I attached a generator with some supercaps and an inverter to a stationary bicycle a few years ago, and even though I mostly use it as a way to feel less guilty watching Youtube videos, it does give me a quite literal feel for some of the items on the lower end of the scale.
- Anything even even halfway approaching a toaster or something with a heater in it is essentially impossible (yes, I know about that one video).
- A vacuum cleaner can be run for about 30 seconds every couple minutes.
- LED lights are really good, you can charge up the caps for a minute and then get some minutes of light without pedaling.
- Maybe I could keep pace with a fridge, but not for a whole day.
- I can do a 3D printer with the heated bed turned off, but you have to keep pedaling for the entire print duration, so you probably wouldn't want to do a 4 hour print. I have a benchy made on 100% human power.
- A laptop and a medium sized floor fan is what I typically run most days.
- A modern laptop alone, with the battery removed and playing a video is "too easy", as is a few LED bulbs or a CFL. An incandescent isn't difficult but why would you?
- A cellphone you could probably run in your sleep
Also gives a good perspective on how much better power plants are at this than me. All I've made in 4 years could be made by my local one in about 10 seconds, and cost a few dollars.
Where I am at least, people using less power because power because power need to profit more, is wild.
They literally had record profits the last few years, rather than being forced to lay down solar. I think power should be a global endeavor, not some local for profit business with complete regulatory capture that makes competition illegal.
Yes I'm angry, because I pay more in electric than most anywhere in the world. If I charge my care with LEVEL 2 using city provided charges, during the day, it's more expensive than gas.
Solar has one of the lowest capital costs [1] so the discounting works in it's favor. And then the non-discountable operating costs also works in its favor since the fuel supply (light) is free.
Yup. It's why even in fairly red states like my own (Idaho) solar, wind, and battery are going up everywhere. Even without significant subsidies the economics are really good for renewables.
They'd be even better if we didn't have extreme tariffs on China.
That's actually what's convinced me that renewables are a better choice than nuclear. I still like nuclear, but renewables are just so much easier and faster to deploy while being a lot cheaper. To make nuclear competitive requires regulatory changes along with a government that's simply willing to tell it's NIMBY citizens YIMBY.
Government literally has to get in the way of renewable deployments at this point to stop them.
Huh! I'm kind of stunned that we only use ~30x the power that we did back then. If I'd been asked to guess I would have added another 0 or even two of them.
Yeah we had an exponential jump when we discovered oil but we maxed that out and the growth has been linear since (and paying for it environmentally too).
I’m waiting for the next big major discovery in energy generation.
We’re always on the verge of fusion… fusion will be like the discovery of oil. Humanity will jump forward… well, technologically at least.
Nuclear is low carbon, it’s fine we lose heat to extract that energy versus stationary and mobile combustion generation, as there is no other effective way to extract that energy at this time.
> Batteries are now cheap enough to unleash solar’s full potential, getting as close as 97% of the way to delivering constant electricity supply 24 hours across 365 days cost-effectively in the sunniest places.
What does this mean? It means we are most of the way there with solar and batteries alone, even if we need a bit of carbon based generation to bridge the gap while solar and battery deployments scale globally. Solar and batteries will only continue to get less expensive and better.
> Something wind, solar and batteries for the next 50 years aren't.
False. If you'd stopped before the "and" you would have been correct, though.
Batteries are really cheap now, and supply of batteries is growing basically as fast as people can get the investments and permissions for the inputs and the factories.
Pumped hydro energy storage relies on the cheapness of water and existing geology. If you have to build the chambers instead of damming a river it's too expensive. Most of the good spots to have a reservoir are already used. If you have to manufacture the bulk media instead of just using water it's too expensive.
The argument against lifting concrete is that you can dig a hole in the ground an pump water in/out of it for more reliability and lower cost than having a crane lift and lower concrete, and it's easy to make it much bigger both horizontally and vertically, so why bother.
But it does appear to be economical even with that, and water is cheaper.
We make lots of holes in the ground on a regular basis, including for extracting fossil fuels. Here's two, note scale bar, though I have no idea what the rock around it is like regarding water losses: https://www.google.com/maps/@50.9063171,6.4418046,17655m/dat...
There are exactly zero economically viable pumped water storage systems where water towers are involved. If you do the math for the amount of a mass of water, you'll see why! It's not feasible.
> They literally had record profits the last few years, rather than being forced to lay down solar. I think power should be a global endeavor, not some local for profit business with complete regulatory capture that makes competition illegal.
Sounds more like you guys should be lowering barriers to entry, not setting up a global non-profit cartel.
True. I suggested global because it allows for scale with copy/paste designs, where things like nuclear could actually become viable.
Where I am, we have a solidly aligned state government. There's no concept of consequences for anyone in power. They're paid by the local companies to pass laws to make competition legal. Some are investors. All corrupt. That's what you get with a solid political alignment.
Vote with your feet (and wallet), and support anything that makes people voting with their feet easier.
It's the McDonald's theory of policy: you don't vote on their burgers at the ballot box, you just go to Burger King or get a doner kebab, if you don't like it.
Once I did a little bike training and looking at my power curve, I was incredibly impressed by how cheap energy is. 100W is an all day number, 200W less so, 300W is exactly 20 minutes when I do an FTP test. 400W is 4x Tour de France winner Tadej Pogačar for an hour and he's a mutant. 1 horsepower is under a minute iirc, definitely under 2. 1kW is maybe 10 seconds. So I could keep my laptop and phone charged probably indefinitely as long as I have food, but not a ton more than that.
Did you try charging an e-bike with your contraption?
I don't know what you can take of this, maybe you can see it as advance pedaling, or to get a feel for energy conversion losses. Anyways, it is the kind of harmlessly stupid idea that I would want to try just because I could.
Amazing stuff, have you written up a blog post? I could see a video being a fun format for this as well. Might help people develop the intuition for watts/power consumption in a different way
I did do a video back then going against the infamous "bicycle toaster challenge" video (in which I determined it was probably less real than they made it out to be). I'm nowhere as fit as those guys, so in my attempt I was only able to turn a bagel into a dry crouton over the course of an hour.
I'm as curious as you to be honest - putting a strain gauge on the pedals for measuring mechanical power has been on my list for quite a while. My own (probably inaccurate) measurements right after the generator says I can get 60-70Wh in an hour, but I can get to 100Wh if I try harder. I have reason to believe my setup underestimates power because my ammeter clamps at 5A and I know I can peak over that on the down stroke of the pedal.
I've seen numbers like 250W mechanical power for an average trained cyclist, so either my setup is rather inefficient, my measurements are off, or I'm going to find out that I'm nowhere near as strong as a real cyclist.
On the other hand, the stationary bike I got originally had a rubber belt, which it would chew excessively and I eventually swapped it for a chain because it kept slipping in spite of tensioning it more, suggesting I'm hitting the thing harder than it was originally designed for (how that translates into power I'm not sure).
> I've seen numbers like 250W mechanical power for an average trained cyclist, so either my setup is rather inefficient, my measurements are off, or I'm going to find out that I'm nowhere near as strong as a real cyclist.
Cyclists' power output is sometimes reported as a 'power curve' - a chart with power on the vertical axis, and duration-of-that-power on the horizontal axis.
For example, a cyclist might be be able to produce 500W for 15 seconds; 350W for 1 minute; 270W for 10 minutes; 200W for 1 hour; and 150W for 5 hours.
The author Hannah Ritchie works on Our World In Data and also publishes the fantastic Sustainability by Numbers substack. It's in the same vein as the late, great David MacKay's Sustainable Energy Without the Hot Air.
This tool has its own recent substack post. See the comments too, especially the one by Chris Preist that contextualizes the energy usage of streaming video (a topic that has also been discussed on HN before).
She's employed by Our World In Data. She is also a published author of printed books. Her substack does not have paid subscriptions enabled (or at least it did not as of last summer; see this post [1]). Our World In Data is funded by donations:
I think stuff like this really crystalises how people misunderstand how much energy stuff uses.
My parents for example sweat the small stuff and go around the house turning LED driven lights off to "save electricity" even though it would barely make a dent in their bill.
Granted, they come from a time of incadescants burning 60-100w at a time so I can see why that habit might be deeply ingrained.
The ridiculously dramatic drop in power we dedicate to lighting is one that is just tough for folks to internalize. As you said, used to, you could have ~10 lights in your house that would add to upwards of 1kw. Nowadays, you can have 50 lights and barely hit 500w. Just mind blowing how far we dropped energy on those.
Same goes for televisions. Your modern TV is probably closer to the basic light bulbs before LEDs.
I'm assuming the general trend is true for all things solid state. That said, lighting is by far the biggest drop for most houses. Remarkably so.
> I'm assuming the general trend is true for all things solid state. That said, lighting is by far the biggest drop for most houses. Remarkably so.
For commercial and industrial installations, VFDs have probably been the biggest efficiency gain, even moreso than lighting. Half of all electricity consumed is used by motors. Thank goodness for solid state power electronics!
I turn LED lights off because of the difference in operational life, and I don't like changing bulbs. M GE bulbs say they have a rated lifetime of 13 years......at 3 hours of usage per day. So if they don't get turned off, then that 3 hours can very easily become 12, and now you are at a rated lifetime of ~4 years instead.
A ‘standard’ (A19 shape, E26 base) 8W 800 lumen LED lamp costs around $5 and will use about $20 of energy over a 15,000 hour lifespan, assuming $0.15/kWh.
That works out to around $0.035 per day for the lifespan of the lamp if you run it constantly for 24 hours a day, I wouldn’t waste time thinking about it. It’s an extra $10 over 12 years, you’re still using the energy.
Investing in occupancy or vacancy sensor wall switches at $25 a piece would be the best option, then you don’t need to remember to turn the lights off!
It's not a cost thing, I just don't like changing bulbs. I find it annoying, and with enough bulbs, when the lifetime is down to 4 years, you are doing one every few months on average.
But yes, I have thought about presence sensors. I'd really only need 3-4 to cover the primary areas where lights get turned on and not off (if I don't do it). I just haven't gotten around to it
Yup. In my experience, average non-nerd folk very very little feel for this stuff. I suspect some believe energy consumption of phone vs car is basically a toss up.
It was genuinely a surprise to see how much relative energy petrol cars use (and shame on me - I'm an electrical engineer). I mean I think I knew it intuitively, but this simple chart blew my mind.
When one gets in the weeds on EVs or ICE cars two things become shockingly clear: internal combustion is hilariously inefficient YET gasoline is hilariously energy dense. Most people's intuition is wrong on both of these points but then they cancel each other out.
Edit: another important point is that the "cost" to acquire gasoline is only the very end of the process. The energy has already been gathered, stored, and most of the processing is complete. Our cost (in money and energy) to "make" gasoline is really just gathering it. This is why the comparison to renewables is often a hard sell, it's just apples to oranges. Gasoline started on third base, renewables are batting from the plate. Some of the internal combustion enthusiasts are holding up e-fuels or synthetic fuels as the solution but then we have to pay for the entire energy gathering and processing pipeline and still be using a conversion method that's not at all efficient. It's the worst of both worlds.
It's inefficient but not hilariously so. Modern ICE are quite amazing technology.
Combined gas turbines (you know, the energy source that powers your electric car) are about 60% efficient for the really good ones, minus 5-7% transmission losses, minus 10-12% charging losses, minus 20% loss in cold climates, lands you at around 35-40% efficiency from fuel source to the wheel.
The Atkinson-cycle engine in the Toyota Prius gets around 40% give or take some losses in the drivetrain. Electric have plenty of upsides, but for some people with cheap gas+high electric costs+cold climate you would honestly be better driving a hybrid.
This is something that always gets lost in these conversations.
Whenever you do the real world calculation for what an electric cars CO2 profile looks like it turns out to be the same as a gasoline car unless your country is majority nuclear.
> Whenever you do the real world calculation for what an electric cars CO2 profile looks like it turns out to be the same as a gasoline car unless your country is majority nuclear.
Maybe today, but basically everywhere south of Canada solar is so high ROI that it's just a question of time before sunlight hours have electricity so much cheaper that the primary daytime parking locations become the favored (slow-)charging spots.
