At some stage I wonder if the UK will need to regulate the charger industry. The price gouging is wild in places. If we look at the energy content of petrol, a litre of gas contains about 9kwh of energy, or at average pump prices 1.58/9 = ~18 pence a kwh.
For sure, EVs are far more efficient at converting a kwh of energy into forward motion, but if we assume 35 mpg (9.25 miles/litre) for the gas car, we need about 970wh to travel 1 mile. A modern EV can manage a mile on ~260wh, almost a quarter of the gas requirement.
There are public charging networks in the UK averaging 92p/kwh - we know we need much less energy to move the more efficient EV, but even with this adjustment fuel cost per mile looks like:
petrol at UK average today: 17p/mi
Electric at very expensive public charger: ~24p/mi !!
At many chargers, there are no savings at all. For comparisons sake, that 92p kwh would be just 28.6p on the most expensive domestic electricity supply, and charging at home would be ~8p per mile on the worst possible tariffs.
I've probably done some bad math somewhere here, but I think the broad picture is correct.
The market should sort this out by itself, not saying regulators shouldn’t watch closely, but competition should be enough to do its thing. Cartel formation especially should be watched for vigilantly.
Of course, thats why I've been clear all my assumptions are for 260wh/mi, which I think is a very fair middle ground figure to compare to a 35mpg car - one can pick far more fuel efficient gas cars for this comparison too, the possibilities are endless.
I think your numbers still illustrate the same point though; if you can't charge at home, an EV is not necessarily cheaper to fuel, and the gap between the public charger price and the cost to a private consumer with home charging is still far too big. 98p vs 15p is staggering.
But compared to the US home charging via a mains outlet is much more viable because it's 240v vs 110v. If you plug you car overnight you'll typically have enough charge to last you the next day.
Almost everyone I know with an EV charging at home just reused the 240v dryer socket to avoid paying for a dedicated fast charger. It's often cheaper too to have an electrician fit a new 240v socket instead of the dedicated charger as well.
Home chargers with dedicated sockets is three phase 400v actually over here in the EU and every single home, and even relatively new apartments have that because of induction stoves.
Three phase power is definitely not 100% in the EU. Not even in Germany, though adoption does tend to be higher than neighboring countries.
And FWIW, I find that my induction cooktop works wonderfully on plain old 240V 40A, so I do not think it is a requirement to get three-phase for that ;-).
The US is 240V. We split it into two 120V legs for some sockets, and not for others. Some people do choose to get by on 120V, true, but they are the minority. Usually people who do not drive often.
Most? You mean all. 240V [0] as been the standard in the US basically since electrification started in the late 19th century. 120V has for all practical purposes never been a thing, it has always been an artifact of split-phase 240V. A deliberate choice to offer two voltages to every consumer.
[0] Okay, technically 240V did not become official until around 1967, but the split-phase design was there from the beginning. They capped it at 240V to stop the creeping up that had been going on in the earlier part of the century. This is why you still have a lot of people (not all of them old enough to have been alive in 1967, oddly enough) that refer to 240V as 220.
Sure, but that is the exception proving the rule. Not quite urban legend, you can find people on mikeholt.com who have actually seen one in the wild. Usually because of some shenanigans the local power company pulled to directly connect more houses by giving each one a phase of a three-phase feed.
Even in Wales, 25% can't. This isn't a figure you can ignore.
And that's a hypothetical, it relies on landlords playing ball etc. then there's the social issues. On the north of England we have lots of terraces built for mill workers, these aren't owned by the richest on society. So then you're in the situation of charging the poorest more for transport. And these are necessarily on towns with good transport links (think 1 bus and hour).
Many of those "can't" won't have cars. 20% of households in Wales have no car.
Now clearly that 25% and 20% won't overlap exactly, but they will overlap a lot
When 80% of cars in the UK are electric and 20% are those households who rely on public streets to store their belongings for 23 hours a day, then sure that will be a problem
Given that there's only about 2 million electric cars in the UK, yet 18 million homes which can charge electric cars, that's a long way off.
We don't use unit of measurements.
We use metrics because we have a lot more context.
Rps, requests per second is a commonly used unit but it has no defined standard, you could and often do average it over time for reporting but no one says you have to. For scaling however you'll probably want to use the max not the average, because no one wants a web application where in business as usual 60% of the time it works every time.
Does the existence of that knowledge make a slight bias lowering the odd on no?
