Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Tesla, TSLA & the Investment World: the Perpetual Investors' Roundtable

This site may earn commission on affiliate links.
TL;DR - (But I'd recommend the article) - Traditional auto makers don't like to innovate in things they don't understand like 'computery-thingys' and it is going to keep hurting them for at least another 6 to 12 months.

Longer version - This article sums up pretty well why traditional auto is really hurting now as these ancient chips used to be high gross margin items and when they stopped ordering, due to the pandemic, it appears that the chip makers shut down and/or moved the capacity to other chips (like IoT devices for home electronics) from the old fabs to save money.

So they are forced to do super scary things like redesign a new circuit board (oh the humanity!) and recertify that new board design (OMG!). Most likely these new boards won't be redesigned by the auto makers themselves as they don't employ those kinds of engineers, and yes, they are hard to come by even if you know what you are looking for. They'll need to outsource the changes and I wonder how those 3rd parties are going to be able to react? Actually I don't wonder, I know, and it isn't pretty. Massive delays could be greater than 1 year for these critical parts. I know it sounds long, but that is **IF** the 3rd parties move quickly.

And this is the icing on the cake...

"As a result, as many as 9.4 million cars, or more than a tenth of the industry’s pre-pandemic output, could be eliminated from production plans, according to market research firm AutoForecast Solutions.

“Because of a 50-cent chip, we are unable to build a car that sells for $50,000,” said Murat Aksel, head of procurement for Volkswagen Group, during a press briefing in Munich last week.

If semiconductor suppliers like Intel and Qualcomm have their way, however, the days of the auto industry relying on these cheap commodity chips are numbered."

Fortune: Chipmakers to carmakers: Time to get out of the semiconductor Stone Age.
Ahem. Speaking as a practicing EE design-type:
These days, I mainly pick up the smoking bits of failed boards and then am involved in redesigning them or their brethren so they don't do that again, whatever "that" happens to be, this time around. Having said that, I've designed boards and such for 30-odd years now.
And there's something serious to say about practice, practice, practice.
Nobody expects, or should expect, a newbie straight out of college to sit down and design a circuit board or system from scratch without a ton of help from the more experienced members in the local design community. The best of the old farts realize that the newbies got spiffy new ideas that, when implemented, can save on costs, reduce time to market, and reduce errors in design and implementation. And the reason one keeps the old farts around is that they have scars on their bodies for every time they, or somebody they knew, or information on a paper they read, etc., that provides information on Here's The Wrong Way To Do It, in Three Part Harmony.
And good design shops worry about tool chains, improvements in tool chains, and how to keep on top of the technology. There will be groups of people, typically, beating their heads against the wall to make sure that what goes out the door works, works well, and doesn't die in three months.
It's possible, as in, "Theoretically Possible" for one person to pull all that off. In practice, if one wants to get stuff out in six months to a year, the minimum size of the hardware team is 5 to 15 people, depending upon the complexity of the task, and it could be easily 3 or 10 times that.
There's organizational dangers in this. It's possible to pluck some 45-year-old out of the middle of such a design group, where everybody's been sharing the various tasks, and put that person out of the organization and into some new one to roll out two or three fast, new designs. Such an engineer, if he/she's a good one, may be able to pull it off; but won't be able to keep it up - no support, and nobody to help with inspections to catch the inevitable mistakes. But let some upper management, non-engineer see that person manage to pull it off once, and that pointy-haired boss will somehow think that they're geniuses for organizing it once and will continue to push that process forward. And this is how one kills design teams: Remove the effectives to somewhere else for, "Just this quick fix!", scant on the engineering and software support (Hey! Engineers are simply a cost to be reduced - they don't actually bring in any cash!), and then wonder why outside companies can come up with better stuff. (Mainly, because those outside companies are smaller, and the CEO's and such can directly see where bad management practices lead. And, if they don't, the outside company goes under, replaced by another with better smarts.)
The other point about having working design teams: They're continuously designing. In a continuously designing shop, the engineers within are used to the idea of incremental, new technology designs, and how to do them, complete with tool sets. Give them a new job, they use all the stuff from the last job, fold new stuff into the new job, and hit the ground running. Worst thing a management team can do: If, for some reason, new jobs aren't present for six months, then fire all the engineers and re-hire them in six months. (Think: Aerospace, outside of SpaceX.) At some level in a large company there's always some idiot who thinks that engineers are interchangeable parts. Yeah.. that almost might be true, if one considers calling somebody "interchangeable" who's barely experienced in some field and having to wait six months while they get up to speed.
It's a failure of management to run out of work for design teams to do. If there's no spanking new stuff to build, then one assigns them work on redesigning (and cost reducing) the old stuff. Yeah, it might not be cost effective, paying all those loaded salaries just to knock $15 off on old board; but it's far cheaper to do it this way than lose the opportunity savings when one can't design the New Thing when it comes up due to a lack of human and infrastructure resources.
So, redesigning a circuit board to go from one hard-to-get processor to one that's easier to get isn't exactly a major design change. Unless one fired all the people who did the original board because their services weren't required, in which case restarting the whole design community required just can't be done in zero time since none of those people are in the organization and/or are retired, and no new people have been brought on board. And if one is waiting for a 3rd-party supplier to do it for one.. Lotsa luck keeping up.
Any questions about those high-faluting newbie design engineers graduating from college all wanting to go to work at Tesla? Imagine going on an employment tour at Detroit and seeing nothing but white-haired, near-retirees doing no circuit design (that's all outsourced) but just running tests on stuff from vendors? And compare that to walking into Tesla, where everybody is bumping elbows and going, "Dude!"
 
