 1955 to attend the first International Conference on the Peaceful Uses of Atomic Energy. There, he made a case for small reactors, arguing that because the “number of small units…is many times the number of large units,” there could be “economy in development and repetitive manufacture” of the small units.
> But nothing in the history of small nuclear reactors suggests that they would be more economical than full-size ones. In fact, the record is pretty clear: Without exception, small reactors cost too much for the little electricity they produced, the result of both their low output and their poor performance.
For me, SMRs don't pass the smell test. Buying one big plot of land for a reactor, building one power interconnect, having a localized water impact is way easier and cheaper than many. Even if its cheaper to build small reactor vessels, MANY of the other costs become more expensive in larger numbers. Costs that you can ignore while you're just building a prototype.
> Even if its cheaper to build small reactor vessels, MANY of the other costs become more expensive in larger numbers.
There is no way SMR could beat solar+battery power cost even now. With sodium and other batteries projected to reduce storage cost, it is even more unlikely that SMR could be price competitive in future grids. SMR is the only way that dying western nuclear industry could attempt to deal with ballooning compliance and finance costs and hope for revival. China or India doesn't have this compliance cost and still build conventional nuclear.
Even for the SMR usecase that got funding recently, mega datacenter electricity, there is new geothermal power companies getting funding and are transitioning from pilot projects to production. Some are using tech that is already being used in oil fracking industry for decades, so new geothermal tech scaling up has far less roadblocks technology and compliance wise. So other than military and mobile civil applications like nuclear icebreaker, I don't see the chance of SMR succeeding anywhere else.
To a first approximation, China does not build nuclear power. Nuclear's share of power production in China is 4% and dropping. It seems to me that the main reason China keeps building a relatively tiny amount of nuclear is strategic. Reasons like ensuring they never forget how and ensuring they always have buildable designs so that if something changed they could ramp up the program. And probably even more likely: nuclear expertise is important to the country's military.
But coal is still 49%. The whole point of nuclear in a carbon-free, mostly-renewable grid is to avoid needing enormous amounts of storage and overproduction. That's a need that doesn't come up when you're still half fossil.
If you want to decarbonize as fast as possible, it makes sense to focus on rolling out wind/solar/battery as fast as possible for now, but keep developing nuclear technology to cover the last bit where it starts getting especially expensive to replace fossil with renewables without losing reliable power. That's exactly what China appears to be doing. They're not just building a few reactors, they're also the world leaders in developing various GenIV designs, like molten salt reactors.
The most cost efficient new-build grid is 90-98% solar/wind/battery depending on your locales insolation & wind coverage. (Source Ember Energy) The remaining 2-10% is nat gas, if you have a piped source, or coal if you don't.
But 98% renewable doesn't mean you can hit 100% by adding 2% more of something else, it means your something else has to supply ~100% of the power 2% of the time.
If China is aiming for that cost-optimal 90-98% renewables, that means they need to build a lot of coal.
It would also be better if you ate only vegetables and drove an ultralight (ebike) good luck. I personally like being able to turn on my lights 100% of the time
It's not about lights, it's about electricity always being available.
And well, always having power available (for domestic, but also for expensive production facilities that looses money while idle) using renewables is a lot more expensive than generating most of your power with renewables
Solar battery is not by any stretch of any possible definition the cheapest generation to deploy. Not even close. We do a disservice to the entire conversation saying easily falsifiable things like that. At 90% penetration solar + battery has a LFSCOE somewhere around $800/mwh depending on the study. Natural gas is something like 30-60. Solar panels are cheap. Building the required transmission and storage to service those panels is far more expensive than the panels themselves.
Increased renewable generation is an important goal, but we need to let the facts guide the path, not the other way around.
A) I am an electrical engineer who makes a living designing control systems for renewable generation. I know more about most of this than just about anyone in this conversation.
B) the post I am replying to is specifically calling for a 98% solar grid.
You replied to: "The most cost efficient new-build grid is 90-98% solar/wind/battery depending on your locales insolation & wind coverage. (Source Ember Energy)"
There's a big difference between "letting the facts guide the path" and tilting at strawmen built out of straw nobody else even brought into the discussion.
If you'd like to make the point that "cost of panels" can be misleading when supporting components and interconnects begin to take up the lion's share of costs then that sounds like a noble thing to remind folk of, but ideally not as a segway into shilling more fracking which includes untold environmental costs as externalities.
That's a fundamental problem with nuclear plants. Their classic "baseload" niche is gone. Like the intermittent sources, wind and solar, they need storage to hold what they overproduce during peak periods.
This is also true of geothermal, which, like nuclear, is almost all fixed capital cost.
It's only a problem because of capitalism. Every plant needs to independently make money, so nobody wants to be the ones spinning down because that's their profit margin dwindling.
But society would be better off with excess generation. So we either need a system where some are incentivized to spin down, perhaps by being paid a consistent amount regardless or something like that, or we need publicly owned power generation that doesn't care about profit.
This argument assumes we must build nuclear power.
