Model 3 "Production Bottlenecks" lead to lower-than-anticipated Q3 delivery #s

[calisnow]

:thumbup: Then I'm sorry for the implication that you had less skin in the game

No need to apologize - I didn't take offense, but thanks. I have spoken little about Model 3 because frankly I am so delighted with my AP2 S that I'm not sure I'm going to get a 3 - so it wouldn't be obvious I have reservations. Do wish I had more range though - should've bought the 100D.

In 48k miles, 2.5 years of ownership, I've been at the service center 11 times. 5 door handles, 2 12Vs, 1 leaky sunroof, GPS broke twice (biggest issue was the limited speed of AP when it thought I was on a neighboring road), etc.

During my last GPS issue, Tesla had my car for nearly a week. I had a new 100D loaner (by new, I mean it had 46 miles on the odo), they sure made HUGE progress with the cabin noise, the rattles, the seat comfort, etc. since my 2015 70D car.

YIKES! Wow. I would expect that's typical for a 2012/13 - but for a 2015 I did not realize they were still that buggy. I understand your wife's hesitation then.

Audi is releasing L3*...Even cars like the new Honda CRV (a friend of mine got one, I got to play with it this weekend) has L2...I can name a lot of other cars that have L2...I agree that Tesla is likely still doing better in the number of disengagements per mile driven (I know that's not recorded for L2 stuff). But when they released L2 2-ish years ago, there weren't a lot of manufacturers doing it. Now? Whole different ballgame.

Fair enough and I should have said simply that I am not sure others are catching up in the overall race to full autonomy. I think that question is undecided and the wildcard is whether or not the massive fleet learning uploads will give Tesla a big advantage - or not. As an aside in a different thread there is some discussion which makes the Audi system appear to almost be simply a marketing coup to be able to say "first to L3." Not only is it 37 mph limited but I read that it also disengages if it does not sense cars to its sides. So a gap in traffic to your right and you are hosed. In my opinion then, Audi has crippled it to the point of being a useless gimmick.

Tesla's AP1 with current firmware sounds more capable than this 2018 Audi that doesn't arrive for another 6 months in dealers - by which time AP2 will have had probably 6-12 more releases and almost 700 million more driven miles to improve system performance. Current pace of AP2 releases is now down two 2 week separations based on the last 3. Here's a chart I made the other day. Yeah I know it isn't scientific:

Full explanation here: Frequency of AP2 updates is steadily rising (chart inside)

It's like growth spurts in kids, right? At one point they're the same height, next morning one is a head taller . I'm sure Tesla will make the big jump soon, but I don't see them getting light years ahead in the near future. The Mobileye seperation cost them a lot of development time. I hope to be proven wrong.

You could be right - my hope is simply that the training video from these uploads will accelerate progress.

Porsche said they're coming out with a charging network at some point in time. Will it actually happen? Who knows. I hope so, not because I want to get a Porsche, but because I hope it spurs more competition about EV manufacturers and pushes Tesla to be better.

I definitely hope they do build a network - competition is good. Tesla needs competition.

 

[calisnow]

FWIW I just checked and my estimates have not moved back yet.

 

[Max*]

Not only is it 37 mph limited but I read that it also disengages if it does not sense cars to its sides. So a gap in traffic to your right and you are hosed. In my opinion then, Audi has crippled it to the point of being a useless gimmick.

Hmm... I didn't know about that. Thanks for the info.

Here's a chart I made the other day. Yeah I know it isn't scientific:

As an algorithm geek, I like the error bars on the chart

 

[electracity]

Don't be discouraged by the state of the software and the forum hand-wringers. As my signature says I have both AP1 and AP2. AP2 is improving faster than AP1 did and the releases are coming fast and furious - people have the tiniest memories around here. AP2 is fantastic and getting better. If anybody happens to be in Wisconsin (where I am at the moment) and wants a ride/demo of AP2 on various roads I'm more than happy to show you. BTW I do not own one share of Tesla stock - I merely give my opinion. I drive a Tesla but that stock is waaaaayyyyyyy too stratospherically priced for my taste. I'm buying GM stock.

I wasn't referring to AP, but the driver interface appears incomplete. This is mostly straight forward application programming. How could they not have it done?

 

[Topher]

Far fewer U.S. model 3 buyers may receive a tax credits than expected. This is probably part of Tesla's choice to lie about being in production.

You do realize that how many Model 3s Tesla makes NOW has no effect on how many people get the tax credits, right?

Plonk.

 

[voip-ninja]

You do realize that how many Model 3s Tesla makes NOW has no effect on how many people get the tax credits, right?

Plonk.

How many Model 3s Tesla makes NOW partially determines WHEN they deliver their 200K vehicle, just like any other vehicle they build.

The 200K tax credit number is cumulative, not by week/month/quarter.

 

[calisnow]

You do realize that how many Model 3s Tesla makes NOW has no effect on how many people get the tax credits, right?

Plonk.

Apparently it is not so obvious to me - I'm a bit slow. Tell me why Tesla's production rate of Model 3's will not affect how many Model 3 buyers receive tax credits.

 

[calisnow]

How many Model 3s Tesla makes NOW partially determines WHEN they deliver their 200K vehicle, just like any other vehicle they build.

