Why Carmakers Can’t Transition to Newer Chips • Hacker News Board

Re: Why Carmakers Can’t Transition to Newer Chips

#111

Post by tremon » Sat, Oct 02, 2021, 5:26 PM UTC

finnh wrote: post #99

rectang wrote: post #58

Earlier quoted context omitted.

From earlier discussions here ( https://news.ycombinator.com/item?id=9643551 "10,000 global variables"), Toyota's conservative, arguably antiquated approach to software architecture wouldn't necessarily prove reliable as that software evolves. Trying to assess whether thousands of global variables are still playing nice during a major rewrite to accommodate a new chip would be definitely be difficult and time consumi…

10K global variables sounds insane, until i realize "oh, huh, we have that too, we just call it _config_". Almost every place we would have a constant (tuning parameters, etc) we instead have a configurable value with the likely default declared in code, but all overridable in config. Managing config can be a hassle, but the number of times we've merely had to tweak a value rather than roll a new build pays for itsel…

I always interpreted the "10K global variables" to mean 10K directly accessible, heap-allocated symbols. It's a code reek. If you use a configuration object or other abstraction, the number of accessible variables may not change, but their scope and access method does.

It's a matter of interpretation I guess, but I would not classify a wrapper object (like .Net's ConfigurationManager singleton) as "10K global variables" even if the accompanying config contained 10K items, or if the ConfigurationManager backing store was preallocated in the data section of the binary.

Re: Why Carmakers Can’t Transition to Newer Chips

#112

Post by coryrc » Sat, Oct 02, 2021, 5:26 PM UTC

lucidguppy wrote: post #5

This feels like lack of planning. The semiconductor industry was a known entity for many years before this level of integration came to vehicles. They should have known this would happen. Ultimately the industry will have to maintain their own production for consistent silicon.

Texas power outages took down one of the few major US plants and a random fire took down another, at the same time as a global pandemic. We're almost at the level of "imports banned from China"-levels of implausibility here.

Sometimes shit happens.

Re: Why Carmakers Can’t Transition to Newer Chips

#113

Post by SkyPuncher » Sat, Oct 02, 2021, 5:30 PM UTC

baybal2 wrote: post #57

SkyPuncher wrote: post #43

Earlier quoted context omitted.

It's really easy to over react because of the silicone shortage, but in many cars these chips are replacing physically manufactured parts or significantly increasing efficiency in a way that isn't possible with physical based components. Electronic fuel injection is a _great_ example. Much of modern fuel efficiency stems from the fact that an engine can accurately control fuel in the cylinder based on a bunch of fact…

I think you are misunderstanding me. What I mean under a computer is a Turing complete machine. A circuit just doing some LP solving, or PID is not necessarily a computer.

By that explanation, I actually 100% confident I'm understanding you correctly.

Most of the functionality of these chips can be represented in some form of hardware. However, that hardware is often much more expensive, significantly less flexible, and likely significantly bulkier.

Re: Why Carmakers Can’t Transition to Newer Chips

#114

Post by AlfeG » Sat, Oct 02, 2021, 5:30 PM UTC

radicalbyte wrote: post #88

shiftpgdn wrote: post #38

Earlier quoted context omitted.

Toyota accelerator-gate is largely a farce. Almost all of the people involved with Toyota unintended acceleration were over the age of 65 and had the poorly designed loose floor mats in their cars. Even if they had perfect readable John Carmack tier coding they still would have lost.

It literally took me years to understand it - how can you accelerate by accident? Until someone mentioned that all of these cars were automatics. But every automatic I've ever driven (not many - I prefer a clutch) moves forward unless you're pushing the break in. In the rest state, it's in motion. The problem isn't Toyota, the problem is a broken system.

Not sure how is this possible. Automatic cars almost always require You to push brake constantly if you not driving. You can't turn car on without pressing brake, can't change to drive mode from parking state without pressing brake. There should be something very broken in Toyota case.

Re: Why Carmakers Can’t Transition to Newer Chips

#115

Post by pwg » Sat, Oct 02, 2021, 5:36 PM UTC

baybal2 wrote: post #14

schiffern wrote: post #7

> “We were able to substitute alternative chips, and then write the firmware in a matter of weeks,” Musk said. “It’s not just a matter of swapping out a chip; you also have to rewrite the software.” https://www.theverge.com/2021/7/26/22595060/tesla-chip-short...

ARM M family cores can run each other's code more or less. Only the libraries for I/O need to be rewritten. In some cases, there are drop in replacements (Gigadevices STM32 clones)

Except that in some cases, the automakers are not using a CPU with external interface hardware, they are using microcontrollers similar to the PIC series, example here: https://ww1.microchip.com/downloads/en/DeviceDoc/40300C.pdf (Note, I'm not suggesting they use these exact chips, this is just an example of the level of onboard integration that is available for micro-controller chips).