For you to prefer charging there your employer only has to charge you less during the day than your utility charges during the night, so the day/night rate arbitrage can easily pay for the metering hardware and installation (at the next opportunity to install without having to dig the parking lot up just for the chargers), with the rest being profit to incentivize the managers to install/offer this.
That means relatively dirty combustion near where people live. The population density around fossil fuel power plants tends to be pretty low in wealthy countries.
You can't pump hot water the same distance you can transmit electricity on HVDC towers.
Train locomotives have used diesel powered generators that then powers electric motors. Would this be less efficient than battery powered EVs? Or better asked, what would be the most efficient use of gasoline?
> Would this be less efficient than battery powered EVs?
Measured in terms of mass * distance, trains with steel wheels will beat anything with rubber pneumatic tires.
Part of the magic of hybrid trains is that you can have multiple generation units that can be turned on or off as needed.
---
Efficiency is just one consideration for a power plant.
Historically, reliability has been more important than efficiency, especially for industrial applications like locomotives. In other words, locomotives are probably not as efficient as they could be. For instance, you could use a lower viscosity engine oil for lubrication, but that would reduce reliability as engines fail due to friction.
In the auto industry these are usually called a series hybrid and there have been a handful. The Chevy Volt (though it had the ability to directly connect the engine to the wheels at highway speeds), and the BMW i3 and i8, the Fisker Karma/Karma Revero. The new Ram Ramcharger truck and the second gen Ford Lightning will also be series hybrids.
It's a really good drivetrain that was unfortunately made untenable for a long time by a combination of regulation and market forces.
When it comes to the environment the most efficient use is to leave it in the ground.
Hybrids work for trains because they are so large and don't need big swings of acceleration or to climb steep grades. They can run the diesel generators at maximum efficiency.
Battery power would be better, because you can build even larger power plants running at higher heats and not have to haul them with you, but the costs of sufficient battery is too large, so far. That is changing.
"Maintenance costs of the lines may be increased by electrification, but many systems claim lower costs due to reduced wear-and-tear on the track from lighter rolling stock. There are some additional maintenance costs associated with the electrical equipment around the track, such as power sub-stations and the catenary wire itself, but, if there is sufficient traffic, the reduced track and especially the lower engine maintenance and running costs exceed the costs of this maintenance significantly."
We once dammed basically every river in the nation because it was in vogue at the time.
Maybe building overhead power lines for rail infrastructure should be the "hip" thing right now instead of AI. Maybe building oodles of solar power farms and batteries should be "hip"
We built electrical infrastructure to the most remote residences just because we could and because it was an investment in our people. We directly funded our massive and formerly world class rail network because we could, and because it would pay off. We built a world class road network half as a make-work project, and it still pays dividends. We purchased Alaska, with no obvious reason. We built a space program to have slightly better nuclear weapons, and it's part of the reason we were so dominant in computer chips for so long.
We have spent something like 40 trillion dollars over the past 25 years, and almost none of it on anything of real value. More than a little of that debt is just handouts to already rich people.
We can build new electric transmission lines and I'm so tired of things that we absolutely 100% can do if we just demand it be done being somehow treated as a problem. America can afford infrastructure.
That's a complicated question that unfortunately has quite a bit of "well it depends" in the answer. I worked in the auto industry for a long time - both doing engine development and EVs - so my opinions here are well-informed but not world expert.
From a pure energy efficiency perspective you can't beat economies of scale. A stationary power plant (even ones that are just big gasoline engines) run at a constant load and RPM so they can be optimized for pure efficiency, they rarely have to start, warm up, and shut down, and they can use larger and more expensive exhaust aftertreatment systems. Most energy conversions grow more efficient with scale and this is no different. The locomotive powertrain works for a handful of reasons but one of them is you can build much more efficient engines that are optimized for a single constant speed and load. But most of the advancements in internal combustion engines over the last 20-30 years don't increase peak efficiency but increase the conditions in which they're efficient. Variable valve timing and lift are probably the most underrated and overpowered technologies that have transformed engines from having one narrow regime of high efficiency to running well over a huge range of the map. But turbocharging, variable intake geometries, 7+ speed transmissions, and mild hybrid systems like belt-starter-generators get honorable mentions here. However we're not talking about anything close to EV-levels of efficiency. I think the cutting edge research engines are running in the mid to high 40s for thermal efficiency (percentage of fuel energy captured as useful work), most passenger car engines probably peak in the mid 30s.
So while there is some efficiency to be gained by a more locomotive-style system it's not as much as you would hope. In the industry that's called a series hybrid system, vs a parallel hybrid system where either ICE or EV power can go to the wheels. The benefits of a series system are more emissions and product features. You can get the full torque and power of an EV, you can start and stop the IC engine in a more emissions optimized way, and and you can filter load spikes to use a small engine that meets average not peak load.
From a more pragmatic perspective, with the energy density of gasoline and other liquid fuels it's probably best to use it in applications for which you just can't use full electrification. Planes are currently the best example of this. It's also worth noting that passenger cars benefit massively from strong hybridization because of the uneven load cycles so that's a technology where you can deploy a gasoline engine but then claw back a lot of the efficiency losses with hybrids. That's not always true, for example boats don't really have a regen cycle so hybridization just doesn't get much.
In Japan, my country, this looks a bit different. A lot of electricity still comes from oil- and gas-fired plants. The mechanics differ (gas turbines vs. car engines), but in both cases we’re still relying on combustion. I suppose some countries have the same issue.
The presentation is nice, but some of the conversions are questionable.
For instance: The cost section, wherein 1kWh in the US is figured as having a cost of 9.7 cents.
In reality, it's not that way at all. Unless we're fortunate enough to live in an area where we can walk over to the neighborhood generating station and carry home buckets of freshly-baked electricity to use at home, then we must also pay for delivery.
On average, in 2025, electricity was 17.3 cents per kiloWatt-hour -- delivered -- for residential customers in the US.
I looked at the electric car example for the United States. It has 3 kilowatt hours priced at $0.51, 17 cents per kilowatt hour, which seems about right. The "petrol car" example at the top of the chart isn't powered by electricity so its cost number is not directly comparable to the things that consume electricity.
The electric shower also seemed pretty optimistic. I live in an area with about 50°F/10°C ground temperature and my 14.4 kW water heater can just keep a relatively efficient shower head flowing at a comfortable temperature.
This one is resistive (tiny and cheap to purchase) but will be just an emergency-backup shower once my home renovations are done.
The house is getting a split-system air-to-water heat pump with an indirect tank for domestic hot water, so it should cut that down substantially (the unit maxes out at around 3kW input but likely will run longer to recover/preheat).
Interestingly, the latest generation of LED lights are so efficient that it makes little financial sense to bother having a light switch. 330 lumens per watt from recent Phillips bulbs!
The cost of having an electrician wire that switch is probably more than a little 2.5 watt light will use in it's lifespan - particularly when you account for the fact lights in hallways are probably in use most of the time anyway.
Add in the effort of switching the light switch a few times a day for many years and it's certainly the case! Or the risk of fumbling in the dark for a light switch at the far end of a room or in a house you aren't familiar with.
Obviously you might still want to turn it off for maintenance, but you have the breaker for that.
They are direct AC bulbs, no inverter needed. They do have smoothing capacitors so they don't flicker, and I think a slight redesign of the drive circuitry could add ~10% more efficiency...
> So, if I wanted to analogize the energy usage of my use of coding agents, it’s something like running the dishwasher an extra time each day, keeping an extra refrigerator, or skipping one drive to the grocery store in favor of biking there. To me, this is very different than, in Benjamin Todd’s words, “a terrible reason to avoid” this level of AI use. These are the sorts of things that would make me think twice.
I end up shrugging. For a Claude Code power user, today, a day's use uses less electricity than a morning commute in an electric car. To say nothing of the costs to keep your workstation running, your building heated or cooled, etc. Not quite a rounding error, but a relatively minor component of overall usage.
At least for programming usage the power usage seems worth it. For starting up 1 million bots to argue with each other on facebook it's obviously a total waste.
At any rate, the power usage will become more apparent when these products stop being subsidised, where power usage is being charged to the end user.
I also was under the impression that queries cost were mostly meaningless, but it seemed only is true for fresh sessions and short queries. I have to say, the result is less dramatic than I expected but still more significant for heavy users (such as myself).
I'm not sure it's even a particularly relevant comparison to an hour of use of various other electronic devices. I'm sure the median user is running a lot fewer queries than a Claude Code power-user, but I would guess it's still more than one in a typical session.
I wanted to write an app like this. Here's what I think would be useful for some people to compare: more checkboxes that represent typical daily activities like "how much energy does driving to the grocery store (x miles away) use" or "how much energy does idling in my car for 30 mins use", "how much energy does riding the bus/trolly/etc use", "how much energy does keeping my place warm vs kind of chilly", and also "how much energy does a cargo ship use to cross the pacific" or "how much energy does it take to source, manufacture, package, transport, and stock a pack of bubblegum". I think that would help people realize the largest components of their footprint and also how it relates globally.
One thing missing but important to understand is the energy embodied in buying 'stuff'. At a very rough approximation, the cost of stuff, especially consumer goods manufactured cheaply, is quite a high percentage energy.
When you look at people's energy usage, quite a lot of it ends up being the embodied energy in the stuff they buy. For quite a lot of people, it's probably the largest category of energy consumption. I once had a very rough go at calculating this here: https://www.robinlinacre.com/energy_usage/
One gram of finished 3nm packaged semiconductor is roughly equivalent to half a kilogram of refined aluminum in terms of energy cost. If you want to spend a lot of energy for not much mass, photolithography is fantastic.
Doesn't show the comparative energy waste of bitcoin?
This source[0] says
> One Bitcoin now requires 854,400 kilowatt-hours of electricity to produce. For comparison, the average U.S. home consumes about 10,500 kWh per year, according to the U.S. Energy Information Administration, April 2025, meaning that mining a single Bitcoin in 2026 uses as much electricity as 81.37 years of residential energy use.
I find this at best non-enlightening and at worst misleading because there is no consistent unit of comparison: 10 minutes here, 1 hour there, 10 miles somewhere else. That renders the visual comparison of a bar chart pretty uninteresting, no?
1 chatgpt query is a little misleading though. Let's see an 8 hour full bore claude code agent session. Or maybe running 3 agents for several hours a day.
It also doesn't include the amortized cost of training the models, as far as I can tell. I believe I heard that training the models took more energy than total queries against that model, but I could be mistaken.
I believe training currently costs significantly more than inference to all the current vendors, so I'd be surprised if it doesn't also use more power.
And by the look of it, that'll be the norm pretty much forever - unless something fundamental about how models can be trained/updated, an "older" model loses value as it's knowledge becomes out of date, even if we no longer get improvements from other sources or techniques.
But other things likely change based on "lifetimes" and usage patterns too - e.g. a large battery for an electric car may have a higher upfront energy cost in manufacturing than a small ICE + fuel tank, but presumably there's a mileage that the improved per-mile efficiency overcomes that, and then continues to gain with each additional mile.
For reference it would be good to have per-passenger numbers for "sitting on a diesel bus", "sitting on an electric bus", "sitting on a tram", "sitting on a commuter train" as well.
The marginal cost of one extra passenger is going to be very nearly zero. The vast majority of the cost is just moving the bus / train / plane, and the overhead / inefficiencies in the system. I've seen somewhere the numbers for one passenger on trains and planes but I can't remember where that was. Just know it is a very very small amount for the added weight of one more passenger.
Yes, I know, and marginal numbers would also be great to have. But just the naive cost of energy per passenger-kilometer averaged would be good to see, given that it too would be much less than that of a private car.
Right, a lot of discussion about the energy economy of transportation would probably be better framed using units of people-miles or cargo-miles, per unit energy.
How does it make sense to compare renewable vs non-renewable energy sources?
Where does that electric car's energy come from?
