The public never usually hears about it, that's why. There are entire companies that specialize in tearing down products and selling reports to customers.
It's not the total price that's a shocker rather they needed to pay $55,000 over the list price. That's way above average even if it's chump change for a major car company.
This is a company with a $54B market cap and a 2015 pre-tax net income of $10B out of $150B in revenue. Yeah, I think they can spare an extra $55,000 now and again.
They didn't even need to, they just didn't feel like waiting in line, or locating a non-Founder's unit to purchase. They must have enough cash allocated for buying cars that they just don't care. And of course there's nothing wrong with that.
This makes me think of the iPhone + RIM (Blackberry) and what a shift in thinking they had after pulling apart the first iPhone they could get hold of.
"RIM was even in denial the day after the iPhone was announced with all hands meets claiming all manner of weird things about iPhone: it couldn’t do what they were demonstrating without an insanely power hungry processor, it must have terrible battery life, etc. Imagine their surprise when they disassembled an iPhone for the first time and found that the phone was battery with a tiny logic board strapped to it. It was ridiculous, it was brilliant."
Legally - you can do some things you figure out in reverse engineering that you could not do if you had the engineer.
Plus, does 1 engineer really know all parts of the Tesla? Can he tell you about the types of lugs used on the wheels? Can he also tell you about the type of glass used in the windows, or the type of leather used in the seats? If cars are anything like software, each engineer knows like 5% of the total product..
Are you trying to say that there is no human endeavor which cannot fit within a single person's mind, or is there some relationship between calculus and a car here?
I'm saying that understanding things thought up and formulated by others is orders of magnitude easier than doing them from scratch. This simple property gets most of us through the schools and to the workplaces.
It is certainly possible to fully understand a car down to minute detail. It's even easier if you get briefed on design decisions.
Do you think it's possible to, say, understand the entire Linux kernel? Because these cars have that, and other kernels besides, just as one example of complexity.
Sure, a single human can have a decent understanding of everything in it, but not down to the deep details. There just isn't enough time to learn all of that.
Since the topic here is the teardown, thought we were discussing the mechanical side of it. As of the Linux kernel there are actually a few people who have very, very thorough understanding of near every design aspect there, even if they didn't memorize every line of code. Linus Torvalds is certainly one.
That's excessive for automotive software, however; you do some device drivers here and there but ain't going to modify the memory allocator or process scheduler. How much detail Musk keeps on to here is open to speculation. On my part, I can say the industrial products I developed I understand really well. Starting from where and why ejector pins on diecasting molds go, the particular conductor plate used for spot welding, wire gage in the cable bundles, through kernel drivers, FPGA programming, EMC compliance issues in PCB layout to how many units fit on a euro pallet.. and am no Elon Musk. Don't sell your ability short.
If you use broad enough definition, where making a car entails mining your own iron ore and building lathes, presses and milling machines from scratch, sure. Otherwise plenty of people have built cars (including engines) and other vehicles from start to finish.
The question was not about building however, but understanding the design of the whole thing down to a fine detail.
I'd bet that Ford and the other big automakers have comparable or even better engineers and designers than Tesla. It's just that their management doesn't tell them to do quite the right things. Seriously, I think the designers of the Chevy Bolt show their genius, but they were instructed to build an econobox/wagon/hot-hatch mashup. The Tesla 3 is just the thing to capitalize on their existing brand.
Because Tesla's engineers all live in California, and Ford's engineers all live in Detroit, and Tesla's engineers really don't want to live in Detroit.
The CEO of Ford made over $17 million last year in compensation, but it was worthwhile to write an article about a 200k R&D expense, with a senior industry analyst giving a quote seriously questioning the $50k spent over sticker price in order to get the vehicle immediately instead of waiting in line.
Is this really a story? Or just clickbait because it has Tesla in the title?
I think it's just supposed to be noteworthy because it is unusual for them to pay that much, and it was specifically to get an early iteration of this particular model.
