Hacker Newsnew | past | comments | ask | show | jobs | submit | stephencanon's commentslogin

What sort of engineering standards are these Cybertrucks built to?

Oh, very rigorous engineering standards. The wheels aren't supposed to fall off for a start.


Can’t be made out of cardboard either.

The Front Fell Off: https://youtu.be/3m5qxZm_JqM?si=DprOulmmDK-H76LX


I saw the title of the post, and I knew somebody would have referenced it.

We’ve taken it *outside* the environment

Same standards as e.g.

2026

Audi Q8 e-tron:

"Popular electric car recalled due to brake pedal problem" [1]

A problem with a "screw connection" (unclear whether this is a mounting screw or it serves some other purpose) can cause the brake pedal to malfunction.

or, in 2024

Audi Q4 e-tron, Volkswagen ID.3, ID.4, ID.5 and ID.7:

"Dangerous error in popular electric cars: brakes can cease functioning" [2]

It says that the ABS pump could drop off which would cause brake fluid to leak out which in turn causes the brakes to cease functioning.

[1] https://carup.se/popular-elbil-aterkallas-for-fel-pa-bromspe... (Swedish)

[2] https://nyheter24.se/nyheter/motor/1296418-farliga-felet-i-p... (Swedish)


Don't forget the doors opening by themselves...

https://www.autoevolution.com/news/vw-id4-recalled-over-door...


> Audi Q4 e-tron, Volkswagen ID.3, ID.4, ID.5 and ID.7:

> "Dangerous error in popular electric cars: brakes can cease functioning" [2]

> It says that the ABS pump could drop off

Using a mechanical ABS in an electric car might be part of the problem


As opposed to thoughts and prayers-based ABS?

Some cars are going with entirely electrically actuated brakes, either inboard on on-hub, compared to the E-Tron which uses traditional hydraulically actuated brakes. One uses an electric motor to wind something to tighten the spring clip by pulling it that then pushes the pads to the rotor and the other uses pressure to overcome the spring by pushing the spring to compress it and push the pads to the rotor. I'm guessing Audi didn't go with entirely electric brakes because they have a reputation for being harsh and difficult to modulate with the pedal, and Audi is supposed to be both a luxury and sport brand where pedal feel is important.

Are talking about brake-by-wire? Where brakes are controlled by electric only, and if electrics die, no brakes?

These are dangerous. Cars are not maintained to aircraft standards and will never be.


With electrically actuated brakes the default power off state is fully engaged. Meaning if the power dies the brakes lock up. That causes it's own issues, obviously, but a sudden deceleration is better than no deceleration at most road speeds.

edit: as formerlyproven below states, the ones currently for sale also have a hydraulic backup.


Insane take. They will be vastly more reliable than hydraulics.

Brake by wire passenger car brake systems are still hydraulic... and all of them have a mechanical backup. There is not a single car on the market today using electromechanical brakes.

Unless you're talking about electric parking brakes in a thread about ABS.


Look up cybercab. Their new disassembled manufacturing method cannot support hydraulic lines.

Just remember to pack a bottle of spare magic smoke incase something leaks!

Which cars?

No, just any combination of electric regenerative braking combined with electrically controlled brakes.

It is an electric car after all


Well, the wheels may fall off, the body panels may fall off (weak glue), but the rest of it is OK right? Well, apart from the bulletproof glass?

So worst case you're rolling down the road on a chassis with no body panels, except you're not really rolling if the wheels fall off.

Hmm.. good job we're not letting in those cheap Chinese EV's and sticking to this top quality homemade stuff.


And the hitch might fall off when towing over a pot hole.

The same engineering standards as other Teslas are.

Meanwhile, about 63% of Tesla Model Ys failed their first mandatory inspection in Finland. The Tesla Model 3 did a bit better at 59% of cars failing their first inspection for the same model year. However, they're faring a lot worse than the third worst car, the Dacia Duster, at 23%, or other EVs like the Volkswagen ID.4 at 6%.

https://www.hs.fi/visio/art-2000011988306.html


To the ones of people who like to move fast and break things.


The original vibe engineering

> What sort of engineering standards are these Cybertrucks built to?

'Vibe-Engineering'


And they’ll probably just tow the recalled trucks outside the environment.

Into another environment?