Suppose the electric car's energy is solar. Then driving the electric car consumes zero barrels of oil, and so is infinitely more renewable than the petrol car. Or how efficient is the electric car at capturing available energy? Maybe the petrol car is 20% efficient at capturing available oil reserve energy. But the electric car is like 1e-30% efficient at capturing available solar output. Incomparable.
Why does their air con take 1kW to run? I wonder how much below ambient they are setting their temperature, and how old that aircon is and how bad the insulation?
The washing machine seems also inefficient.
My dryer is also a lot more efficient than theirs, at least if Miele's app is to be believed about how much it uses each cycle.
The author is in the UK, so they probably looked at a product like [1] which is "ideal for rooms 16-26m²" and has cooling power consumption of 1005W.
Residential air conditioning in the UK often involves small, noisy units that are only used for a few weeks at the height of summer. They'll have a thermostat built in, sure, but the user will eagerly turn them off when the room's cool just to get some quiet.
I can't find a github or email for Hannah - if you're reading this i'd like to add Australian energy price data via Open Electricity[0] to the data (reach out via my profile)
> Gas heating (Single room) (1 hour of use) -- 2700Wh.
Wow dude, your room is very-very poorly insulated. Having a 2700W heater turned on constantly just for one room is a lot. If it's 0C outside and 20C inside, a typical room should not lose more than 500W, but better be in 200--400W range .
Kind of a stretch to suggest that an internal combustion vehicle requires 3x more "energy" to move it than an equally physics-burdened (weight, friction, etc) electric vehicle...
This is only "true" if the energy stored in the vehicle's battery got there without any relevant conversion inefficiency; If those joules came from a gas-fired plant, overall efficiency is only about 35-40%: comparable to a typical internal combustion powered-automobile or actually worse than a diesel automobile.
The upstream energy costs of supplying fossil fuels can also be counted. They add about 1/4 or 1/5th of the total in what are called "Well to wheel comparison". The comparison in the article is "tank to wheel'
It's not a direct analogue to energy but CO2 emissions per km are roughly 4 times lower for an EV charged on the EU grid in 2025, well to wheel.
We need an equivalent for water as well! Almost none of the public has any idea about water use across the industrial, agricultural, and residential sectors...
Its always interesting how complex electronics uses little energy. But its the simple stuff like heating water or turning a motors that use magnitudes more.
Physically, it makes sense, but its sort of counter intuitive in terms of utility. You might imagine the tools with the most utility use the most energy.I think thats how many people think of it.
I'm surprised that cooling takes less energy than heating. I imagine that depends a lot on the temperature range; they only need so much to cool a room even on a "hot" day in the UK.
Still... AC still feels like magic. I know how it works and understand the over-unity factor. But it feels like it ought to take enormous energy for it to work at all.
I think specifically it's comparing gas heating vs AC. Heat pump heating would probably do better. In other words, it takes less energy to move heat inside/outside than to "create" it
(With caveats like heat pumps are much less effective in extreme cold)
You can reject 4 watts of heat with 1 watt of electricity using vapor-compression refrigeration.
You can get that up to 7 or 8 watts (or more!) per watt with evaporative cooling towers and vapor-compression combined.
AFAIK you can’t move heat into somewhere using cooling towers, they only increase cooling efficiency.
Heat pump heating is limited to around 4W of heat moved for every 1W of electricity, with the efficiency dropping as delta T drops (aka as it gets colder outside)
You can generate 1 watt of heat with 1 watt of electricity and a resistive heater, they’re more or less 100% efficient.
This is neat. I think I'm actually more interested in avg ChatGPT query than median single query so that I can enter a large query # and be confident in the associated energy cost for that larger number (e.g. what's the energy cost for 1,000 chat gpt queries)
I don’t think the problem
Is how many joules of energy are used. But the cost/burdnen to produce them. The costs and forms for each of these examples is very different making their energy use incomparabale.
In my petrol-powered Prius I average 52 MPG, so at the current price of gas ($2.60/gallon) I pay $0.50 to drive 10 miles - less than the listed cost to drive an electric car the same distance.
This needs a phantom power feature. How much does your phone charger consume when its it charging your phone but still pligged in.
Whats the cost of your TV when its "off?"
Switching mode power supplies use milliwatts when there's no load. Appliances are usually 3W or less when off as they need to keep the IR sensor turned off. The best appliances are <1W.
not sure I understand the petrol car using 3x as much energy as an EV... wouldn't it make sense to convert gasoline to electricity then? I presume that must not be as efficient as other ways to convert fuel to gasoline? (I understand the math is there but... I'm temporarily failing to get it)
I think lists like these might be useful for energy audits and thinking about ways to make better use of energy
70% of the energy in a petrol car is lost as heat. Only around 30% or less of the energy actually propels the car. I imagine that's why there's a big difference.
Gasoline engines are around 30-35% efficient, the rest is lost as heat. That goes for whether they’re spinning an alternator as part of an engine-generator set or just moving a vehicle.
You can’t get more than 35-40% efficiency so converting to electricity is a wash, you lose the energy to heat no matter what.
Also, the chart does not take into account how the electricity for the EV is generated, it would be just as inefficient as the gasoline car if the electricity was generated by burning hydrocarbons, but that detail is left out.
Yeah I'm not sure that EV number maths out. That's 18kw at 60mph for 10 minutes. It sounds really low to me. Going 45mph on a 110lb eMoto with 180lb rider takes ~6kw.
If everyone had their own turbine to generate electricity from gasoline then it would be. But from a central power plant there are transmission losses which cuts out a lot of the potential efficiency increase. The obvious upfront and maintenance cost of everybody owning a personal turbine though makes that method kind of iffy.
Wow, putting everything in the same units is really informative. Running my 450 watt gpu for a day is approximately equivalent to driving a car 10 miles.
I think you simply can't predict any number without really measuring it.
Ryzen + GPU + 1h of gaming, but 3 screens attached.
still a "desktop computer" - stuff in the browser can spike GPU wattage, but I don't think it will be meaningfully different over several hours of true idle..
My problem is more like: I could measure that, but I'd have to recable everything to have the power meter behind the power strip with just PC + screens... which I will probably do at some point, but it involves redoing 50% of the cables in this room.
OP says one query uses 0.3 Wh. Driving an electric car for 10 miles = 3,000 Wh which is roughly 10,000 Wh per hour.
I'm not sure how many queries is equivalent to an hour of Claude code use, but maybe 5 seconds, which means an hour of continuous use = 216 Wh, or ~50x less than an electric car.
A coding agent runs near-constantly, so of course it'd require a lot more compute than running even, say, a multi-minute query with a thinking model every hour. How much exactly is pretty hard to calculate because it requires some guesswork, but...
For a long input of n tokens from a model with N active parameters, the cost should scale as O(N n^2) (this is due to computing attention - for non-massive n, the O(N n) term is bigger, which is why API costs per token are fixed until a certain point and then start to rise). From the estimates from [1], it's around 40Wh for n=100k, N=100B. I multiply by 2.5 to account for Opus probably being ~2.5x larger than gpt-4o, and also multiply by 2 to pessimistically assume we're always close to Opus's soft context limit of 200k (it's possible to get a bigger context for extra cost, but I suspect people compact aggresively to not have to use it). That gets me 7.2J/t, which at a rough throughput estimate of 20t/s gives me power of 144W. Like a powerful CPU or a mediocre GPU, and still orders of magnitude lower than a car.
It is not only about raw power consumption. Comparing driving an electric car with using AI only in kW hides a major point: Hyperscale datacenters are massively centralised, which brings it's own problems; a lot of energy is used for cooling, and water consumptions is enormous. Charging electric cars at home is distributed and does not suffer from the same problems as the centralised hyperscalers do. Also, running AI models at home is not much different than a gaming session :)
This is an incredible sequence of assertions, every single one of which is very incorrect.
"A lot of energy used for cooling": hyperscale data centers use the least cooling per unit of compute capacity, 2-3x less than small data centers and 10-100x less than a home computer.
"Water consumption is enormous": America withdraws roughly 300 billion gallons of fresh water daily, of which IT loads are expected to grow to 35-50 billion gallons annually by 2028. Data center water demands are less than a rounding error.
"distributed and does not suffer from the same problems": technically correct I guess but distributed consumption has its own problems that are arguably more severe than centralized power consumption.
I clicked-through on one of the sources (a blog post on "ohmaticelectrical"), and the website is either hacked or is just really scummy. Here's what I was greeted with:
Thought the same thing. I think I measured my gaming PC to idle at like 70–100 watts. Obviously an average office PC would be lower, but you also need to add on the energy for the monitor, which seems to start at around 15 watts for some basic monitors, and I imagine that's not at high brightness.
Nice! A small note: airfryer tend to cycle on/off much less then slow-air/static ovens so they consume way more per unit time, gas ovens never fully stay "off" when operating, they just reduce the flames.
My electricity provider has a rough breakdown. Which I was kind of incventivized to look at because of money. Heating is the big one. The electricity bills are seasonal.
Next.
I can do my laundry at night because the electricity is cheaper.. oh wait I can’t. That’s apparently unsafe. So I have to do it in the evening. Okay. I’m not going to move my whole small freetime evening around to save a buck on the half-evening long wash cycle. So nevermind that.
The nagging about turning off all the lights were always a consumer blaming ritual that doesn’t matter.
I like the comparison concept. It's like that "order of magnitudes every programmer should know" list, but applied to anyone who cares about energy.
That said, and hot take: people shouldn't worry about energy independent of what they pay for it. The whole point of a price is to fold a complicated manifold of scarcity-allocation into a set of scalars anyone can rank against each other. Appealing to people's sense of justice or duty to get them to use less energy than they'd otherwise be willing to buy is just asking them to lead a less utility-filled life than they can because you think you can allocate scarcity better than the market. I can't, and you can't either. Nobody can.
If you claim that people should listen to moralized pleadings and not the market because prices don't internalize certain externalities, duty is on you to get those externalities accounted so they can properly factor into prices, not apply ad-hoc patches on top of markets by manipulating people's emotions.
As for getting externalities internalized: as a society, we call the procedure for updating rules "politics", and it's as open to you as to anyone else. If you propose policy X and you can't get X enacted, perhaps it's because X is a bad idea, not because the system is broken.
Not everyone anyone claims is an externality is, in fact, a cost we must account. We should have a prior that costs are accounted and need evidence to rebut it --- and any such rebuttal must involve numbers, not emotional appeals. What specific costs are unaccounted? How large are these costs? Through what specific mechanism are they escaping existing accounting mechanisms? "I feel like we're using too many electrons for X" is not a valid argument for the existence of an unaccounted externality.
That is, unless there's some specific reason to believe otherwise, we should believe market get it right, especially with fungible commodities like kWh.
How do you propose to convince people to get those externalities accounted without emotions? How do you convince people of the value of externalities that are far away in place or time (but not less real)?
The idea that you would be worried about how many electrons you use and it's relationship to climate change is on its face kind of ridiculous.
It's like worrying about how many times you personally ordered Chinese food affects the price of Diesel fuel in India. It's an absurd leap of logic, and the parent is right to call out these arguments which are almost always emotional.
> As for getting externalities internalized: as a society, we call the procedure for updating rules "politics", and it's as open to you as to anyone else.
Ok so I do need to worry about energy so that I can identify these unaddressed externalities and work towards updating the rules. You can to care before you can get involved in this stuff. You can't tell me not to worry about it and then also say that it's basically my fault for not getting involved if the price is wrong.
> any such rebuttal must involve numbers, not emotional appeals
Who are you arguing with? You're commenting about a website that has strictly numbers and nothing else.
Your dismissal of moral concerns is not convincing.
Imagine a world where the only energy you do is use was generated by a stationary bike you had to ride yourself. You would, generally speaking, use that energy differently than energy you would pay for--you would generally reserve your effort for worthwhile things, and would be averse to farming energy yourself just to power frivolity or vice. How you determine what to put your energy into would explicitly be a moral question.
Instead in our world we an abstractions conceals the source of the energy. But if the moral concerns from the first world had any weight, they haven't lost it now; if energy is anything short of completely free we should by the same logic be averse to expending energy on worthless work or vice. The human being is not a utility monster, but something very different, and moral questions of this sort are central to how it navigates the world, they should not be dismissed.
