It's also how LLMs thing (not accusing the author of slop here, just making a slightly unrelated observation).
If I ask my Claude (Opus 5.5) about how some Claude feature works, its natural inclination is to start reversing its own binary with strings instead of looking at Anthropic's documentation. It will, entirely unprompted, try to figure out how its environment works in Claude Code cloud and reverse–reverse-engineer Anthropic's Rust binaries to figure out why things happen the way they do, and what it needs to do about it.
I asked Claude (Sonnet 5.5) how to turn off the diffs side panel and it told me that it does not know. I asked it to Google AI mode on the search page and it told me to type /diff.
However Claude and I did write code all the day long and a long job went down from 2+ hours to 30 minutes, same final output.
That's because it was trained to waste as many tokens as possible. If you would've tell it to look for a Linux exploit, would looking at the binaries of Linux also would be a wise choice instead of looking in src?
Yep. I didn't get into it in the article, but in reality the previous owner had already (accidentally) thrown the SD card in the trash, and was pretty much asking me if they want me for them to go digging through their trash to find it, and I told them not to worry about it.
I’m a little confused by the description of that interaction - why wouldn’t you take the SD card? Did they want to charge extra for it? Seems like even a 10 year old map (plus apparently other functionality) is something you’d want.
In any case, glad it took you on the wild goose chase that led to the write up!
Because OP refused to believe that he, a technically literate person, wouldn’t be able to find the correct settings page and assumed that the previous owner was just not the brightest bulb. Which, in my opinion, is reasonable, because a reasonable person wouldn’t expect infotainment systems settings being only available with maps SD card.
I started using an "engineering notebook" for things like that. They are called other things in other lines of work or school (lab notebook, captain's log, ...). In an actual engineering notebook, you would use a sewn binding where you can't add or remove pages, the pages are numbered, you have an index in the front, entries are made in strictly chronological order, you'd have someone else witness and sign pages, draw a diagonal line across empty areas so you can't remove things, only cross things out with a single line, etc. All that because you might need to prove you invented something and other legal CYA reasons. Most of these requirements, you don't need for personal projects, though. But the things that are useful: keeping track of projects, what settings you've tried, what the results were, and so on. That way you don't have to reproduce it from scratch again later. "What settings did I need use to create the gerber files in kicad for manufacturing? How did I set ___ in the oscilloscope? What parts did I use to make that thing? What things did I wish I had checked last time? Where did I put those files?" You can also print things out and tape them in for reference.
I just use composition notebooks which are like $1-$5 depending on the quality, store, sales, etc. That's also what we used as a lab notebook in chemistry class when I was in high school, which follows a similar discipline, so to me it's fitting.
Easy fix: drive old garbage like me, lol. Or grab an old classic, throw some money at it, and you've got something cool that costs less than some new POS. My '78's clock sets with a knob.
I've heard that some Subaru models have 3 clocks in the view of the driver, all on the CAN bus, one being the nav and another being a remote display for the nav, and one that's just there, of which none sync to each others. I've looked up how it happened, and discovered that to just to add insult to injury, at least one uses a uC that has RTC feature but no XTAL pin for 32.768kHz or one of any sane base 2 divisions and uses CPU clock of 15MHz or something for it, so it just drifts away.
I don't understand how Subaru stays in the business. Just how. It's a 4 digit time. Just broadcast it on CAN. Why not.
Also: a lot of these older, completely rectangular sat nav units actually follow what is colloqially known as "2DIN" standards. (cheap import Amazon)Aftermarket units are available and those can be swapped in with just a bit of work. That's what most people do when they have issues with the stock unit installed for their markets.
My smart fortwo has two clocks, one on the radio, one in the driver’s dashboard;
Both go wrong over time, but both in two different directions, so the trick is to calculate the „average“ between both, and you end up with roughly the correct time.
My first car, a Hyundai i10 I bought back in 2010, was very basic and had few features. It only had a simple radio with an FM/AM tuner and a CD player, no navigation system.
What surprised me most was that the clock adjusted itself automatically whenever daylight saving time started or ended.
There was no calendar or date setting, so I assume it must have received the time either via RDS (I don't know whether RDS supports this) or through a built-in DCF77 receiver.
All my subsequent cars had far more features, yet none of them had this one.
I also wonder why cars with built-in GPS navigation don't synchronize their clocks using the time from the GPS signal. Heck, the car could even determine your time zone from your GPS coordinates, so there'd be no need to configure anything manually.
Checking wikipedia, there is optional support for transmitting time and date in RDS, with an explicit usecase of updating the clock in cars.
> I also wonder why cars with built-in GPS navigation don't synchronize their clocks using the time from the GPS signal.
GPS doesn't transmit timezone information, nor does it transmit leap seconds. It transmits a very, very accurate UTC minus 18 seconds, but without access to the internet for timezone database updates or (until recently) leap seconds, it's a bit useless for updating a clock.
I'm not sure if you guys read the article, but OP's car does use GPS for the time, applying a user-settable offset (or in OP's unfortunate situation, a fixed offset).
It did in the 1999 Mercedes-Benz S-Class I purchased after 15 years of insane depreciation.
I thought I broke the clock with a DIY repair because it too couldn’t be manually set (aside from enabling a +30 min offset). Luckily it fixed itself after driving out of the garage when it found a GPS signal.
MOE: How long has that been in the soup, froghead?
CURLY: About uh--- [Looks at the three watches on his wrists.]
MOE: Hey! What’s the idea of the three watches?
CURLY: That’s the way I tell the time.
MOE: How do you tell the time?
CURLY: This one runs ten minutes slow every two hours. This runs twenty minutes fast every four hours. The one in the middle is broken and stopped at two o’clock.
MOE: Well, how do you tell the time?
CURLY: I take the ten minutes on this one and subtract it by the twenty minutes on that one. Then I divide it by the two in the middle.
MOE: Well, what time is it now?
[Curly grabs a clock from the inside of his jacket pocket.]
CURLY: Uh, ten minutes to four.
MOE: Hmm so it is!
[Moe grabs the clock from Curly and hits Curly on the head with it. The clock breaks.]
My '97 Range Rover has an analogue clock with a quartz-controlled movement, and a button to set the time.
It gains anything up to an hour a day, particularly when my six-year-old is helping me clean the car and he pushes the button to watch the hands go round really fast.
It's fine. Time is time, we get there when we get there.
Who said anything about getting there faster? What if you are stuck in traffic and need to know if you will be on time or late? What if you see a fuel station and want to know if you have enough time to top up your tank? There are actually quite a few valid reasons that you need to know the time while travelling in a vehicle
If you’re stuck in traffic and you know you are goung to be late, you can phone ahead to let them know! This won’t be possible without knowing the current time, something which a reasonably accurate timepiece will inform you of
But I don't need to know the time when I'm going somewhere. I need to know the time before I set off, so I set off at the right time, and then I get there at the right time.
It's not just cars. Our house alarm had two clocks, one in the alarm itself, one in the display/control panel, and the two didn't talk to each other, not just for that but everything else as well, for example motion sensor names.
It's now in an e-cycling waste bin somewhere, replaced by an ESP32 I/O board and Alarmo in Home Assistant, at 1/20th the cost of the original PoS alarm.
There's a certain sense in which competence in some domains repels competence in other domains. If Apple made a car the rear fender would come off just as you opened the drive side door.
If I had to hazard a guess as to the mechanism as it would turn up in an automotive company it is that automotive engineers have the SAE to set professional standards and that software "engineers" have nothing [1]... And if an organized group has a fight with a disorganized group the result is predictable.
[1] the ACM is a professional associate for computer science professors.
> There's a certain sense in which competence in some domains repels competence in other domains. If Apple made a car the rear fender would come off just as you opened the drive side door.
This feels like a logical fallacy but I'm having a terrible time disproving it at anything near large company scale
But then the base model with only one clock would need a clock that reads the can bus, and something putting the time on the CAN bus. Instead they just put a cheap ass standalone clock in the base model. And this clock trickles up to the models with more features.
IIRC at least one of the two other displays optionally take turn-by-turn remote guidances from the nav. So it can easily have a time data type added to that data, and they didn't. Though, all I researched for was for an outdated model of a car, and it seems one of three desynched clocks appear to be not be present anymore, and the desynced clock count must have gone down by one, unless they compensated for the loss by adding one.
And their engines have struggled with gasket issues and oil consumption for a long time. People here love the AWD, though (my state enshrines AWD as a special class in law, artificially driving some demand).
This is a pretty common misconception (it's worse with 4WD, since at least with AWD you sometimes have torque vectoring and other features which do improve your ability to stay on the road while accelerating), and it gets a lot of people hurt. In most reasonable scenarios (not doing anything dumb like accelerating through the curve), you're more likely to fly off a winter curve with AWD than FWD.
The big advantage is when accelerating. It's a nice quality of life feature. I typically switch to AWD when I move to winter climates. It's not as safe though, _especially_ if you have little enough training that you're relying on intuition to make that conclusion that AWD must be safer.
Sorry, it's hard to capture all the nuance a conversation like this needs without writing so much that the length keeps people from capturing the important parts.
If you want to maximize speed through a curve plus straightaway, you should pick a path through that curve minimizing curvature at the end so that you are able to accelerate out the end of the curve. As a reasonably good approximation for normal driving, your tires have the same traction regardless of the direction of the load being placed on them. That gives you a budget of net force you're able to apply. If you're near the sideways force limit going around a curve and add acceleration (or braking -- always do your braking before you enter the turn, or gradually reduce it as the curvature increases), the wheels with those extra forces will break traction. The solution is to apply that acceleration at a point of milder curvature.
Taking that idea back down to safe driving speeds, the same idea exists when considering crossing a margin of error rather than the ultimate slip threshold. All else equal, accelerating as you come out of a curve, if you've chosen your path through the curve correctly (really important to get that right; there's a curve back near my hometown which has taken out half a dozen people over the years because its curvature increases at the end and they haven't planned for it), will minimize your time through the curve and the following straightaway at your chosen safety threshold.
However, all of that is predicated on minimizing time assuming perfect knowledge and perfect driving. The acceleration itself is actively harmful from a safety perspective. In FWD, the wheels losing traction are the ones connected to your steering wheel. That's not ideal. In RWD you're prone to oversteer and spinning out of control. In AWD you can get either or both of those failure modes. Moreover, navigating typical curves, you're so far from slippage that the technique of accelerating at the end just means you'll go from the speed limit to something greater. For winter curves, the margin of traction uncertainty is so great that you can't gain much from accelerating out the end anyway (and have a pretty bad risk profile -- accelerate on a straightaway and slip a bit and with a bit of training you're basically fine, slip off a curve and you have a really bad rest of your short life potentially).
Safely accelerating in a curve is a learned skill though, and most people don't care that much about driving. Unless it looks like they're interested in more depth, my advice is always to brake before the curve and accelerate afterward. It's safe, fast enough, and easy to remember.
A powered rear axle tries to overcome the front. This why you can do 'drifting' on a RWD. And when you are in the curve there are different forces and road resistance on your front and rear axles (your outer side is loaded and inner is not) so if you do accelerate on a 4WD/AWD you are asking your car to spin out.
I think this is true, but there are still reasons to like RWD:
- Smaller turning circle because the front wheels can turn further without complicated engineering.
- More traction when accelerating on a wet surface, due to weight transfer backwards. (1)
- Usually better weight distribution due to engine layout.
There are other downsides in addition to what you mentioned:
- Any scenario where you will definitely lose traction, then rear-wheel drive becomes harder to control. There was an old Top Gear episode with a steep grassy hill, and drivers of the RWD cars could not get up it. (2)
- The transmission tunnel required for RWD
- Reduced fuel economy due to having to turn the power 90 degrees (assuming the engine is mounted in the front). (3)
1) This one may sound like unnecessary acceleration, and it probably often is, but there are situations where it can arise, especially if you prefer older cars. There is one particular hill I used to encounter frequently, where you are trying to turn onto a main road, so you want to accelerate fairly briskly to avoid waiting for a huge gap in traffic. On economy tires, in the wet, you can easily reach the limits of FWD. I would see many cars time a gap, and then have to rapidly decide to give up on it when their acceleration was much slower than they expected due to wheel spin. I think there may have been bus lane paint on the road too, but I'm not certain. Obviously AWD is even better than RWD for traction, but at that point, that is usually only when requiring rapid acceleration.
