Did Tesla lose focus by making the Model 3 an autonomous car instead of a great EV?

[xav-]

I might not be the norm, but I almost never wear sunglasses. Perhaps my eyes work differently, but I find sunglasses to be mostly pointless. Really just for fashion, if the mood hits me right.

Your suspicion is correct. You are definitely not the norm

 

[eisbock]

I agree. I don't want to rush to judgment but I want a drivers car that can drive autonomously. Elon's "taxi" talk and single screen do not bode well IMO. I'm going to wait till the final reveal though and some test drive reports.

Until then I am going to be biting my fingernails, looking out the front window like a worried parent waiting for a late kid to come home, lol. Jk.

You and me both. I have high hopes that this was just a PR stunt to sell more S's and that Elon has something else up his sleeve, but these comments are extremely worrying considering they coincide with the release candidate video. The car is done. There's no turning back now!

 

[Bokonon]

I have high hopes that this was just a PR stunt to sell more S's and that Elon has something else up his sleeve, but these comments are extremely worrying considering they coincide with the release candidate video.

As I see it, this whole marketing push has been a lot more about the S than the 3, with the 3 merely serving as a foil to the more advanced S. So, in that respect, the "PR stunt" label seems appropriate.

As to whether Elon "has something up his sleeve" with respect to the 3... here's how I've been thinking about that question within the context of this anti-selling blitz (and the polarizing reaction it has provoked).

 

[Kant.Ing]

Well if it's so autonomous, the seats are designed wrong... They need to put a convertible bed in the car! If Model 3 will not have any new tech in it, how about any old tech? If it's so autonomous (probably with AP3), it should also render Model S/X obsolete.

Or Model 3 is really a Yaris EV that we've tried hard to avoid.

 

[Wolverinegeoff]

Maybe your eyes are different, but anywhere there's snow cover and bright sun around here, sunglasses are a necessity.

Florida = also

 

[AnxietyRanger]

More musings about Germany and Sweden - the primary brand value of BMW, MBZ and Volvo has gone out the window. There MUST be a lot of moments of existential angst happening among German and Swedish execs right now.

• BMW - handling superior to the rest of the world. In a world of autonomous cars this is now irrelevant.
• MBZ - Durability at the top of the chain. In a world of EV's with few moving parts - this is now irrelevant.
• Volvo - Crash safety superior to the rest of the world. In a world of autonomous cars crash rates drop 90+% due to software/cpu/sensor suites that are available now - and will cost only a few dollars in less than 5 years. You see insecurity ALREADY out of Volvo - an engineer saying AP 1 is an "unsupervised" toy - and then we see AP 1 reduces crashes 40%.

There is a very real existential crisis going on with European auto. In addition to the entire mission statement of these brands being tossed in the garbage - the franchised dealer networks are another problem. Can they survive in a world of EV's which require little service? How much longer will premium buyers tolerate going to a service center for updates? Yet how will franchised dealers learn to tolerate direct-from-mothership updates to customers that remove the dealer from the loop?

To those who say "Look to Germany" for the EV future - I say - look to Lucid Motors. This is further indication that Germany is out of touch, when the most tangible competition for Tesla is yet another California based startup. The Air is the car that MBZ, BMW and Audi - SHOULD be showing us. It addresses all of the Model S's weak points - and it isn't just a concept car. So why aren't they? Lucid has $130M in funding, a finalized design and a crack executive team, including the Model S's original designer. All the Germans can do is pray nobody injects Lucid with a few billion - which seems likely to happen. @AnxietyRanger - right now, looking to what could make me defect in 24 months from Tesla - Lucid is by far the sexiest option.

Further thoughts on Lucid's odds of success - Tesla has blazed the trail and done the hard work. Tesla has destroyed the idea that consumers put more trust in legacy automakers than startups. Tesla has proven that EV's can be sexy. Tesla has shown the world that an automotive startup can succeed (something thought impossible by many even 3 years ago, a couple years into the Model S's life. Notice even famous Tesla FUD spreader Bertel Schmitt of Forbes has gone oddly quiet lately - and the media FUD spreaders seem to be finally losing steam). And Tesla has shown that the "desire" factor of legacy luxury makers can be rendered impotent frighteningly fast. Finally - Tesla has already built out a nationwide rapid charge network and is inviting everyone to use it. The Germans are, thus far - too proud. But a scrappy startup like Lucid might not be.

If Lucid can execute? It will be far easier for them to get a foothold than it was for Tesla.

@calisnow Just as a mid-conversation comment: A good, mature discussion with nice progression of mutual understanding and entertaining of ideas. Thank you. I appreciate it. I wish more Internet conversations were like this. I will get back to the thick of things later.