Doesn't this argument hinge on equivocating between two different definitions of aversion, though? I'm averse to bananas, but that doesn't mean I think it's immoral to eat them. The moral dimension kicks in if somebody else had to ride that stationary bike for you, because then you'd be wasting their time on frivolities.
Of course I'd use energy differently if it cost more. If I had to generate energy by pedaling a bike, I'd consider it costly indeed. So what? Energy doesn't cost as much as it would if I had to manually generate it, and who are you to say allocation decisions made under that regiment are good and ones made under ours are bad?
Wouldn't your argument also compel us to use steel as if it were gold? Salt as if it were saffron?
Aa Digital ID is getting forced on the public, it is dangerous, almost traitorous to demand/suggest ordinary citizens need to care about energy efficiency to the point of giving up their freedoms/conveniences!
Focus your thinking on solving/promoting energy production instead!
What’s startling to me is how many comments in this thread just take the provided values as gospel without asking questions that methodology answers either in the abstract or barely describes. Also going giving a cost for “United States” is absolutely nonsense - electricity, gas and gasoline prices vary widely across the country. There is no one cost for each, and the average is worthless for this kind of thing (especially since the average of each - gas vs. electric vs. gasoline cost - are independent variables that have no relation to each other on a region by region basis).
My first question was: "Is this whitewashing LLM energy usage?"
And yes, that seems to be the undercurrent here. Complete with linking to themselves to validate the data they used to make their estimates.
Either these companies need to build these massive data centers that consume massive amounts of electricity OR these LLMs don't use a lot of electricity.
You don't get both. If LLMs don't require a lot of electricity, then why are we building so much more capacity? If all of that capacity is required, then what is the real cost of sending a query to these LLMs?
Hannah Ritchie is a quite well reputed writer and data scientist squarely in the climate field. She's written two books on climate and I found the one I read (Not The End of the World) was quite good and data-driven.
I'm not sure I understand where the issue is here - something can use a small amount of energy per use but a large amount in aggregate because of lots of use.
You have set up a false conflict. The data centers are "huge" and they also consume about the same power as 1 airplane. These things are both true.
It is also not really true that they are huge, it is a misconception driven by biased reporting about facilities that really aren't very remarkable compared to material distribution warehouses, beverage bottling plants, and suchlike.
> You don't get both. If LLMs don't require a lot of electricity, then why are we building so much more capacity?
A small number times a large number is often a large number. Have you heard of the concept called "per capita"? In any case, electricity is going towards data centers in proportion to the degree to which these data centers do useful work. AI companies buy the electricity fairly on an open market, sometimes even subsidizing this market by funding new generating capacity.
If all these people and companies are making electricity allocation decisions that make sense to them with their own money, who are you to stop in and say that their voluntary transactions are incorrect? Who died and made you the king?
Useful work is debatable here, a lot of people just talk to the thing or use it instead of searching the internet.
The owners surely think, or at least want us to think that it is very useful indeed, otherwise we'd see no point in burning through piles of investors cash to buy overpriced ram, storage, gpus, cpus, nics, secure the power to run it and then subsidise the users to use it.
I do think that transaction is wrong and it's going to bite them in the ass in the long term, but I don't have the money to outbid them for the power. I do get to see them crash and burn when the investors get impatient.
They’re not even saying they shouldn’t do it or that they’re not useful or not worth it but you Cannot logically say both “these things do not use a lot of power” and “we need to build more power plants to handle these things”
It isn't all new capacity. The popular discourse hardly ever mentions it but AI is a small fraction of why we need new datacenters and the bulk of the demand is driven by general IT needs, particularly consolidation of small, grossly wasteful corporate data racks into vastly more efficient cloud services.
Edited to answer:
The question has also been addressed by the same author as the article: USA spent a quarter century not building generators and that negligence has finally caught up to us, despite objectively heroic efficiency efforts on the part of the IT sector.
LLMs don't use a lot of electricity per user. Why should the fact that the energy usage happens in data centers instead of each user's house be an important moral factor?
Indeed, looking at a "single median query" totally disregard the fact that:
- first, those queries are mostly useless and we could totally do without them, so it's still a net pollution
- they are being integrated everywhere, so soon enough, just browsing the web for a few hours is going to general 100k+ such equivalent "small queries" (in the background, by the processes analyzing what the user is doing, or summarizing the page, etc). At that time, the added pollution is no longer negligible. And most of this will be done just to sell more ads
1. Your prediction is that soon browing the web for a few hours is going use >30,000+ Wh (based on the "equivalent" you mentioned)? (For comparison to that 30,000+ figure, the energy use from using a laptop for a few hours is 75 Wh, all per OP source.)
2. > most of this will be done just to sell more ads
Are you predicting that the value of ads is going to increase by a number of magnitudes? Because 30,000+ Wh of electricity has a quite significant cost, and even a video ad currently only earns pennies, so I'm trying to imagine how the math would work in this scenario?
I attached a generator with some supercaps and an inverter to a stationary bicycle a few years ago, and even though I mostly use it as a way to feel less guilty watching Youtube videos, it does give me a quite literal feel for some of the items on the lower end of the scale.
- Anything even even halfway approaching a toaster or something with a heater in it is essentially impossible (yes, I know about that one video).
- A vacuum cleaner can be run for about 30 seconds every couple minutes.
- LED lights are really good, you can charge up the caps for a minute and then get some minutes of light without pedaling.
- Maybe I could keep pace with a fridge, but not for a whole day.
- I can do a 3D printer with the heated bed turned off, but you have to keep pedaling for the entire print duration, so you probably wouldn't want to do a 4 hour print. I have a benchy made on 100% human power.
- A laptop and a medium sized floor fan is what I typically run most days.
- A modern laptop alone, with the battery removed and playing a video is "too easy", as is a few LED bulbs or a CFL. An incandescent isn't difficult but why would you?
- A cellphone you could probably run in your sleep
Also gives a good perspective on how much better power plants are at this than me. All I've made in 4 years could be made by my local one in about 10 seconds, and cost a few dollars.
Where I am at least, people using less power because power because power need to profit more, is wild.
They literally had record profits the last few years, rather than being forced to lay down solar. I think power should be a global endeavor, not some local for profit business with complete regulatory capture that makes competition illegal.
Yes I'm angry, because I pay more in electric than most anywhere in the world. If I charge my care with LEVEL 2 using city provided charges, during the day, it's more expensive than gas.
Energy security is national security.
Cheap electricity means you can do things that made "no sense" with expensive electricity. (e.g. smelt aluminum)
Cheap electricity means you can underbid regions that have expensive electricity...
As Technology Connections said, "Panels that cover your electrical needs for the next 25+ years? In the Midwest, we call that a good deal!"
I love Technology Connections, but he has no idea what discounting is in economics. Or at least he writes his videos as if he doesn't.
What discount rate are you using?
Solar has one of the lowest capital costs [1] so the discounting works in it's favor. And then the non-discountable operating costs also works in its favor since the fuel supply (light) is free.
[1]: https://en.wikipedia.org/wiki/Cost_of_electricity_by_source#...
Yup. It's why even in fairly red states like my own (Idaho) solar, wind, and battery are going up everywhere. Even without significant subsidies the economics are really good for renewables.
They'd be even better if we didn't have extreme tariffs on China.
That's actually what's convinced me that renewables are a better choice than nuclear. I still like nuclear, but renewables are just so much easier and faster to deploy while being a lot cheaper. To make nuclear competitive requires regulatory changes along with a government that's simply willing to tell it's NIMBY citizens YIMBY.
Government literally has to get in the way of renewable deployments at this point to stop them.
[flagged]
Where does that 0.3 TW figure come from? That seems awfully high.
Oops it’s actually 0.6 TW.
https://ourworldindata.org/energy-production-consumption
Convert TWh to TW for a year average.
Huh! I'm kind of stunned that we only use ~30x the power that we did back then. If I'd been asked to guess I would have added another 0 or even two of them.
Yeah we had an exponential jump when we discovered oil but we maxed that out and the growth has been linear since (and paying for it environmentally too).
I’m waiting for the next big major discovery in energy generation.
We’re always on the verge of fusion… fusion will be like the discovery of oil. Humanity will jump forward… well, technologically at least.
You're trying to converse with a LLM. It's made up.
Nope it’s 100% legit and I even remembered it wrong. It’s actually 0.6 TW.
You can get the number of TW from this report https://ourworldindata.org/energy-production-consumption
Convert TWh to TW for a year average.
jfc. What is the point? What do people get out of doing that?
I don't know, but HN in particular has an AI-sycophancy problem where I see this most common versus other link aggregator sites.
is that a sustained 20TW? Absolutely crazy that we're generating 60kwh per person daily. Where does it all go?
Lots of it is lost to heat with legacy fossil generation.
You have pretty much the same heat losses with nuclear, or anything else where you heat water to turn a turbine.
Nuclear is low carbon, it’s fine we lose heat to extract that energy versus stationary and mobile combustion generation, as there is no other effective way to extract that energy at this time.
Quantification of global waste heat and its environmental effects - https://www.sciencedirect.com/science/article/abs/pii/S03062... - Applied Energy Volume 235, 1 February 2019, Pages 1314-1334
* 49.3–51.5% of global energy use would end up as waste heat in 2030.
* Transport sector accounts for the largest (43%) recoverable waste heat in 2030.
To note, we are almost at installing 1TW of solar PV every year globally.
Most of those technologies also need uninterrupted power supplies. Something wind, solar and batteries for the next 50 years aren't.
Ember Energy: Solar electricity every hour of every day is here and it changes everything - https://ember-energy.org/latest-insights/solar-electricity-e... - June 21st, 2025
> Batteries are now cheap enough to unleash solar’s full potential, getting as close as 97% of the way to delivering constant electricity supply 24 hours across 365 days cost-effectively in the sunniest places.
What does this mean? It means we are most of the way there with solar and batteries alone, even if we need a bit of carbon based generation to bridge the gap while solar and battery deployments scale globally. Solar and batteries will only continue to get less expensive and better.
Our World In Data: Installed solar energy capacity - https://ourworldindata.org/grapher/installed-solar-pv-capaci...
Solar PV go brrr.
> Something wind, solar and batteries for the next 50 years aren't.
False. If you'd stopped before the "and" you would have been correct, though.
Batteries are really cheap now, and supply of batteries is growing basically as fast as people can get the investments and permissions for the inputs and the factories.
Pumped hydro is one solution. You bank the excess wind/solar using gravitational potential energy and then draw on that whenever you need to.
https://en.wikipedia.org/wiki/Dinorwig_Power_Station
Yes, we just need to build the mountains first.
Yeah, or water towers. No need to play god here.
Pumped hydro energy storage relies on the cheapness of water and existing geology. If you have to build the chambers instead of damming a river it's too expensive. Most of the good spots to have a reservoir are already used. If you have to manufacture the bulk media instead of just using water it's too expensive.
Pumped hydro doesn't need a river, "just"* rock which isn't water porous and some nearby body of water (lake, sea, whatever).
The economics works out even if you were lifting concrete blocks rather than water, hence why you get pictures like this: https://en.wikipedia.org/wiki/File:Energy_Vault_Test_Tower_2...
The argument against lifting concrete is that you can dig a hole in the ground an pump water in/out of it for more reliability and lower cost than having a crane lift and lower concrete, and it's easy to make it much bigger both horizontally and vertically, so why bother.
But it does appear to be economical even with that, and water is cheaper.
We make lots of holes in the ground on a regular basis, including for extracting fossil fuels. Here's two, note scale bar, though I have no idea what the rock around it is like regarding water losses: https://www.google.com/maps/@50.9063171,6.4418046,17655m/dat...
* it's never "just" with things on this scale
There are exactly zero economically viable pumped water storage systems where water towers are involved. If you do the math for the amount of a mass of water, you'll see why! It's not feasible.
Indeed, you can get a sense of the scale on the Dinorwig wikipedia page and the pages it links to.
It has a storage capacity of about 9.1GWh.