2) I suppose with perfect/computer assisted driving, RWD would out-perform FWD.
3) I once searched, and had a very hard time quantifying this, but it is almost certainly a few percent, and perhaps more. There are also weight differences involved because of the long drive-shaft, but I think the effect is larger than that. It is very hard to find an apples-to-apples comparison.
RWD cars also feel much nicer to steer, even when just driving around town. The fact that the steering axle doesn't receive engine torque makes working the steering wheel while moving feel smoother and less jerky.
AWD provides more traction while accelerating, which makes it safer, which is why people use it in winter... Saying it's less safe is ridiculous.
Now if you happen to be really good at driving, it's even safer because you can recover from a slide faster, again because of the increased traction during acceleration. Once your tires have traction again the additional traction moves you out of the slide faster.
The only problem here is people going too fast which can happen with any car.
Why does the additional traction from accelerating meaningfully increase safety?
I'm struggling to come up with a reason why the faster acceleration itself would be a safety issue (maybe trying to get out of the way of a truck that can't stop or something? it seems niche), so let's instead look at the increased control during acceleration (from less force on the steering tires).
When does that matter from a safety perspective?
In a straight line, it doesn't. Let off the gas and try again. No matter what you do, you're going to go in your intended direction of travel or stay put.
So that brings us to accelerating around a curve. Don't do that on public roads in normal cars in winter conditions. The risk profile is terrible. If you misjudge it just once you're in a ditch.
Even your slide example seems strange. Could you describe the conditions of a slide where applying power helps you gain traction faster than not applying power? The only thing immediately coming to mind is reducing the yaw force in certain oversteer scenarios with a counter-steer and a bit of power, but in that case whether AWD helps or not is extremely situational and depends on the exact build of the car.
I think I agree with you regarding AWD generally not increasing safety except allowing accelerating earlier out of curves - which implies pretty hard (unsafe?) driving. To be fair to people who thinks AWD helps, some people do drive this way. Also, there's the usually-unspoken implication that sports cars are created with this sort of driving in mind, even on public roads - no matter how often the put fine-print about "controlled roads" at the bottom of their ads.
But aside from all that, I think there is a scenario where applying power can help gain traction faster. In a FWD, if someone has gone into a corner too fast, or encounters a more slippery surface than expected, and the vehicle starts to understeer straight-ahead, I have heard of the idea of accelerating to get the vehicle to continue around the corner. The turned front wheels, while still not gripping properly, spin and pull the front in the direction of the corner more than would happen from just braking or maintaining speed. I'm probably not explaining that well, and it doesn't sound safer than just braking! I am fairly certain I remember seeing something which I think is related, in the Owner's Manual for some old front-wheel drive Japanese sports car. Something like an old Honda Prelude or a Toyota Celica. They played up the safety of FWD "pulling you through" the corner. None of this makes the case for AWD though. I can only see the advantage I wrote above about accelerating earlier/harder, and I think you already addressed that.
But that's now AWD, that's the drivers. I lived in CO so know how to drive in the snow. I have AWD, but most important, I have snow tires.
I love driving up to Tahoe in a blizzard. I'll happily be doing 25 mph going uphill as it's roughly the safe speed for the conditions. I'll constantly be passed by AWD SUVs going 40-50, while going uphill.
Then we summit Donnor Pass. I'm still cruising along at 20-25. You can see the fear in these SUV drivers' eyes as they start sliding down with their "all season" tires. Invariably, they'll then slow to 5mph with their flashers on in the right lane and I can cruise down t 25mph and pass all the cars that flashed at me or gave me a look going up.
It is baffling to me that CA loves to enforce chain control but exempts AWD without actual winter tires from it. It's almost never the 2wd drive cars with chains that pile up, it's always the AWD SUVs. If you're going to enfore chains, at least do it usefully and only exempt actual winter tires. Then it would be useful, not just a massive slow down.
In a very long winded way, couldn't agree with you more!
I can vouch for this: whenever there's heavy snowfall (in Indianapolis, so less common and severe than Michigan, but same ballpark), nearly all cars I see stuck in snowbanks are AWD/4WD.
Personally, I'm comfortable driving anything but a RWD pick-up truck with no payload in snow and ice, try to drive at an appropriate speed for weather conditions, and rarely have trouble.
Even the RWD is fine with some training. I used to deliver pizza in the Minneapolis suburbs in a Miata. It was fine. I had to get a running start to pull out of a neighborhood in a steep depression once during a snow storm since I didn't have snow tires, and anything more than 4-5" on the road would mean I had to stay home (too much force pushing on the front bumper), but the RWD was fine. It was nice having such a small interior space because the heater worked instantly.
Yep. I grew up in Wisconsin, and I saw the same thing all the time. My dad used to quip that those drivers figured that because they had 4 wheel drive, they must also have 8 wheel brake.
And on gravel / dirt roads in the country - our P10 Primera's AWD helped my ex-wife avoid a nasty crash on one such road with our kids in the back, if she'd been in a FWD they likely would have collected a tree.
The marine corps traffic safety training teaches that many more accidents are avoided by the steering wheel and gas pedal than the brake. At high speeds, the brake can only avoid things many yards away while the steering wheel (and gas in winter for traction) can avoid obstacles just feet away.
It makes slides more controllable since you will spin around the middle of the vehicle rather than the front or back. All four tires interacting under power with snow also helps to restore traction. It isn't a magic wand of course that subverts the laws of physics.
I hadn't heard this before, so just trying to think it through.
Are you saying that when when cornering at a constant speed, if a slide starts, the vehicle will rotate differently to a front-wheel-drive or rear-wheel-drive vehicle?
Or that when cornering, if you accelerate enough to spin the wheels, then an AWD vehicle will rotate around the middle?
For the 1st case, when you slide because there is not enough traction available for that speed, on that surface (e.g. driving too fast, or encountering ice), I wouldn't have thought the axles that power is delivered will make a large difference to how the vehicle rotates. Either the front loses traction, or the rear, or both. Wouldn't the weight distribution be more relevant than the driven axles?
For the acceleration example, again I think of either the front losing traction, the rear, or both. Front-wheel-drive would lead to the front losing traction, RWD the rear losing traction, and AWD ... I'm not sure. With some systems, it's plausible that neither the front or rear would spin until both lost traction simultaneously. I tend to think a 4 wheel slide would be similar to understeer in a front-wheel drive car, but perhaps a little more like sliding sideways out of the lane, rather than plowing straight ahead. This doesn't seem to me like rotating around the middle, so I feel like I'm missing something.
To be honest, the idea of AWD making the vehicle rotate around the middle (and rear-wheel-drive making the vehicle rotate around the...front?) seems like the persistent and widespread idea of AWD allowing a vehicle to corner faster at a constant speed. My thought is that it's a misconception, but I'm happy to be wrong!
There are many types of AWD systems and they don't all behave the same, but generally speaking an AWD car will still power understeer in a corner, but less so than a FWD car.
Are you referring to descending long hills using engine braking rather than brakes to avoid brake fade?
I can see AWD allowing better engine-braking than FWD in some low-traction situations, but I don't think it would make much difference if trying to stop quickly. I think for modern cars, using engine braking would almost never provide quicker stopping, or more controlled stopping than brakes alone.
Regarding three pedals, you could downshift in an two-pedal automatic too couldn't you?
And I've noticed a ton of Subarus doing stupid stuff like cutting people off, which I attribute in part to their EyeSight ADAS system marketing. (Which is nothing special anymore)
10yr ago I might have bought this BS but these days I own AWD and FWD of the same car and live where it's all hills so you're not pulling that fast one on me.
The "weakest links in the chain" when it comes to snow driving are largely centered around putting motive power to the ground without breaking traction. Think like steep hills, crossing the piled up slush between lanes, getting moving from a stop, etc. Except inn rare circumstances, your "this feels decent speed" is attainable regardless of AWD so it's not like you go any faster.
>The results for roads covered with ice or snow showed that injury crashes increased by 19-31% with AWD. Similar results were found for head-on and single-vehicle crashes. No significant difference was found between Permanent and Automatic AWD. On icy or snowy roads, AWD cars had a 13-15% higher impact severity than 2WD cars, which corresponded to an 8-10% increase of impact speed for AWD cars. On dry or wet roads, no differences were found between AWD and 2WD. Although based on a limited material, the survey indicated that AWD and 2WD cars had similar distributions of winter tires. The results suggested that AWD may lead drivers to underestimate the level of available friction on icy or snowy roads and therefore to drive at faster speeds than they would do with a 2WD car
Meh. Tires matter more than drive layout. Best winter car I ever owned was a 1997 Miata with Blizzaks. Beat the pants off our AWD Lexus RX on all-seasons.
Every manufacturer is struggling with oil consumption. They switched to too soft piston rings (to try and reduce friction), which causes more oil to end up in the combustion chamber.
The FB engine that was introduced _12 years ago_ no longer has the gasket and oil burn issues that the EJ engines had. Instead you can talk about the belt-driven CVTs that supposedly never need an oil change which is true, if you like them to explode. The planetary eCVT on new hybrids should be fine though.
In western Washington State everything is on a hill and it feels as if the road is always wet all the time due to rain. Subaru is the most popular car brand in the whole state because it handles these conditions well.
OTOH, if the WA temp is below freezing, then - to get up the next hill when wearing summer tires -- you may have to back up a block or two to build enough momentum.
If the previous hill was too close to the next, you've got a quandry.
Wet all the time isn’t a serious loss of traction. Wet after a dry spell is, and everyone there forgets it every time. You will have delays on the highways due to wrecks.
Nobody has the slightest clue how to deal with ice.
I would counter that it depends on the kind of tires you are running and anyone selling tires in Cascadia that aren’t wet rated should be locked up for reckless endangerment.
Seattle isn’t that bad in wet pavement, unless it hasn’t rained. Which only happens about four to six times a year but that’s still enough that people should pay attention.
Even ultra high performance summer tires can tolerate wet. You have to deliberately choose to run slicks (or bald tires) to get something that can't handle rain.
People who make fun of greater Seattle for freaking out over snow generally don't understand just how extreme the topography is out here, even downtown and on the highways. It's not Ohio or the Northeast with at most gently rolling hills. It's snow PLUS ridiculously steep terrain.
As someone that grew up in Pittsburgh and lived in the Seattle area for 5 years, Seattle's topography isn't that extreme, the drivers are just awful. My favorite example is this interstate exit: https://www.youtube.com/watch?v=Tij3SRs_5wg&list=PL_ZOHdoqrC...
To be fair, SF is the hilliest city and Seattle is #2 last I checked. But nobody in Seattle complaining about everyone else in Seattle is talking about people sliding into intersections downtown, they’re talking about the number of completely avoidable ditches and accidents in places that are inexcusable.
I lived in Seattle longer than I lived in the Midwest. I’m not talking about people spinning out on big hills, and I never got the impression other people were. It was seeing cars spun out on curves or gentle inclines that rated bagging on Seattleites for being dumb as a box of hammers on the road.
I’ve seen two cars on their roofs on an uphill in Seattle. How the fuck do you even DO that?
Tires are the big thing. The vast, vast majority of Seattlites aren't swapping out for a winter tires set in winter, and they aren't running a 3PMSF-rated all-season / four-season / all-weather tire, either.
The first time I went out of my way to research tires I was shocked how much braking distance typical tires lose at 50% wear. I would have expected 50% to be when things first start getting bad, not 25%.
These days I aim to replace my tires at about 50% wear and no later than 60%. New tires make you corner so well it almost feels like a slightly newer car.
It kinda feels like insurers should give you a discount for running good/great tires. It's such a safety benefit. The externality of crappy tires isn't captured in their sale price.
If it mattered enough to be commercially viable, they totally would be. Actuaries are ruthless.