Beware the investors dressed in frippery and low quarterly expectations, but inwardly are ravening jaguars!
 

Attachments

  • 3_MPM_JAGUAR_DIVE_05-650x1024.jpg
    3_MPM_JAGUAR_DIVE_05-650x1024.jpg
    95.1 KB · Views: 63
Just came back from Tesla service, they were great. Control arm on the car was replaced under 24h and I got a loaner for the first time.

My mind however got blown by the guy in front of me who came in with his used Model X he purchased 2 weeks ago and complained that the nearby superchargers were full. He wanted the service center to charge his car. The service rep said it'll take a hour because there are two cars being charged. The guy said "I am going to sell this car in 2 weeks because the service has been terrible! You tell me I have to wait a hour?". The rep was like "sorry sir but I have to bring your car in, then there's the charging time".....

I wonder if my wife can go into volkswagen tomorrow and complain about her empty gas tank...

Did you get his number by any chance? I'll buy him out of his misery for about 75% of his purchase price - and suffer through the "bad service".
 
Senator Manchin drops another turd in the punch bowl - doesn’t just want to significantly cut down the $3.5T spending bill (which contains the EV tax credit changes), but ALSO is wanting a vote on it delayed until sometime in 2022.

Important to remember this doesn’t just Impact Tesla, but screws GM as well which I assume is a giant part of the UAW workforce. So who knows, maybe a prolonged delay could see the EV tax credit moved into a separate standalone bill for quicker passage given its importance to Biden’s and the dems union base in the swing states.

 
@Tronguy @Discoducky
Risk v reward
I was at at a Tier 1 on the other side of electronics design. A stopped OEM factory costs millions per hour in penalties. How do you stop an OEM factory? Find a problem on a released module.
Production Validation alone is months of testing (once the test systems (ideally reuse on micro change) and test harnesses (reorder, but high labor time) exist. And don't get me started on integration of required 3rd party CAN drivers for the new micro where you pay them to fix their bugs (eventually).

Tesla can iterate because it's all in house and they don't fine themselves if something ends up not working. One big team (AFAIK).
 
I expect more free advertising on CNBC tomorrow morning due to Elon's multiple feisty tweets towards the president. (I like feisty Elon, especially when he's 100% correct)
Although, I disagree with Biden's Tesla policy or lack of one. I can't see how this helps Elon Musk/Tesla. Some cooler heads need to speak to each of them about their common interests. After all, Biden does not need to antagonize 60 million plus Elon Musk twitter followers.
 
@Tronguy @Discoducky
Risk v reward
I was at at a Tier 1 on the other side of electronics design. A stopped OEM factory costs millions per hour in penalties. How do you stop an OEM factory? Find a problem on a released module.
Production Validation alone is months of testing (once the test systems (ideally reuse on micro change) and test harnesses (reorder, but high labor time) exist. And don't get me started on integration of required 3rd party CAN drivers for the new micro where you pay them to fix their bugs (eventually).

Tesla can iterate because it's all in house and they don't fine themselves if something ends up not working. One big team (AFAIK).
Yep, nail on the head!
 
Although, I disagree with Biden's Tesla policy or lack of one. I can't see how this helps Elon Musk/Tesla. Some cooler heads need to speak to each of them about their common interests. After all, Biden does not need to antagonize 60 million plus Elon Musk twitter followers.
There are 7.2 million union workers in the U.S. Musk has 60 million+ followers. Think they are not playing the right cards. Also it's very easy to not antagonize Musk but the left decided to target him which is suicidal for elections considering his followers are probably mostly blue.
 
Senator Manchin drops another turd in the punch bowl - doesn’t just want to significantly cut down the $3.5T spending bill (which contains the EV tax credit changes), but ALSO is wanting a vote on it delayed until sometime in 2022.

Important to remember this doesn’t just Impact Tesla, but screws GM as well which I assume is a giant part of the UAW workforce. So who knows, maybe a prolonged delay could see the EV tax credit moved into a separate standalone bill for quicker passage given its importance to Biden’s and the dems union base in the swing states.


Given most tesla models are sold out now until way into 2022 already and GM dont have an EV to sell until LG figure out the whole non exploding car thing I'm not sure how much this hurts tesla or GM.
 
There are 7.2 million union workers in the U.S. Musk has 60 million+ followers. Think they are not playing the right cards. Also it's very easy to not antagonize Musk but the left decided to target him which is suicidal for elections considering his followers are probably mostly blue.
Plenty of his followers are foreigners and play no part in our elections.
 
Once again bragging rights about being *first* is out done by the guy and company who have the sincerest and purest motives.
Indeed.

What's unfortunate is that the average person doesn't really appreciate that the orbital trip those 4 folks is a couple orders of magnitude more difficult than the Karman hops the other two trips were...