The problem is that electricity is fundamentally priced on the margin. Now that we are moving beyond a purely centralized design.
Think about a homeowner or factory with their own renewables and storage.
They fundamentally operate on marginal price by choosing when they use their own system and when they buy from the grid.
If you add state owned reliable power to this mix they will cherry-pick. Their own cheap electricity when it delivers and the states subsidized reliable electricity when it doesn’t.
We can try add all manner of markets, fixed connection costs and what not to this but all it does is introduce strange arbitrage possibilities in the market.
Which is why we have settled on net energy markets and then the lowest possible amount of ancillary markets to shore up any gaps which would cause problems for the larger society.
We have needed to build nuclear power since the day it was invented. The amount of fossil fueled power generation is indisputable proof of that.
The only way we don't need nuclear power is if we have some other way of producing the energy we need without dooming the planet, and obviously we do not as we're still burning billions of tons of coal every year. Not to mention gas.
We could have made power generation fully fossil free many decades ago if we actually took climate change seriously. It would have been fine. Sure fossil fuels may have been cheaper, but if you include the damage they're already causing and will cause in the future you could multiply the cost of nuclear by orders of magnitude and it would still come out on top. Plus higher investment would lead to faster improvement in the technology, mass production of components and lower cost etc.
I agree that renewable with storage is cheaper of you price in destruction of habitable zones of the planet. Most people don't, and the dollar cost of the storage is a problem over fossil.
The coal argument about China doesn't hold water when you look at what they're actually building. Around 80% of their new power sources are renewable, and 10 to 15% is coal. They're aggressively focusing primarily on building renewable -- so to say that their coal is 49% is either misleading or entirely missing the point.
This assumes that the renewables built are actually contributing to the economy, which we don‘t actually know for sure. In fact there was an article stating that they discard a lot of the energy produced because the grid cannot deliver. [1]
> China, the world's top producer of solar power, rejected 360 terawatt-hours (TWh) of clean power from January to June, up 49% from the same period a year earlier, according to a report this month by Global Energy Monitor (GEM) and the Center for Research on Energy and Clean Air (CREA).
which is great, but as i understand it will still lead to an overall decline in share of nuclear power. The scale of their power production is unfathomable
No, almost every one is above 1GW - divide the net capacity by the number under construction to see the power [1]. They appear to be almost all CAP1400 (1.4GW), Hualong One (1.0GW), or CAP1000 (1.0GW). I see a single Linglong 1 (100MW) under construction. Proposed are all above 1GW.
You need to look at where those nuclear plants are being built, in the east, and where the new energy is out west that makes it appear that nuclear is falling behind. Having power generation that doesn’t need to be transported via UHV lines is useful, it also acts as a nice baseline that keeps the grid stable.
China is building more new nuclear power than the rest of the world combined. They’ll soon produce as much renewable electricity as the United States produces electricity
> There is no way SMR could beat solar+battery power cost even now.
Not a particular fan of this paper, but it does illustrate a point ... https://ieeexplore.ieee.org/document/8867359 showed that for the UK to go 100% solar it would need an energy capacity of 1/3 of the total grid energy demand over the year. Figure 6 A shows the big trend of 6 1/2 months of discharge and 5 1/2 months of charge (i.e. cycle perhaps once a year). Seasonal variations do matter.
It also seems like they are constraining the system to have no overproduction.
It’s like assuming that a fossil based system has all its producers generating the expected capacity factor and then smoothing out the season and daily demand changes with storage. Due to the difference between summer and winter demand such a fossil system would also need to have months of storage to compensate.
Which of course is absolute stupidity. When you can just overbuild production capacity and leave a far simpler problem to solve.
The underlying demand and production has not changed so much since then. The requirements for storage still exist, and strongly depend on when the power is delivered as well as needed.
> a fossil system would also need to have months of storage
Coal and gas also get produced in the winter at a relatively constant rate. Plus we know how to handle piles of coal, caverns full of gas, tanks full of LNG, and linepack for shorter duration gas storage.
> you can just overbuild production capacity and leave a far simpler problem to solve
Sure, you then have an economic problem. The effective capacity factor of the intermittents get driven down. How are they going to be paid for, if much of the time the market is saturated?
I find it telling that you call it an ”economic problem” and ”intermittents”. It seems like you have an axe to grind, but not much backing your standpoint anymore. So you’ve fallen to using derogatory.
Those same fossil fuels have the same economic ”crowding” out problem when cheaper sources in the same class delivers.
A single cycle gas turbine would love to get paid running at 100% all year around. It doesn’t because CCGT plants with higher efficiency undercut it.
Just like what happens in renewables. They start crowding out each other. Storage steps in and solves the peaks. More renewables come online until they ”crowd each other out” and around we go.
That’s called being a market. Which you nuke fans seems deathly afraid of given the economics of new built nuclear power.
It is interesting to see how large-scale nuclear is handled.
120% and 60% increase in cost for FOAK and NOAK (Table 2-1), plus no learning rate for nuclear construction beyond that (Table C.2).