I wonder if there's some reason they specifically went for one of the earliest ones sold. Maybe as one of the first ones made, it would be more in "prototype" mode versus one of the latest versions.
The news is not about the amount indeed, it's basically pocket money for Ford. However, they go in great lengths to acquire a Tesla through unusual channels. The founders edition are very limited and rare and only given to close friends & board members. They wouldn't do this with a new Honda I imagine.
What makes it interesting is to see how the car industry reacts to Tesla's disrupting the industry, which in this case is one reaction (of many). Thats what I think is fascinating.
>They wouldn't do this with a new Honda I imagine.
Long waiting lists aren't the norm for most cars. I have very little doubt that Ford routinely does timely teardowns of major vehicles that it competes with.
> They wouldn't do this with a new Honda I imagine.
Appropriately enough Honda acquired something like 20 Porsche 911s when developing the S2000, all individual purchases to hide the fact.
Cessna routinely did the same with competitors aircraft, back when they actually developed aircraft, but to save money they only leased them and had to ensure they were put back together.
> They wouldn't do this with a new Honda I imagine
I worked in an advanced vehicle design skunkworks at GM Design Staff in the late 1980's (a time when the quality of Japanese cars was a top threat to US car makers). While there I had access to a wide range of top of the line Japanese-domestic-market-only vehicles imported into the US specifically for teardowns just like this, frequently before Japanese buyers could obtain them locally domestically. In those days, it wasn't electric power plants that were the issue, it was (among other things) the density and efficiency and downright beauty with which the engine compartment contents were packed. I still remember standing around an open engine with a bunch of hotshot car guys, all of us in awe at the look of the engine compartment in front of us. There was nothing on the US market packed anywhere near as tightly, yet everything was completely accessible and downright beautiful. In hindsight I strongly suspect this was a sign they started using CAD packing tools long before we did (laser scanning of hand-sculpted car designs was still a few years away in the US, to say nothing of actual CAD-first designs). If the packing was done by hand (as it might well have been) I've genuinely no idea how they managed the complexity of that design task. Sometimes you need to see that someone has completely outclassed you in order to break free and take your own thinking to the next level. So yes, when there are Japanese or German or any other cars that stand out as major design advances, you can be sure Ford and GM and all the other major manufacturers have them in review and teardown the day they start coming off the assembly line.
I suppose if I were carrying your particular axe I would have to take this story as a grindstone but since I'm not, I didn't get any outrage out of this puff piece. I have to imagine I'm lucky in that regard. I'd hate if something so innocuous were able to stoke unrelated class anger so easily. It would mean I'm cheaply manipulable.
For instance, if I were carrying that particular grindstone, I would feel immediately compelled to downvote any comment that pointed out my emotional shackles in plain language
We've asked you before not to be personally abrasive on HN. We ban accounts that do this repeatedly. Please post civilly and substantively, or not at all.
Also, please re-read the HN guidelines (https://news.ycombinator.com/newsguidelines.html, specifically the two at the bottom) and don't post any more comments about getting downvoted. Those are out of bounds for good reason.
Why is is that reverse engineering gets people cease and desist letters, especially in the software and electronics community? Is the difference that Ford keeps the findings internal and doesn't share them with the public? If so, that's pretty bad, because someone dissected a product and now everyone else should do the same instead of sharing the result. It's a waste of money and work force, all in the name of trying to hide stuff even though patents and copyrights are abused already for similar purposes.
It's not the act of disassembling that's important, it's what you do with the knowledge afterwards. For example, the team doing the Tesla teardown is now poisoned; they can write up their findings from the clean room and provide them to other engineering teams for implementation (IP issues aside), but they are now barred from working on any derivative technologies themselves. It's a bit cumbersome, but this is how engineering espionage is done.