The upper reservoir (https://en.wikipedia.org/wiki/Marchlyn_Mawr) holds 9.2 million cubic meters of water.
So 1 million cubic meters of water provides ~1GWh.
We can see how that compares in terms of raw GPE (Gravitational Potential Energy):
1 million cubic meters of water = 1E6 * 1E3 kg = 1E9 kg
There's roughly a 500m vertical drop between the upper and lower takes at Dinorwig so:
1E9 kg * 500 m * 9.8 m/(s^2) = 4.9E12 J =~ 1.36GWh
As for water towers, if you look at something like the Roihuvuori tower in Helsinki (from https://en.wikipedia.org/wiki/Water_tower) which is one of the largest:
Height: 52m, Capacity: 12000 cubic meters
If we are generous and say that all of the water is stored at the maximum height then:
12000 * 1E3 kg * 52m * 9.8 m/(s^2) =~ 1.7MWh
You'd need over 5000 of them to match what Dinorwig can provide.
Have you heard of batteries?
> They literally had record profits the last few years, rather than being forced to lay down solar. I think power should be a global endeavor, not some local for profit business with complete regulatory capture that makes competition illegal.
Sounds more like you guys should be lowering barriers to entry, not setting up a global non-profit cartel.
True. I suggested global because it allows for scale with copy/paste designs, where things like nuclear could actually become viable.
Where I am, we have a solidly aligned state government. There's no concept of consequences for anyone in power. They're paid by the local companies to pass laws to make competition legal. Some are investors. All corrupt. That's what you get with a solid political alignment.
Vote with your feet (and wallet), and support anything that makes people voting with their feet easier.
It's the McDonald's theory of policy: you don't vote on their burgers at the ballot box, you just go to Burger King or get a doner kebab, if you don't like it.
Australia I assume?
Once I did a little bike training and looking at my power curve, I was incredibly impressed by how cheap energy is. 100W is an all day number, 200W less so, 300W is exactly 20 minutes when I do an FTP test. 400W is 4x Tour de France winner Tadej Pogačar for an hour and he's a mutant. 1 horsepower is under a minute iirc, definitely under 2. 1kW is maybe 10 seconds. So I could keep my laptop and phone charged probably indefinitely as long as I have food, but not a ton more than that.
https://velo.outsideonline.com/road/road-racing/tour-de-fran...
Did you try charging an e-bike with your contraption?
I don't know what you can take of this, maybe you can see it as advance pedaling, or to get a feel for energy conversion losses. Anyways, it is the kind of harmlessly stupid idea that I would want to try just because I could.
What a ridiculous idea, I love it.
Amazing stuff, have you written up a blog post? I could see a video being a fun format for this as well. Might help people develop the intuition for watts/power consumption in a different way
Kind of, it's in bits and pieces here:
https://hackaday.io/project/191731-practical-power-cycling
and is also a few years out of date.
I did do a video back then going against the infamous "bicycle toaster challenge" video (in which I determined it was probably less real than they made it out to be). I'm nowhere as fit as those guys, so in my attempt I was only able to turn a bagel into a dry crouton over the course of an hour.
https://www.youtube.com/watch?v=AcNXp86BJ-o
Any sense what the efficiency ratio was for your setup?
I'm as curious as you to be honest - putting a strain gauge on the pedals for measuring mechanical power has been on my list for quite a while. My own (probably inaccurate) measurements right after the generator says I can get 60-70Wh in an hour, but I can get to 100Wh if I try harder. I have reason to believe my setup underestimates power because my ammeter clamps at 5A and I know I can peak over that on the down stroke of the pedal.
I've seen numbers like 250W mechanical power for an average trained cyclist, so either my setup is rather inefficient, my measurements are off, or I'm going to find out that I'm nowhere near as strong as a real cyclist.
On the other hand, the stationary bike I got originally had a rubber belt, which it would chew excessively and I eventually swapped it for a chain because it kept slipping in spite of tensioning it more, suggesting I'm hitting the thing harder than it was originally designed for (how that translates into power I'm not sure).
> I've seen numbers like 250W mechanical power for an average trained cyclist, so either my setup is rather inefficient, my measurements are off, or I'm going to find out that I'm nowhere near as strong as a real cyclist.
Cyclists' power output is sometimes reported as a 'power curve' - a chart with power on the vertical axis, and duration-of-that-power on the horizontal axis.
For example, a cyclist might be be able to produce 500W for 15 seconds; 350W for 1 minute; 270W for 10 minutes; 200W for 1 hour; and 150W for 5 hours.
Oh don’t sell yourself short. It can certainly be both! (:
Thanks for sharing the details.
All in good fun of course, it has to be healthier than watching Youtube just sitting around normally.
The author Hannah Ritchie works on Our World In Data and also publishes the fantastic Sustainability by Numbers substack. It's in the same vein as the late, great David MacKay's Sustainable Energy Without the Hot Air.
This tool has its own recent substack post. See the comments too, especially the one by Chris Preist that contextualizes the energy usage of streaming video (a topic that has also been discussed on HN before).
https://hannahritchie.substack.com/p/does-that-use-a-lot-of
And wrote a great book: Not the End of the World
Yep. It's a very good book and well worth a read.
It's interesting to see how upset people are on Goodreads about that book:
https://www.goodreads.com/book/show/145624737-not-the-end-of...
The top reviews are mostly people angry with Ritchie for not being a catastrophist.
Who pays for their research?
She's employed by Our World In Data. She is also a published author of printed books. Her substack does not have paid subscriptions enabled (or at least it did not as of last summer; see this post [1]). Our World In Data is funded by donations:
https://ourworldindata.org/funding
[1] https://hannahritchie.substack.com/p/reflections-on-substack
What narratives and framings does a blog post or “visualization tool” serve? What does their overall work? What’s their recurring ideological slant?
Could be wholesome and altruistic. Or it could be something else.
Someone can be an honest ideologue (useful idiot) without being directly funded by someone shady.
I think stuff like this really crystalises how people misunderstand how much energy stuff uses.
My parents for example sweat the small stuff and go around the house turning LED driven lights off to "save electricity" even though it would barely make a dent in their bill.
Granted, they come from a time of incadescants burning 60-100w at a time so I can see why that habit might be deeply ingrained.
The ridiculously dramatic drop in power we dedicate to lighting is one that is just tough for folks to internalize. As you said, used to, you could have ~10 lights in your house that would add to upwards of 1kw. Nowadays, you can have 50 lights and barely hit 500w. Just mind blowing how far we dropped energy on those.
Same goes for televisions. Your modern TV is probably closer to the basic light bulbs before LEDs.
I'm assuming the general trend is true for all things solid state. That said, lighting is by far the biggest drop for most houses. Remarkably so.
> I'm assuming the general trend is true for all things solid state. That said, lighting is by far the biggest drop for most houses. Remarkably so.
For commercial and industrial installations, VFDs have probably been the biggest efficiency gain, even moreso than lighting. Half of all electricity consumed is used by motors. Thank goodness for solid state power electronics!
I turn LED lights off because of the difference in operational life, and I don't like changing bulbs. M GE bulbs say they have a rated lifetime of 13 years......at 3 hours of usage per day. So if they don't get turned off, then that 3 hours can very easily become 12, and now you are at a rated lifetime of ~4 years instead.
A ‘standard’ (A19 shape, E26 base) 8W 800 lumen LED lamp costs around $5 and will use about $20 of energy over a 15,000 hour lifespan, assuming $0.15/kWh.
That works out to around $0.035 per day for the lifespan of the lamp if you run it constantly for 24 hours a day, I wouldn’t waste time thinking about it. It’s an extra $10 over 12 years, you’re still using the energy.
Investing in occupancy or vacancy sensor wall switches at $25 a piece would be the best option, then you don’t need to remember to turn the lights off!
It's not a cost thing, I just don't like changing bulbs. I find it annoying, and with enough bulbs, when the lifetime is down to 4 years, you are doing one every few months on average.
But yes, I have thought about presence sensors. I'd really only need 3-4 to cover the primary areas where lights get turned on and not off (if I don't do it). I just haven't gotten around to it
But now you’re using up switch cycles!!!
Yup. In my experience, average non-nerd folk very very little feel for this stuff. I suspect some believe energy consumption of phone vs car is basically a toss up.
It was genuinely a surprise to see how much relative energy petrol cars use (and shame on me - I'm an electrical engineer). I mean I think I knew it intuitively, but this simple chart blew my mind.
When one gets in the weeds on EVs or ICE cars two things become shockingly clear: internal combustion is hilariously inefficient YET gasoline is hilariously energy dense. Most people's intuition is wrong on both of these points but then they cancel each other out.
Edit: another important point is that the "cost" to acquire gasoline is only the very end of the process. The energy has already been gathered, stored, and most of the processing is complete. Our cost (in money and energy) to "make" gasoline is really just gathering it. This is why the comparison to renewables is often a hard sell, it's just apples to oranges. Gasoline started on third base, renewables are batting from the plate. Some of the internal combustion enthusiasts are holding up e-fuels or synthetic fuels as the solution but then we have to pay for the entire energy gathering and processing pipeline and still be using a conversion method that's not at all efficient. It's the worst of both worlds.
> internal combustion is hilariously inefficient
It's inefficient but not hilariously so. Modern ICE are quite amazing technology.
Combined gas turbines (you know, the energy source that powers your electric car) are about 60% efficient for the really good ones, minus 5-7% transmission losses, minus 10-12% charging losses, minus 20% loss in cold climates, lands you at around 35-40% efficiency from fuel source to the wheel.
The Atkinson-cycle engine in the Toyota Prius gets around 40% give or take some losses in the drivetrain. Electric have plenty of upsides, but for some people with cheap gas+high electric costs+cold climate you would honestly be better driving a hybrid.
> Combined gas turbines (you know, the energy source that powers your electric car)
Not everywhere. My car charges off an average of 80% renewables (mostly hydro and geothermal), right now it's 95%.
But it is definitely something you need to take in to account when purchasing, an EV isn't right for everyone.
This is something that always gets lost in these conversations.
Whenever you do the real world calculation for what an electric cars CO2 profile looks like it turns out to be the same as a gasoline car unless your country is majority nuclear.
> Whenever you do the real world calculation for what an electric cars CO2 profile looks like it turns out to be the same as a gasoline car unless your country is majority nuclear.
Not at all true: https://www.carboncounter.com/
US-specific but you can even pick a state and it will use the generation mix of that state
You charge your car at night.
At night the sun doesn't shine.
The mix is mostly coal or if you're lucky mostly gas.
This is the type of bullshit I mean by doing real world calculations.
Maybe today, but basically everywhere south of Canada solar is so high ROI that it's just a question of time before sunlight hours have electricity so much cheaper that the primary daytime parking locations become the favored (slow-)charging spots.
For you to prefer charging there your employer only has to charge you less during the day than your utility charges during the night, so the day/night rate arbitrage can easily pay for the metering hardware and installation (at the next opportunity to install without having to dig the parking lot up just for the chargers), with the rest being profit to incentivize the managers to install/offer this.
The wind blows, water flows, nuclear glows. Further, demand's at lows.
In my state (idaho) at night the power mixture is primarily renewable/clean because of this.
I charge during the day, from my rooftop PV panels. Over a year we are net negative on grid consumption.
This means you don’t take your car to work, which isn’t typical
I produced over 400kWh last week, in February.
This is wrong, and by a long way.
Or majority renewable.
The tyranny if the rocket/horse equation: You need energy to carry the energy you need to move.
There's a good reason so many sprawling civilizations of the past involve leveraging wind-power for transport.
Every single ICE car driving down the highway is throwing away enough waste heat to heat a small apartment building on a freezing cold day.
IF the joules of energy in your EV battery came from gas-fired or coal-powered generation, a similar amount (~60%) was simply dumped somewhere else.
I wish we did more e.g. district heating with that waste heat in the US.
That means relatively dirty combustion near where people live. The population density around fossil fuel power plants tends to be pretty low in wealthy countries.
You can't pump hot water the same distance you can transmit electricity on HVDC towers.
Exactly. The environmental/social burden isn’t just the energy used in the raw physical form, but the cost to acquire and make it useable.