Interest rates during construction unfairly penalise nuclear as "GenCost uses the simplest way which is to increase the capital cost by the assumed discount rate raised to the power of the construction time" (page 97)." This results in ~20% increase in capital costs against other simple scenarios like equal construction costs across each year.
30 year plant lifetime, rather than say 60 years. That results in ~10% increase in capital costs.
> Those same fossil fuels have the same economic ”crowding” out problem when cheaper sources in the same class delivers.
That line of reasoning only works if there is something to make one plant more expensive to produce electricity than another. For natural gas the cost of fuel is far greater than the CAPEX. For intermittents only the variable OPEX can distinguish between generators, which is mostly for wind and I guess most severely for offshore wind. Cannibalisation is the big problem for intermittents. The notion of succession doesn't work for them.
Now you’re desperately trying to rationalize new built nuclear power.
GenCost has an amazing FAQ section you evidently either ignored, or did not peruse.
For example here they discuss economic life vs operational life, when you live in reality rather than grasping for straws:
> Why is the economic life used in LCOE calculations instead of the fulloperational life?
> The LCOE calculation converts all upfront and ongoing costs to annual costs which is then divided
by annual production. The capital cost component of a technology is converted to an annual
repayment to the debt and equity providers. The annual repayment amount is determined using
the economic life and the weighted average cost of capital. The economic life is shorter than the
asset life for some technologies such as coal, nuclear and hydro. Some stakeholders have queried
why this is so.
> Debt and equity providers require a shorter payback period than the total asset life for some
technologies to avoid the risk that part of the equipment might fail or might need new investment
(sometimes called refurbishment or extension costs) to keep operating safely and reliably. To
determine the economic life, debt and equity providers might look to the warranties provided
with the equipment. They might also look at the typical timing of refurbishments or life extensions
for that technology. The economic life is an input provided by the engineering firm that AEMO
commissions each year as an input to GenCost.
> Some stakeholders suggested that coal and nuclear could access special financing arrangements to
move the economic life closer to the asset life. However, our preference is not to introduce special
arrangements for technologies where there is limited Australian evidence. A common approach to
the LCOE calculation is important to maintain comparability. The 2024-25 report does explore the impact of longer capital recovery periods in Section 2. It finds there is no significant benefit from
the longer operational life of nuclear relative to shorter-lived technologies whose costs have been
falling over time.
Even looking at China and South Korea they see essentially zero learning effects across plants after the FOAK build. Small ones at the same plant.
Crying about FOAK vs NOAK is not even close to solving the absolutely stupidly large subsidies new built nuclear power needs.
Again with the loaded terms. Sad. The market is limited until for example Jevons paradox expands it. Which will never happen with new built nuclear power due to how expensive the electricity is, that leads to energy poverty for generations instead. But I digress.
Look at Texas or California. About all new renewable projects in those markets are coupled with storage.
What you call cannabilisation, and try to paint like the end of the world, is simply the market working. Now pure renewable projects aren’t enough, instead you need to sell the electricity when the consumers demand it.
In just a year or two storage has massively smoothed out the price swings in Texas.
But again, that would require curiosity rather than desperately trying to poke holes the study already answered.
2.2 * 1.25 * 1.1 = 3.025x LCOE for the first nuclear power station.
1.6 * 1.25 * 1.1 = 2.2x LCOE for the second plant.
1.0 * 1.25 * 1.1 = 1.375x LCOE for remaining plant.
Wind/solar is heavily dependent on the required storage (which depends on matching supply and demand), additional transmission and backup generation prices.
Finding the price tipping points is the point of the exercise. When you catch someone's fingers on the scales you realise what the game is.
>For example here they discuss economic life vs operational life, when you live in reality rather than grasping for straws:
It's essentially just saying the market can't think long term enough leading to drastic differences in your calculations. But hence it's typically governments pushing these projects forward.
>The market is limited until for example Jevons paradox expands it. Which will never happen with new built nuclear power due to how expensive the electricity is
But it did happen for nuclear power in the past.
Now you see essentially the opposite.
>Why are you so afraid of renewables and storage?
I'm not. I think it's a great set of technologies. But I think if one tries to get to 100% everywhere one's going to stub their toes on the scenarios where it's not all roses.
Why? Because the storage part is hard at scale in a lot of places and the intermitency more pronounced.
In Texas and Cali are incredibly sunny places in the south of the US.
Silicon valley where it's already an issue gets 3 times as much sunshine hours during winter as let's say berlin and those hours are far less usefull.
>Crying about FOAK vs NOAK is not even close to solving the absolutely stupidly large subsidies new built nuclear power needs.
Is this not intentionally sidestepping the ludicrous amount of subsidies that have been handed out to renewables in aggregate?
Berlin would need to overproduce insane amounts in summer to handle it's winters.
The flip side is that what you are saying is that nuclear power will be commercially viable into the 2100s. And betting the house on that. While knowing that the electricity they provide is expensive enough to lead to energy poverty for generations.
