Due to it being electric or due to the specific design? EVs are generally much easier to design for water crossings. I actually drove an electric motorcycle across a river fully submerged, which it wasn't even designed for (had to do a thorough check afterwards but it was completely fine). This is not even remotely possible with the bike I normally ride (Africa Twin).
Proper sealing, mostly. The bike I was riding is a custom-built enduro, the electric part is fully sealed up to the handlebars but the river turned out to be a bit deeper, as it often happens. Electric drivetrains are much simpler. They aren't running as hot as ICE, don't need outside air, have less vibration and fewer moving parts... you can make it a proper submarine if you desire. In fact, certain 2WD electric mopeds are rated for underwater riding.
It's possible to use a normal motorcycle fully submerged as well [1], but designing for that is way harder due to the exposed engine, you need a ton of things and not just a snorkel.
"Safe" is relative, but I've taken older Honda civics through water part-way up the doors. When you're in the middle of nowhere it's nice to have options. Do you run the risk of major electrical faults if you run this through water?
The Honda Fit is great. You can probably squeeze an extra decade out if you're willing to swap out the motor or transmission (used, 100k miles or so, if you shop around $2k-$3k should be doable), and if you're using it heavily then you have the advantage that most cara on the market take less abuse, so you can maybe grab a decade beyond that by picking up somebody else's used Fit when you're done repairing yours.
If you stay on top of fluid filming or wool waxing you can largely avoid this. I've got 7 winters of WI slush and salt on my truck and just a few nickel sized spots of surface rust on the frame so far.
> used, 100k miles or so, if you shop around $2k-$3k should be doable
Where are you finding a 100k mile Honda Fit for $3k? Before I bought my current daily driver, Honda Fits were on my list to look out for and in the central NJ area I never saw one in decent condition around that mileage for less than $5k. Even looking now I see people trying to part out theirs for $2k or looking for $4k for a 200k mile one. I messaged someone on FB Marketplace that had a 2013 with 65k miles on it to try and bring down their $11k asking to $8k and just got ignored.
NJ is probably on the higher end of the market but the deviation can't be that big.
Sorry I wasn't clear. You can get a motor with 100k miles from a totaled car for $3k, including the labor to replace it.
To your actual question, I bought mine (2008, manual) in 2018 for $5k with 100k miles in The Bay, and it took about a month of waiting for a good deal to crop up. I've put another 100k on it without issue and plan to drive it a long time. Inflation and the chip shortage have roughly kept up with depreciation, so I'm currently seeing some good options in the $6k range and similarly expect that $5k is around the bottom of what you can pay for a nice vehicle with 100k miles on it.
Also, deviations can absolutely be that big. It's more prevalent at the top of the market, but there are big differences in Subarus and Civics, for example, in different parts of the country, even in the sub-$5k range. It's often worth a flight and driving back to purchase a car (if you value your time at $0 or have other things to do while you're there).
Damn, that's a great deal. And yea, $5k seems to be about the bottom of the barrel in terms of getting a decent car.
For my "daily" driver (I drive a few times a week and it's rarely more than 20 miles), I ended up buying an imported WRX on an auction site. Cost more than a used Honda Fit but it's a ton of fun to drive.
Japanese cars, particularly cars that have been orphaned, keep their value at high mileage.
If I had to get a high mileage car in a hurry in upstate NY with some expectation that my acquisition + repair costs would be reason I'd go looking for a 2005 Buick. Maybe half of that is getting older, the other half is that my son drives a '96 Buick which has needed some creative maintenance but has been rock solid reliable after a flurry of work where we replaced aging parts.
Why a Buick? Which 2005 Buick? You can probably, find a simpler or less expensive doppelgänger if you look at the same car but with a Chevy or an Olds badge. I don’t mean this as a dig, simply curios.
Buick is gonna be less ragged out than an Impala or Olds of same age (you see the same with the Grand Marquis vs the Crown Vic). An 05ish one will come with a 3400 or 3800v6 which was pretty solid and cheap to own by then. The rest of the car is nothing special, just keep oil and coolant in it and drive it.