The problem with gas is not that burning it doesn’t maximally capture all energy, but that there are externalities to doing so.
Train locomotives have used diesel powered generators that then powers electric motors. Would this be less efficient than battery powered EVs? Or better asked, what would be the most efficient use of gasoline?
> Would this be less efficient than battery powered EVs?
Measured in terms of mass * distance, trains with steel wheels will beat anything with rubber pneumatic tires.
Part of the magic of hybrid trains is that you can have multiple generation units that can be turned on or off as needed.
---
Efficiency is just one consideration for a power plant.
Historically, reliability has been more important than efficiency, especially for industrial applications like locomotives. In other words, locomotives are probably not as efficient as they could be. For instance, you could use a lower viscosity engine oil for lubrication, but that would reduce reliability as engines fail due to friction.
Nissan makes a range of these under the e-power branding:
https://www.nissan-global.com/EN/INNOVATION/TECHNOLOGY/ARCHI...
In the auto industry these are usually called a series hybrid and there have been a handful. The Chevy Volt (though it had the ability to directly connect the engine to the wheels at highway speeds), and the BMW i3 and i8, the Fisker Karma/Karma Revero. The new Ram Ramcharger truck and the second gen Ford Lightning will also be series hybrids.
It's a really good drivetrain that was unfortunately made untenable for a long time by a combination of regulation and market forces.
When it comes to the environment the most efficient use is to leave it in the ground.
Hybrids work for trains because they are so large and don't need big swings of acceleration or to climb steep grades. They can run the diesel generators at maximum efficiency.
Battery power would be better, because you can build even larger power plants running at higher heats and not have to haul them with you, but the costs of sufficient battery is too large, so far. That is changing.
Isn't it better for trains to just to draw from the electric grid?
Do you have to run new electric transmission lines? Will you have to maintain those power lines?
Possibly, yes. But that seems to be worth it:
https://en.wikipedia.org/wiki/Railway_electrification
"Maintenance costs of the lines may be increased by electrification, but many systems claim lower costs due to reduced wear-and-tear on the track from lighter rolling stock. There are some additional maintenance costs associated with the electrical equipment around the track, such as power sub-stations and the catenary wire itself, but, if there is sufficient traffic, the reduced track and especially the lower engine maintenance and running costs exceed the costs of this maintenance significantly."
We once dammed basically every river in the nation because it was in vogue at the time.
Maybe building overhead power lines for rail infrastructure should be the "hip" thing right now instead of AI. Maybe building oodles of solar power farms and batteries should be "hip"
We built electrical infrastructure to the most remote residences just because we could and because it was an investment in our people. We directly funded our massive and formerly world class rail network because we could, and because it would pay off. We built a world class road network half as a make-work project, and it still pays dividends. We purchased Alaska, with no obvious reason. We built a space program to have slightly better nuclear weapons, and it's part of the reason we were so dominant in computer chips for so long.
We have spent something like 40 trillion dollars over the past 25 years, and almost none of it on anything of real value. More than a little of that debt is just handouts to already rich people.
We can build new electric transmission lines and I'm so tired of things that we absolutely 100% can do if we just demand it be done being somehow treated as a problem. America can afford infrastructure.
Someone could graph the cost/benefit ratio on putting the batteries on the trains vs putting wires up everywhere.
That's a complicated question that unfortunately has quite a bit of "well it depends" in the answer. I worked in the auto industry for a long time - both doing engine development and EVs - so my opinions here are well-informed but not world expert.
From a pure energy efficiency perspective you can't beat economies of scale. A stationary power plant (even ones that are just big gasoline engines) run at a constant load and RPM so they can be optimized for pure efficiency, they rarely have to start, warm up, and shut down, and they can use larger and more expensive exhaust aftertreatment systems. Most energy conversions grow more efficient with scale and this is no different. The locomotive powertrain works for a handful of reasons but one of them is you can build much more efficient engines that are optimized for a single constant speed and load. But most of the advancements in internal combustion engines over the last 20-30 years don't increase peak efficiency but increase the conditions in which they're efficient. Variable valve timing and lift are probably the most underrated and overpowered technologies that have transformed engines from having one narrow regime of high efficiency to running well over a huge range of the map. But turbocharging, variable intake geometries, 7+ speed transmissions, and mild hybrid systems like belt-starter-generators get honorable mentions here. However we're not talking about anything close to EV-levels of efficiency. I think the cutting edge research engines are running in the mid to high 40s for thermal efficiency (percentage of fuel energy captured as useful work), most passenger car engines probably peak in the mid 30s.
So while there is some efficiency to be gained by a more locomotive-style system it's not as much as you would hope. In the industry that's called a series hybrid system, vs a parallel hybrid system where either ICE or EV power can go to the wheels. The benefits of a series system are more emissions and product features. You can get the full torque and power of an EV, you can start and stop the IC engine in a more emissions optimized way, and and you can filter load spikes to use a small engine that meets average not peak load.
From a more pragmatic perspective, with the energy density of gasoline and other liquid fuels it's probably best to use it in applications for which you just can't use full electrification. Planes are currently the best example of this. It's also worth noting that passenger cars benefit massively from strong hybridization because of the uneven load cycles so that's a technology where you can deploy a gasoline engine but then claw back a lot of the efficiency losses with hybrids. That's not always true, for example boats don't really have a regen cycle so hybridization just doesn't get much.
But that's not an apple-to-apple comparison.
Like, if you "save energy" by not driving a petrol car, you can't "use the same energy" on electric car, or lighting.. not even prower a generator.
They are not interchangeable.. But this chart encourage us to think them as the same.
In Japan, my country, this looks a bit different. A lot of electricity still comes from oil- and gas-fired plants. The mechanics differ (gas turbines vs. car engines), but in both cases we’re still relying on combustion. I suppose some countries have the same issue.
Yes, similarly in the US: I think the largest portion of the energy in the US is produced with gas fired power plants.
It gets across just how ridiculously energy-dense liquid petroleum fuels are.
And then look at how little distance you need to travel to produce 1kg of CO2.
The presentation is nice, but some of the conversions are questionable.
For instance: The cost section, wherein 1kWh in the US is figured as having a cost of 9.7 cents.
In reality, it's not that way at all. Unless we're fortunate enough to live in an area where we can walk over to the neighborhood generating station and carry home buckets of freshly-baked electricity to use at home, then we must also pay for delivery.
On average, in 2025, electricity was 17.3 cents per kiloWatt-hour -- delivered -- for residential customers in the US.
https://www.eia.gov/electricity/monthly/epm_table_grapher.ph...
I looked at the electric car example for the United States. It has 3 kilowatt hours priced at $0.51, 17 cents per kilowatt hour, which seems about right. The "petrol car" example at the top of the chart isn't powered by electricity so its cost number is not directly comparable to the things that consume electricity.
On the energy tab: It says that driving a petrol car 10 miles uses 10,000 Watt-hours, eg 10 kWh.
On the costs tab, for the United States: It says that this has a cost of $0.97.
97 cents ÷ 10kWh = 9.7 cents per kWh
(I didn't look further than that. Perhaps I should have.)
---
edit: I now see a note at the very bottom stating that it is using an assumed "$0.17 for electricity".
$0.17 per kWh is plenty close enough for rough figurin', so I'd like to take this opportunity to retract my previous complaint.
Yes, for the petrol car it's showing primary energy from petrol (gasoline).
https://en.wikipedia.org/wiki/Gasoline_gallon_equivalent#Gas...
Assuming 33.41 kWh/gallon it takes about 0.3 gallons to get 10 kWh, which costs $0.97 at a pump price of $3.23 per gallon.
The electric shower also seemed pretty optimistic. I live in an area with about 50°F/10°C ground temperature and my 14.4 kW water heater can just keep a relatively efficient shower head flowing at a comfortable temperature.
I had this problem once with a water heater: Get in shower, and things are nice and hot. But the temperature decreased rapidly, and immediately.
It turned out that it had been plumbed backwards.
Heat pump or resistive?
This one is resistive (tiny and cheap to purchase) but will be just an emergency-backup shower once my home renovations are done.
The house is getting a split-system air-to-water heat pump with an indirect tank for domestic hot water, so it should cut that down substantially (the unit maxes out at around 3kW input but likely will run longer to recover/preheat).
Interestingly, the latest generation of LED lights are so efficient that it makes little financial sense to bother having a light switch. 330 lumens per watt from recent Phillips bulbs!
The cost of having an electrician wire that switch is probably more than a little 2.5 watt light will use in it's lifespan - particularly when you account for the fact lights in hallways are probably in use most of the time anyway.
Add in the effort of switching the light switch a few times a day for many years and it's certainly the case! Or the risk of fumbling in the dark for a light switch at the far end of a room or in a house you aren't familiar with.
Obviously you might still want to turn it off for maintenance, but you have the breaker for that.
What if you want to turn it off to have darkness?
Classic example of “we figured out we could, didn’t consider if we should”
engineer vs end user
Do you have a link to the 300 lumens per watt lighting? I'm curious.
I think that mostly only applies bulbs without inverters.
Philips UltraEfficient LED | Energy-saving lighting solution | Philips lighting https://share.google/lptcUJwHZWOQP1JMJ
They are direct AC bulbs, no inverter needed. They do have smoothing capacitors so they don't flicker, and I think a slight redesign of the drive circuitry could add ~10% more efficiency...
Which of these bulbs? The highest efficiency I'm seeing poking around is 230L/W (Don't get me wrong, still very impressive)
I see it has a ChatGPT median query, but for those of us using coding agents this isn't so relevant.
Here's a post that makes an estimate:
https://www.simonpcouch.com/blog/2026-01-20-cc-impact/
> So, if I wanted to analogize the energy usage of my use of coding agents, it’s something like running the dishwasher an extra time each day, keeping an extra refrigerator, or skipping one drive to the grocery store in favor of biking there. To me, this is very different than, in Benjamin Todd’s words, “a terrible reason to avoid” this level of AI use. These are the sorts of things that would make me think twice.
I end up shrugging. For a Claude Code power user, today, a day's use uses less electricity than a morning commute in an electric car. To say nothing of the costs to keep your workstation running, your building heated or cooled, etc. Not quite a rounding error, but a relatively minor component of overall usage.
At least for programming usage the power usage seems worth it. For starting up 1 million bots to argue with each other on facebook it's obviously a total waste.
At any rate, the power usage will become more apparent when these products stop being subsidised, where power usage is being charged to the end user.
Thanks for sharing this.
I also was under the impression that queries cost were mostly meaningless, but it seemed only is true for fresh sessions and short queries. I have to say, the result is less dramatic than I expected but still more significant for heavy users (such as myself).
I'm not sure it's even a particularly relevant comparison to an hour of use of various other electronic devices. I'm sure the median user is running a lot fewer queries than a Claude Code power-user, but I would guess it's still more than one in a typical session.
I wanted to write an app like this. Here's what I think would be useful for some people to compare: more checkboxes that represent typical daily activities like "how much energy does driving to the grocery store (x miles away) use" or "how much energy does idling in my car for 30 mins use", "how much energy does riding the bus/trolly/etc use", "how much energy does keeping my place warm vs kind of chilly", and also "how much energy does a cargo ship use to cross the pacific" or "how much energy does it take to source, manufacture, package, transport, and stock a pack of bubblegum". I think that would help people realize the largest components of their footprint and also how it relates globally.
One thing missing but important to understand is the energy embodied in buying 'stuff'. At a very rough approximation, the cost of stuff, especially consumer goods manufactured cheaply, is quite a high percentage energy.
When you look at people's energy usage, quite a lot of it ends up being the embodied energy in the stuff they buy. For quite a lot of people, it's probably the largest category of energy consumption. I once had a very rough go at calculating this here: https://www.robinlinacre.com/energy_usage/
One may look at aluminum as a solid form of energy. In fact, https://en.wikipedia.org/wiki/Aluminium%E2%80%93air_battery
One gram of finished 3nm packaged semiconductor is roughly equivalent to half a kilogram of refined aluminum in terms of energy cost. If you want to spend a lot of energy for not much mass, photolithography is fantastic.
