5.4 is great. I use it for python professionally and for typescript/front-end games and educational apps recreationally. In my experience it's roughly as good as opus, just a lot cheaper. It's amazing how much usage you get for $20/mo
15kWh 48V LFP battery around $1800 with low quality battery management system in metal box on wheels. Car batteries need more expensive inverters if you want to fast charge them (150kW-950kW) and super fast discharge them while driving fast (>100 kW). Thus my 600kW extender comes to almost $62000 for vans and small trucks. Cheaper if installed as house battery. The Mercedes eSprinter 56kW van costs around $80000 new but we sell 3 year old vans like this for $4000 without battery. So refurbished and converted to eCamper with 1800 mile range you pay $6700. You can drive 3000 km (almost 1860 miles) with this battery in the eSprinter and eCamper. A normal size car would go twice as far with this battery but it's big and heavy enough that you need to tow it in a trailer.
The crucial point though is the charging/discharging inverter (converter) that I purpose built (printed circuits boards) and a change to the car firmware. Without it the car will reject the battery, your acceleration would be less and it also would not last the same amount of discharge cycles. My battery electronics works fine for cars, trucks, boats, house and neighborhood batteries (up to 6mW per shipping container).
eSprinter 2022 is 56kWh. In Europe I'm limited in the size of a battery by the total legal weight of van and it's trailer combined. So I can not tow more weight than 600KWh LFP batteries with this particular van. But with $0.01 cost per kWh it only cost $60 (52,08 Euro) for a full charge, good for 2000 km (Amsterdam to southern Spain). So even though I carry 5.5-6 times as much weight as a small city car around, it cost me a lot less than having a tiny battery and charging at commercial chargers with $0.40-$0.90 fees per kWh. And a lot less than gasoline (benzine) or diesel.
Also the larger battery means the individual cells can be pulse charged much slower and each cell individually at the rate where it doesn't damage that much. I measure the temperature, voltage and current of each cell so they never overheat. This is how I get many more cycles out of each cell so they last 50 years. It is also safer, with thousands of temperature measurements several times per second not a single sell gets warm, and if they ever do it is because it is damaged and we can immediately disconnect it and tell the driver where to locate it and remove it.
For a truck these thousands of battery cells discharging slowly in parallel becomes the reason all trucking companies will be forced to switch from diesel to electric, it is several times cheaper per mile or km. Lower energy cost, lower maintainace, lower downtime, longer life. The only thing you would want is that the maximum weight limit per truck goes up so you can ship more per trip. Right now you ship little kilo's if you carry a heavy battery. But charging with your own solar at home base is so much cheaper that it is worth to do two trips versus 1 trip with diesel.
The reason electric trucks are not yet everywhere is that the truck makers ask ridiculous amounts for battery cells that are still wired in series and discharged too fast to last long. Simply bad design. We need a disruptive electric truck startup and we need a disruptive battery startup. Investors welcome...
But 600 kWh is about ~4k kg, no? Isn't that like the max hauling a sprinter can do? So doesn't this just get you a bunch of range at the "cost" of not being able to haul anything or am I missing something?
7000 kg is the maximum a van and its trailer can way by law in Europe. It can haul a lot more.
No, my eSprinter camper is a small room with kitchen, bed and shower. The trailer ways 3500 kg, the eSprinter 2670. I could haul at least 889 kilo more. If I had the bigger motor I could haul twice as much.
Thank you. I hardly get to explain the techology I 'invent' (power chips, power router, parallel battery charger, car firmware, charging (station) software, simulation software) because the investors customers only want to hear that its cheaper or sells better (then Tesla). Or that besides going from 4000 to 20000 dis/charge cycles you also prevent any li-ion fires and have fire alarm sensors on every battery cell. The main thing I would like to shout from the rooftops is: Not a single battery on the planet charges their battery cells in parallel as we do, they all shorten their cell lifetime by charging/discharging them to fast in series, what will damage all battery cell types but especially the li-ion.
It is the same with the article we are commenting on here: if people just listen to the statistics, the simulations and the actual market developments they would see that 100% solar+battery is the cheapest energy.
The simple message is Solar is by far the cheapest energy: below 1 dollar cent per kWh and that will fall a lot more in the next decade until we get to 'a squanderable abundance of free and clean energy' as Bob Metcalf puts it https://www.youtube.com/watch?v=axfsqdpHVFU
Batteries still double the cost of that solar but these prices are falling rapidly too. It is already cheaper to have solar nearby than transmit it over a distance of a few miles.