That seems absolutely insane.
Jevons Paradox did not happen for nuclear power. What happened was crazy cost overruns, cancellations and the industry collapsing into its current state.
Who cares if we get to 95%, 97, 99% or 100% carbon neutral electricity when we still need to decarbonize agriculture, aviation, chemicals, industry, construction and so on?
Don't let perfect be the enemy of good enough. Transition that final firming to whatever carbon neutral sources we land on when their emissions matter in the late 2030s and 2040s.
We need to optimize decarbonization per dollar spent with the shortest time to market.
You do realize that renewable subsidies are being phased out all over the world? They aren't needed anymore. Complaining about "equality" because your desired solution didn't deliver in time is a kindergarten level argument.
We've spent the past 70 years subsidizing nuclear power. It just never delivered on its promise.
The relevant question is: Where does Germany spend the next €100 billion today to avoid the most emissions?
And that is certainly not new built nuclear power.
>While knowing that the electricity they provide is expensive enough to lead to energy poverty for generations.
What are you on about?
>Jevons Paradox did not happen for nuclear power. What happened was crazy cost overruns, cancellations and the industry collapsing into its current state.
And the opposite happened in the past whilst europe was building bunch more of them.
>Who cares if we get 95%, 97, 99% or 100% when we still need to decarbonize agriculture, aviation, chemicals, industry, construction and so on?
Funny you say that when some of those are significantly harder. Are you going to turn on steel plants only in summer? Do you think that might have some effect?
>You do realize that renewable subsidies are being phased out all over the world? They aren't needed anymore. Complaining about "equality" because your desired solution didn't deliver in time is a kindergarten level argument.
You should tell my government and the neighbouring governments.
>We've spent the past 70 years subsidizing nuclear power. It just never delivered on its promise.
It did exactly that?
>The relevant question is: Where does Germany spend the next €100 billion today to avoid the most emissions?
it has spent €700 billion to $1 trillion on the energywende (a metric which doesn't properly include a lot of private investment) and ends as one of the worst emitters in europe whilst deindustrialising due to high energy prices.
They jerk eachother off about how high their renewables share of the economy is every summer and then import and fire up the browncoal and gasplants again troughout the other seasons. When that solar panel is producing a 10th of it's average summer output (already below the capacity everyone loves to roll with for articles) in winter there's only one big winner possible.
People like to point out how france's electricity production is subsidised when they come out looking good even when the gov intentionally puts a stick in it's wheels by essentially forcing it to subsidise pricing to competition.
> You do realize that renewable subsidies are being phased out all over the world?
CfD is going strong in the UK; it seems about the only way of deploying it, given the revenue uncertainty associated with market saturation/cannibalisation.
Of course, other subsidies exist. Feed in tariffs. Transmission charges. Capacity and ancillary service payments to provide services that solar/wind do not.
This tells me you are not serious. The easiest nuclear reactors to restart, as per that same nuclear lobby group, were those in the north.
Northern Germany is already overproducing electricity, leading to curtailment of renewables.
What problem are you solving with even more overproduction in the north leading to re-dispatch?
> CfD is going strong in the UK; it seems about the only way of deploying it, given the revenue uncertainty associated with market saturation/cannibalisation.
I love you now are trying to smear off-shore wind costs on everything. Of course not lookign at solar, storage or on-shore wind.
Lets lookat those. In Germany, which still do CFD bids for solar, even though a ton get built on pure market value, the CFD bids today are below the market price for the capture-rate of their electricity.
The CFDs are also structured to not pay out when electricity is zero or negative. They trade money for a tiny bit of certainty. That is how far we've come.
> Of course, other subsidies exist. Feed in tariffs. Transmission charges. Capacity and ancillary service payments to provide services that solar/wind do not.
Who pays when half the French nuclear fleet is offline? Who pays when the majority of the eastern european nuclear fleet is offline? Who pays for the N+1 requirements coming from nuclear power being enormous single points of failures leading to large reserves being necessary?
People like you love to complain about these things, but you can never formulate a solution where nuclear power is required to pay for the problem large single points of failures cause in the grid.
> Northern Germany is already overproducing electricity, leading to curtailment of renewables.
Overproducing some of the time. And pulling electricity from Sweden and Norway at other times.
> I love you now are trying to smear off-shore wind costs on everything.
Why is Denmark running the CfD auction if everything is so rosy for onshore wind and solar (plus batteries)?
> the CFD bids today are below the market price for the capture-rate of their electricity.
OK, they are betting on the future value of the electricity they produce being lower than today.
> The CFDs are also structured to not pay out when electricity is zero
So existing generators with older CfDs have priority over new generators with CfDs? Interesting.
> Who pays when half the French nuclear fleet is offline?
Why were they offline?
> Who pays for the N+1 requirements coming from nuclear power being enormous single points of failures leading to large reserves being necessary?
