Is the Mach-E really 7 years behind Tesla?

[timbop]

LOL, you make it sound like it's an option. yeah, if you had a 3, you'd be concerned about it's charging rate, not some theoretical maximum. Just more proof all of your knowledge comes from the internet, not experience.

you guys need to stop feeding the troll... All the nonsense is burying what could be interesting and useful

Sponsored

 

[ab13]

Third party analysis of HW 3.0 suggest that the Tesla designed computer is 6 to 7 years ahead of legacy automakers, at least. The engineer involved in the teardown in Japan stated that he did not believe Tesla's feat could be duplicated by Legacy automakers--ever. He state that he believed this due to the structure of the companies and their supply chains.

This is largely to do with the time and resources needed. Tesla spent about 4 years
from idea to production build to get the computer made. They hired off top people from Apple, AMD, etc... to work on the project. If you are an existing company with many suppliers, spending 4 years on a project that others may already be doing wouldn't be a good use of resources. Both Mobileye, Nvidia and others are suppliers to the automakers. At this time it's the software that's the issue, automakers are not mass releasing software that would need highly integrated hardware yet. Nvidia's chip that will be in automakers vehicles in 2022 is about 5x faster than the Telsa chip. That aligns when level 3+ software has been tested out by companies. This year Lexus and Mercedes plan to release level 3 capable vehicles, but these are high end models, so limited volumes first. Then GM has many more Super Cruise models coming in the next couple of years.

As far as the usability of FSD, we'll know more sometime soon. The HW 3.0 computer is so much faster than 2.5 that the software is not capable of taking advantage of it. Tesla is currently doing a complete foundational re-write of the AP software. When that re-write is complete, we'll know what the AP can really do. Musk has stated that the advance will be revolutionary as opposed to evolutionary compared to the current software. Additionally, advances will happen at a much accelerated rate. Right now, 99% of the effort is being put in to developing the new software. Once the new software is done and released to the public, all of that effort that is currently invisible to the public will be put in to advancing the software. I agree that the $7k charged for FSD is not currently worth it. The $7K is an investment in what Elon Musk says will come. If he delivers FSD, that $7K will be a huge bargain. If he does not, it is an investment in a dream. Believe him or not, that is your choice.

NTSB and NHTSA are likely to have issue with their lack of driver monitoring. So I wouldn't expect it too soon. In Europe I read that they are having to limit some AP features due to the safety issues.

https://arstechnica.com/cars/2020/02/ntsb-blasts-tesla-caltrans-and-nhtsa-for-autopilot-death/

In the linked NTSB report it says.
"Risk Mitigation Pertaining to Monitoring Driver Engagement. The Tesla Autopilot system did not provide an effective means of monitoring the driver’s level of engagement with the driving task, and the timing of alerts and warnings was insufficient to elicit the driver’s response to prevent the crash or mitigate its severity. Requirements are needed for driver monitoring systems for advanced driver assistance systems that provide partial driving automation (SAE Level 2 systems), and Tesla needs to develop applications that more effectively sense the driver’s level of engagement and that alert
drivers who are not engaged."

Comparing Waymo to Tesla is like comparing bread to steak. Waymo relies on incredibly detailed GPS maps and comparatively simple software to drive in a pre-defined and limited area. (Downtown Portland for example.) Tesla is using artificial intelligence and standard GPS maps to enable a car to drive in a myriad of locations and conditions. The promise is to get in your car, tell it where to go--be it across town or across the country--and go to sleep. Every HW 2.5 Tesla "watches" the road and sends driving information back to the mothership. Tesla's are literally "learning" how to drive themselves. The AP driving data collected by Tesla is estimated to be worth $50B all by itself, as no one else has this data, or the means to collect it.

Google was testing millions of miles a day in simulation years ago. Any major company is going to do that, every day millions of miles of simulation tests, in addition to road tests. It's good to have on the road data, but simulations can be done thousands and millions of times faster and in greater variety. Toyota and others are using Nvidia's out of the box simulator, Constellation. Though companies have been using their own systems internally, even Tesla. All of their systems are similar in that they learn to drive themselves by AI, it's a question of if the company does the testing itself or they are expecting paying customers to do their testing.

