Zhelko Dimic
Careful bull
Really Professor? Sugar happens so much to him, I'm afraid he'll become diabetic ...
Last edited:
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
- Status
Really Professor? Sugar happens so much to him, I'm afraid he'll become diabetic ...
As I mentioned in a previous post of mine. Part of the production (tooling to be exact) depreciates over units but everything else except labor and parts depreciates overtime and is independent of units. So if they ordered a big pressing machine designed to produce 200k a year to fill up those veiled area but only produced 30k, the depreciation of that big machine will be shared by the 30k units, increasing the COGS significantly before ramping up. I believe this is what we are seeing for the around 0 gross margin of current TE.
Also in that previous post of mine, I don't think we can rule out the possibility that they thought they got it, hired labor to go for 200k a year, but due to lack of sufficient time to check and re-check, ended up with 30k and the labor cost per car again would be significantly higher than expected. Very similar to the Model X's case you present here.
Due to the worries of no mass production, I am worried about the gross margin of Model 3 in 2017 if no significant amounts were produced. It would reinforce the perception of Tesla loses money on each car they make.
The perfect car for Norway is a wagon, especially with AWD and tow hitch. People gladly pay up towards 100k USD for a Volvo V90 or V90 Cross Country (including taxes). Over 500 people have preordered the V90 Cross Country, and they expect to sell around 1300 in 2017. That's a lot for a relatively small country.
RobStark
Well-Known Member
The Volvo's are not Lucid/FF bulbous aerodynamic ugly.
We shall what sales are vs Model S if and when they reach Scandinavia.
A lot of people had to made big leaps of faith buying Tesla in the bigginging. Trusting in battery degradation. Overall safety. Battery integrity in a crash.
Tesla has established that it's cars are safe and secure. Not sure how many people are willing to go out and take that risk on another company when soon after the 3 the model y will come out, and the pickup and the semi. I think it might be awhile and a lot of investment before another company can successfully sell electric cars. I don't find it likely any others are going to get the benefits of being the only option that Tesla got.
I tend to agree they probably won't be as good-looking as the Volvos. (They are however still camoflaged, so it's hard to tell exactly.) And Volvo does have a strong brand here. Like Green Pete I'm also a bit doubtful about whether the FF or Lucid Motors offerengs will be very compelling coming from very new startups.
As a competitor to Tesla, the Jaguar I-Pace is likely the top contender here in Norway, with 741 preorders as of three days ago. If it has a tow hitch, access to Tesla superchargers and the right price, it's not impossible for sales to exceed those of the Model 3 here. (We really do hate sedans.) In fact, I might get one, if Jaguar executes it well.
Last edited:
Access to supercharger? I find the to be about as likely as Tesla making a ICE
Tesla has said they'll allow access to other car companies, as long as they shoulder their share of the costs. They have also confirmed they have been approached by a european car company: Tesla “Totally Cool” With Rivals Using Supercharger Network, Already In Talks With European Firm
One thing is certain, the advertized 50 kW charging doesn't cut it. That is one area where Jaguar needs to over-deliver.
Ulmo
Active Member
Agreed on both points. (I don't see Jaguar also benefiting from (non-existent) in-transit charging.)
Notice how Tesla has steadfastly expanded their proprietary network of SuperChargers and Destination Chargers, while saying "sure others could share our network if they negotiate proper support"? Tesla's attitude all along has been to jump-start the competition, and not look behind. Today, new standardized charging standards are barely on the scene yet (CCS?), but meanwhile Tesla is going full speed ahead with what they are already selling and what already works. This helps Tesla expand their real customer base and all the while the proprietary nature of it makes the competition have to overcome that hurdle by being inspired to do something about it. At some point in the future, the competition will look at what Tesla has built and try to adopt something compatible, although right now the competition is probably looking at the "problems" the original first to market proprietary first-comer is experiencing, and thinking they'll wait until they know they can work out some of those problems (hey, is Tesla dragging its feet on fixing SuperCharger issues on purpose to propel its proprietary advantage like this?)