@calisnow

Again than you for the thoughtful comments during the week. Some musings based on them, in no particular order:

- No doubt the traditional automotive industry, sales, service and energy networks included, faces a massive disruption and transition. Not all will make it and some will change nature of their business dramatically in the process. Though who can adapt and who can not, or how long a time does this evolution span, remains to be seen.

- The dealership is a good question. I am not fully convinced yet the nature of car as a sales item supports Tesla's approach universally. The nature of large volume, large size, large price items distributed over vast geographical areas with trade-ins involved is not as simple equation as selling phones. The secondary market alone is a big difference. Tesla's cumbersome and not necessarily well suited for cars CPO process is one question mark here.

- The way traditional brands can transcend to the new reality is likewise probably a more complex question. Their ability or imability to re-invent themselves will probably matter more than past strengths.

- Don't discount brand loyalty completely. Cars are not consumer electronics. People may be less fickle.

@AnxietyRanger - on the luxury interior aspect. Musk says S/X switch to 2170 cells before 2017 close. Another poster here calculates that will save over 280 lbs (140+ kilos) for the same kwhr in the 100 kwhr battery. Musk also says Tesla gets - what - 35% cost reduction?

So there you go - the money and the weight savings to increase the luxury materials and the NVH of the interior - and they already hired the designer.

Possible, though I would expect experience plays more role in Tesla's current luxury predicament than cost or weight. When and how they get this part of their business together remains to be seen. I agree the Volvo interior hire was a promising move.

Just some general thoughts. Let's keep talking.

 

[electracity]

Your suspicion is correct. You are definitely not the norm

There is substantially nueroadaptation to bright light. If you always wear sunglasses you will need to always wear sunglasses.

This can be a problem for actors who always wear sunglasses in sunny LA. They can't shoot outdoor scenes without squinting. The smart ones prepare for outdoor scenes by reducing sunglass use to prepare for that kind of shoot.

Personally I avoid sunglasses in the early morning except for driving into the sun. Bright light exposure early in the day is good.

 

[boaterva]

Just as a mention, Transitions Drivewear sunglasses work great, and don't seem to have the polarizing issue. I've not tried them on HUDs or other like displays, but they don't have any effect on my iPhone, or similar displays. They work great in the car and outside, also. (Have to add, great prices on all Transition lens at Costco!)

 

[zenmaster]

Just as a mention, Transitions Drivewear sunglasses work great, and don't seem to have the polarizing issue. I've not tried them on HUDs or other like displays, but they don't have any effect on my iPhone, or similar displays. They work great in the car and outside, also. (Have to add, great prices on all Transition lens at Costco!)

I think the relative angle of polarization for screen and glasses is what matters, and these aren't standardized.

 

[Reciprocity]

You don't think a center instrument cluster is a "realistic expectation"? At this point, we just want features that make the car functional. It's not unrealistic to expect that.

Not a "realistic expectation" in a car that is 1/2 of the price of the single other great looking electric car with 2 screens. You cant have everything. You cant have S60 Range, S60 screens and everything else and still have it cost 1/2 as much. Sorry for trying to be realistic. I am sure designers sat in a room and discussed how they would keep the price down and something had to go. I am sure they wanted a spaceship HUD and 3 screens, but they just couldn't manage it and keep costs down while delivering on time. Do you think we are smarter then them? I mean, we all think a HUD or a second screen or instrument cluster is mandatory, but it clearly isn't.

Poor analogy. Apple took the keyboard and offered an arguably better alternative.

Tesla took the instrument cluster and replaced it with... nothing?

Not Poor, because many didn't think the fat fingered touch screen was going to work at all, much less for typing. It took a while for the average person to rap their heads around. I am not talking about the fanbois that lined up. I am talking about the 70 year old guy I saw the other day on the plane who was tooling around on his iPhone like it was second nature. It took those folks a while to come to understand the elegance, as I am sure you will find a why to struggle through with the 1 screen. Or you can go buy an Awesome Bolt! I hear they are offering as much as $5k off and I hear it has an instrument cluster.

So yes. I think its unrealistic to expect 100 features in a car priced to have 50 features. I am sorry that the designers didnt keep the features you wanted and got rid of the features you didnt, but hard choices have to be made. My guess is they err on the side of easiest to produce over cost specifically. They also clearly wanted to be a Tesla and not look like every other car.

 

[Reciprocity]

There will never be enough data to solve all of the corner cases. The number of corner cases is infinite.

You are correct, I was referring to just the 100 or so most common cases. Clearly it will require machine learning and AI to get to full Autonomy, The key is the data. The machine needs to see every possible issue 100 times or 1000 times so it can recognize the patters of how the human solved the problem. This requires millions or billions of miles of driving data.