 

[mark360]

you guys need to stop feeding the troll... All the nonsense is burying what could be interesting and useful

nak is a troll

 

[Nak]

you guys need to stop feeding the troll... All the nonsense is burying what could be interesting and useful

Agreed. I apologize for helping to keep this going. I'll put the troll on ignore now.

 

[MadScientist]

Country lanes in the UK are an example of continuous odd ball situations and I don't believe we'll see autonomous driving on those sort of roads for a very, very long time.

There are some You Tube clips by 'Tesla Driver' which illustrate this.

Tesla Driver

 

[dbsb3233]

Only if each guest is getting up at all hours of the night to get out of bed, throw on clothes, walk out to the parking lot in the cold, move their car to the charger at 2am... then repeat it all again to unplug at 2:40am.

Sorry, that just ain't gonna happen. No way. Hotel guests are gonna want to plug in when the arrive at the hotel, and unplug when they leave the hotel the next morning. Just like they do at home in their garage. That's the service many hotels are likely to offer for their customers, because that's what their customers will want (as well as it being far more affordable for them to offer than L3).

 

[dbsb3233]

No, all those scenarios are long-term. Homes, apartments, hotels, and workplaces are ALL 8+ hour scenarios for the typical resident/employee/guest. Thus good fits for L2.

It's high volume/high traffic retail charging stations (like EA) that are appropriate for L3. And best placed in spots where users have a good 30-minute task. Like eating (restaurants), or shopping (grocery stores/malls). And of course along highways since it's mostly road-trippers that need them.

Hotel usage is not a 30-minute task... unless it's across the street from where the prostitutes hang out, that is.

 

[jlauro]

Although I could see the valet method working, in most cases I would personally prefer to have a room at $10 or so extra a night more that came with it's own dedicated L2 charger for the night. They could remotely shutdown the charger for those not wanting the service, or activate it with your room key. If I was there for a week, I might only want the service for the first and last night.

 

[dbsb3233]

Correct. Same way with that outlet in one's home garage. And nobody is going to turn their garage into an L3 retail charging station open to the public either. Because that's not it's purpose.

The purpose of a hotel charger is for the hotel's guests to charge their parked car overnight while sleeping. The purpose of an office building charger is for the employees to charge their parked car during their work shift. Just like the purpose of the charger in a home garage is to charge their parked car while home overnight.

I'll try to avoid the politics of it, other than to say such decades-off mandates aren't worth the paper they're written on. Reality will dictate what happens by then. But you're right that the US needs a more robust charging network to meet the demand of a growing market share of BEVs.

That doesn't mean "only L3" though. It means the right types to meet the variety of demand. L3 charging is only PART of the solution. Yes, we need more L3 chargers, as is happening with EA and Tesla and others that will likely pop up. What we need even more is L2 chargers at the logical places. Homes are the obvious first place, but that's a homeowner's responsibility. Those are easy for anyone with a garage to put in on their own. Next is apartment/condo buildings, where millions of Americans live. Also many workplaces. And hotels. Those are all ideal for L2 chargers, for the reasons I stated (to serve those business' residents/employees/customers).

L3 chargers are the equivalent of retail gas stations. They generally go in high traffic routes like highways and city street corners. But unlike gas stations, they WON'T be the primary way people fuel their BEVs. They'll be the exception rather than the rule. Because frankly, it's a pain in the *** to sit there for 30 minutes waiting on a car charge. So for most BEV drivers, L3 charges will be the rare exception, not the rule. L2 while your car is parked will be the primary method.

 

[dbsb3233]

I don't care what time in the middle of the night I'd need to get up (twice). The point is having to get up in the middle of the night PERIOD. Sorry, that ain't gonna fly.

As for a valet charging service, seems unlikely. Maybe at some high-end hotels where people valet park already. But that's not very mainstream. Most people don't like turning the car over to a valet.

Sponsored