By the way, whatever happens with charging standards, Tesla already has a network of DC chargers. What if? What if? What if: Tesla upgrades their specifications on their car receptacle, across its new fleet (thereby upgrading out of most of the problematic areas of the old proprietary standard)? Then: it would be not a huge cost (compared to a whole SuperCharger) to just upgrade the end of the cord (or replace each cord and recycle them in a central location somehow) --- approximately $5 Million - $10 Million network wide (more if doing Destination Network). Actually, that would be the cheap part. There would be about $500 per Tesla for the following: Every Tesla owner would get a new converter plug for their car: a converter plug from new to old (for old cars and new charger cords -- these would probably have to be given out for free, and could even have a built in cooling system to aid in their lack of overheating?), and a converter plug from old to new (for new cars and old charger cords) that would probably be included in sale price of whole new cars for the first 6-18 months and then as soon as 100% of destination chargers are converted, the old adapter would not be included and would be an extra cost ($500), for people with old HPWC's who don't want to rewire them (there should be an easy way to recycle their cords for upgrades too). This would cost about $500 per existing car (so, around $100 Million -- not chump change). Probably, Tesla is weighing its options now, considering a CCS compatible plug, or other superior options, because of the total commitment that would cost; they should thoroughly find the existing list of inadequacies and thoroughly recreate those situations and thoroughly test a new standard to see if it is 100% fixed. That's probably what is taking Tesla so long, too. So, while this is happening, Tesla is dragging its feet on making its SuperChargers not a competitive advantage for itself (i.e., keeps the status quo of having a proprietary advantage), and the future is still good for the entire Electric Vehicle industry (one way or another they will get a long distance network), and Tesla gets to achieve all its goals: making money, spurring innovation with electric vehicles, and getting other companies on board. So, in the meanwhile, Tesla will make an extra how many sales because of this advantage, with its accompanying profit? Let's say it's 20,000 cars when Model 3 is full out (including all models). Their extra profit from that would be about $140 Million. Even if they only sell 10,000 extra cars because of the proprietary advantage and convert over to a newer better standard, they'd only experience a cost of $30 Million to keep the best Electric Vehicle fleet and network going strong.
I'm very excited to see what FF comes up with. People have written them off after their disastrous FFZero demo, hopefully they show something better next month. That said, I don't think the next practical EV will come from a startup. It has to be from an incumbent. The Tesla startup story is too good to be repeated in my opinion.
Market needs to understand that creating a single demo model car and building 1000's of units of it are two completely different ball games. We all know the pains Tesla had to go through to reach the level of production they are at now. I remember Elon saying a few years back he had no idea how difficult it would be to mass produce a physical product and his respect level for manufacturing companies went up significantly as he was working on Tesla.
I dont think a lot of entrepreneurs and investors would be willing to go through that level of pain. Yes launching new EV's and Tesla killers is glamorous but there is a HUGE non glamorous aspect of it that people miss. Tesla was not only blessed to have a gritful CEO who could rally his troops on command, but also committed investors like Jurvetson and Thiel who are personally aligned with the CEO's vision. Same is valid for SpaceX.
Even if the FF/Lucid cars look beautiful and practical, I really doubt they can achieve volume production. E.g Fisker, which looks so beautiful and aerodynamic, but was ridden with quality and production issues when manufactured on a volume scale.
But goodluck to FF, I am excited
That's reasoning by (bad) analogy again.
With an automated production line, it basically either runs full speed or not at all. If it's designed to run at that speed, they can get it up to that speed for a week once they've debugged it. It is, in actual fact, designed to run at that speed. The only ways it won't are:
(1) lots of bugs
(2) design failure
(3) non-automated sections of the line
Even #3 isn't critical if the non-automated sections can be taken off-line and parallelized, which is true for probably everything but wiring harnesses.
But it isn't.
There's been a lot of effort spent developing automated assembly lines starting in the 1980s -- with GM even. Sure, none of that effort was entirely successful, but each attempt picked off and solved another aspect of the problem. When I run down the pieces which need to be automated to make a fully automated assembly line, I'm not seeing very many unsolved problems.
Wiring harnesses, as I pointed out years ago, and as Hochholdinger pointed out earlier this year, are perhaps the most major issue. But I can think of two methods of doing them off the top of my head, and if the top tier automation experts are focusing on this and this alone for over a year, I really figure they can do it.
Seat fabric sewing is hard or impossible to automate but can be done entirely off the assembly line.