So what does this mean for the career of one of these engineers? This sounds like a pretty ratty gig: take apart someone else's product, write up a report, and now you don't get to do any actual engineering of your own unless you quit and find a new job.
It makes me wonder how useful the information will be, since it'll likely be done by the worst engineers in the company. Why would a good engineer sign up for this job?
If it can be enforced in court (or has) it's most likely for a limited time, or you would pretty much preclude someone from applying what they've learned through investigation.
I'm not talking about legality (at least not directly), I'm talking about how working on this project will affect your career at this company. If the company puts you to work taking apart someone else's product and documenting it, but then relegates you to that role forever because they don't want you "contaminating" their IP with anything that might be patented, that means that you can never go back to an engineering development role at that company. You'll always be stuck just taking stuff apart and documenting it, and when you get tired of that, you'll have to quit and find yourself a new job at a different company.
So I'm questioning what kind of quality work you're going to get. It doesn't seem like top engineers are going to want to take that job, and in fact it'll probably be the bottom-of-the-barrel engineers who'll want this job because it's a stable paycheck and there's zero room for creativity or inventiveness. So what kind of results are you going to get, by not having really talented and knowledgeable engineers looking at it?
Okay, but I've seen my share of various engineers and different people are motivated by different things. That means, I know people who love taking things apart but don't really have much interest in building stuff from scratch. They love to take apart and modify mostly. And I wouldn't classify those I've met as lower-skilled.
Maybe, but I have a very hard time buying the idea that there's engineers out there who don't want to do any engineering, and just want to take things apart and document them, and are also very talented engineers.
I do realize that not all engineers want to do development: that's why we have test engineers. I've dealt with test engineers before; they're a special breed, and have a real talent for finding weaknesses and enjoy that challenge a lot. That's rather different than just disassembling something and documenting it. There's no challenge there, like there is in testing something to its breaking point and finding its weaknesses so you can make it better.
Citation please. This is completely different from implementing a BIOS from a set of specs created from original source code.
To the degree there is protected IP, it would be covered by patents. Which are public. That's the point of a patent. And it doesn't matter if you were aware of the original art or not if you violate a patent.
If there are merely trade secrets/techniques that Tesla or anyone else simply hoped no one would be able to deduce but weren't otherwise protected, those are fair game.
Personal experience dealing with a team that reverse-engineered industrial hardware. They would acquire competitors' products and hand them off to teardown teams who weren't allowed direct communications with the rest of engineering. These strictures may have been a legal artifact of the company's significant background in software as well as hardware dev, or perhaps to deal with contractual issues, but they were firmly enforced.
Yeah - this type of stuff is what I would expect Ford to be doing. It might be different if they bought a gasoline Subaru but this is a Tesla which is a brand that is supposed to be revolutionizing the automotive market. If they paid twice or three times the price to get something as early as possible I still wouldn't be surprised.
In the book "Soul of a new machine" there is described a moment in 1978 when an Data General engineer (Tom West) gets to go look at and take apart a newly installed DEV VAX at a friends company.
The comments here are really interesting. I agree that I'm not sure why this is news worthy since it seems like companies spend much more than this doing the exact same thing in other verticals. I'm sure Google and Samsung have spent hundreds of thousands, if not millions, doing the same thing to Apple products. R&D includes trying to understand how others do it. Spending nearly $200K to tear apart a Tesla seems pretty reasonable to me.
Assuming that Apple has 10 products on the market at any time, and you buy one per month at $1,000 each, that's $10,000 dollars a month, so for $120,000 you could fund the teardowns for a year.
When Lexus launched the LS400 two percent of the sales that year came from auto manufacturers. They could not believe how the car was so good. They still haven't caught up in many ways.
OT: the letters PR have a primary association with Public Relations, i.e. people who are paid to spin the truth on the behalf of others. I see from your profile that those are your real-life initials... but it's not a great first impression. You might want to open an account as Pablo-asimuv, or, really, anything else.