So the 1.X GW of standby is bad, but the whole-system sized standby required for intermittents is fine? Also let's not confuse the capacity factor of nuclear (including scheduled maintenance) with the odds of nuclear being offline unexpectedly.
> you can never formulate a solution where nuclear power is required to pay for the problem large single points of failures cause in the grid.
Wait, is this where I am meant to talk about SMRs? Or nuclear peaker power stations?
Namely, It's very easy to 'design to load' where you don't need to worry about the next shipment of Coal, or an extended weather event causing excessive cloud cover and depleting your reserves.
Heck, even as far as compared to LPG, you don't have to worry as much about disruptions or possible cost shifts around LPG supply.
Yes, I'm possibly tongue-in-cheek handwaving specific types of 'weather events' here and potential impacts, i.e. Tsunamis... OTOH it's a lot easier in current gen designs to make something that would have minimal risk for something, say, in the middle of nowhere Texas.
To be specific, solar with 4hr storage is price competitive with grid right now. Storage becomes expensive after that at 8hrs even with projected sodium or lithium prices. For a datacenter usage they probably need couple of days storage at 50hrs like the planned google datacenter. Only alternate chemistry batteries that could be scaled up easily like flow, iron air, zinc etc. could be competitive for multi day storage, though most of these are still at experimental or pilot stages.
> Google’s new Minnesota data center comes with the world’s largest battery—and won’t raise electric bills
> The tech giant says it will fund enough new wind, solar, and long-duration storage to cover the project’s power demand and avoid shifting costs to ratepayers.
It works better on a geographically spread-out grid, and the data centers are running into the same limit that renewable plants are, which is that the grid is struggling to keep up (in large part, in many places, due to a history of underinvestment).
Data centers with good load balancing between regions do not need exactly the same power 24/365 so the latest solar-wind-storage is already cheaper in theory. But it would take few years before production of sodium and other batteries will scale up and the mean time a turbine powered by natural gas wins.
> there is new geothermal power companies getting funding and are transitioning from pilot projects to production.
If geothermal works economically (outside of volcanic zones where it already does) then it's game over for any other power source except existing dams though because it solves all problems at once.
That'd be a great thing for humanity but I don't think it's worth stopping pursuing alternatives already, because it's still a big "if".
I'm not a geologist, but for comparison, the Siberian Traps[1] erupted for 2 million years, and there have been other large events on that scale. Any heat we release will be insignificant in comparison.
Technically it would, but the same way an individual ant's breath contribute to global warming.
Note that the thermal energy contained in the inner earth isn't a fixed amount: the earth constantly generates new heat from friction and radioactivity, and that new heat is radiated away in space at night, alone the heat captured from the sun during the day. So technically geothermal will increase the efficiency of this heat transfer a tiny bit, but it's really a negligible amount. (In reality the ant's contribution to global warming is probably orders of magnitude than the phenomenon we're talking about here).
Nope. It is very far away from core. Core is very big. And energy is already flowing outwards. So just using it really have not any significant impact.
Real issue is that it is not actually usable everywhere. Again Earth is big and going deep in some places is very hard.
SMR is driven by funding, insurance, and regulation. China doesn't have these issues and builds conventional at a great price. China is building 37 reactors, only 4-5 are SMR-type units, and are larger at 600MW. The rest are standard PWR, and the 40+ in planning are dominating by standard PWR.
Regulatory risk is too high, building too infrequent to understand costs, funding for first-of-its-kind is too hard for nuclear in most countries. SMR makes it fundable. New designs give the hope that regulatory burdens can be lowered.
China's nuclear production is only growing in an absolute sense. As a share of production, it's dropping, from a very low level of 4%. China is building other power types far faster than nuclear.
This is an important point: in a lot of ways, SMRs are a strictly worse than 1000MW plant. The potential financial benefits come from all the other factors you mentioned. I would rather have a "worse" nuclear power plant that actually gets built.
The hope of SMR is that it can be cost-effective at the same level that other factory-produced power sources are. If that's not on the table, then large-scale nuclear deployment mostly isn't going to happen.
> If it's so much cheaper to build multiple small reactors, just build one big plant with 24 small reactors.
That is literally the plan with several designs like NuScale (and I think TerraPower). The plan with NuScale is to ship the reactors on rail or barge, and then truck it in the last few miles. So they cost savings is in not having to custom desgin the actual components for each site, and build them on site. Standard reactor, standard monitoring systems, standard control room able to monitor multiple reactors, etc.
Plus, when you have 12 of them onsite in one large area, you can take one offline for refueling, and still produce power with the rest of them.
NuScale infamously failed to get their reactor funded in Utah (UAMPS and the CFPP), it was just too expensive. Costs kept rising, large utilities declined to sign up, and in the end those utilities remaining were going to hit the contractual off ramp so it was just cancelled. I have some links to minutes of municipal utility meetings in Idaho Falls that showed the wheels coming off (even though there was great local support for the effort.)
Their design requires considerably more steel and concrete per MW(e) than a large conventional PWR power plant. You don't do civil construction in a factory, and that's where much of the cost is. Their design appears to have it roots in the (false) idea that what was holding back nuclear was perception of safety, rather than cost.