I could fork this and with a 1 line change earn dozens of dollars as long as I don't tell anyone what that secret change is.
No, I think the bias is really just a reflection of how propositions are phrased. We could imagine a mirror-world prediction market that offers all the same propositions, but phrased oppositely: e.g. a market in "person X will die by Y date" becomes a market in "person X will survive until Y date". And in that market, we would see a bias towards propositions resolving as Yes.
Looking at the amount of wires going into this, my instinct is that this cannot scale, in 5-10 years this won't be doable for a Pentium chip, at least not as an at home hobby project. But I actually think it could go the other way, and in 5-10 years you'll be able to do this at home for far more sophisticated kit, unlocking crazy amounts of reverse engineering possibilities that were once thought of as near impossible, or at least only possible for a nation state scale setup.
I'm pretty sure if you really want to, you could do something like this as a hobbyist with a Pentium right now.
Instead of futzing with wires on a breadboard you could simply designing a PCB up front, throw the design over the fence at JLC or PCBWay, insert coin, wait patiently at the mailbox, solder your scavenged Socket 7 onto the board.
The days of toner transfer and aquarium pumps are already long gone. Getting production quality, one-off, multi layer PCBs done as a hobbyist is dirt cheap these days, no government budgets required.
> you could simply designing a PCB up front, throw the design over the fence at JLC or PCBWay, insert coin, wait patiently at the mailbox
It blows my mind that I can use free-as-in-beer Free-as-in-Speech software to design a PCB, email it to a dude in China, and get a finished working professional-looking PCB back in my hand within a week, for the price of a couple of coffees. And if I want the components stuck on too, it'll cost a little extra, maybe three coffees it costs now.
If I want it really quickly then for the price of a decent takeaway curry I can have it flown over next day. What the actual hell?
Edit: the slowest part of "next day" is when it hits the UK, and if I could guarantee it just gets delivered to DHL's Edinburgh depot I could drive down there in two hours.
The reality is that's just how cheap commodity circuit construction is, and even small shops in the US sometimes approach that low cost, and we've been paying crazy markup on electronics for decades. That dirt cheap board is so profitable it is using air freight to get back to you. It is literally burning money just for convenience, yet that is the "cheap" option.
Electronics cost a lot to manufacture in the 70s, but is entirely automated now but for "reasons" we have only seen a small part of that savings.
When you buy the "Cheap" version on AliExpress, they are still making a healthy margin, yet Americans will happily buy the exact same product off Amazon for next day shipping for 10x the cost and think they are getting a "deal"
This extends to cars as well, with the F150 costing as little as $20k to build, even with "Expensive" very unionized and well compensated labor. The higher market trims only cost a little more to make but take in far higher profit margins. How much of China's supposedly "Subsidized" car price (as if the US doesn't do anything to subsidize cars) is just a lower profit margin?
Things should be way cheaper to western consumers. Where does all that extra money go? "Marketing and administration", basically bloated executive suites, bloated middle management, and the pockets of Meta, Google, AWS, and Apple. Oh gee, those exact companies seem absurdly wealthy and are basically responsible for all economic growth in the past few decades.
> even small shops in the US sometimes approach that low cost
It's ludicrously expensive to ship things to and from the US though, and since they're now paying some insane markup because no-one understands what tariffs are the prices have got even sillier.
It's not impossible that the Aws charges were wrong, it's pretty unheard of.
I don't understand why the details of the charges aren't mentioned in the post.
If you think it's unlikely you could have a $1500 bill because you 'barely use' it then that's just wrong. In the cloud single unoptimised choices can cost thousands if you don't keep an eye on your costs, you need to look at the charges.
I think folks are mis-interpreting the story. To me it sounds like AWS was doing everything fine (and OP agrees), but he wanted to re-negotiate the $1500/month fee because of his under usage (maybe a cheaper tier?), so he sent an email. They didn't respond, so he stopped paying to get their attention and entice them to respond. But, intead of responding, they just terminated his account.
I read it otherwise. This is the smoking gun, to me:
> So I stopped paying. Why keep paying charges I believe are wrong when the company won't discuss them?
That sounds to me like he thought they were mis-billing him, to the tune of $18K per year, not that they were billing him correctly but he wanted a better price.
Seems more of a productivity killer rather than an aid but cats are great marketing. I see no reason not to submit this for YC funding in the next round
reply