Basically he's picking a very well sorted platform of a vehicle and then choosing the brand that most correlates with buyers who'll keep it in good order.
When my son got his first job and needed a car in a hurry to commute I helped him get his first car and in the search we wound up looking at a lot of Buicks that we liked and were at a good price and in good condition, particularly circa 2005 though we wound up with a '96 Park Avenue. It was my first experience owning an American car (title in my name for the cheaper insurance) and my own experience plus what I read indicates I probably would have done well with one of the 2005s.
My take is that at that age you don't pay that much more for the upbadged car but you're likely to find it in good condition but you get to enjoy the bling (the '96 is ahead of its time with traction control) and Buicks of that vintage have one of the best engines GM ever made.
A mandatory part of today’s safety features is a digital rear-view camera. Typically, this view pops up on a modern car’s central infotainment screen, but the Slate doesn’t have one of those. It makes do with just a small display behind the steering wheel as a gauge cluster, which is where that rearview camera will feed.
There's a tendency sometimes to leave certain words implicit. "It does not [even] take 21 days to form a habit." Some of those linguistic habits never leave their birthplace, giving you colloquialisms like "do you want to come with [us]" in the Midwest. Most don't have any regional association though, and sentences with dropped fragments sound perfectly normal, with your mind filling in the gaps.
In this case, the sentence is correct (though ambiguous) as written, but it has the same words you often see from a longer phrase with an omitted word implied through tone and other vocal cues, a pattern used commonly enough that I'd wager most people interpret it the same as you.
I've tried a few times (whenever somebody sent that sort of gift card my way), and it was always >$10 in explicit fees, plus all the menu items being silently marked up 20%.
Mind you, that's still not exactly "expensive" for delivery [0], but I can make better food both faster and cheaper than waiting, I can pick it up and actually use an insulated bag faster than waiting for delivery, and pretty much any other food strategy at least guarantees I'll actually have the expected meal and not have to waste my time with customer support (or money if I decide it's not worth the hassle).
[0] Imagine you're a driver, you incur $3 in actual expenses, the delivery takes 15min, and they have to wait 15min for the food to be ready (this is the thing that makes pizza delivery more efficient -- as soon as you get to the store there's another pizza waiting). Suddenly that's a $14/hr gig without any benefits and where you need to purchase a special insurance on top of things, assuming the only part Uber keeps is the 20% fee they're scamming you out of. Beyond that, you're at a much higher risk of bodily harm than doing something like construction, and if those aren't good drivers I don't really want to be encouraging more of them to be on the street, especially with time pressures (and if they are skilled ... that's less than McDonald's pays even before you factor in benefits).
The normal use case for `inline for` is when you have to close over something only known at compile time (like when iterating over the fields of a struct), but when your behavior depends on runtime information (like conditionally assigning data to those fields).
Unrolling as a performance optimization is usually slightly different, typically working in batches rather than unrolling the entire thing, even when the length is known at compile time.
The docs suggest not using `inline` for performance without evidence it helps in your specific usage, largely because the bloated binary is likely to be slower unless you have a good reason to believe your case is special, and also because `inline` _removes_ optimization potential from the compiler rather than adding it (its inlining passes are very, very good, and despite having an extremely good grasp on which things should be inlined I rarely outperform the compiler -- I'm never worse, but the ability to not have to even think about it unless/until I get to the microoptimization phase of a project is liberating).
1. That roofline curve idea applies to multiple processes, computers, and data centers just as well. If you have enough "cache" (disk, RAM, whatever), you can do a distributed matmul and actually effectively use every coprocessor at nearly 100% efficiency.
2. If you need f32 intermediate precision, you can approximate that with Kahan-like ideas and still take advantage of the f16 core, at somewhere in the 25%-50% efficiency range (still much better than the <10% you get by ignoring the tensor core).
reply