(I checked because this comment sounded an awful lot like a PR exercise for Lexus.)
I thought about it when creating the account, but decided to go with it. Since everything I post represents ASIMUV as well its kind of a weird PR kind of thing too... I do appreciate you making the check. HN is a very important community for me and following its rules is crucial.
The post does sound like a PR exercise for Lexus, but do a quick web search and you will be impressed with how good the Lexus was back then. They still are great cars. I have never nor do I have any future projects or connections with Toyota/Lexus or any of its affiliates. I just think they build good stuff. :)
Good for Ford. They'd be negligent to not try their hardest to get some competitive advantage any above board way possible. I don't see what the fuss is about. OMG AMD buys Intel chips and checks them out!!! Sometimes they pay over market price.
Carmakers continually buy all sorts of competitors' models for comparison testing and R&D purposes. From what I remember, the company I interned for (one of the big 3 in Detroit) bought Ferraris, Porsches, BMWs, etc to compare drive quality and for R&D. Additionally, sometimes they don't want a new car, they want one that is 10 years old and has real world wear and tear. They essentially have a broker who will find any car they ask for. I worked on the alloy design team, and one of the more unique stories I heard was about 10 years ago, they bought a 20 year old Mercedes to essentially cut into pieces. Why? The model they bought was one of the first cars with a unique alloy in the body IIRC, and they wanted to run mechanical as well as characterization tests on it to see how it aged.
69 comments
[ 3.3 ms ] story [ 143 ms ] threadMy guess is that this was the first one Ford was able to buy at any price....
"RIM was even in denial the day after the iPhone was announced with all hands meets claiming all manner of weird things about iPhone: it couldn’t do what they were demonstrating without an insanely power hungry processor, it must have terrible battery life, etc. Imagine their surprise when they disassembled an iPhone for the first time and found that the phone was battery with a tiny logic board strapped to it. It was ridiculous, it was brilliant."
From: http://www.edibleapple.com/2010/12/28/rim-was-in-disbelief-f...
This is what innovation looks like - its so absurd that it might actually work !
Plus, does 1 engineer really know all parts of the Tesla? Can he tell you about the types of lugs used on the wheels? Can he also tell you about the type of glass used in the windows, or the type of leather used in the seats? If cars are anything like software, each engineer knows like 5% of the total product..
It is certainly possible to fully understand a car down to minute detail. It's even easier if you get briefed on design decisions.
Sure, a single human can have a decent understanding of everything in it, but not down to the deep details. There just isn't enough time to learn all of that.
That's excessive for automotive software, however; you do some device drivers here and there but ain't going to modify the memory allocator or process scheduler. How much detail Musk keeps on to here is open to speculation. On my part, I can say the industrial products I developed I understand really well. Starting from where and why ejector pins on diecasting molds go, the particular conductor plate used for spot welding, wire gage in the cable bundles, through kernel drivers, FPGA programming, EMC compliance issues in PCB layout to how many units fit on a euro pallet.. and am no Elon Musk. Don't sell your ability short.
Now a car? NO WAY, not even Musk.
http://freakonomics.com/podcast/i-pencil/
The question was not about building however, but understanding the design of the whole thing down to a fine detail.
Is this really a story? Or just clickbait because it has Tesla in the title?
I wonder if there's some reason they specifically went for one of the earliest ones sold. Maybe as one of the first ones made, it would be more in "prototype" mode versus one of the latest versions.
What makes it interesting is to see how the car industry reacts to Tesla's disrupting the industry, which in this case is one reaction (of many). Thats what I think is fascinating.
Long waiting lists aren't the norm for most cars. I have very little doubt that Ford routinely does timely teardowns of major vehicles that it competes with.
They do the same with all their competitors. (Yes even hondas)
Get one early, measure tolerances, get one 6 months later, see if their manufacturing practices got better.
Appropriately enough Honda acquired something like 20 Porsche 911s when developing the S2000, all individual purchases to hide the fact.