Your last point is entirely backwards and missed the scaling concept entirely. I suggest you watch a documentary on the Model T. The point being that these aren't civil projects once it's modular. The prices and material are what's being optimized. To compare to large RPVs and BWRs efficiency is idiotic as that's the only place they beat SMRs and the known downside to SMRs which is the point of scaling it.
It has always been a regulatoryu issue. As given by the fact Valar has a microreactor currently running just to disrpove your thesis.
> The point being that these aren't civil projects once it's modular.
But that's simply wrong. All the projects (except for silly microreactors which don't have a prayer of being competitive) involve substantial civil engineering.
Large scale civil engineering projects will always cost more than simply installing stuff that comes almost fully assembled from a factory (solar, wind, battery or small/medium gas turbines). Opex will always be much higher for a generation facility which requires 850 FTE specialist employees to operate (US average per plant) compared to the minimal requirements for on-site employees for utility scale solar or wind facilities. Generation tech which requires no fuel or hazardous waste handling and storage also has an obvious cost advantage. None of these factors have anything to do with “need to believe”.
I don't think I get your point. A project will not be approved and funded if the local population does not believe it is safe, so safety must be demonstrated through a variety of means, including some you mentioned. This is directly tied to the high costs involved.
> To sum up, since the early 1970s, the cost of constructing nuclear power plants in the U.S. has been steadily rising. This can be traced to a constantly shifting regulatory environment, which has continuously changed plant design requirements, and added more and more safety features, which often were required to be implemented on plants under construction. The regulatory environment is partially a reflection of the fact that nuclear power and the risks of radiation had become increasingly controversial, and that early understanding of the likelihood of a nuclear plant accident was often inadequate.
Nuclear in this millennium is not competing with nuclear or other tech from the 1970s - it’s competing with newer technologies (solar, wind, batteries and gas turbines) which do not require 5 to 10 year huge civil engineering works. All these alternative technologies benefit from mass production in factories. While civil engineering does not get cheaper over time - not just for nuclear plant but all large infrastructure construction.
And that’s just capex - even if a nuclear plant could be built for free, the cost of operation (huge head count, fuel and waste handling), means it cannot compete with newer technologies. Indian Point shutdown years before end-of-life because it was too expensive to operate.
Nuclear plants are large and complex and expensive to build and run. I’m not sure why it should be surprising that electricity generation technology has advanced sine the development of the PWR - it’s been more than half a century. The world has moved on from electricity generation using a huge steam engine attached to an alternator with a fiddly and complex firebox.
It’s economics and newer technologies that have made nuclear power obsolete - not public opinion.
Oh ok, yeah I think we agree. the original quote I was responding to was "the (false) idea that what was holding back nuclear was perception of safety, rather than cost." I was simply trying to say that safety and cost are not independent. Safety is a massive reason why nuclear costs so much.
NuScale's putative safety improvement came from considerably larger cost. Safety and cost may not necessarily be related, but in NuScale's case they certainly were.
> If it's so much cheaper to build multiple small reactors, just build one big plant with 24 small reactors.
The biggest costs to nuclear are associated with each of them being unique snowflakes. They need to be standardized and mass produced to bring down costs.
So the dream is many big plants (eg starting 10+ per year), which is what France did and China does, but since we can’t seem to have that here, small reactors are an attempt to solve that.
The technology was given almost a century to prove itself cost-effective. It did not succeed in the market and only ever worked with truly massive amounts of government subsidies.
High fixed costs + more reactor sites -> more expensive, more nuclear waste sites for taxpayers to clean up
Each of those existing plants were competitive against coal at the time without having the negative externalities of widespread radiation exposure (uranium in coal just released to the atmosphere); mercury contamination in all our freshwater lakes and the ocean, leading to strict limits on consuming fish for children and pregnant women; and a half-century of CO2 emissions.
Orders of magnitude more people are killed by rooftop solar, but we haven't raised safety standards on all other sources of electricity to be the same level we require for nuclear.
NuScale is a terrible example. Their flagship project collapsed before it even started because of cost escalation and their other project was some sort of crypto scam.
> For me, SMRs don't pass the smell test. Buying one big plot of land for a reactor, building one power interconnect, having a localized water impact is way easier and cheaper than many.
Some things would depend on the specific location and the grid around it:
The Point Lepreau Nuclear Generating Station in New Brunswick may be 'overkill' because when it goes down every few years for inspections/retooling, there's little other redundancy available. SMRs would be useful for that regional grid: install 3-4 and one can go down with less fuss.
In (e.g.) Poland there were small/medium coal-fired generation stations near coal mines. If the mines are now empty (or retired for climate change), then the grid connections could be reused for SMRs on an existing generation site: less need to find a new site and build new power pylons, etc.
Security costs are also an obstacle. Any nuclear reactor requires 24×7 armed security and that's largely a fixed cost regardless of power output. You need almost as many guards for a small reactor as a large one.