Doesn't show the comparative energy waste of bitcoin?
This source[0] says
> One Bitcoin now requires 854,400 kilowatt-hours of electricity to produce. For comparison, the average U.S. home consumes about 10,500 kWh per year, according to the U.S. Energy Information Administration, April 2025, meaning that mining a single Bitcoin in 2026 uses as much electricity as 81.37 years of residential energy use.
[0] https://www.compareforexbrokers.com/us/bitcoin-mining/
About 1,200 kWh per transaction, currently[0]. I wrote about this back in 2022, when it was about 2,200 kWh per transaction[1].
Edit: made a chart with this data, but adding in a bitcoin transaction[2]
[0]: https://digiconomist.net/bitcoin-energy-consumption
[1]: https://rollen.io/blog/crypto-climate/
[2]: https://imgur.com/a/ggAGylW
I find this at best non-enlightening and at worst misleading because there is no consistent unit of comparison: 10 minutes here, 1 hour there, 10 miles somewhere else. That renders the visual comparison of a bar chart pretty uninteresting, no?
A bicycle continues to be the most efficient means of transportation:
https://bsky.app/profile/davidho.bsky.social/post/3mga7uhxnd...
Even an eBike is way, way more efficient than a gas-powered car, and causes orders of magnitude less wear and tear on the road.
The number I guesstimated for my e-scooter is about 500 miles per dollar of electricity.
Glad it has AI. AI has nothing on cars. Save a car trip a week even if electric is way more than 10k queries.
1 chatgpt query is a little misleading though. Let's see an 8 hour full bore claude code agent session. Or maybe running 3 agents for several hours a day.
It also doesn't include the amortized cost of training the models, as far as I can tell. I believe I heard that training the models took more energy than total queries against that model, but I could be mistaken.
I believe training currently costs significantly more than inference to all the current vendors, so I'd be surprised if it doesn't also use more power.
And by the look of it, that'll be the norm pretty much forever - unless something fundamental about how models can be trained/updated, an "older" model loses value as it's knowledge becomes out of date, even if we no longer get improvements from other sources or techniques.
But other things likely change based on "lifetimes" and usage patterns too - e.g. a large battery for an electric car may have a higher upfront energy cost in manufacturing than a small ICE + fuel tank, but presumably there's a mileage that the improved per-mile efficiency overcomes that, and then continues to gain with each additional mile.
Or even a 5 minute shower. About 2500 queries.
chatgpt use should be in the default set since energy use of ai is so often in the news now - and more often in social media
For reference it would be good to have per-passenger numbers for "sitting on a diesel bus", "sitting on an electric bus", "sitting on a tram", "sitting on a commuter train" as well.
The marginal cost of one extra passenger is going to be very nearly zero. The vast majority of the cost is just moving the bus / train / plane, and the overhead / inefficiencies in the system. I've seen somewhere the numbers for one passenger on trains and planes but I can't remember where that was. Just know it is a very very small amount for the added weight of one more passenger.
Yes, I know, and marginal numbers would also be great to have. But just the naive cost of energy per passenger-kilometer averaged would be good to see, given that it too would be much less than that of a private car.
Right, a lot of discussion about the energy economy of transportation would probably be better framed using units of people-miles or cargo-miles, per unit energy.
How does it make sense to compare renewable vs non-renewable energy sources?
Where does that electric car's energy come from?
Suppose the electric car's energy is solar. Then driving the electric car consumes zero barrels of oil, and so is infinitely more renewable than the petrol car. Or how efficient is the electric car at capturing available energy? Maybe the petrol car is 20% efficient at capturing available oil reserve energy. But the electric car is like 1e-30% efficient at capturing available solar output. Incomparable.
Why does their air con take 1kW to run? I wonder how much below ambient they are setting their temperature, and how old that aircon is and how bad the insulation?
The washing machine seems also inefficient.
My dryer is also a lot more efficient than theirs, at least if Miele's app is to be believed about how much it uses each cycle.
> Why does their air con take 1kW to run?
The author is in the UK, so they probably looked at a product like [1] which is "ideal for rooms 16-26m²" and has cooling power consumption of 1005W.
Residential air conditioning in the UK often involves small, noisy units that are only used for a few weeks at the height of summer. They'll have a thermostat built in, sure, but the user will eagerly turn them off when the room's cool just to get some quiet.
[1] https://www.argos.co.uk/product/7623899
Yeah, looking at the text at the bottom, I looks like they took the peak power usage of their AC, not the steady state.
I can't find a github or email for Hannah - if you're reading this i'd like to add Australian energy price data via Open Electricity[0] to the data (reach out via my profile)
[0] https://explore.openelectricity.org.au/
Her github is here: https://github.com/HannahRitchie
thank you!
> Gas heating (Single room) (1 hour of use) -- 2700Wh.
Wow dude, your room is very-very poorly insulated. Having a 2700W heater turned on constantly just for one room is a lot. If it's 0C outside and 20C inside, a typical room should not lose more than 500W, but better be in 200--400W range .
Kind of a stretch to suggest that an internal combustion vehicle requires 3x more "energy" to move it than an equally physics-burdened (weight, friction, etc) electric vehicle...
This is only "true" if the energy stored in the vehicle's battery got there without any relevant conversion inefficiency; If those joules came from a gas-fired plant, overall efficiency is only about 35-40%: comparable to a typical internal combustion powered-automobile or actually worse than a diesel automobile.
The upstream energy costs of supplying fossil fuels can also be counted. They add about 1/4 or 1/5th of the total in what are called "Well to wheel comparison". The comparison in the article is "tank to wheel'
It's not a direct analogue to energy but CO2 emissions per km are roughly 4 times lower for an EV charged on the EU grid in 2025, well to wheel.
We need an equivalent for water as well! Almost none of the public has any idea about water use across the industrial, agricultural, and residential sectors...
I guess that's much harder: Those who measure correctly are likely also interested in saving.
Its always interesting how complex electronics uses little energy. But its the simple stuff like heating water or turning a motors that use magnitudes more.
Physically, it makes sense, but its sort of counter intuitive in terms of utility. You might imagine the tools with the most utility use the most energy.I think thats how many people think of it.
I'm surprised that cooling takes less energy than heating. I imagine that depends a lot on the temperature range; they only need so much to cool a room even on a "hot" day in the UK.
Still... AC still feels like magic. I know how it works and understand the over-unity factor. But it feels like it ought to take enormous energy for it to work at all.
I think specifically it's comparing gas heating vs AC. Heat pump heating would probably do better. In other words, it takes less energy to move heat inside/outside than to "create" it
(With caveats like heat pumps are much less effective in extreme cold)
You can select gas heating vs electric heat pump. The heat pump looks to be about a third the cost of gas.
You can reject 4 watts of heat with 1 watt of electricity using vapor-compression refrigeration.
You can get that up to 7 or 8 watts (or more!) per watt with evaporative cooling towers and vapor-compression combined.
AFAIK you can’t move heat into somewhere using cooling towers, they only increase cooling efficiency.
Heat pump heating is limited to around 4W of heat moved for every 1W of electricity, with the efficiency dropping as delta T drops (aka as it gets colder outside)
You can generate 1 watt of heat with 1 watt of electricity and a resistive heater, they’re more or less 100% efficient.
Yeah without knowing the climate, temperature delta and insulation these values don't really mean much.
This is neat. I think I'm actually more interested in avg ChatGPT query than median single query so that I can enter a large query # and be confident in the associated energy cost for that larger number (e.g. what's the energy cost for 1,000 chat gpt queries)
I don’t think the problem Is how many joules of energy are used. But the cost/burdnen to produce them. The costs and forms for each of these examples is very different making their energy use incomparabale.
In my petrol-powered Prius I average 52 MPG, so at the current price of gas ($2.60/gallon) I pay $0.50 to drive 10 miles - less than the listed cost to drive an electric car the same distance.
This needs a phantom power feature. How much does your phone charger consume when its it charging your phone but still pligged in. Whats the cost of your TV when its "off?"
Switching mode power supplies use milliwatts when there's no load. Appliances are usually 3W or less when off as they need to keep the IR sensor turned off. The best appliances are <1W.
0.3Wh for the median ChatGPT query seems optimistic to me. It looks like it’s based on a quote from Sam Altman.
The GPU alone is probably pulling close to 0.2Wh per second.
not sure I understand the petrol car using 3x as much energy as an EV... wouldn't it make sense to convert gasoline to electricity then? I presume that must not be as efficient as other ways to convert fuel to gasoline? (I understand the math is there but... I'm temporarily failing to get it)
I think lists like these might be useful for energy audits and thinking about ways to make better use of energy
70% of the energy in a petrol car is lost as heat. Only around 30% or less of the energy actually propels the car. I imagine that's why there's a big difference.
Gasoline engines are around 30-35% efficient, the rest is lost as heat. That goes for whether they’re spinning an alternator as part of an engine-generator set or just moving a vehicle.
You can’t get more than 35-40% efficiency so converting to electricity is a wash, you lose the energy to heat no matter what.
Also, the chart does not take into account how the electricity for the EV is generated, it would be just as inefficient as the gasoline car if the electricity was generated by burning hydrocarbons, but that detail is left out.
To get those types of numbers you would have to be charging from a grid that is nearly 100% coal. Real grids heavily favor EVs.
Yeah I probably should’ve mentioned that the more solar/wind/hydro you use to charge an EV, the more the efficiency goes up, thanks.
Yeah I'm not sure that EV number maths out. That's 18kw at 60mph for 10 minutes. It sounds really low to me. Going 45mph on a 110lb eMoto with 180lb rider takes ~6kw.
If everyone had their own turbine to generate electricity from gasoline then it would be. But from a central power plant there are transmission losses which cuts out a lot of the potential efficiency increase. The obvious upfront and maintenance cost of everybody owning a personal turbine though makes that method kind of iffy.
Wow, putting everything in the same units is really informative. Running my 450 watt gpu for a day is approximately equivalent to driving a car 10 miles.
I doubt your 450w gpu runs at that wattage 24 hrs, unless you're mining with it.
My (admittedly old) gpu+CPU idles around 50-75w.
I think you simply can't predict any number without really measuring it.
Ryzen + GPU + 1h of gaming, but 3 screens attached.
still a "desktop computer" - stuff in the browser can spike GPU wattage, but I don't think it will be meaningfully different over several hours of true idle..
My problem is more like: I could measure that, but I'd have to recable everything to have the power meter behind the power strip with just PC + screens... which I will probably do at some point, but it involves redoing 50% of the cables in this room.
During training it’s running at around 410. But yeah idling at 450 would be pretty crazy.
Most clothes washer cycles are about an hour, is the motor really averaging 800 watts over that time?
The EU Energy label on the first (more expensive) washing machine I looked at said 440Wh for a cycle, so 800Wh for an average one isn't unreasonable.
One hour of Claude code— well, I’d guess it would be comparable to an hour of driving an electric car. How to know?
OP says one query uses 0.3 Wh. Driving an electric car for 10 miles = 3,000 Wh which is roughly 10,000 Wh per hour.
I'm not sure how many queries is equivalent to an hour of Claude code use, but maybe 5 seconds, which means an hour of continuous use = 216 Wh, or ~50x less than an electric car.
OP has a longer article about LLM energy usage: https://hannahritchie.substack.com/p/ai-footprint-august-202...
My point is that Claude might easily be about 50x more energy intensive than normal ChatGPT prompting.
A coding agent runs near-constantly, so of course it'd require a lot more compute than running even, say, a multi-minute query with a thinking model every hour. How much exactly is pretty hard to calculate because it requires some guesswork, but...
For a long input of n tokens from a model with N active parameters, the cost should scale as O(N n^2) (this is due to computing attention - for non-massive n, the O(N n) term is bigger, which is why API costs per token are fixed until a certain point and then start to rise). From the estimates from [1], it's around 40Wh for n=100k, N=100B. I multiply by 2.5 to account for Opus probably being ~2.5x larger than gpt-4o, and also multiply by 2 to pessimistically assume we're always close to Opus's soft context limit of 200k (it's possible to get a bigger context for extra cost, but I suspect people compact aggresively to not have to use it). That gets me 7.2J/t, which at a rough throughput estimate of 20t/s gives me power of 144W. Like a powerful CPU or a mediocre GPU, and still orders of magnitude lower than a car.