Cessna routinely did the same with competitors aircraft, back when they actually developed aircraft, but to save money they only leased them and had to ensure they were put back together.
Edit: finished the anecdote about Cessna
I worked in an advanced vehicle design skunkworks at GM Design Staff in the late 1980's (a time when the quality of Japanese cars was a top threat to US car makers). While there I had access to a wide range of top of the line Japanese-domestic-market-only vehicles imported into the US specifically for teardowns just like this, frequently before Japanese buyers could obtain them locally domestically. In those days, it wasn't electric power plants that were the issue, it was (among other things) the density and efficiency and downright beauty with which the engine compartment contents were packed. I still remember standing around an open engine with a bunch of hotshot car guys, all of us in awe at the look of the engine compartment in front of us. There was nothing on the US market packed anywhere near as tightly, yet everything was completely accessible and downright beautiful. In hindsight I strongly suspect this was a sign they started using CAD packing tools long before we did (laser scanning of hand-sculpted car designs was still a few years away in the US, to say nothing of actual CAD-first designs). If the packing was done by hand (as it might well have been) I've genuinely no idea how they managed the complexity of that design task. Sometimes you need to see that someone has completely outclassed you in order to break free and take your own thinking to the next level. So yes, when there are Japanese or German or any other cars that stand out as major design advances, you can be sure Ford and GM and all the other major manufacturers have them in review and teardown the day they start coming off the assembly line.
Also, please re-read the HN guidelines (https://news.ycombinator.com/newsguidelines.html, specifically the two at the bottom) and don't post any more comments about getting downvoted. Those are out of bounds for good reason.
We detached this subthread from https://news.ycombinator.com/item?id=11534726 and marked it off-topic.
There are patents.
This would be a form of white box reverse engineering.
It makes me wonder how useful the information will be, since it'll likely be done by the worst engineers in the company. Why would a good engineer sign up for this job?
So I'm questioning what kind of quality work you're going to get. It doesn't seem like top engineers are going to want to take that job, and in fact it'll probably be the bottom-of-the-barrel engineers who'll want this job because it's a stable paycheck and there's zero room for creativity or inventiveness. So what kind of results are you going to get, by not having really talented and knowledgeable engineers looking at it?
I do realize that not all engineers want to do development: that's why we have test engineers. I've dealt with test engineers before; they're a special breed, and have a real talent for finding weaknesses and enjoy that challenge a lot. That's rather different than just disassembling something and documenting it. There's no challenge there, like there is in testing something to its breaking point and finding its weaknesses so you can make it better.
To the degree there is protected IP, it would be covered by patents. Which are public. That's the point of a patent. And it doesn't matter if you were aware of the original art or not if you violate a patent.
If there are merely trade secrets/techniques that Tesla or anyone else simply hoped no one would be able to deduce but weren't otherwise protected, those are fair game.
[0] http://www.wired.com/2014/07/go-inside-the-lab-where-gm-tear...
Making them public is part of patenting anythong I think.
Oddly VAX cost $200,000.
Very short exert http://kottke.org/14/07/the-soul-of-a-new-machine
Long exert of the book which continues the story of the VAX disassembly was published in the Atlantic called "flying upside-down".
http://www.theatlantic.com/magazine/archive/1981/07/flying-u...
Soul of a new machine is a really good book. highly recomended.
*Edit:
Here is more about the history of the car itself: http://ateupwithmotor.com/model-histories/lexus-ls400/
(I checked because this comment sounded an awful lot like a PR exercise for Lexus.)
The post does sound like a PR exercise for Lexus, but do a quick web search and you will be impressed with how good the Lexus was back then. They still are great cars. I have never nor do I have any future projects or connections with Toyota/Lexus or any of its affiliates. I just think they build good stuff. :)
Strikes me that a reasonable sized team would cost that per-day.
What if it is to try to find patents they can sue over?