There's a political issue. Just look at how hard it's been to protest all the new gas plants that came up over the last ~20 years. The reason is that we built many small gas plants, instead of a handful of large ones. It divided up the protesters to the point where they couldn't work together.
When we built less, but larger, nuclear power plants, they're easier to protest. If we can make SMRs economical, it makes it significantly harder to protest.
The people protesting nuclear honestly shouldn't be allowed to vote if they claim they're clean energy due to the apparent lack of critical thought. Seriously, if you're an actual engineer or scientist, this statement is self evident.
> The case is an example of Flock cameras being used to arrest a wanted alleged criminal suspected of committing a violent crime. But it is also an example of police surveillance overreach *in that the cameras were specifically used to create a pretext to tack on additional petty crimes* based primarily on where a person’s vehicle had traveled.
I think everyone agrees. Where the disagreement lies is if this is how the technology is actually used, and the answer seems to be: no.
Regardless, even if the technology was used currently purely for good, when you’re “pro surveillance” you’re actually arguing it will always be used for good, even in some hypothetical future with hypothetical politics. That’s a very hard sell.
I'm pro technology too, but being pro surveillance is orthogonal to being pro technology. Surveillance is an application of technology, not technology itself. It's the difference between being pro-science and being pro-dropping-the-atomic-bomb. The atomic bomb suffers the same problem, too. It only works if used for "good", which... I mean, let's be real.
People don't appreciate the actual argument against surveillance. It's not that it's bad, or evil. It's that it's fundamentally infrastructure which will be used against innocent people.
We have seen surveillance employed to the maximum degree in every totalitarian society which has ever existed. If we're going based purely off the track record, it doesn't look good. Ultimately it's all risk analysis.
Is preventing, like, a car theft or something worth the risk of Nazi Germany x10? Probably not I'd say. And you might say, well, that's hyperbolic. But given a long enough timescale, I don't think it is. I mean, imagine how many more Jews could have been killed had Hitler had access to Flock. And yes, Hitler was an anomaly. But anomalies happen if you wait long enough. We'd have to be foolish to think there won't be a second Hitler. Not that we even need one - we're already seeing citizens crushed under the surveillance apparatus in the US. Today. Right now. With our "just a little bit" authoritarian government.
does this work for any allegation? if it was a serious one surely they could get a warrant. if it's not enough for that it might as well be gossip imo.
I watched them happily train officers from Agencies on doing things they were not allowed to do by law, that Flock knew they weren't.
That "Chinese wall" of separation is nowhere near as concrete as you like to act.
Flock might not set the laws, but when the Flock CEO says he would prefer you to be collateral damage on his company's mission for "a world without crime, thanks to Flock" than dare risk someone not get caught for a crime or alleged crime, via Flock, then sorry, they absolutely are participating in things, and complaints against them are entirely appropriately targeted.
Flock provides tools and information that influence and encourage bad behavior from police.
Yes, we need laws to help stop this. Ideally, those laws should entirely destroy Flock's business model and lead to its bankruptcy.
The issue is not whether cameras are useful for law enforcement. The issue is how that camera information is managed and overseen. What Flock is doing would fit right in in a cartoon authoritarian state, but it doesn't belong in the real world.
dendrites are not really a significant problem in popular batteries. It's associated with lithium metal, vs lithium in normal batteries is in the form of salts. Solid state lets you use metal, which is much more energy dense since you don't need the salts.
The most common lithium battery failure mode is that you have a hole in the plastic separator between the +/- sheets inside the battery, which shorts and causes a hotspot that eventually starts a fire. Dendrites cause the short by growing across the gap. In normal batteries it is caused by a manufacturing defect. The outcome is pretty similar.
The energy "stored" in the light oil electrolyte of a battery is >10x more than the electrical energy or the energy released by reacting lithium alone.
An 18650 battery weighs ~50g and stores ~10 watt-hours. 10 watt-hours is 8,604 calories, enough to heat 50g of water by 172 C or 310 F. The battery would not even burn without a liquid electrolyte to ignite.
The difference between gpt 2 and 3 was insane. 2 could generate limericks when it wasn't repeating a word 300x. 3 could actually do some things. By comparison gemini robotics has hardly changed at all.
I will also point out that slow, non-fluid robotics is on a totally different level of difficulty from fast fluid motion. Asimov could walk pretty smoothly, but it didn't fall over because it used a very careful sequence that was never unbalanced; you could pause at any point without falling over. Move faster, like boston dynamics, and you need to account for the change in balance from your arms swinging... or rather, you need to be able to account for the rotational inertia etc from moving multiple masses along complex paths with multiple points of articulation at hundreds or thousands of times per second.
An algorithm to fold tshirts 90% of the time is easy. The cloth hangs down by gravity and you can just look for right angles (corners), find their coordinates with binocular matching, and move them to meet each other. Getting 99%, or folding them quickly, so that the fabric is actually moving instead of just hanging still- incredibly, incredibly more complex.