No, no, no. it’s been towed beyond the environment, it’s not in the environment

Rush of Australians to the comments section.


IDK but millions of other vehicles are recalled every year and never mentioned here.

At least they aren’t using Full Self Engineering (yet).

Blame it on a loose nut behind the steering wheel.

Indeed. The insane part is not that Tesla built an absolute dumpster fire of a vehicle, that was something that should have been obvious at every point.

The insane part is the number of people who were somehow able to put up $120k for one, and proudly boast how awesome their new car was even though it spent most of its time in the repair shop or breaking doing very basic things, and failing to do "Truck" things that even my hatchback can manage.

Presumably it's not a coincidence that so many of them were bought by brand new weed shop owners.


I once saw a Ferrari trying to negotiate speed bumps and a roundabout. On a road that has a 30 kilometre speed limit.

For me a car is essentially a tool so it needs to be practical. But for others it's a hobby.


It looks like they were designed by a disruptive startup unburdened by the history and experience of designing and building cars.

It was super delayed and I think that's because they couldn't execute in all the ways they promised they would. The final product is very rushed and pretty different from the initial promises. I think they got into "Let's just ship SOMETHING" mode as the delays were getting insane.

‘We threw the rule book out of the window’

Also worked very well for the Oceangate Titan submersible.

> “brake rotor stud holes may crack and allow the stud to separate from the wheel hub.”

Possible

While mechanical failures can happen in all companies, that do sounds like an inexperienced design (maybe from Tesla, maybe from a partner?)


I can't find pictures online but I'm assuming since it only affects the 2wd and it says if the rotor cracks the stud might leave that the rotor is also the hub.

Doing a half baked job on a part for your super low volume "we only make this to advertise a low starting price" model is something just about any OEM would do.

I bet their supplier just took whatever Chevy Van rotor they had that was close and modified it to fit and as a result it got a little thin somewhere.

Edit: Nope, I couldn't find a picture but I found pictures of big brake kits for the 2wd and clearly it's not an old (read: cheap) integrated hub and rotor.


A 23 year old startup.

Tesla standards?

We have a ground-source heat pump for our ADU. We did it because we were curious about just how efficient we could make the house, but I don't expect that it will ever break even financially vs a modern air-source system with resistive backup in our climate (northern New England, typically very few –20˚ nights, –10˚-0˚ more common with daytime highs in the single digits).

It works great, but it's hard to see a way to it making sense for most folks here.


You might still get the most out of it when the AMOC collapses.

I'm not aware that AMOC collapse is going to significantly alter New England's climate as it will Europe's.

Prior to the current generation Intel designs, Apple’s branch predictor tables were a good deal larger than Intel’s IIRC, so depending on benchmarking details it’s plausible that Apple Silicon was predicting every branch perfectly in the benchmark, while Intel had a more real-world mispredict rate. Perf counters would confirm.

Must be the water.

No Kimi You will not have the drink.

The CORE-MATH project authors, most of whom are French academics (including the author of the linked paper).

I don’t know of any interesting work in this space that came out of Red Hat, why do you suggest them?


GNU libc


"I'd like to thank my mother, Ayn Rand, and God" is the usual example.

Yes, you can reorder the list to remove the ambiguity, but sometimes the order of the list matters. The serial comma should be used when necessary to remove ambiguity, and not used when it introduces ambiguity. Rewrite the sentence when necessary. Worth noting that this is the Oxford University Press's own style rule!


I always heard this one...

We invited the strippers, JFK, and Stalin to the party. [three groups invited - strippers, a president, and a premier]

We invited the strippers, JFK and Stalin to the party. [the president and premier are strippers]

Very different visual conjured by those two sentences.


"John helped his uncle, Jack off a horse"

"John helped his uncle Jack off a horse"

Two very different outcomes...


shouldn't there be another comma after Jack?

John helped his uncle, Jack, off a horse.

Because while speaking it I only pause after uncle and "Jack off a horse" together next. feels like there should be another pause after Jack?


I'd prefer:

We invited the strippers, JFK and Stalin, to the party [two strippers, named JFK and Stalin]

if the goal is to minimize ambiguity.


I can see it being tiresome to read text where the author is continuously interjecting clarification with brackets.


The square-bracket clarifications here are meta-text designed to absolutely clarify the intended reading of the preceding text, so that the reader can contrast their understanding with the intended one.