[1] https://epoch.ai/gradient-updates/how-much-energy-does-chatg...
Beside the point, but 10,000 Wh per hour is kind of an insane unit. It's 10,000 watts. Or 10 kW if you're really into the whole brevity thing.
It is not only about raw power consumption. Comparing driving an electric car with using AI only in kW hides a major point: Hyperscale datacenters are massively centralised, which brings it's own problems; a lot of energy is used for cooling, and water consumptions is enormous. Charging electric cars at home is distributed and does not suffer from the same problems as the centralised hyperscalers do. Also, running AI models at home is not much different than a gaming session :)
This is an incredible sequence of assertions, every single one of which is very incorrect.
"A lot of energy used for cooling": hyperscale data centers use the least cooling per unit of compute capacity, 2-3x less than small data centers and 10-100x less than a home computer.
"Water consumption is enormous": America withdraws roughly 300 billion gallons of fresh water daily, of which IT loads are expected to grow to 35-50 billion gallons annually by 2028. Data center water demands are less than a rounding error.
"distributed and does not suffer from the same problems": technically correct I guess but distributed consumption has its own problems that are arguably more severe than centralized power consumption.
I clicked-through on one of the sources (a blog post on "ohmaticelectrical"), and the website is either hacked or is just really scummy. Here's what I was greeted with:
https://imgur.com/a/https-ohmaticelectrical-co-uk-scams-OmLx...
(The scammer didn't actually manage to put anything in my clipboard, so I'm not sure what they were hoping I'd run.)
Some more things I’d like to see:
- AI related values of course
- elevators, escalators
- traffic signals, street lights
> Desktop computer - 1 hour of use - 50 Wh
That seems low...
50W average doesn't seem absurd, peak power is going to be an order of magnitude higher, but computers are often running pretty close to idle...
Thought the same thing. I think I measured my gaming PC to idle at like 70–100 watts. Obviously an average office PC would be lower, but you also need to add on the energy for the monitor, which seems to start at around 15 watts for some basic monitors, and I imagine that's not at high brightness.
Nice! A small note: airfryer tend to cycle on/off much less then slow-air/static ovens so they consume way more per unit time, gas ovens never fully stay "off" when operating, they just reduce the flames.
My electricity provider has a rough breakdown. Which I was kind of incventivized to look at because of money. Heating is the big one. The electricity bills are seasonal.
Next.
I can do my laundry at night because the electricity is cheaper.. oh wait I can’t. That’s apparently unsafe. So I have to do it in the evening. Okay. I’m not going to move my whole small freetime evening around to save a buck on the half-evening long wash cycle. So nevermind that.
The nagging about turning off all the lights were always a consumer blaming ritual that doesn’t matter.
I often do my laundry at night, what's the problem? I set it to finish roughly when I intend to get up.
Turning the lights off made a lot of sense in the 1990s and earlier, when each bulb was 60W.
Switch to heat pumps for everything
I like the comparison concept. It's like that "order of magnitudes every programmer should know" list, but applied to anyone who cares about energy.
That said, and hot take: people shouldn't worry about energy independent of what they pay for it. The whole point of a price is to fold a complicated manifold of scarcity-allocation into a set of scalars anyone can rank against each other. Appealing to people's sense of justice or duty to get them to use less energy than they'd otherwise be willing to buy is just asking them to lead a less utility-filled life than they can because you think you can allocate scarcity better than the market. I can't, and you can't either. Nobody can.
If you claim that people should listen to moralized pleadings and not the market because prices don't internalize certain externalities, duty is on you to get those externalities accounted so they can properly factor into prices, not apply ad-hoc patches on top of markets by manipulating people's emotions.
As for getting externalities internalized: as a society, we call the procedure for updating rules "politics", and it's as open to you as to anyone else. If you propose policy X and you can't get X enacted, perhaps it's because X is a bad idea, not because the system is broken.
Not everyone anyone claims is an externality is, in fact, a cost we must account. We should have a prior that costs are accounted and need evidence to rebut it --- and any such rebuttal must involve numbers, not emotional appeals. What specific costs are unaccounted? How large are these costs? Through what specific mechanism are they escaping existing accounting mechanisms? "I feel like we're using too many electrons for X" is not a valid argument for the existence of an unaccounted externality.
That is, unless there's some specific reason to believe otherwise, we should believe market get it right, especially with fungible commodities like kWh.
How do you propose to convince people to get those externalities accounted without emotions? How do you convince people of the value of externalities that are far away in place or time (but not less real)?
The idea that you would be worried about how many electrons you use and it's relationship to climate change is on its face kind of ridiculous.
It's like worrying about how many times you personally ordered Chinese food affects the price of Diesel fuel in India. It's an absurd leap of logic, and the parent is right to call out these arguments which are almost always emotional.
How am I as an individual supposed to get externalities priced in?
And given that right now they are clearly not, what’s your plan until then?
Which specific externalities are you concerned about? Do they affect you directly?
Climate change and pollution, and yes they do.
Sure, by your own argument, you should somehow increase the price of people telling other people what to avoid spending money on
> As for getting externalities internalized: as a society, we call the procedure for updating rules "politics", and it's as open to you as to anyone else.
Ok so I do need to worry about energy so that I can identify these unaddressed externalities and work towards updating the rules. You can to care before you can get involved in this stuff. You can't tell me not to worry about it and then also say that it's basically my fault for not getting involved if the price is wrong.
> any such rebuttal must involve numbers, not emotional appeals
Who are you arguing with? You're commenting about a website that has strictly numbers and nothing else.
Your dismissal of moral concerns is not convincing.
Imagine a world where the only energy you do is use was generated by a stationary bike you had to ride yourself. You would, generally speaking, use that energy differently than energy you would pay for--you would generally reserve your effort for worthwhile things, and would be averse to farming energy yourself just to power frivolity or vice. How you determine what to put your energy into would explicitly be a moral question.
Instead in our world we an abstractions conceals the source of the energy. But if the moral concerns from the first world had any weight, they haven't lost it now; if energy is anything short of completely free we should by the same logic be averse to expending energy on worthless work or vice. The human being is not a utility monster, but something very different, and moral questions of this sort are central to how it navigates the world, they should not be dismissed.
Doesn't this argument hinge on equivocating between two different definitions of aversion, though? I'm averse to bananas, but that doesn't mean I think it's immoral to eat them. The moral dimension kicks in if somebody else had to ride that stationary bike for you, because then you'd be wasting their time on frivolities.
Of course I'd use energy differently if it cost more. If I had to generate energy by pedaling a bike, I'd consider it costly indeed. So what? Energy doesn't cost as much as it would if I had to manually generate it, and who are you to say allocation decisions made under that regiment are good and ones made under ours are bad?
Wouldn't your argument also compel us to use steel as if it were gold? Salt as if it were saffron?
Aa Digital ID is getting forced on the public, it is dangerous, almost traitorous to demand/suggest ordinary citizens need to care about energy efficiency to the point of giving up their freedoms/conveniences!
Focus your thinking on solving/promoting energy production instead!
What’s startling to me is how many comments in this thread just take the provided values as gospel without asking questions that methodology answers either in the abstract or barely describes. Also going giving a cost for “United States” is absolutely nonsense - electricity, gas and gasoline prices vary widely across the country. There is no one cost for each, and the average is worthless for this kind of thing (especially since the average of each - gas vs. electric vs. gasoline cost - are independent variables that have no relation to each other on a region by region basis).
Why are people so gullible?
My first question was: "Is this whitewashing LLM energy usage?"
And yes, that seems to be the undercurrent here. Complete with linking to themselves to validate the data they used to make their estimates.
Either these companies need to build these massive data centers that consume massive amounts of electricity OR these LLMs don't use a lot of electricity.
You don't get both. If LLMs don't require a lot of electricity, then why are we building so much more capacity? If all of that capacity is required, then what is the real cost of sending a query to these LLMs?
Hannah Ritchie is a quite well reputed writer and data scientist squarely in the climate field. She's written two books on climate and I found the one I read (Not The End of the World) was quite good and data-driven.
https://hannahritchie.com/
You're going to have to make a stronger case that this data is biased towards LLM than that.
I'm not sure I understand where the issue is here - something can use a small amount of energy per use but a large amount in aggregate because of lots of use.
You have set up a false conflict. The data centers are "huge" and they also consume about the same power as 1 airplane. These things are both true.
It is also not really true that they are huge, it is a misconception driven by biased reporting about facilities that really aren't very remarkable compared to material distribution warehouses, beverage bottling plants, and suchlike.
> You don't get both. If LLMs don't require a lot of electricity, then why are we building so much more capacity?
A small number times a large number is often a large number. Have you heard of the concept called "per capita"? In any case, electricity is going towards data centers in proportion to the degree to which these data centers do useful work. AI companies buy the electricity fairly on an open market, sometimes even subsidizing this market by funding new generating capacity.
If all these people and companies are making electricity allocation decisions that make sense to them with their own money, who are you to stop in and say that their voluntary transactions are incorrect? Who died and made you the king?
Useful work is debatable here, a lot of people just talk to the thing or use it instead of searching the internet.
The owners surely think, or at least want us to think that it is very useful indeed, otherwise we'd see no point in burning through piles of investors cash to buy overpriced ram, storage, gpus, cpus, nics, secure the power to run it and then subsidise the users to use it.
I do think that transaction is wrong and it's going to bite them in the ass in the long term, but I don't have the money to outbid them for the power. I do get to see them crash and burn when the investors get impatient.
It’s new capacity!
They’re not even saying they shouldn’t do it or that they’re not useful or not worth it but you Cannot logically say both “these things do not use a lot of power” and “we need to build more power plants to handle these things”
It isn't all new capacity. The popular discourse hardly ever mentions it but AI is a small fraction of why we need new datacenters and the bulk of the demand is driven by general IT needs, particularly consolidation of small, grossly wasteful corporate data racks into vastly more efficient cloud services.
Edited to answer: The question has also been addressed by the same author as the article: USA spent a quarter century not building generators and that negligence has finally caught up to us, despite objectively heroic efficiency efforts on the part of the IT sector.
https://hannahritchie.substack.com/p/usa-electricity-growth
If so, why do they need to build new power plants for it?
[flagged]
I am a cloud capacity planner and you're just another dog on the internet, "but do go off" as they say online.
Yeah you can, though to be fair its referred to as jevons paradox because it is counterintuitive.
I’m not saying it’s inefficient. I’m saying cloud computing uses a lot of power.
What about it is whitewashing? This seems like it would be a great resource if you wanted to contextualize the argument you're gesturing at.
LLMs don't use a lot of electricity per user. Why should the fact that the energy usage happens in data centers instead of each user's house be an important moral factor?
War is Peace,
Freedom is Slavery,
Facts are Whitewashing.
Much like when people discuss whether these companies are profitable: training costs don't count.
Indeed, looking at a "single median query" totally disregard the fact that:
- first, those queries are mostly useless and we could totally do without them, so it's still a net pollution
- they are being integrated everywhere, so soon enough, just browsing the web for a few hours is going to general 100k+ such equivalent "small queries" (in the background, by the processes analyzing what the user is doing, or summarizing the page, etc). At that time, the added pollution is no longer negligible. And most of this will be done just to sell more ads
1. Your prediction is that soon browing the web for a few hours is going use >30,000+ Wh (based on the "equivalent" you mentioned)? (For comparison to that 30,000+ figure, the energy use from using a laptop for a few hours is 75 Wh, all per OP source.)
2. > most of this will be done just to sell more ads
Are you predicting that the value of ads is going to increase by a number of magnitudes? Because 30,000+ Wh of electricity has a quite significant cost, and even a video ad currently only earns pennies, so I'm trying to imagine how the math would work in this scenario?
LLM calls are getting cheaper every day.
Are they getting cheaper faster than their usage is increasing?