Even being able to simply do the task isn't enough here. It also has to do the task while also never, ever, ever accidentally folding your toddler as though it were a towel. Having industrial-strength robots around inherently unpredictable humans in a fail-safe way is an extremely difficult problem that will not be addressed any time soon.
This is not the bar for societal adoption, and we can already see a demonstration that it is false.
A waymo is basically an industrial strength robot. It operates among inherently unpredictable humans already. There is no guarantee that it will never, ever hurt a human.
And yet in many cities around the world, you can call a waymo and ride it and society accepts it.
I think that's not a bad point but it ignores the different safety standards that most people accept when it comes to a) driving and b) their toddlers. Like a household appliance that kills 20-40k toddlers every year would be a major scandal, but it's OK for automobiles.
> folding them quickly, so that the fabric is actually moving instead of just hanging still- incredibly, incredibly more complex.
TBF fabric is much more difficult to simulate than a bipedal body is.
As I recall openai had mujoco playing soccer nearly 10 years ago. Obviously real world bodies are much more difficult but I'd be curious to learn why that is.
I don't think that there's any inherent difficulty to do with physical bodies, it's just that the learning techniques used by the big labs, that is to say, various forms of Reinforcement Learning, don't transfer to the real world. Because in the real world if you try stuff at random until you figure out the right policy, you end up destroying too many robot bodies. Imitation learning that went wild a couple years ago with ALOHA looks like a great way to get around this on paper, but then you realise that you have to train your robot for hours on every single task you want it to perform and it still won't generalise even between similar tasks. The sim-2-real gap also remains a big obstacle and you can't just train in simulation and transfer to the real world seamlessly; and then the seams end up being more like chasms.
Then there's the problem of generalisation. RL is insane in figuring out the dynamics of any arbitrary environment you drop it in at any time. Unfortunately once you have a policy trained in one environment (or one task) you have to train again for the next environment (or task) you want to deal with. The number of different environments, situations or tasks in the real world ends up being overwhelmingly large.
So for example DeepMind has been trying for ever to train various robots to do stuff like grasp arbitrary objects by training in the real world or in simulation and still grasping is an unsolved problem, in the general case; and in fact in most cases.
With LLMs it was different in that there was a www of text to train on and the stakes are lower because there's no sim-2-real gap and there's no need to deal with the real world, in training or deployment, it's just text on a screen. That's very hard to replicate in robotics.
There are major breakthroughs still to be made and they aren't currently even on the horizon. And no, I don't have the answer with my super secrete robotic AGI homebrew project either, just saying :P
Scale obviously does not make this more okay; watching many intersections or cell towers is obviously less reasonable than a single intersection or tower.
> In comparison: license plates, when in public, are always visible, and very easy to discern from one-another (different state-unique numbers); so in my mind the expectation of privacy is far lower.
Legally, overhearing a conversation is not different. If you are loudly talking about your drug deals in public in front of an officer, they can use that as evidence. A police department could hire officers to stand everywhere in public and listen to every conversation nearby.
Or, more specifically, they could stand conspicuously close to every payphone and listen. The phones are in public; the officers don't need to be uniformed. Practically speaking this is not at all different from wiretapping. That's what the police did in Katz- wiretap a public payphone.
Flock cameras are not in any meaningful sense different from having officers follow around everyone and record everywhere they go. Its irrelevant whether one officer is following one person or if many people are following them, each within their own small area. That would absolutely not be legal without a warrant. The only difference is a private company is doing it and selling it to the police, something that should clearly not be legal.
The constitution does not really care about scale, though, and that’s my point. It’s a reason why the legislature should care about Flock, but not why the judicial should.
"some caching gains" is a pretty huge understatement- snipping something out of the middle of the window requires rebuilding the entire context. Thats a shitload of tokens.
Afaik messing with the context also pretty reliably degrades performance still. The model responses reference things that no longer exist to it and it becomes more chaotic.
The real usefulness of parallel or sub-agents is not that they run at the same time, its that they isolate noisy or self-contained context away from the main window.
I still feel that during agentic workload sometimes it would be nice to have the model identify it is veering off the main track, send out a "keep the cached states and tell me which they are" command to the inference server, do the side thing (such as handling an error that plopped up that has not that much to do with the main task) and return back to the cached state with just a comment tacked at the end to say "oh and btw I fixed DNS" instead of having the DNS debugging inside the context window now. Maybe other harnesses just steer the models more towards using subagents for such tasks and my pi is misconfigured. I can use the tree feature, but having insight into what's cached would be nice there.
You can structure your context window to minimize the amount of editing you do further back. You usually only need to edit and correct the most recent response. It's little different from forking the conversation at an earlier point, and nobody warns about that being a sneaking footgun. There is still a prefix to cache.
browsers aren't common either. Standards, formats, and interfaces are, which is exactly what WASM is and what this demonstrates. Native apps don't need a common operating system or even a common core like nix. They just need to support a common interface, like browsers do.
reply