Since this is not happening, we can infer that insurance companies believe the benefit to their risk profile is less than the cost to administer the program and issue discounts deep enough for customers to participate.
It seems counterintuitive that stopping distance is not that big a deal, but it must be true.
Certainly it would be cheaper for them to run a program where you take a photo of your tires every few months, vs. the program they all have where you plug a surveillance device into your car. So the true benefit from good tires must be far less than driving style.
Hard to square this tire company "change them ASAP" marketing stuff with the tread depths that new performance tires (think like "street legal" race tires) come with.
Performance tires aren't going for economy. They come with less tread height because it has better behavior under side loads (turning). And less rolling resistance and marginally less weight.
Wet is ~40-50% reduction in traction on asphalt, and it's even worse for paint. That seems pretty serious. (Of course, surface oils can be even worse.)
Typically stopping traction is only mildly worse on wet pavement, but lateral traction takes a pretty big hit. It's a serious loss in any case, but the limits for good tires on dry pavement are so high that even at 50% of that you are totally fine if you just drive normal. With modern tires if you're driving more than like 3/10 the ability of the car you're probably driving like an asshat.
uhhh If one of your tyres is going bad then at least two need to be changed on all vehicles, and, more importantly, some investigation needs to take place why only one wheel is going bad (poor rotation strategy, wheel alignment, bad brakes... etc)
Punctures and pinches can take out one tire. As can psycho exes, teenagers whose parents didn’t raise them right, manufacturing defects. And the passenger side tires are more susceptible to sidewall damage from poor parallel parking.
The sting of losing two tires due to a mistake is a lot less than that of losing all four when three are still good. It’s one extra versus three extra tires.
What benefit do you see? I live in Idaho and go skiing frequently. I’ve driven, up and down the windy snowy icy road, every vehicle I’ve ever owned - RWD truck, FWD minivan, and now an AWD Model Y.
They’re all a little different, but I’ve not once found myself in a situation where I noticed AWD really making a difference. Maybe in parking lots? Certainly not enough that I would ever make a purchase decision based on drivetrain.
Feels like everything I’ve driven has been modern enough to have great stability control, and that’s all that matters.
It is easy. I drove for several years a car with permanent AWD (Audi A6 Quattro), then moved to one that can switch between front drive and AWD. I don't remember the last time I switched to AWD (mileage is better if I leave it on front drive only).
In my experience you still have to have at least M+S tires with AWD. You can't try the pass with summer tires or plain all season (non M+S, non 3 mountain peak) tires just because you have AWD.
As someone who traverses the Cascades on a regular basis during winter months, I've found all-weather tires like Michelin CrossClimate 2 to be phenomenal for year-round driving.
Yep, I use CC2s as well, year round. They are good tires, and 3PMSF. Paired with AWD, Oregon is fine with them being used in lieu of chains when going over the Cascades.
Michelin CrossClimate 2 are M+S/mud & snow. Or was that your point? The snow tyre lines are mostly due to the rubber compound which is far more effective <-4c, but wear quickly >7c (ideally don't use until cold)
Of the EJ20 just to keep it easy to play with the blocks without needing a modified type certs right? None of my business but they seem to be a bit unserious somewhat within bounds of a car company.
> I don't understand how Subaru stays in the business. Just how.
Not sure if relevant, but Toyota has been buying parts of Subaru over the years, in 2020 ~ they have roughly 20% of Subaru. I'm not sure if its one of those "Microsoft investing into Apple" or "Intel investing into AMD" type of moments (which tbh, I wonder if Nvidia should be trying to invest in Intel or AMD since they're doing so freaking well its horrifying).
I've heard that the redesigned 2026 Outback has a much nicer interior. Lots of physical buttons and supposedly more stable infotainment software.
We've been pretty happy with our 2020 Outback so far (bought it just before the chip shortage). My only complaint is the stability of the infotainment software. The big portrait screen is actually quite nice, and with CarPlay I don't really care about the rest of UI and software. Except that the stupid thing just crashes all the time.
yeah, I've heard that too - I have a new forester - it's def slow/finicky
maybe I'll get a outback instead next time, they are kind of the same to me though the hybrid on the forester I have def makes it a more pleasant less CVT'y ride...
I'm hoping they bring the new Outback design (both exterior and interior) to the Ascent. The 2027 Ascent is already out and unfortunately it's the same old styling and infotainment.
I can confirm. My Subaru 2023 forester's infotainment unit got replaced after 1.5 years due to it randomly crashing, and Carplay being unstable. On a long road trip, if you go to get gas, unplug your phone, and try to replug, just wouldn't connect. Even rebooting it. Nothing.
Also had my 2 front speakers stop working, which mean handsfree didn't work since phone calls are routed to front speakers only. Had to unplug the car battery, wait, and then the speakers worked.
Does anyone know why the mini display on the top, lags so much when changing music tracks or even Radio stations? It even sometimes doesn't update the radio station, and just mixes up data from 2 different stations. Replacing the infotainment unit didn't fix that or the lag. It's like purposely designed to be annoying.
That small display on top of aircon vents is called Multi Function Display and it's its own thing, IIRC getting information from CAN bus from supoorted nav units. If that one is dropping messages, maybe it's a grounding or a harness issue?
I will never buy Subaru because their infotainment garbage had a phantom power draw causing dead batteries left and right. Subaru refused to acknowledge it, so I spent $1k out of pocket to get it fixed to have a usable car.
Then I got a notice in the mail that Subaru agreed to replace and compensate people for the phantom power draw issue, except I wouldn’t be eligible because I couldn’t prove it since I already fixed my issue…even though Subaru knew all the VINs with the problem.
indeed, i have had several subarus in my life from a bone stock 98 impreza RS through to the newest STI, and while i love them and will probably never buy another brand, the infotainment (and speakers) absolutely suck in all of them.
Subaru's bluetooth implementation adds multiple seconds of delay, so your navigation instructions will be so late that you'd miss turns if you relied only on audio.
I don't know, my Volvo supports both and I pretty much never bother using Android Auto, it's just such a faff. The built in interface(sensus) works great and the navigation is really good. Like, what do you use either of those two systems for?
Who cares about carplay or androidauto when plain ass bluetooth does most everything you should be caring to do while operating your vehicle (which you should in fact be 100% focused on and not touching your damn phone or other screens or anything else.)
At least Android Auto supports integration with steering wheel buttons and center console display on many cars — I imagine CarPlay too — this means you can can keep your music and navigation (favorites and such) in one place.
The Legacy is famous for going fast (especially the twin turbo one, when the second turbo kicks in, it feels like when a plane applies take-off thrust), and having great handling, they're still frequently used in rally racing in my country for that reason.
They're also famous for being incredibly expensive to replace the spark plugs in due to the engine's boxer layout meaning you need to remove the engine to get at the plugs. But so much fun to drive.
Depends on which Legacy you mean. I had an '05 Legacy GT wagon (EJ255 STI engine with smaller turbo / intercooler). I did plugs myself maybe 4-5 times. It was always a pain, taking 2-3 hours, but I always did it with the engine in..
Think we'll need to be doing something similar soon to remove ads from our cars. Read of somebody in an older Volkswagen getting an ad for the new model when they turned the car on one morning.
My car came with a cheap low-quality aftermarket combination car stereo and GPS. It would only play music from discs (no USB support), didn’t even support variable bitrate MP3s, and the GPS didn’t know the car’s location half the time.
I finally replaced the head unit with a plastic bracket and a standard old school in-dash stereo system. Much nicer setup.
It’s the cars which can’t have their head unit replaced with a simple stereo which worry me.
> It’s the cars which can’t have their head unit replaced with a simple stereo which worry me.
I agree. There's a lot of room for innovation, consumer choice, and competition that was previously guaranteed to be available in those one or two DIN units of space.
I haven't seen them present in any new car I've looked at in nearly ten years, but they were in every car I saw from the 50 years prior.
Denso head units for Toyotas can play MP3s, but the order of the files is when when they were copied to the SD drive. And you can't sort them any other way.
Yes, I can move the files, but it's also ridiculous in the 2020s that I have to resort to doing this workaround instead of the software reading ID3 metadata and sorting by one of the well-known fields.
Thanks for the tip on fatsort, it does save me a bit of trouble.
Embedded controllers of that class are usually not able to do such things. And if they are, it becomes a bottleneck when you plug in say a 64GB device. Five minute drive? No music because the stereo takes six minutes to sort the files on startup.
Another drawback of the death of DIN aftermarket stereos is that they stopped innovating there, I last updated my stereo in ~2015 and not found a more advanced replacement since. So never got one to accept a >256GB drive, though trivial elsewhere today. Raspberry Pi level tech.
My car stereo, in theory, cannot accept a drive over 32gb in size. In practice, I have a 128gb drive in there. I used a third party utility to format it as a FAT32 filesystem to be 128gb in size. The car stereo only claimed a 32gb limit only because that’s the largest fat32 filesystem Windows will make from the GUI.
Yeah, that's often the case. It's not really the size of the filesystem I think, my description was simplified. Believe when I got over N thousand songs it stopped working, which defeated the whole purpose of having my entire music collection on a drive. Still works on the home stereo though.
Fun fact: the FAT32 limit - which, by the way, was literally just added by Dave Plummer because he thought anything larger would have too much wasted slack space - made its way into the SD card spec. It is perfectly valid behavior for a device to refuse to read a >32GB card with a FAT32 file system, or a <32GB card with exFAT, or a card with multiple partitions, etc. None of those are actually SD cards as they are formatted.
In practice, nobody in their right mind would design a device's firmware that poorly, unless they were deliberately trying to force users to stay within the bounds of the device's QA test procedures. I don't have any examples of this. There are devices that will misbehave with such a card, however: for example, the Nintendo 3DS will happily use an oversize FAT32 card, but the Data Management screen will take way longer to load for some reason.
Also fun fact: the first time you plug a correctly formatted SDXC into a Nintendo Switch, it "downloads an update" and restarts. Presumably Nintendo wanted to save money on exFAT licensing fees so Nintendo is only paying for each unit that actually uses the exFAT library?
3- Yup, nailed it, same reason WMA and Flash require "activation" on the PSP and why newer MPEG codecs like H.264 and AAC require an "optional media update" on the Xbox 360
2- The slowdown usually comes from scanning the FAT to count the free space, with the PSP being notorious for this
Another related point is the varying support for the "semi-official" 64K cluster size, which gives a curious effect on the 3DS: in GBA mode (officially only used for 10 games only distributed as compensation to those who bought the original model at full launch price), a strip of "junk" (coming from the virtualized GBA's VRAM, IIRC it's just wraparound) will be shown at the bottom of the screen...
The counter argument to that is well designed modern cars have great info-tainment systems which get regular updates. My Tesla gets regular software updates, has upgradable hardware, and won't be subject to these kinds of issues for decades. Old cars encouraged exchanging and needed aftermarket parts because they didn't care or try, so the products were low-quality and often failed.
> My Tesla gets regular software updates, has upgradable hardware, and won't be subject to these kinds of issues for decades.
Teslas from four years ago with Intel Atom processors and HW3 are already getting different and lesser updates than new Teslas, without any hardware upgrade plan in place.
A Tesla you buy right now is running HW4 which is inferior to current Cybercabs and HW5 has already been announced.
Legacy automakers are still using the "ship it and forget it" approach. If I want my 2023 Hyundai to have the same features as a 2025 (manual battery preconditioning, Google Maps directions on HUD, wireless Android Auto, etc), I need to root the head unit and install a bunch of vibe coded packages from randos on GitHub.
Although to be fair the vibecoded packages are probably higher quality than what Hyundai would've shipped.
Correct. Without root, the only way to heat the battery was if you use the cars built in nav to get directions to a fast charger
The problems being:
1. It didn't always recognize a destination as a fast charger
2. The built in nav is missing recently opened fast chargers
3. It only does this if you have the "preconditioning" setting enabled, but the car has a bug where if you use the (maybe only US version) Hyundai app to change the AC or DC charge limits it unchecks that setting.
Ah, I have an Equinox EV and it will automatically precondition based on the on-board maps, which, in my case, is Google Maps. I don't have the luxury of Android Auto or Carplay unless I jump through some hoops and self-publish an app in the Play Store. And even then it is janky.