The 200K tax credit number is cumulative, not by week/month/quarter.

damnit you beat me to the punch - i was doing the slow build to then drop the anvil on @Topher who was condescending to @electracity - the real point of course is that if Model 3 production is significantly delayed then more and more Model S/X's will occupy that 200K slot. So of course the rate of Model 3 production affects how many model 3 buyers get the credit.

 

[calisnow]

I wasn't referring to AP, but the driver interface appears incomplete. This is mostly straight forward application programming. How could they not have it done?

ah, ok. I haven't seen much info about the state of the UI.

 

[Topher]

Apparently it is not so obvious to me - I'm a bit slow. Tell me why Tesla's production rate of Model 3's will not affect how many Model 3 buyers receive tax credits.

Because until the 200,000 all cars get a tax credit, that number is completely fixed. Once that number is exceeded, what counts is the number produced from that day on until the date limit is reached. The total number of Tesla owners that get the full tax credit is strictly 200,000 plus number produced in the quarter when they reach that limit, and the following quarter.

As you can see production NOW does not affect that number.

Thank you kindly.

 

[omgwtfbyobbq]

What Topher said. I could even see Tesla limiting US sales of the S/3/X so they hit their 200,000th US delivery early in Q1 or Q2 of next year.

 

[Big-T]

FWIW I just checked and my estimates have not moved back yet.

Perhaps they're also 3 weeks delayed on updating that part of the website

 

[run-the-joules]

I don't remember, about 2 hours after reservations started. I drove by in the morning, there was a line, I was lazy so I went to work (I used to work like a mile from the service center), came back 2 hours later, no line.

Noon EST maybe?

I think that explains it. You’re likely very early in the queue and my being in CA is the only reason I’m still in that window at all.

 

[gregincal]

damnit you beat me to the punch - i was doing the slow build to then drop the anvil on @Topher who was condescending to @electracity - the real point of course is that if Model 3 production is significantly delayed then more and more Model S/X's will occupy that 200K slot. So of course the rate of Model 3 production affects how many model 3 buyers get the credit.

Yes it does, but it's quite possible a change in the ramp could even give more Model 3 owners the credit, if the slower initial ramp delays the quarter in which they reach 200K but then they reach full production earlier in the following quarter. It's really hard to say at this point how much of an effect it will have.

 

[electracity]

Perhaps they're also 3 weeks delayed on updating that part of the website

Tesla is never late until they are late

 

[electracity]

Yes it does, but it's quite possible a change in the ramp could even give more Model 3 owners the credit, if the slower initial ramp delays the quarter in which they reach 200K but then they reach full production earlier in the following quarter. It's really hard to say at this point how much of an effect it will have.

More likely the opposite if S/X production pushes Tesla over 200K before the model 3 is ready to be made in large numbers.

The ideal situation for model 3 buyers is a steep M3 production ramp pushing the company past 200K. The forecasts for large numbers of M3 buyers receiving the full credit assumes Tesla building tens of thousands of M3 per quarter in the critical time around expiration. In a long, slow M3 ramp the recipients of the full credit will be Tesla/SpaceX employees and some current Tesla west coast owners.

When would Tesla hit 200K with just S/X production?

 

[gregincal]

More likely the opposite if S/X production pushes Tesla over 200K before the model 3 is ready to be made in large numbers.

The ideal situation for model 3 buyers is a steep M3 production ramp pushing the company past 200K. The forecasts for large numbers of M3 buyers receiving the full credit assumes Tesla building tens of thousands of M3 per quarter in the critical time around expiration. In a long, slow M3 ramp the recipients of the full credit will be Tesla/SpaceX employees and some current Tesla west coast owners.

When would Tesla hit 200K with just S/X production?

With just S/X? Not until Q2 of next year. So sure, if they aren't making the Model 3 in large numbers by then it will be a disaster. The point was that a slower ramp this fall won't make much difference as long as they are at full production by early next year. They will still have almost six months of full production at full credit, and will have only missed a month or so. If they are producing a thousand cars a month next June, then yes, less people will get the credit.

 

[voip-ninja]

More likely the opposite if S/X production pushes Tesla over 200K before the model 3 is ready to be made in large numbers.

The ideal situation for model 3 buyers is a steep M3 production ramp pushing the company past 200K. The forecasts for large numbers of M3 buyers receiving the full credit assumes Tesla building tens of thousands of M3 per quarter in the critical time around expiration. In a long, slow M3 ramp the recipients of the full credit will be Tesla/SpaceX employees and some current Tesla west coast owners.

When would Tesla hit 200K with just S/X production?

I believe with or without M3 they will hit it between December and March.

You hit it on the head. M3 being dribbled out is the worst scenario from a tax credit perspective since there is no cap on how many get the credit once the 200k number is hit, Tesla would want to be cranking out as many as possible to not only hit sales objectives but keep customers happy.

 

[kengchang]

I believe with or without M3 they will hit it between December and March.

You hit it on the head. M3 being dribbled out is the worst scenario from a tax credit perspective since there is no cap on how many get the credit once the 200k number is hit, Tesla would want to be cranking out as many as possible to not only hit sales objectives but keep customers happy.

Not a chance they are hitting it before 2018 with just S/X. That would be 100% production going to the US under your scenario and I believe someone estimated they are at 150K~160K in the US.

 

[alloverx]

We need someone at Tesla to run a database script to update the estimated date. Should take someone about 60min to figure out. 60min to test. 2 days to get approval and then 10min to run.