Single chip system, with built in analog comparators, timers, signal capture and PWM generation module, serial UART, etc. I.e., a whole "system on a chip" with the external world interfaces already built into the chip. Changing one of these for an ARM CPU with external analog comparators, timers, PWM modules, serial UART's, etc., is a redesign effort, not just a "change the chip" effort. If the clock speed possible via the internal clock oscillator is sufficient, then this chip needs only +5V and ground (plus programming) to be able to interface to and control some external analog or digital equipment.

Re: Why Carmakers Can’t Transition to Newer Chips

#116

Post by coryrc » Sat, Oct 02, 2021, 5:37 PM UTC

dale_glass wrote: post #37

Wouldn't the solution be planning better? Sure, I understand that making a new car in combination with making a new ECU or brake controller on a new process is going to be stupidly dangerous and troublesome. So just don't. Rather than doing it as a part of a single project, there should be a department or separate company making the ECU/controller. This way when the semiconductor company moves forwards, ECU Group sta…

Your comment displays a breathtaking amount of ignorance.

Re: Why Carmakers Can’t Transition to Newer Chips

#117

Post by delfinom » Sat, Oct 02, 2021, 5:38 PM UTC

tremon wrote: post #111

finnh wrote: post #99

Earlier quoted context omitted.

10K global variables sounds insane, until i realize "oh, huh, we have that too, we just call it _config_". Almost every place we would have a constant (tuning parameters, etc) we instead have a configurable value with the likely default declared in code, but all overridable in config. Managing config can be a hassle, but the number of times we've merely had to tweak a value rather than roll a new build pays for itsel…

I always interpreted the "10K global variables" to mean 10K directly accessible, heap-allocated symbols. It's a code reek. If you use a configuration object or other abstraction, the number of accessible variables may not change, but their scope and access method does. It's a matter of interpretation I guess, but I would not classify a wrapper object (like .Net's ConfigurationManager singleton) as "10K global variabl…

>I always interpreted the "10K global variables" to mean 10K directly accessible, heap-allocated symbols. It's a code reek.

Which to me on an old _accelerator_ design is suspect because that's alot of heap for the micros that would have been used back in the day. This isn't a Linux system. It was at best a micro with like 8KB of RAM. (I'm not actually sure what it is but there's no way it was impressive).

Re: Why Carmakers Can’t Transition to Newer Chips

#118

Post by useful » Sat, Oct 02, 2021, 5:39 PM UTC

Ekaros wrote: post #96

cma wrote: post #69

Earlier quoted context omitted.

> risk-taking guts that Tesla exhibits to push new, better, electronic parts into test and production. After their delaminationg non-automotive grade screens I'd be wary. Their solution to it was to make the air conditioner always run in the sun when parked and promote it as a feature, wasting untold megawatt hours of electricity.

That sounds like possible death trap. Drive Tesla to some place without range, stay out there for week. Then come back an notice car is dead and you can't reach next charging station. Maybe this is why Musk is making Starlink...

you can tow charge an electric car with another car or truck

Re: Why Carmakers Can’t Transition to Newer Chips

#119

Post by thoughtstheseus » Sat, Oct 02, 2021, 5:41 PM UTC

01100011 wrote: post #53

shiftpgdn wrote: post #38

Earlier quoted context omitted.

Toyota accelerator-gate is largely a farce. Almost all of the people involved with Toyota unintended acceleration were over the age of 65 and had the poorly designed loose floor mats in their cars. Even if they had perfect readable John Carmack tier coding they still would have lost.

Sure but that's not related to the point I'm making. The point is that not following software design best practices can leave you open to liability, or at least put you in such a weak legal position that you are compelled to settle. If/when Tesla gets hauled into court over its software flaws, any lack of adherence to best practices will make the company appear negligent.

This is a risk adverse culture then. Tesla is spending gobs of money specifically on building safer if not the safest cars out there.

Re: Why Carmakers Can’t Transition to Newer Chips

#120

Post by syntaxing » Sat, Oct 02, 2021, 5:42 PM UTC

mwint wrote: post #110

csours wrote: post #79

Earlier quoted context omitted.

> "I wonder if the balance is shifting towards taking action." I can't find a public source, but some actions are being taken. Bear in mind that any major shifts would happen in a new product, that is, something 5-7 years out.

Fundamentally, what is it that prevents the established auto makers from making changes mid lifecycle? Seems like Tesla keeps pulling it off; why can’t GM? Is it that the component specification+acquisition cycle you describe is optimized to take as long as the development cycle of a new car?

Disclaimer: I work and worked for subsidiaries of big automakers but this opinion is of my own.

I would guess it’s a mixture of culture, cost, and scale. Culture wise, Tesla is extremely vertically integrated so that gives them a lot more breathing room. Cost wise, Tesla pretty much still sells car at a loss and relies heavily on carbon offset subsidies for income. When your revenue is established, trying to change margin or anything is probably much more difficult. Lastly, it’s scale, while Tesla is trying to catch up, the throughput of the major automakers is absolutely mind defying. The whole system is an oil machined that any sort of downtime is detrimental and difficult once the line is established. To put into perspective, Tesla “monumental” Q2 quarter shipped 200K cars or so. GM in the US only sold 200K per month.