Even so, I have a button in the battery settings which can start the preconditioning whenever I want to. It's been a fantastic car, but I am envious of your DCFC capabilities. But not of your ICCU. :-)
> well designed modern cars have great info-tainment systems which get regular updates. My Tesla gets regular software updates
And tracks you everwhere you go, resulting in that data being available to hackers, marketers, and any sleazeball that wants to extort you by suing you.
Not unique to tesla, of course, but a property of all those 'well designed modern cars'.
My 2014 vehicle stopped updating it's time automatically a few years ago. I think some cellular technology was phased out or perhaps the cellular contract GM had for vehicle connectivity expired.
Either way I just manually set the time twice a year now - I guess I'm thankful it still has that option.
When I was playing around with the radio in my girlfriends vehicle there was no option to set manually on the clock screen. Seems like a pretty shortsighted move to not allow that.
My 2014 Mazda 3 has two buttons directly under the clock: “HH” and “mm”. To reset the clock during spring DST I have to press “HH” once. In fall I have to press it 23 times. The clock is so accurate I think I realistically only hit the “mm” button when the battery dies.
Funny to see how even in 2014 things were diverging so greatly
Probably the 2G network, with the 3G network to follow; I've had my electricity and gas meters replaced the other day because they used the 2G network still to communicate.
Older versions of the GM OnStar hardware used 2G cellular radios and carriers have decommissioned those networks. Adapters are available for some vehicles to use newer networks.
Considering that OnStar spies on you even if you don't subscribe to the service it's much better to just leave it broken, or even to actively disconnect it in newer vesicles.
Your tech expectation may be way too modern, could be some terrestrial radio clock being phased out. Back then some AM/FM alarm clocks would sync time over low freq radio. For FM, RDS can do it too.
It seems absurd to me that cars needs a cellular connection to automatically set time, meanwhile 10$ clocks can get the time correctly from the radio signal that are broadcasted. More insulting the fact that the car itself has an FM and also an AM tuner that are capable of receiving time signal and decoding them while the radio is not in use. And that modern cars that use digital radio (DAB) have a standard to get the time. Come on, car infotainment is so outdated...
I had a luxury car from 2001 where the time could not be manually changed, only set to specific time zones. Of course, all the time zone rules had long diverged from its rule set, and it had to be changed four times a year to show the correct time.
This car also inexplicably forced you to manually choose a country, and then a time zone, despite having a GPS that already knew exactly where you were and already had chosen a previous time zone.
Carplay is nice but I would have settled for a non-luxury "clock set" button.
My 2019 Subaru has a red 7-segment LED clock below the touchscreen that has 2 physical buttons right there that change the hour and minute. It's a luxury I don't think I'll soon take for granted. I changed my battery recently (the original OEM battery finally got too tired), so I've set my clock twice in the last week, taking about 10 seconds and 1/2 second to accomplish respectively.
I can't wait until folks start pulling firmware out of head units, using AI to remove the crap, and putting it back in (akin to the TV revolution happening now)
I'm not using that setup, but I just don't use remote control at all. The TV control is done through the "single-button interface". I guess you could say XBox controllers are remote controls.
> Carplay is nice but I would have settled for a non-luxury "clock set" button.
The clock set button is obviously the right solution, but this is actually why I demand CarPlay on any infotainment system in a car. I don't trust the manufacturers will keep it up to date, but I know my phone software will be updated/updateable/replaceable.
I live in a region that changed time zones just 2 days ago and my brand new 6-month old car now as an incorrect rule set and presents the wrong time when synced with GPS.
It has no over-the-air update capability so it seems like this sync-with-GPS feature is now broken for me unless the manufacturer puts out an update to be done at the dealership.
(in NZ) I have a 2005 import from Japan with time set by GPS, but apparently no way of changing the time zone away from Japan. The navigation still thinks I'm in Japan too but randomly flying around at high speeds too - even the compass calibration is way off. I too would rather have a manual clock set.
As an Arizona resident, my recent Honda CRV's clock is off by an hour for half the year. Since usually the carplay display overrides the clock, it's no big deal, but it cracks me up that, despite being a professional programmer, I have yet to figure out how to set it correctly
The difficulty of setting a VCR's clock was a meme in the 1990s, and I can attest, justifiably so.
But at least it was possible. The clock setting being impossible without a proprietary and easily lost or destroyed SD card is pretty good, but we can do better. It could be just impossible to fix a wrong clock. Truly that would be the apotheosis of user unfriendly design.
My very technical and hands on father used to have us set the VCR clock, the controls were in a tiny compartment under a flap that required either small fingers, or very careful manipulation with the point of a protractor or something.
We went from difficult to manipulate buttons through to multi-function buttons where controls get hidden through some incomprehensible maze of menu options, to now where it sometimes feels lucky if you even get an option to manually set the time at all.
I'm reminded of an old commercial depicting kids taunting another kid by asking "What time is it at Billy's house? 12 O'Clock! 12 O'Clock! 12 O'Clock!" Might have been a Blockbuster commercial? I can't remember.
The Internet Helpdesk sketch, by Three Dead Trolls In A Baggie, had the helpdesk guy describe the hapless user he's trying to help as a "12 o'clock flasher". "You know, every appliance in his house is always flashing 12:00".
Oh man, this reminds me of an episode of a show I watched when I was a kid in the 90s, where the running joke was all the characters, at different points, saying something like "I tried setting up the VCR to record it, but every time I press the button the number 12:00 starts flashing on the screen and I panic." I haven't been able to find it again. I think it was https://en.wikipedia.org/wiki/Mother_Goose_and_Grimm_(TV_ser...
In 1974 when I was an medical intern at Los Angeles County-University of Southern California Hospital, ALL the round analog wall clocks with hands in all 26 operating rooms were correct six months a year: there was NO provision for automatically advancing/moving them back when DST started/ended, nor did anyone do it manually.
I was told by our school electrician that long ago the clocks were adjusted by the power company. So around 2am DST they would either send power at 59Hz* to fall back or 61Hz* to spring forward. This was from the era when power grids were separate and analog. With newer national grid-tie it might not even be possible to do now.
> I was told by our school electrician that long ago the clocks were adjusted by the power company. So around 2am DST they would either send power at 59Hz* to fall back or 61Hz* to spring forward. This was from the era when power grids were separate and analog.
That wouldn't have worked.
Grid-tied synchronous clocks (https://en.wikipedia.org/wiki/Electric_clock) were once ubiquitous and did derive their timekeeping stability from the grid frequency, but that is only suitable for keeping time once set. It works because electric grids are frequency-controlled and stabilized, where deviations from the reference frequency are corrected over long intervals.
In principle you could send power at 59 or 61Hz to adjust these clocks, but there are two problems:
1. The grid frequency isn't arbitrary, it's also directly coupled to the rotary inertia of every turbine that powers the grid. This is why power fluctuations cause frequency changes: temporary oversupply or undersupply causes these turbines to slow down or speed up respectively, using their inertia as power storage. Potentially screwing up every other piece of mechanical equipment tied to the grid for the sake of clock changes is... a poor optimization.
2. A one-Hertz difference isn't enough to change the clocks in a timely way. To change time by an hour you need to gain or lose 3600 * 60 = 216k cycles, but a 1Hz change gains or loses one cycle per second. It'd take about 2.5 days to complete a one-hour change via this mechanism.
I don’t know about that, but I was told by someone who worked at a power generation plant in the 1970s that they had a clock in the plant, that they would compare to the time broadcast on the radio frequency that carried the official time, and any time it diverged they would slightly adjust the frequency of the alternating current until it matched.
I was a kid but I do know the clocks in my elementary, middle, and high school were accurate. Near end of day, we would impatiently stare at the clock waiting for the release bell. Being a public school, the clocks were bolted securely to the wall. Being little shits, students would probably steal them if they could, LOL. So how did they keep 100s of clocks on time? I asked and was told by the school maintenance that it was done through power.
I remember in my elementary school the clocks were all synchronized to a system somewhere, presumably in the head office. The day after going back to school after a DST change the clocks would all be off by an hour and sometime in that morning they would all suddenly freeze, then fast forward or rewind an hour and then restart. In retrospect it is amusing that whomever was in charge of updating the clocks didn't do it when they first got in but waited until the school day had started.
What are the odds they did have a centralized system for setting the clocks, but nobody knew how to operate it or cared enough to find out? Maybe someone set a password and then lost it. And of course if you have a centralized system it means you can't manually fix the individual clocks.
I worked in a Network Operations Center in the early 1990s, and there was a large projection display with four other monitors around it. Across the top were four round analog clocks, representing the time zones of the United States. I was one of four operators, all in our 20s.
So, Daylight Savings Time ended, and the early-morning operator had all the clocks set by the time I arrived, and the third operator came in, looked at them, and said "Great! So, did you take all four of them down and adjust the time individually?" "Yeah" "Didn't think about taking one off the end, adjusting it, and moving the others down the row?" *facepalm*
A design rule for the display of any appliance with a clock: It must be possible to remove or hide the clock. I don’t need 10 devices to keep track of time for me.
Yes, please. I don't want a clock on my microwave, or my oven. If I'm not interacting with it, it should produce zero light.
(Some people use oven features that involve knowing an absolute time, such as "turn on at this time". That seems to be the uncommon case, and meanwhile everyone has to deal with a clock that's rarely right.)
My microwave defaults to no clock after a power outage. I prefer it that way.
I used to have a range, then microwave, then toaster oven, then coffee maker, all with different times. I tried so hard to get them in sync, but it was just not possible. Plus, the coffee maker clock would drift over time. Since that point in my life the microwave, toaster oven, and coffee maker have been replaced and only the microwave has an optional clock.
Oddly enough, the newer appliances are all higher quality than the ones they replaced. I guess clocks are added to things to give them more perceived value.
Eh, if you only have a clock on only the microwave or oven then you lose access to the current time when you're cooking something or convert it over to timer mode. Having some redundancy isn't a bad thing in this instance.
Or simply have a dedicated clock in there and turn off the display on all of the appliances.
There should be two kinds of devices in my home: those allowed to connect to the Internet and keep their time up to date, and those that are not allowed to care what time it is. My microwave and my oven are in the latter category.
Many Jewish people use the feature while home. There are prohibitions on lighting fire on the sabbath, which has been interpreted more broadly as a prohibition on "turning stuff on". That's why you can set your oven to be on all day on Saturdays, for example, but they are usually home while it's happening. You can't turn the oven on, but you can use it while it's on. Similarly, some elevators are configured to stop on every floor on the way up and down on Saturdays.
Do "many" Jewish people do this? It sounds like something only religious nutters would do, wasting energy etc. (Doesn't the bible/torah say people should take care of the earth?)
The Torah is against harming nature (such as cutting down trees), and wasting things that are good.
However, climatism has devolved into its own, irrational, new religion. Followers of it believe, for example, that the apocalypse is nigh unless they drink from paper straws, while the very rich can fly in private jets so long as they pay a tithe to the priests of climatology.
I don't have any stakes or interest in a debate about the validity of religious beliefs or the prevalence of believers. Suffice it to say it is a practice that exists, and technology exists to facilitate it. My oven has a sabbath mode, so it's not something you necessarily need to go out of your way to have either.
Best they can do is blinking "12:00". Actually I encountered a kitchen appliance doing almost the opposite, after power off/on it didn't function before confirming any time.
My 4 year old Siemens oven is like this, as is the one (can’t remember the brand) in the previous house we rented for a short bit.
I wonder if it’s a kind of child safety thing. It’s possible on my current oven to enable a child lock by holding the clock button for 5 seconds, preventing the turning of the knobs from having any effect.
After a power failure the oven similarly will not react to the knobs being turned until the clock button is pressed, but a mere second or so is good enough.
Amen to that, I have a pair of built-in IKEAs, a microwave and convection oven, both with clocks when they're idle. But of course they're separate devices so they drift separately, which is obnoxious-looking. Happily the microwave did have a hidden setting to just display - - - - instead of the clock (though for no apparent reason the convection oven does not have that option, sigh).
My previous BMW was set to Polish, but showed some Dutch city names in the satnav in Cyryllic transcription so I would be driving to Эйндховен, for example. But only some; the other ones were in Latin script.
It's been awhile since I looked, but for at least some Mazda's you can use a usb stick with mp3s that have certain names to get access to the linux system. Miata owners use it to enable the touchscreen when moving.
The whole article is ludicrous. Why do you even need a car with an SD card slot in order to change the time? I reset all of my clocks this weekend and I don't even have a car.
> Why do you even need a car with an SD card slot in order to change the time?
The idea was that time came from GPS and you had a SD card with the update files for the maps and the current tzdata information. Timezones worldwide change ridiculously often and tying that together with the regular map updates makes sense.
Remember these things are from an era where smartphones were barely a thing and so was mobile phone connectivity in cars.
I think there are two aspects:
1. like no I don't want my microwave sending random blips of location data to a server and no I don't want hyper-personalised ads on my coffe machine
2. I want my machine to do what I want and not be limited by random limitations set in the software, either intentionally or via incompetancy.
This is really hillarious to me since I have the exact same problem in my Renault Scenic, there's just no way to change it. But half of the year it's correct, the other half I just automatically deduct 1 hour. It's one of those problems that are on the to do list but realistically I'm never gonna spend that much time on it. Not really a good engineer answer but oh well. Nice writeup!
I never understood why car companies don't put more effort and resources into their MMI systems. Audi is a big offender. I've never seen a more thoughtfully designed car with an MMI interface that's more frustrating and buggy.
First is to bring the hardware and software design in-house. Except auto companies don't do well at software, they're used to letting OEM vendors write the engine control firmware etc etc. There's also the problem of the very long design cycles in car manufacturing. The new car platform being built today started its design four to five years ago. And that's when the hardware was locked down.
Volvo's platform was so old at launch that they threw in the towel and are now offering to upgrade your hardware if you complain hard enough.
The other choice is to shop it all out to a consulting firm or OEM vendor. Then you have cultural conflicts between the car maker and the programmers, communication overhead, choice of hardware conflicts, etc. And you know it's going to be shopped out to the cheapest bidder. And you're still at the mercy of launching with hardware that's way older than the smartphone in the driver's pocket.
I took a similar approach to this article on the MMI software for my 2021 Q5.
partial emulation of the user space on a near equivalent kernel. java backend w/ angular frontend for the UI.
I was exploring ripping out this terrible workflow of not being able to change the user profile unless the car is started, which means I'm often stuck craming myself into my wife's seat preset to star the car lol.
the f/w uses dm-verity however so I'm not sure what flashing would entail. I'm also afraid to try.
I'd rather try to get privilege escalation on the car itself.
Almost all hardware companies don't understand or care about software at all. They see "software" as just another line item on the BOM, like a screw or a gasket. You source it from the cheapest supplier and then attach it to the product somewhere on the assembly line. The idea that software is the interface to the user, a signal of quality, part of product differentiation, an ecosystem in which future products live... none of this even remotely occurs to most hardware companies, especially car manufacturers.
Isn't it because the nav feature is market specific? People generally don't seem to like nav software brought in from different continents, across the planet.
I'm european, recently went to Korea. The nav system was almost perfect, only complaint I have is that it signalled every single speed bump (I can see them, thanks). There probably was a setting somewhere to tell it not to, but I didn't care enough to try to find it.
I was just wondering the same thing for a 2013 Honda Civic - has the screen's (the "head unit"?) firmware been fully dumped and reverse engineered?
How realistic would it be to run fully custom firmware there, and even communicate over usb - for example, if I had a pi running in the car, and it used the usb connection to show things on the built-in display?
I'd love to pwn the head unit in my 2014 Honda CR-V just so I can stop it from sending a "Play" command every time a phone associates with the Bluetooth and the audio system was left "on".
I also have a 2014 CR-V and I hate this too. But even moreso, I hate that there isn't a play/pause button at all. If I need to stop audio (e.g. pulling up to a drivethru, or starting a conversation), I gotta hit the phone. Heaven forbid I'm playing a CD, nothing I can do at that point.
Putting up a bug bounty of "ordering you a pizza" if anyone tackles this!
Had a Chevy that was extremely obnoxious about this. If it was playing over USB it absolutely would not allow it to stop. Hitting pause on the phone would have the car immediately hit play to restart it.
It's probably easier to just buy an aftermarket head unit. The only thing you'll need to confirm is that it supports features you care about e.g. steering wheel controls.
Okay, this was gratifying to read, but ... can't you wedge CarPlay onto this thing? I'd want a victory lap with root doing at least something. Bad apple on car boot, there's lots of options here!
Yeah, I'll keep working on this to get it to do more stuff that's at least a little bit more useful, if not just a better UI that isn't 90% a useless "No maps found" screen. It's more complicated than it might initially seem because they actually have a kernel module blocking me from mounting non-signed partitions as exec, so I'll need to work around that.
I'll probably post a follow up if I get anywhere with this!
I've been thinking about this for a 2016 Toyota in the family which has a quirky but decent enough idea for the main screen where a local DB file is consulted and a saved picture of the currently playing artist is displayed.
Sounds great but there are some hilarious examples where the fuzzing of the name (best guess, I figure in case the name is slightly wrongly typed - mp3 era) is far too loose. "Bayonne" is showing a picture of Beyoncé and "Steve Lacy" is showing a bald white guy. It has become one of those things in life where I get a bit excited to see who a brand new artist will get as the matched image, reminds me of childhood, honestly.
I figure with the game decompilation news I'm seeing the reverse engineering of the firmware (or at least the DB format) would not be a significant challenge, and it would mix in some fun other tasks like compiling (and signing?) a new DB with tons of images and names, and then figuring out how to avoid bricking a vehicles computer while doing so. But... like I assume many others here... I've got enough projects on the go right now so maybe in a few months.
They make touchscreen smart mirrors that easily mount on top of your OEM rearview mirror. Half the display shows a rear camera that you mount by your license plate, this functions as your mirror. The other half of the display shows wireless android auto or carplay, so you can browse music and use Google maps nav. Phone connects to Bluetooth of your car, so all steering wheel buttons work. Mine was about $120 and I love it. No need to mess with OEM mmi and no janky screen mounted somewhere that either blocks your OEM screen or awkward position. Very clean and super easy to install and use. The view of navigation is in a better spot than looking down at your mmi screen.
Such a contrast with the mechanical clock in my 1981 Volvo 240. Even my 29 year newer (2010) Toyota Yaris just has a simple 12-hour digital clock (not even AM or PM) with a button to set the time. I really appreciate how simple both are.
How does DST work with GNSS PNT signal? The international atomic time and GPS time does not have DST built in. That offset of 120 was for DST or lack of tz data? Do we still need DST in this AI era?
A cheap RTC clock chip would have costed a few cents, but Car manufacturers depend on GNSS timing for insurance purposes.
Pretty awesome write up to explain their journey in it all.
I suspect a bunch of car head units will face new problems in the spring when many Canadian provinces will no longer change their clocks. The units won't get updated DST rules.
my skoda simply has an entry in the menu: DST? which puts the clock one hour ahead or behind.
This posts remind me that when i will be in the market for a new van i will have to decide to get the trafic or the vivaro, solely on the head unit they use (renault ones have had incredibly stupid design decisions for years, on ergonomics alone, even more in firmware)
Over here in South Africa we have a single time zone and no DST.
This contributed to me in my very early years of dabbling in software development being oblivious to time zone considerations and would do really bad things like make sure the web server was setup with the correct time zone so that the ASP Classic output would show the correct time.
> They mentioned that the car apparently needs this SD card just to change the time on the clock, but I didn't really find that believable (...) After all, videos of similar head units online showed the menu with the clock settings option right there.
I don't get this attitude, I see it in many people to doubt by default when you should bookmark it mentally and think "what's the reason for this?"
I got a new samsung smart tv in April and never connected it to the internet. Just noticed (likely because if the DST change) that the date and time was all wrong when it turns on. Could not unnotice it. I was very pleasantly surprised I could manually set it and no internet was required. I expected to be annoyed at it forever or to go down a rabbit hole like this and fail.
How old was the Tesla? I have a 2015 Tesla S and the problem with these older ones is that the CPU and RAM are just not enough to run a modern browser. In fact even the very out of date browser on mine is occasionally killed by the OS to reclaim resources.
> When I bought the car, the previous owner asked me whether I wanted a "navigation map SD card" with it.
Why would they even ask?
> Since the maps on the SD card were apparently very out of date and didn't function properly, I decided I didn't need it and the previous owner tossed it.
WTF? My 2016 Mazda has an out-of-date SD card too. I removed it because it slightly speeds up how fast the system boots without it, but it's in the glove box for any future owner when I sell the car. I just can't understand tossing it. It's not like it takes up any room.
(My Mazda also sets its time via GPS, but I can get to the timezone offset and DST checkbox regardless of the presence of the SD card.)
They basically asked me if I want them to go digging through their trash because they had accidentally thrown it out just shortly before handing the car over. I said that it's no worries and I'll figure it out. Just didn't expect that this would be the way I'd do it.
Ah gotcha. Yeah the DRM on these cards is also a pain. I vaguely recall I was able to get updated maps from a Chinese supplier on Amazon but I don't recall how it worked around the DRM issue. In the US this was only a $20 or so experiment though.
Edit: available for $30 these days. Apparently the card gets locked to the VIN on first use.
Random rant... kind of triggered by the original post.
I still drive a car from 2011, but my headlights / fog lights / any other front lights that are not my emergency lights stopped working, super expensive to get it all reviewed by Toyota here in Florida (apparently Florida is the worst for these sort of things - already $250 down and they couldn't figure it out), and so left it with a relative who is a mechanic, and electrical problems are up his alley (I believe he studied this in particular too).
Both fuse boxes (driver side) and engine bay side are messed up (not enough voltage going into the engine bay one, after I replaced the driver side one with one found at a junkyard by a buddy for $80 compared to $1200 that Toyota bills for the part mind you).
Anyway, so I'm borrowing my moms much newer 2024 Toyota RAV4 (or whatever year), lots of neat things, but there's this really annoying feature I had no idea new cars have where the engine shuts off when you're stopped, I knew people did this to save on gas, but I always thought it was silly and a waste of time, and it felt really rare, until I realized why I hear engines start back up at red lights more frequently nowadays, if I have to floor it I dont want to wait, so I push the button depending on if I'm in frequent stop and go traffic.
My coworker drives a much newer (2026) SUV from VW... Apparently the button became like 4 screens in the touch-screen tablet the car comes with, just to force people to stop hitting the physical button, and you can't just disable it permanently.
Well.. unless you jailbreak the software, which he will likely do once his warranty runs out. Why do we force these ridiculous things on cars? It's no wonder something as ridiculous as Brabus exists... Since they're not a manufacturer they can remove all these things, and sell you a "performance" car, but its really mostly just the car restored to its normal state as it would have been released 10 years ago? or whatever.
Why do we have to go through so many hoops for things we "own" and why is it that the reliability of cars has gone backwards? Don't get me started on Nissans newer engines.
Who can make me a normal car that isn't a slug on Florida highways or turning off every time I hit the breaks? (Being dramatic, but its exhausting how much nonsense goes into cars now) Please, you'll have a customer for life.
Something to be aware of with that auto-engine stop at the lights, if you ever need to replace the battery, it necessitates a particular kind of battery either an enhanced flooded battery or absorbent glass mat battery - I learned that the hard way >_<
Also - when it comes to getting parts for Japanese cars, I thoroughly recommend rockauto.com, a US colleague put me onto it, and I use it from NZ because a) it has parts I struggle to find in NZ and b) even with the international shipping, parts are cheaper from there than locally.
> My coworker drives a much newer (2026) SUV from VW... Apparently the button became like 4 screens in the touch-screen tablet the car comes with, just to force people to stop hitting the physical button, and you can't just disable it permanently.