$0,01 per kWh from solar, that is the price worldwide on the condition that they sell you the panels at a reasonable price and don't overcharge you on all the other parts like micro-inverters, field or rooftop installation, permitting and labour. That adds up to around 5 cent for rooftop solar in Australia for example (including everything). 1 cent is for solar panels lasting 50 years (with 20% degradation over decades), we refer to such prices as Levelised Cost Of Energy (LCOE) over lifetime. It halved in the last 10 years for solar and it will halve again (20% cost reduction on each doubling of manufactured capacity). Similar for batteries, they also go down around 20% each year.
1 kWh Wind, or Hydro, Thermal and other renewables do not go down as much in cost price because they have mechanical or chemical components that do not last as long as solar cells and need maintenance and repair.
We keep the cost low by group buying in bulk at wholesale prices (a shipping container with 770 panels for 20-30 houses) with our coop instead of premium installer prices by the electrotechnical or building companies.
If you let our Fiberhood coöperative in the US install your solar, batteries, tiny house or eCamper you do not pay these high tariffs, we have enough panels pre-tariff. So you still can hit 1 cent per kWh but only if you get the decent installers and sellers.
Our energy storage solutions have widened. First you timeshift all electricity use of the car, house or neighborhood into daylight hours when the sun shines. This means a bunch of electronics and software changes. Next we build termal storage solutions, you can store heat much cheaper than electricity. You move heat around with a heat pump. Or you heat your water tank with a datacenter computer in your water tank (for free). In summer you store solar electricity in ice. Or you store it in iron, aluminum, glass or silicon by melting ore and purifying, You embodied the solar electricity into the purified ore.
In northern and southern latitudes you need 10 to 50 times more solar panels to heat houses during cold winters. This means you have large overcapacity in summer that you can sell as embodied iron, etc. Batteries are only needed to store for the hours there is no sunlight during 24 hours, no need to store longer. The cheapest place is to store it in the electrical cars in your neighborhood. That is why we install our own brand ev car chargers in the neighborhood of the panels. In contrast, Tesla chargers overcharge you a factor of 34 to 76 and that's partially because its fossil energy and transmitted over hundreds of miles.
Also Trump doubling solar panel prices with tarifs and shutting down subsidies is wrong, it makes it much more expensive. Add an oil third world war however does help, we sold double solar, batteries and evs in the last month.
It's also widely misunderstood. Just because the spot price of electricity is set by the price of gas doesn’t mean the consumer pays that price for all of their electricity.
A lot of wind and solar are on Contracts for Difference. That means when market prices go above the agreed level, the generator pays the difference back through the scheme, which reduces supplier costs rather than the generator simply keeping the whole windfall.
This is particularly relevant when e.g. the price of gas goes way up due to the Iran war, it doesn't mean that the consumer ends up paying more for the energy from wind
> Presumably because the price is a volatile, and storage gives you more flexibility around when you buy.
I will give credit to the person who got there before me. :)
Smoothing out price volatility is a big one.
But also it gives you options:
You can buy it "today" when its cheap and store it for when you need it (e.g. winter months).
You can also trade on that basis too. For example you can make a future-dated commitment to buy gas (knowing you have the storage available to take delivery). But if the situation changes and you later find you don't need it, you can sell that contract to someone else (or you can still take delivery and re-sell it). But you can't do any of that without having the ability to take delivery, because the person who sold you that future-dated contract will want both your money and to get the gas they sold you off their hands.
Because if you have enough renewables and storage to eliminate gas from the mix you are no longer paying gas prices. The more often that happens, the cheaper your bills get.
For me, it's beyond doubt these tools are an essential skill in any SWE's toolkit. By which I mean, knowing their capabilities, how they're valuable and when to use them (and when not to).
As with any other skill, if you can't do something, it can be frustrating to peers. I don't want collegeues wasting time doing things that are automatable.
I'm not suggesting anyone should be cranking out 10k LOC in a week with these tools, but if you haven't yet done things like sent one in an agentic loop to produce a minimal reprex of a bug, or pin down a performance regression by testing code on different branches, then you could potentially be hampering the productivity of the team. These are examples of things where I now have a higher expectation of precision because it's so much easier to do more thorough analysis automatically.
There's always caveats, but I think the point stands that people generally like working with other people who are working as productively as possible.
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.
Love this! Would be super interested in any details the author could share on the data engineering needed to make this work. The vis is super impressive but I suspect the data is the harder thing to get working.
The most time and energy has been getting my head around the source data [0] and industry-specific nuances.
In terms of stack I have a self-hosted Dagster [1] data pipeline that periodically dumps the data onto Cloudflare R2 as parquet files. I then have a self-hosted NodeJS API that uses DuckDB to crunch the raw data and output everything you see on the map.
reply