Robots can even glue fabric in place. The glass roof actually eliminates most of the headliner attachment, and even the non-glass part of the roof can have headliner pre-attached with the advantage of gravity (rather than being glued in from below). Floor fabric is even easier.
Bolting things together was one of the harder problems (hold the nut here and hold it still... now thread the bolt) but we know for a fact that Tesla's solved that already, since they announced that they'd switched from manual to robotic attachment of the battery packs. Without changing the battery pack attachment design.
Most of the subsystems can be very much "clip in", which is easy for robots.
One of the biggest problems for robots is taking components out of a box and getting them oriented correctly. This is what Lego is expert at, and probably the reason Tesla hired the guy from Lego.
In short, I think the obstacles for an automated assembly line have nearly all been solved, and years ago. Tesla has hired the people who know the solutions, even when they come from sources like Lego (outside the car industry). The wiring harnesses are the only one which is actually an unsolved problem, and it shouldn't be a very hard problem.
The car it's going to build, the Model 3, was supposed to be finalized last July (and probably was). It was designed for ease of manufacturing, which means that each part would have a manufacturing plan already associated with it by the time the design was finalized. The factory line is presumably being designed in CAD (that's what they did for the Gigafactory, no reason not to do it if you've already got the CAD software). Tesla's got good relationships with its robot suppliers (and there are multiple robot suppliers, so they're not sole-source limited). From the Q3 report, it sounds like everything has been sourced.
In order to hit the overly optimistic internal target deadline of July 1, they have six months to take delivery of things they've already ordered, install them according to an existing CAD plan, and test them.
That's possible. The suppliers (including the robot suppliers) have said that it's possible but tight.
But yeah, I'm pretty sure it'll be late. But then Elon already said it wouldn't meet the internal target deadline; so assume a standard three months contingency and October 1. But Elon's optimistic; so assume six months and January 1 2018.
But unless there's a critical failure on a very-long-lead-time item or a really serious screwup in the production line design, the extra six months should be plenty and they should be going by January 2018. Most of the pieces of the production lines aren't actually very-long-lead-time, not even the robots; the worst are probably the big stamping machine (which they already have) and the paint shop (which they already have).
You can install a factory in a month, if you've ordered all the equipment already and have a CAD layout for it. It's not a difficult construction project.
I think *all* of us are assuming Model 3 will be late. Even Elon Musk is assuming it'll be late! If it comes out in July 2017, it'll be miraculous. The quesiton is *how* late. I think Jan 2018 is a comfortable assumption; Adam Jonas apparently thinks it will take another entire year, until Jan 2019, and I just can't see where he gets that idea.
Not develop their own GPU, but they are likely to get their own chip.
System-on-a-chip design. You have off-the-shelf components in the form of patterns/designs (I forget what they're called), but you order them assembled together in a specific fashion -- on a single chip.
So it's an off-the-shelf GPU design, but a custom chip.
Though -- a GPU actually has next to no design in it. The damn things are just lattices for doing parallel identical transformations, which are mostly unpatentable, with a thin veneer of control hardware. Most of the hard work has been in the control hardware. Designing their own GPU wouldn't be as hard as you think (though certainly not a "next year or year after" project).
JRP3
Hyperactive Member
Then maybe eliminate sewing entirely, at least for non leather options. Seems as if heating/pressing/forming fibers and plastic into shape could be fully automated and with modern materials should lead to perfectly fine products.
JRP3
Hyperactive Member
I'm assuming unless there is some total disaster they will at least get some cars to customers in 2017. Remember first customers are employees so any issues can be quickly addressed.
Is there any reason for anyone to really not feel confident in a mid 2017 start date? Or is it just because previous product delays?
Properly even more than we give them credit for. Just buying TSLA stock I had an issue years ago as those were the same thoughts I had: how can a little silicon company build a safe, reliable, and compelling car to compete with the likes of Prosche, BMW, Mercedes, etc... then I realized EM also builds space rockets! Now it's more like how can these old car manufacturers keep up with a high paced high tech company with aerospace know how.
JRP3
Hyperactive Member
I've been following Tesla too long to feel confident about mid 2017. I'll be really happy if they start shipping by the end of Q3.
- Status
Similar threads
- Locked