> Well.. unless you jailbreak the software
Maybe this is locked on VW, but given that Audi and Porsche share a lot of the same software, methodology, etc., I feel "decently confident" that this can be disabled via an OBD scanner (obdEleven, for one, was built around the VAG brands).
I was able to disable it in my fiancée's entry trim 2017 A4 in about 20 seconds, and no manual "long codes".
It's not an option needed on my RS 5, but I've done a few other things via the tool, and the dealership didn't have an issue with it - when I brought my car in for some other issue (water ingress into a sensor/light), all they said was "Are you okay with the tech resetting any changes you've made if they need to for diagnostics?" (they didn't), and "Do you accept that there will be a diagnostic fee if the tech doing a reset of changes resolves your problem?"
I wish someone would jailbreak the Nissan Leaf (MY2018) head unit. Every time I turn on the car, it shows a EULA asking if Nissan can please collect every detail of my car at all times and sell it to whoever asks.
Every. Time.
Even if I beat my head with hammers to think like a lawyer, I can't imagine how it would be necessary to ask constantly. Doesn't matter if I accept or decline, it still asks again.
I hate to laugh but the thought of you constantly rejecting the EULA for weeks on end, to finally give in one day, press accept and come to terms with it, only to get in the next time and have it pop up.
I can only imagine the amount of expletives that were said at that exact moment.
Me too, but it's more a love/hate relationship, because aside from this issue, and a few other annoyances, I love this car to bits. I've been driving for over 25 years, and this car is my favourite of all of them, including some newer and fancier electrics. I think it's the throttle response. Nissan just nailed it.
I also hate when my android based dash can't stay when I mute, every time I start it will unmute and with the same volume but FM, thought of doing similar to this.. right time.
I bought a Porsche GT3 coded to Canada. Meaning km/h in speedometer in instrument cluster, etc. Dealers had no clue now to recode. Porsche NA also not helpful. Russian car forums for the win. Recoded myself in a few hours for only the cost of a usb to OBD-II adapter.
My 2008 Honda Civic has a old-fashioned set-it style clock, but from the factory, it's always ran about a minute a week fast. Which of a course is an excruciating kind of boiling the frog.
Also, it now sometimes thinks my headlights are on and dims the display for a few seconds randomly which I guess isn't too bad after 15+ years. It doesn't seem to have any other problems except for worn out suspension and leaking exhaust.
8 inches or more of dashboard real estate to display "maps not found", "don't get distracted" warning, terms and conditions, and blinking "12:00". Having had rented a fair share of subcompact and compact cars in half dozen of countries, the car navigation was updated and working exactly only once.
> The system had DRM checks which rejected SD cards without the correct CID (a unique ID burned into the card) and files with signatures based on that specific CID, so I couldn't even copy the files to a blank SD card.
FWIW, there are grey market sites that will sell you an SD burned with a custom CID so you can clone an SD card with this kind of copy protection. It sounds like in this case maybe he could've gotten away with an empty maps card + menu definitions file.
That must be one of the most iconic HN sentences ever.
If I ask my Claude (Opus 5.5) about how some Claude feature works, its natural inclination is to start reversing its own binary with strings instead of looking at Anthropic's documentation. It will, entirely unprompted, try to figure out how its environment works in Claude Code cloud and reverse–reverse-engineer Anthropic's Rust binaries to figure out why things happen the way they do, and what it needs to do about it.
However Claude and I did write code all the day long and a long job went down from 2+ hours to 30 minutes, same final output.
My takeaway: You should have just gotten the SD card and trusted the previous owner. Sometimes "simple users" do know. ^^
In any case, glad it took you on the wild goose chase that led to the write up!
Man this is too real for personal projects. I really need to write down my findings and thoughts more for future me.
I got a few of them, partially-used, from a canceled project and I really enjoyed using and carrying them.
I don't understand how Subaru stays in the business. Just how. It's a 4 digit time. Just broadcast it on CAN. Why not.
Also: a lot of these older, completely rectangular sat nav units actually follow what is colloqially known as "2DIN" standards. (cheap import Amazon)Aftermarket units are available and those can be swapped in with just a bit of work. That's what most people do when they have issues with the stock unit installed for their markets.
Both go wrong over time, but both in two different directions, so the trick is to calculate the „average“ between both, and you end up with roughly the correct time.
What surprised me most was that the clock adjusted itself automatically whenever daylight saving time started or ended.
There was no calendar or date setting, so I assume it must have received the time either via RDS (I don't know whether RDS supports this) or through a built-in DCF77 receiver.
All my subsequent cars had far more features, yet none of them had this one.
I also wonder why cars with built-in GPS navigation don't synchronize their clocks using the time from the GPS signal. Heck, the car could even determine your time zone from your GPS coordinates, so there'd be no need to configure anything manually.
Checking wikipedia, there is optional support for transmitting time and date in RDS, with an explicit usecase of updating the clock in cars.
> I also wonder why cars with built-in GPS navigation don't synchronize their clocks using the time from the GPS signal.
GPS doesn't transmit timezone information, nor does it transmit leap seconds. It transmits a very, very accurate UTC minus 18 seconds, but without access to the internet for timezone database updates or (until recently) leap seconds, it's a bit useless for updating a clock.
Then just let the user override TZ offsets or the like if they want.
I thought I broke the clock with a DIY repair because it too couldn’t be manually set (aside from enabling a +30 min offset). Luckily it fixed itself after driving out of the garage when it found a GPS signal.
CURLY: About uh--- [Looks at the three watches on his wrists.]
MOE: Hey! What’s the idea of the three watches?
CURLY: That’s the way I tell the time.
MOE: How do you tell the time?
CURLY: This one runs ten minutes slow every two hours. This runs twenty minutes fast every four hours. The one in the middle is broken and stopped at two o’clock.
MOE: Well, how do you tell the time?
CURLY: I take the ten minutes on this one and subtract it by the twenty minutes on that one. Then I divide it by the two in the middle.
MOE: Well, what time is it now?
[Curly grabs a clock from the inside of his jacket pocket.]
CURLY: Uh, ten minutes to four.
MOE: Hmm so it is!
[Moe grabs the clock from Curly and hits Curly on the head with it. The clock breaks.]
It gains anything up to an hour a day, particularly when my six-year-old is helping me clean the car and he pushes the button to watch the hands go round really fast.
It's fine. Time is time, we get there when we get there.
The analog clock is accurate enough that I don't fuss with it other than to reset it for DST (I have no similar children).
My solution to the lack of synchronization: turn off the radio clock (which is fortunately an option).
I don't really care about what time I get where I'm going, I get there when I get there.
How long does it take you to fill your tank?
How does any of this need you to know the time right now?
I've never been "stuck in traffic", so the problem has never arisen.
(theeheehee, I’ve got an old Land Cruiser, there's a bit of a friendly rivalry)
And carry a few spares and tools. I bet you do, too.
It's now in an e-cycling waste bin somewhere, replaced by an ESP32 I/O board and Alarmo in Home Assistant, at 1/20th the cost of the original PoS alarm.
If I had to hazard a guess as to the mechanism as it would turn up in an automotive company it is that automotive engineers have the SAE to set professional standards and that software "engineers" have nothing [1]... And if an organized group has a fight with a disorganized group the result is predictable.
[1] the ACM is a professional associate for computer science professors.
This feels like a logical fallacy but I'm having a terrible time disproving it at anything near large company scale
they sell SKUs with no nav and no remote display for the nav. and they still need a clock.
The big advantage is when accelerating. It's a nice quality of life feature. I typically switch to AWD when I move to winter climates. It's not as safe though, _especially_ if you have little enough training that you're relying on intuition to make that conclusion that AWD must be safer.
Is this dumb in winter, or dumb generally? I was taught to do this.
If you want to maximize speed through a curve plus straightaway, you should pick a path through that curve minimizing curvature at the end so that you are able to accelerate out the end of the curve. As a reasonably good approximation for normal driving, your tires have the same traction regardless of the direction of the load being placed on them. That gives you a budget of net force you're able to apply. If you're near the sideways force limit going around a curve and add acceleration (or braking -- always do your braking before you enter the turn, or gradually reduce it as the curvature increases), the wheels with those extra forces will break traction. The solution is to apply that acceleration at a point of milder curvature.
Taking that idea back down to safe driving speeds, the same idea exists when considering crossing a margin of error rather than the ultimate slip threshold. All else equal, accelerating as you come out of a curve, if you've chosen your path through the curve correctly (really important to get that right; there's a curve back near my hometown which has taken out half a dozen people over the years because its curvature increases at the end and they haven't planned for it), will minimize your time through the curve and the following straightaway at your chosen safety threshold.
However, all of that is predicated on minimizing time assuming perfect knowledge and perfect driving. The acceleration itself is actively harmful from a safety perspective. In FWD, the wheels losing traction are the ones connected to your steering wheel. That's not ideal. In RWD you're prone to oversteer and spinning out of control. In AWD you can get either or both of those failure modes. Moreover, navigating typical curves, you're so far from slippage that the technique of accelerating at the end just means you'll go from the speed limit to something greater. For winter curves, the margin of traction uncertainty is so great that you can't gain much from accelerating out the end anyway (and have a pretty bad risk profile -- accelerate on a straightaway and slip a bit and with a bit of training you're basically fine, slip off a curve and you have a really bad rest of your short life potentially).
Safely accelerating in a curve is a learned skill though, and most people don't care that much about driving. Unless it looks like they're interested in more depth, my advice is always to brake before the curve and accelerate afterward. It's safe, fast enough, and easy to remember.
- Smaller turning circle because the front wheels can turn further without complicated engineering.
- More traction when accelerating on a wet surface, due to weight transfer backwards. (1)
- Usually better weight distribution due to engine layout.
There are other downsides in addition to what you mentioned:
- Any scenario where you will definitely lose traction, then rear-wheel drive becomes harder to control. There was an old Top Gear episode with a steep grassy hill, and drivers of the RWD cars could not get up it. (2)
- The transmission tunnel required for RWD
- Reduced fuel economy due to having to turn the power 90 degrees (assuming the engine is mounted in the front). (3)
1) This one may sound like unnecessary acceleration, and it probably often is, but there are situations where it can arise, especially if you prefer older cars. There is one particular hill I used to encounter frequently, where you are trying to turn onto a main road, so you want to accelerate fairly briskly to avoid waiting for a huge gap in traffic. On economy tires, in the wet, you can easily reach the limits of FWD. I would see many cars time a gap, and then have to rapidly decide to give up on it when their acceleration was much slower than they expected due to wheel spin. I think there may have been bus lane paint on the road too, but I'm not certain. Obviously AWD is even better than RWD for traction, but at that point, that is usually only when requiring rapid acceleration.
2) I suppose with perfect/computer assisted driving, RWD would out-perform FWD.
3) I once searched, and had a very hard time quantifying this, but it is almost certainly a few percent, and perhaps more. There are also weight differences involved because of the long drive-shaft, but I think the effect is larger than that. It is very hard to find an apples-to-apples comparison.
Now if you happen to be really good at driving, it's even safer because you can recover from a slide faster, again because of the increased traction during acceleration. Once your tires have traction again the additional traction moves you out of the slide faster.
The only problem here is people going too fast which can happen with any car.
I'm struggling to come up with a reason why the faster acceleration itself would be a safety issue (maybe trying to get out of the way of a truck that can't stop or something? it seems niche), so let's instead look at the increased control during acceleration (from less force on the steering tires).
When does that matter from a safety perspective?
In a straight line, it doesn't. Let off the gas and try again. No matter what you do, you're going to go in your intended direction of travel or stay put.
So that brings us to accelerating around a curve. Don't do that on public roads in normal cars in winter conditions. The risk profile is terrible. If you misjudge it just once you're in a ditch.
Even your slide example seems strange. Could you describe the conditions of a slide where applying power helps you gain traction faster than not applying power? The only thing immediately coming to mind is reducing the yaw force in certain oversteer scenarios with a counter-steer and a bit of power, but in that case whether AWD helps or not is extremely situational and depends on the exact build of the car.