There is no suggestion that one would do this in "regular" text.


If JFK and Stalin were strippers, there’d be a comma after Stalin to denote the parenthetical clause.


I mean first off: no the exact same image is conjured because we are reading this in context of knowing who jfk and stalin are and we know they aren't strippers and all language is contextual.

That said:

We invited the stripper, JFK, and Stalin to the party.

We invited the stripper, JFK and Stalin to the party.

The supposed ambiguity is back. Although again there is no ambiguity to the reader. The juxtaposition of the two versions wouldn't work as a joke if there was any ambiguity


Fine, I’ll be boring…

We invited the strippers, Krystal and Britney, to the party.

We invited the strippers, Krystal, and Britney, to the party.

And yes, if strippers is made singular, poor Krystal becomes Schrödinger‘s Stripper.


Removing the Oxford comma not only fails to resolve the ambiguity, but introduces yet more ambiguity by implying Ayn Rand to be God.


Just put the colon there if you need to introduce a list, it's one of its functions. "I'd like to thank: my mother, Ayn Rand and God". The same goes for that "two strippers" example: "We invited the strippers: JFK and Stalin, to the party".


I want you to know that I would only write this in a discussion nitpicking about grammar: :)

> "I'd like to thank: my mother, Ayn Rand and God".

A colon should not connect a verb and its objects; generally you need an independent clause before the colon (i.e., a clause that could be a complete sentence). One could properly say,

  I'd like to thank the following: My mother, Ayn Rand and God.
Also, these examples leave ambiguity. Your mom could be Ayn Rand, and if she was, then you might very well think she was God, or be making a joke about it.

> "We invited the strippers: JFK and Stalin, to the party"

Nope. A colon isn't a parenthetical in the middle of a sentence; that is, you can't continue the sentence after a colonic phrase (there's no such thing so I made up that term :D ). And again, the clause before that colon is not an independent clause. One can use parentheses (of course) or em dashes for parenthetical phrases:

  We invited strippers (JFK and Stalin) to the party.
  We invited strippers - JFK and Stalin - to the party.
A proper colon might be as follows:

  We invited strippers to the party: JFK and Stalin!
But I'd put an em dash there (and to heck with LLMs and their em dash overusage).


Enlarging a branch predictor requires area and timing tradeoffs. CPU designers have to balance branch predictor improvements against other improvements they could make with the same area and timing resources. What this tells you is that either Intel is more constrained for one reason or another, or Intel's designers think that they net larger wins by deploying those resources elsewhere in the CPU (which might be because they have identified larger opportunities for improvement, or because they are basing their decision making on a different sample of software, or both).


I mean, he's comparing 2024 Zen 5 and M4 against two generations behind 2022 Intel Raptor Lake. The Lion Cove should be roughly on par with the M4 on this test.


That would fall under "more constrained", due to process limits.


Your students should be able to figure out if a computation is exact or not, because they should understand binary representation of numbers.


For throughput-dominated contexts, evaluation via Horner's rule does very well because it minimizes register pressure and the number of operations required. But the latency can be relatively high, as you note.

There are a few good general options to extract more ILP for latency-dominated contexts, though all of them trade additional register pressure and usually some additional operation count; Estrin's scheme is the most commonly used. Factoring medium-order polynomials into quadratics is sometimes a good option (not all such factorizations are well behaved wrt numerical stability, but it also can give you the ability to synthesize selected extra-precise coefficients naturally without doing head-tail arithmetic). Quadratic factorizations are a favorite of mine because (when they work) they yield good performance in _both_ latency- and throughput-dominated contexts, which makes it easier to deliver identical results for scalar and vectorized functions.

There's no general form "best" option for optimizing latency; when I wrote math library functions day-to-day we just built a table of the optimal evaluation sequence for each order of polynomial up to 8 or so and each microarchitecture and grabbed the one we needed unless there were special constraints that required a different choice.


Ah, I didn't know of either scheme. Still, am I right in that this mainly makes sense for degrees above five or six or so?


You can often eke something out for order-four, depending on uArch details. But basically yeah.


Ideally either one is just a library call to generate the coefficients. Remez can get into trouble near the endpoints of the interval for asin and require a little bit of manual intervention, however.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: