4 ms·

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

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 starts a project targeting the new process, and releases the results when it's ready. And meanwhile what goes into the cars is the ECU built on the previous, well tested process. Also, are modern cars really in much need of custom silicon? Micro-controllers are stupidly powerful now. There has to be off the shelf hardware capable of handling what a car needs. There are long standing architectures like ARM that don't require starting from scratch every time somebody makes a better chip.

coryrc 5y ago
Your comment displays a breathtaking amount of ignorance.
- dale_glass 5y ago
  Then I welcome the opportunity to learn about something new. Please explain what's wrong, I'm most interested.
  - coryrc 5y ago
    Start reading IEEE, SAE, or other trade magazines; get an EE degree in the Midwest; get a job in the automobile supplier industry; go to industry conferences or just read their proceedings...
    - dale_glass 5y ago
      Or you could do me a favor and explain, since you already seem to know.
pas 5y ago
And usually this is what happens. Sort of. Big brands have new cars in the pipeline, and their product teams and thus the whole supply chain has newish stuff in development. [0] But since the market is rather fragmented/niche/specialized (even if the gross volume is large, the number of market participants is small), it simply makes no sense to launch a new component without already having orders for it. The problem is that when COVID struck almost everything was put on hold, scrapped, and ... it simply takes too much time to get things back on track. (As there is basically a full half year of shortage.) > are modern cars really in much need of custom silicon Well, again, because the market is special, there's a lot of custom stuff, even if it's almost the same chip internally. Lack of openness and standardized interchangeable components lead to this. (Yes, there are a lot of aftermarket parts, but they are probably even worse.) Also, probably the big dilemma for automakers is that they are simply not accustomed to this level of transparency in their supply chains. Now they point the blame at chipmakers, but they don't buy chips, they buy components from (the cheapest) vendors. (Who are cheap because they also never had any stored inventory, nor rainy day funds, nor any elegance and care in their designs that would allow this kind of retooling to other chips.) [0] But only very-very-very tiny incremental newishness. Large major version upgrades are seen as something that is too risky. (See this part from the article: "Quigley added that trying to design new chips and vehicles that will use them in parallel often introduces yet more headaches.") And even if the whole industry is a decade behind, there's no time to catch up, because there's just no market need for upgrading that subcomponent to have a better software/CPU platform. Instead automakers launch wholly redesigned lines. Or source completely new components for the new set of functionalities envisioned. (But if some vendor can meet the new specs with again a bit of incremental upgrade, then that'll happen.)

coryrc 5y ago

Your comment displays a breathtaking amount of ignorance.

dale_glass 5y ago
Then I welcome the opportunity to learn about something new. Please explain what's wrong, I'm most interested.
- coryrc 5y ago
  Start reading IEEE, SAE, or other trade magazines; get an EE degree in the Midwest; get a job in the automobile supplier industry; go to industry conferences or just read their proceedings...
  - dale_glass 5y ago
    Or you could do me a favor and explain, since you already seem to know.

pas 5y ago

And usually this is what happens. Sort of. Big brands have new cars in the pipeline, and their product teams and thus the whole supply chain has newish stuff in development. [0] But since the market is rather fragmented/niche/specialized (even if the gross volume is large, the number of market participants is small), it simply makes no sense to launch a new component without already having orders for it. The problem is that when COVID struck almost everything was put on hold, scrapped, and ... it simply takes too much time to get things back on track. (As there is basically a full half year of shortage.) > are modern cars really in much need of custom silicon Well, again, because the market is special, there's a lot of custom stuff, even if it's almost the same chip internally. Lack of openness and standardized interchangeable components lead to this. (Yes, there are a lot of aftermarket parts, but they are probably even worse.) Also, probably the big dilemma for automakers is that they are simply not accustomed to this level of transparency in their supply chains. Now they point the blame at chipmakers, but they don't buy chips, they buy components from (the cheapest) vendors. (Who are cheap because they also never had any stored inventory, nor rainy day funds, nor any elegance and care in their designs that would allow this kind of retooling to other chips.) [0] But only very-very-very tiny incremental newishness. Large major version upgrades are seen as something that is too risky. (See this part from the article: "Quigley added that trying to design new chips and vehicles that will use them in parallel often introduces yet more headaches.") And even if the whole industry is a decade behind, there's no time to catch up, because there's just no market need for upgrading that subcomponent to have a better software/CPU platform. Instead automakers launch wholly redesigned lines. Or source completely new components for the new set of functionalities envisioned. (But if some vendor can meet the new specs with again a bit of incremental upgrade, then that'll happen.)