But aside from all that, I think there is a scenario where applying power can help gain traction faster. In a FWD, if someone has gone into a corner too fast, or encounters a more slippery surface than expected, and the vehicle starts to understeer straight-ahead, I have heard of the idea of accelerating to get the vehicle to continue around the corner. The turned front wheels, while still not gripping properly, spin and pull the front in the direction of the corner more than would happen from just braking or maintaining speed. I'm probably not explaining that well, and it doesn't sound safer than just braking! I am fairly certain I remember seeing something which I think is related, in the Owner's Manual for some old front-wheel drive Japanese sports car. Something like an old Honda Prelude or a Toyota Celica. They played up the safety of FWD "pulling you through" the corner. None of this makes the case for AWD though. I can only see the advantage I wrote above about accelerating earlier/harder, and I think you already addressed that.
In Michigan, like clockwork the AWD SUV’s would go flying off the elevated freeways into the deep snow at 50mph in every storm.
I love driving up to Tahoe in a blizzard. I'll happily be doing 25 mph going uphill as it's roughly the safe speed for the conditions. I'll constantly be passed by AWD SUVs going 40-50, while going uphill.
Then we summit Donnor Pass. I'm still cruising along at 20-25. You can see the fear in these SUV drivers' eyes as they start sliding down with their "all season" tires. Invariably, they'll then slow to 5mph with their flashers on in the right lane and I can cruise down t 25mph and pass all the cars that flashed at me or gave me a look going up.
It is baffling to me that CA loves to enforce chain control but exempts AWD without actual winter tires from it. It's almost never the 2wd drive cars with chains that pile up, it's always the AWD SUVs. If you're going to enfore chains, at least do it usefully and only exempt actual winter tires. Then it would be useful, not just a massive slow down.
In a very long winded way, couldn't agree with you more!
Personally, I'm comfortable driving anything but a RWD pick-up truck with no payload in snow and ice, try to drive at an appropriate speed for weather conditions, and rarely have trouble.
Misconceptions like this are why people are lulled into a false sense of security.
Are you saying that when when cornering at a constant speed, if a slide starts, the vehicle will rotate differently to a front-wheel-drive or rear-wheel-drive vehicle?
Or that when cornering, if you accelerate enough to spin the wheels, then an AWD vehicle will rotate around the middle?
For the 1st case, when you slide because there is not enough traction available for that speed, on that surface (e.g. driving too fast, or encountering ice), I wouldn't have thought the axles that power is delivered will make a large difference to how the vehicle rotates. Either the front loses traction, or the rear, or both. Wouldn't the weight distribution be more relevant than the driven axles?
For the acceleration example, again I think of either the front losing traction, the rear, or both. Front-wheel-drive would lead to the front losing traction, RWD the rear losing traction, and AWD ... I'm not sure. With some systems, it's plausible that neither the front or rear would spin until both lost traction simultaneously. I tend to think a 4 wheel slide would be similar to understeer in a front-wheel drive car, but perhaps a little more like sliding sideways out of the lane, rather than plowing straight ahead. This doesn't seem to me like rotating around the middle, so I feel like I'm missing something.
To be honest, the idea of AWD making the vehicle rotate around the middle (and rear-wheel-drive making the vehicle rotate around the...front?) seems like the persistent and widespread idea of AWD allowing a vehicle to corner faster at a constant speed. My thought is that it's a misconception, but I'm happy to be wrong!
I can see AWD allowing better engine-braking than FWD in some low-traction situations, but I don't think it would make much difference if trying to stop quickly. I think for modern cars, using engine braking would almost never provide quicker stopping, or more controlled stopping than brakes alone.
Regarding three pedals, you could downshift in an two-pedal automatic too couldn't you?
Do you own an AWD vehicle? They outperform 4x4s on everything except rock.
The "weakest links in the chain" when it comes to snow driving are largely centered around putting motive power to the ground without breaking traction. Think like steep hills, crossing the piled up slush between lanes, getting moving from a stop, etc. Except inn rare circumstances, your "this feels decent speed" is attainable regardless of AWD so it's not like you go any faster.
https://trid.trb.org/View/1484056
If the previous hill was too close to the next, you've got a quandry.
Nobody has the slightest clue how to deal with ice.
I'm from the Northeast but happened to be in Seattle on a day when it snowed. Every road was a hill and every hill was full of sideways cars.
Seattle isn’t that bad in wet pavement, unless it hasn’t rained. Which only happens about four to six times a year but that’s still enough that people should pay attention.
I’ve seen two cars on their roofs on an uphill in Seattle. How the fuck do you even DO that?
And don’t get me started on the theft lane.
Past two winters I was in RWD too. Just exercise some throttle control and get proper tires.
(Snow I’m willing to acknowledge is trickier, but modern cars in wet roads shouldn’t be hampering any of you…)
These days I aim to replace my tires at about 50% wear and no later than 60%. New tires make you corner so well it almost feels like a slightly newer car.
Since this is not happening, we can infer that insurance companies believe the benefit to their risk profile is less than the cost to administer the program and issue discounts deep enough for customers to participate.
It seems counterintuitive that stopping distance is not that big a deal, but it must be true.
Certainly it would be cheaper for them to run a program where you take a photo of your tires every few months, vs. the program they all have where you plug a surveillance device into your car. So the true benefit from good tires must be far less than driving style.
The sting of losing two tires due to a mistake is a lot less than that of losing all four when three are still good. It’s one extra versus three extra tires.
They’re all a little different, but I’ve not once found myself in a situation where I noticed AWD really making a difference. Maybe in parking lots? Certainly not enough that I would ever make a purchase decision based on drivetrain.
Feels like everything I’ve driven has been modern enough to have great stability control, and that’s all that matters.
Not sure if relevant, but Toyota has been buying parts of Subaru over the years, in 2020 ~ they have roughly 20% of Subaru. I'm not sure if its one of those "Microsoft investing into Apple" or "Intel investing into AMD" type of moments (which tbh, I wonder if Nvidia should be trying to invest in Intel or AMD since they're doing so freaking well its horrifying).
We've been pretty happy with our 2020 Outback so far (bought it just before the chip shortage). My only complaint is the stability of the infotainment software. The big portrait screen is actually quite nice, and with CarPlay I don't really care about the rest of UI and software. Except that the stupid thing just crashes all the time.
maybe I'll get a outback instead next time, they are kind of the same to me though the hybrid on the forester I have def makes it a more pleasant less CVT'y ride...
Does anyone know why the mini display on the top, lags so much when changing music tracks or even Radio stations? It even sometimes doesn't update the radio station, and just mixes up data from 2 different stations. Replacing the infotainment unit didn't fix that or the lag. It's like purposely designed to be annoying.
Then I got a notice in the mail that Subaru agreed to replace and compensate people for the phantom power draw issue, except I wouldn’t be eligible because I couldn’t prove it since I already fixed my issue…even though Subaru knew all the VINs with the problem.
https://www.subarubatterysettlement.com/
Plus I have had to replace bearings and control arms within 50k miles…which I expect better on even American brand cars much less Japanese.
Uh, presumably you understand that the core value proposition of a car does not hinge on timekeeping, right?
Subarus are famously long-lived and reliable. That's how they stay in business.
The Legacy is famous for going fast (especially the twin turbo one, when the second turbo kicks in, it feels like when a plane applies take-off thrust), and having great handling, they're still frequently used in rally racing in my country for that reason.
They're also famous for being incredibly expensive to replace the spark plugs in due to the engine's boxer layout meaning you need to remove the engine to get at the plugs. But so much fun to drive.
Now, that they are in their 80s, having their clock set correct the entire year is terribly confusing to them.
I finally replaced the head unit with a plastic bracket and a standard old school in-dash stereo system. Much nicer setup.
It’s the cars which can’t have their head unit replaced with a simple stereo which worry me.
I agree. There's a lot of room for innovation, consumer choice, and competition that was previously guaranteed to be available in those one or two DIN units of space.
I haven't seen them present in any new car I've looked at in nearly ten years, but they were in every car I saw from the 50 years prior.
Thanks for the tip on fatsort, it does save me a bit of trouble.
Another drawback of the death of DIN aftermarket stereos is that they stopped innovating there, I last updated my stereo in ~2015 and not found a more advanced replacement since. So never got one to accept a >256GB drive, though trivial elsewhere today. Raspberry Pi level tech.
I seriously doubt compute is an issue for a music player in the last 20-30 years (and more).
My car stereo, in theory, cannot accept a drive over 32gb in size. In practice, I have a 128gb drive in there. I used a third party utility to format it as a FAT32 filesystem to be 128gb in size. The car stereo only claimed a 32gb limit only because that’s the largest fat32 filesystem Windows will make from the GUI.
In practice, nobody in their right mind would design a device's firmware that poorly, unless they were deliberately trying to force users to stay within the bounds of the device's QA test procedures. I don't have any examples of this. There are devices that will misbehave with such a card, however: for example, the Nintendo 3DS will happily use an oversize FAT32 card, but the Data Management screen will take way longer to load for some reason.
Also fun fact: the first time you plug a correctly formatted SDXC into a Nintendo Switch, it "downloads an update" and restarts. Presumably Nintendo wanted to save money on exFAT licensing fees so Nintendo is only paying for each unit that actually uses the exFAT library?
2- The slowdown usually comes from scanning the FAT to count the free space, with the PSP being notorious for this
Another related point is the varying support for the "semi-official" 64K cluster size, which gives a curious effect on the 3DS: in GBA mode (officially only used for 10 games only distributed as compensation to those who bought the original model at full launch price), a strip of "junk" (coming from the virtualized GBA's VRAM, IIRC it's just wraparound) will be shown at the bottom of the screen...
Teslas from four years ago with Intel Atom processors and HW3 are already getting different and lesser updates than new Teslas, without any hardware upgrade plan in place.
A Tesla you buy right now is running HW4 which is inferior to current Cybercabs and HW5 has already been announced.
Except for maybe the '58 Thunderbird, but that was rectified in '61 in my opinion.
Although to be fair the vibecoded packages are probably higher quality than what Hyundai would've shipped.
The problems being:
1. It didn't always recognize a destination as a fast charger
2. The built in nav is missing recently opened fast chargers
3. It only does this if you have the "preconditioning" setting enabled, but the car has a bug where if you use the (maybe only US version) Hyundai app to change the AC or DC charge limits it unchecks that setting.
4. I prefer Google Maps to the built in nav
Even so, I have a button in the battery settings which can start the preconditioning whenever I want to. It's been a fantastic car, but I am envious of your DCFC capabilities. But not of your ICCU. :-)
And tracks you everwhere you go, resulting in that data being available to hackers, marketers, and any sleazeball that wants to extort you by suing you.
Not unique to tesla, of course, but a property of all those 'well designed modern cars'.
Either way I just manually set the time twice a year now - I guess I'm thankful it still has that option.
When I was playing around with the radio in my girlfriends vehicle there was no option to set manually on the clock screen. Seems like a pretty shortsighted move to not allow that.
Funny to see how even in 2014 things were diverging so greatly
2022, I got a letter saying that my car's crash response SOS system will stop working because FCC is retiring the 3G spectrum it is using.
The car clock however is synced to the GPS instead.
If GM used NTP instead of GPS, then it makes sense that it stopped working arbitrarily around 2022.
https://www.onstar.com/support/faq/2g-network-sunset
I'll gladly take setting the clock twice a year in exchange for the OnStar spyware ceasing to function.
I didn't note down exactly when it stopped updating but this seems extremely likely based on my memory.
Interesting about the radio based time though!
https://en.wikipedia.org/wiki/GPS_week_number_rollover
This car also inexplicably forced you to manually choose a country, and then a time zone, despite having a GPS that already knew exactly where you were and already had chosen a previous time zone.
Carplay is nice but I would have settled for a non-luxury "clock set" button.
The Toyota one requires selling your driving data to insurance companies.to update the tzoffset based on location.
And so two clocks on the dashboard are off this week until I can remember where the dst adjustment is in the settings.
The clock set button is obviously the right solution, but this is actually why I demand CarPlay on any infotainment system in a car. I don't trust the manufacturers will keep it up to date, but I know my phone software will be updated/updateable/replaceable.