This is also how they can validate the system by putting it through millions of scenarios based on actual events and no one has to hand code it.

Lets all try not to be so literal, this is a message board not a conference on AI and Machine learning. Dumb it down for us simple folks.

 

[SouthFLGuy]

Your suspicion is correct. You are definitely not the norm

There are actually some HUDs that work with polarized sunglasses.

 

[zenmaster]

You are correct, I was referring to just the 100 or so most common cases. Clearly it will require machine learning and AI to get to full Autonomy, The key is the data. The machine needs to see every possible issue 100 times or 1000 times so it can recognize the patters of how the human solved the problem. This requires millions or billions of miles of driving data.

This is also how they can validate the system by putting it through millions of scenarios based on actual events and no one has to hand code it.

Lets all try not to be so literal, this is a message board not a conference on AI and Machine learning. Dumb it down for us simple folks.

Problem is that if it's dumbed down too much, you get a misleading idea of what autonomy actually means and we won't be referring to the same idea.
For example this statement "The key is the data. The machine needs to see every possible issue 100 times or 1000 times so it can recognize the patters of how the human solved the problem. " Is very misleading, because it assumes that everything was in place to have the potential to eventually learn the correct behavior from analyzing that data. We are simply not there yet - not even close.

 

[FirstSea]

I agree with the idea that Tesla is moving towards an autonomous future, but I honestly think only the 3 will embody that idea. That's, the 3 will be (at some point in time) autonomous, but the 3 will be the only Tesla with the ultra plain interior,since the 3 was designed to be a taxi, just like the Tesla truck will be a truck. I don't see Tesla replacing the entire instrumental cluster in the S, X or the Roadster, since these are sedans, crossovers or sport cars meant to be driven with the added benefit of being autonomous someday. Designing every single Tesla from now on to be solely autonomous like the 3, would make no sense, no one would buy a luxury sport vehicle (roadster) to be driven around in it.

I'm not angry about the information Elon released, rather I'm disappointed, we were told we could finally get a Tesla at an accessible price, but instead of getting a sedan, they gave us a taxi...

 

[zenmaster]

I agree with the idea that Tesla is moving towards an autonomous future, but I honestly think only the 3 will embody that idea. That's, the 3 will be (at some point in time) autonomous, but the 3 will be the only Tesla with the ultra plain interior,since the 3 was designed to be a taxi, just like the Tesla truck will be a truck.

Look at it more like a temporary experiment. Tesla is still in business to make money and will ultimately be corrected by market demands or their product will fail. Competition will eventually ensure desirable design and features find their way into EVs at this price point.

 

[FirstSea]

True, but I honestly don't think Tesla is in the position to be doing trial and error, since they have finally gotten to the point in which they can take on the big boys on the block. Tesla can easily be crushed by their own weight if they make a big mistake, they aren't GM or VW who can just throw money at problems to make them go away.

Yet, even if the 3 fails in the US and Europe, China will eat it; so in the end Tesla will continue...

 

[AustinPowers]

Yet, even if the 3 fails in the US and Europe, China will eat it; so in the end Tesla will continue...

What makes you so sure China will "eat it"? They have so many affordable EVs over there already, why should they be waiting for the Model 3?

 

[zenmaster]

True, but I honestly don't think Tesla is in the position to be doing trial and error, since they have finally gotten to the point in which they can take on the big boys on the block.

It looks like Tesla has a window of opportunity now where certain bad decisions at least hurt them less. What competitive platform changes are really necessary and how could they not afford to do these in time or money?

 

[ratsbew]

In my honest opinion, I think they should just ignore surface street autonomous drive completely until they achieve full no-driver-supervision-required autonomous drive on divided highways without stop signals.

I totally agree. I want "on-ramp to off ramp" autopilot that can drive me on the interstate highway system and divided highways safely. I drive 10 hours to my family quite often and the drive is incredibly dull and easy....something that AP2.0 should be able to handle soon based on what I've seen from 8.1. All that's really needed is the ability to take ramps and to follow navigation.

 

[Skotty]

I totally agree. I want "on-ramp to off ramp" autopilot that can drive me on the interstate highway system and divided highways safely. I drive 10 hours to my family quite often and the drive is incredibly dull and easy....something that AP2.0 should be able to handle soon based on what I've seen from 8.1. All that's really needed is the ability to take ramps and to follow navigation.

I tend to poo poo autonomous driving expectations a lot (because I think fully autonomous driving is a far more difficult challenge than most people give it credit for), but I actually think it's a great idea for interstate travel and much more achievable in that context. Happy I'm not the only one who would like to see them focus primarily on interstate / divided highway first. Cheers!