It has no over-the-air update capability so it seems like this sync-with-GPS feature is now broken for me unless the manufacturer puts out an update to be done at the dealership.
But at least it was possible. The clock setting being impossible without a proprietary and easily lost or destroyed SD card is pretty good, but we can do better. It could be just impossible to fix a wrong clock. Truly that would be the apotheosis of user unfriendly design.
We went from difficult to manipulate buttons through to multi-function buttons where controls get hidden through some incomprehensible maze of menu options, to now where it sometimes feels lucky if you even get an option to manually set the time at all.
P.S. Found the video: https://www.youtube.com/watch?v=k7FYR72mr0E
That wouldn't have worked.
Grid-tied synchronous clocks (https://en.wikipedia.org/wiki/Electric_clock) were once ubiquitous and did derive their timekeeping stability from the grid frequency, but that is only suitable for keeping time once set. It works because electric grids are frequency-controlled and stabilized, where deviations from the reference frequency are corrected over long intervals.
In principle you could send power at 59 or 61Hz to adjust these clocks, but there are two problems:
1. The grid frequency isn't arbitrary, it's also directly coupled to the rotary inertia of every turbine that powers the grid. This is why power fluctuations cause frequency changes: temporary oversupply or undersupply causes these turbines to slow down or speed up respectively, using their inertia as power storage. Potentially screwing up every other piece of mechanical equipment tied to the grid for the sake of clock changes is... a poor optimization.
2. A one-Hertz difference isn't enough to change the clocks in a timely way. To change time by an hour you need to gain or lose 3600 * 60 = 216k cycles, but a 1Hz change gains or loses one cycle per second. It'd take about 2.5 days to complete a one-hour change via this mechanism.
So, Daylight Savings Time ended, and the early-morning operator had all the clocks set by the time I arrived, and the third operator came in, looked at them, and said "Great! So, did you take all four of them down and adjust the time individually?" "Yeah" "Didn't think about taking one off the end, adjusting it, and moving the others down the row?" *facepalm*
(Some people use oven features that involve knowing an absolute time, such as "turn on at this time". That seems to be the uncommon case, and meanwhile everyone has to deal with a clock that's rarely right.)
I used to have a range, then microwave, then toaster oven, then coffee maker, all with different times. I tried so hard to get them in sync, but it was just not possible. Plus, the coffee maker clock would drift over time. Since that point in my life the microwave, toaster oven, and coffee maker have been replaced and only the microwave has an optional clock.
Oddly enough, the newer appliances are all higher quality than the ones they replaced. I guess clocks are added to things to give them more perceived value.
This is a oddly important feature for Asian market rice cookers, which have a backup battery to retain the time and is not user-replaceable.
https://www.ifixit.com/Guide/Zojirushi+Rice+Cooker+Model+NS-...
Or simply have a dedicated clock in there and turn off the display on all of the appliances.
Indeed. "Please create fire hazard exactly at this hour when I'll not have time to do it personally".
However, climatism has devolved into its own, irrational, new religion. Followers of it believe, for example, that the apocalypse is nigh unless they drink from paper straws, while the very rich can fly in private jets so long as they pay a tithe to the priests of climatology.
https://en.wikipedia.org/wiki/Sabbath_mode and https://en.wikipedia.org/wiki/Shabbat_elevator
I wonder if it’s a kind of child safety thing. It’s possible on my current oven to enable a child lock by holding the clock button for 5 seconds, preventing the turning of the knobs from having any effect.
After a power failure the oven similarly will not react to the knobs being turned until the clock button is pressed, but a mere second or so is good enough.
I'm finally putting 2 years of undergraduate German courses to use... Yay?
https://github.com/ea/bosch_headunit_root
If anyone feels like using that access to throw some tokens at various systems.
It can be changed without the SD card in the car radio.
The idea was that time came from GPS and you had a SD card with the update files for the maps and the current tzdata information. Timezones worldwide change ridiculously often and tying that together with the regular map updates makes sense.
Remember these things are from an era where smartphones were barely a thing and so was mobile phone connectivity in cars.
https://phrack.org/issues/7/hackers-manifesto
"Hold on babe, the christmas lights have to finish installing a firmware update", is a real sentence that actual people are saying in 2026.
First is to bring the hardware and software design in-house. Except auto companies don't do well at software, they're used to letting OEM vendors write the engine control firmware etc etc. There's also the problem of the very long design cycles in car manufacturing. The new car platform being built today started its design four to five years ago. And that's when the hardware was locked down.
Volvo's platform was so old at launch that they threw in the towel and are now offering to upgrade your hardware if you complain hard enough.
The other choice is to shop it all out to a consulting firm or OEM vendor. Then you have cultural conflicts between the car maker and the programmers, communication overhead, choice of hardware conflicts, etc. And you know it's going to be shopped out to the cheapest bidder. And you're still at the mercy of launching with hardware that's way older than the smartphone in the driver's pocket.
TDD was an invention of Chrysler.
partial emulation of the user space on a near equivalent kernel. java backend w/ angular frontend for the UI.
I was exploring ripping out this terrible workflow of not being able to change the user profile unless the car is started, which means I'm often stuck craming myself into my wife's seat preset to star the car lol.
the f/w uses dm-verity however so I'm not sure what flashing would entail. I'm also afraid to try.
I'd rather try to get privilege escalation on the car itself.
this was maybe a few hours one evening.
My Model 3 is showing the wrong time and this requires an OS update to fix.
Suffice to say I would prefer an option to update it myself.
https://driveteslacanada.ca/news/tesla-fixes-bc-clock-bug-up...
How realistic would it be to run fully custom firmware there, and even communicate over usb - for example, if I had a pi running in the car, and it used the usb connection to show things on the built-in display?
Putting up a bug bounty of "ordering you a pizza" if anyone tackles this!
I'll probably post a follow up if I get anywhere with this!
Sounds great but there are some hilarious examples where the fuzzing of the name (best guess, I figure in case the name is slightly wrongly typed - mp3 era) is far too loose. "Bayonne" is showing a picture of Beyoncé and "Steve Lacy" is showing a bald white guy. It has become one of those things in life where I get a bit excited to see who a brand new artist will get as the matched image, reminds me of childhood, honestly.
I figure with the game decompilation news I'm seeing the reverse engineering of the firmware (or at least the DB format) would not be a significant challenge, and it would mix in some fun other tasks like compiling (and signing?) a new DB with tons of images and names, and then figuring out how to avoid bricking a vehicles computer while doing so. But... like I assume many others here... I've got enough projects on the go right now so maybe in a few months.
A cheap RTC clock chip would have costed a few cents, but Car manufacturers depend on GNSS timing for insurance purposes.
https://www.youtube.com/watch?v=oC6AiLqjtRg
I suspect a bunch of car head units will face new problems in the spring when many Canadian provinces will no longer change their clocks. The units won't get updated DST rules.
my skoda simply has an entry in the menu: DST? which puts the clock one hour ahead or behind.
This posts remind me that when i will be in the market for a new van i will have to decide to get the trafic or the vivaro, solely on the head unit they use (renault ones have had incredibly stupid design decisions for years, on ergonomics alone, even more in firmware)
(edit to be clear: not all places use 1 hour offsets)
This contributed to me in my very early years of dabbling in software development being oblivious to time zone considerations and would do really bad things like make sure the web server was setup with the correct time zone so that the ASP Classic output would show the correct time.
I don't get this attitude, I see it in many people to doubt by default when you should bookmark it mentally and think "what's the reason for this?"
And as usual, someone got bit ;]
How can I find the time to go down rabbitholes such as this in my own life (and not get bored and get distracted in the process)?
The Web is a universal platform very capable in 2026.
> When I bought the car, the previous owner asked me whether I wanted a "navigation map SD card" with it.
Why would they even ask?
> Since the maps on the SD card were apparently very out of date and didn't function properly, I decided I didn't need it and the previous owner tossed it.
WTF? My 2016 Mazda has an out-of-date SD card too. I removed it because it slightly speeds up how fast the system boots without it, but it's in the glove box for any future owner when I sell the car. I just can't understand tossing it. It's not like it takes up any room.
(My Mazda also sets its time via GPS, but I can get to the timezone offset and DST checkbox regardless of the presence of the SD card.)
Edit: available for $30 these days. Apparently the card gets locked to the VIN on first use.
https://www.amazon.com/Navigation-KR9D66EZ1B-Select-Mazda-20...
I still drive a car from 2011, but my headlights / fog lights / any other front lights that are not my emergency lights stopped working, super expensive to get it all reviewed by Toyota here in Florida (apparently Florida is the worst for these sort of things - already $250 down and they couldn't figure it out), and so left it with a relative who is a mechanic, and electrical problems are up his alley (I believe he studied this in particular too).
Both fuse boxes (driver side) and engine bay side are messed up (not enough voltage going into the engine bay one, after I replaced the driver side one with one found at a junkyard by a buddy for $80 compared to $1200 that Toyota bills for the part mind you).
Anyway, so I'm borrowing my moms much newer 2024 Toyota RAV4 (or whatever year), lots of neat things, but there's this really annoying feature I had no idea new cars have where the engine shuts off when you're stopped, I knew people did this to save on gas, but I always thought it was silly and a waste of time, and it felt really rare, until I realized why I hear engines start back up at red lights more frequently nowadays, if I have to floor it I dont want to wait, so I push the button depending on if I'm in frequent stop and go traffic.
My coworker drives a much newer (2026) SUV from VW... Apparently the button became like 4 screens in the touch-screen tablet the car comes with, just to force people to stop hitting the physical button, and you can't just disable it permanently.
Well.. unless you jailbreak the software, which he will likely do once his warranty runs out. Why do we force these ridiculous things on cars? It's no wonder something as ridiculous as Brabus exists... Since they're not a manufacturer they can remove all these things, and sell you a "performance" car, but its really mostly just the car restored to its normal state as it would have been released 10 years ago? or whatever.
Why do we have to go through so many hoops for things we "own" and why is it that the reliability of cars has gone backwards? Don't get me started on Nissans newer engines.
Who can make me a normal car that isn't a slug on Florida highways or turning off every time I hit the breaks? (Being dramatic, but its exhausting how much nonsense goes into cars now) Please, you'll have a customer for life.
Also - when it comes to getting parts for Japanese cars, I thoroughly recommend rockauto.com, a US colleague put me onto it, and I use it from NZ because a) it has parts I struggle to find in NZ and b) even with the international shipping, parts are cheaper from there than locally.
> Well.. unless you jailbreak the software
Maybe this is locked on VW, but given that Audi and Porsche share a lot of the same software, methodology, etc., I feel "decently confident" that this can be disabled via an OBD scanner (obdEleven, for one, was built around the VAG brands).
I was able to disable it in my fiancée's entry trim 2017 A4 in about 20 seconds, and no manual "long codes".
It's not an option needed on my RS 5, but I've done a few other things via the tool, and the dealership didn't have an issue with it - when I brought my car in for some other issue (water ingress into a sensor/light), all they said was "Are you okay with the tech resetting any changes you've made if they need to for diagnostics?" (they didn't), and "Do you accept that there will be a diagnostic fee if the tech doing a reset of changes resolves your problem?"
Every. Time.
Even if I beat my head with hammers to think like a lawyer, I can't imagine how it would be necessary to ask constantly. Doesn't matter if I accept or decline, it still asks again.
I can only imagine the amount of expletives that were said at that exact moment.
God, I hate modern cars.
Me too, but it's more a love/hate relationship, because aside from this issue, and a few other annoyances, I love this car to bits. I've been driving for over 25 years, and this car is my favourite of all of them, including some newer and fancier electrics. I think it's the throttle response. Nissan just nailed it.
Also, it now sometimes thinks my headlights are on and dims the display for a few seconds randomly which I guess isn't too bad after 15+ years. It doesn't seem to have any other problems except for worn out suspension and leaking exhaust.
https://www.josephecombs.com/2026/09/08/coding-an-audi-batte...
FWIW, there are grey market sites that will sell you an SD burned with a custom CID so you can clone an SD card with this kind of copy protection. It sounds like in this case maybe he could've gotten away with an empty maps card + menu definitions file.