Tesla vs. Waymo (Employee Perspective) + 400 Mile Range Tesla Model 3? Not Yet.

Hey everyone, Rob Man, you were here and today we were talking about the long

range

single motor

model

three potentially made in China, and then we also have a recap of an interview on the Tesla podcast in the third row with a former Tesla autopilot engineer. Strong day for Tesla stock today, ending up five point seven percent at five hundred and forty-five dollars and forty-five cents which, compared to the Nasdaq, was down 0.3% to start. There has been a lot of news today about a single engine made in China,

model

three with more time.

range

than the current standard range that Tesla is producing at the Shanghai gigafactory currently, all the news about this today was sparked by a Bloomberg article this morning, so we'll talk about that in a second, but this topic as a whole is not new to us.

We knew about a month ago that Tesla had requested approval to manufacture that variant of the Model 3 in China, but this news has resurfaced today thanks to that Bloomberg article titled quote Tesla is close to the debut of the longer-range Model 3 in China and when Bloomberg The article says that Tesla actually plans to start offering this version of the Model 3 with pricing as soon as this week, according to people familiar with the matter, and cites what really seemed to grab people's attention from this story was that Bloomberg also said the vehicle would have a range of more than 650 kilometers (404

mile

s) on a single charge compared to approximately 450 kilometers for the current most basic final version, so a Model 3 with a range of more than 400

mile

s, which without It's certainly an attention-grabbing headline, as Tesla is actually about to start making a model three four with a hundred-mile range, well, not yet, at least in the EPA cycle, so let's talk about this.

Different countries use different test cycles to measure vehicle range, so there's the epa in the US, there's the w LTP and the NEDC internationally and of those three tests. cycles, the epa will generally give you the lowest range and generally give you the most realistic real world range, while the w LTP and NEDC test cycles will generally show a higher range for the same vehicle as us. This can be easily seen by comparing the US design study with the China design study for the current versions of the Model 3 we can see that in the United States the dual engine Model 3 has an estimated range of 322 miles while in China the same vehicle has an estimated range of 590 which converts to just under 367 miles or about 13.9% more range reflected in the design study in China than in the US.

If we take that to the article from Bloomberg that it has a range of over 650 kilometers or 404 miles and we used that same ratio, then you could estimate that the epa range for this version of the model 3 would be approximately 355 miles. The other thing to consider here is that Bloomberg only says above 650 kilometers, so it could be a little higher than that and we have someone on Twitter. on D Couric posting a report from News 18 com which I have no idea about the validity of that source but it says 668 kilometers of range for this version so it's obviously a bit speculative but if it were 668 kilometers it would be 415 miles sooner . our epa and then conversion method would be just 365 miles, so each vehicle will convert differently, but 355 to 365 miles EPA range for this variant seems like a pretty good estimate, the midpoint of that range would again become about 579. kilometers, so it's still not 400 miles for the Model 3, but it's still a big number, especially considering that when Tesla first launched the Model 3 long-range signal engine in the US. , which was mid-2017, you might remember that was the first one.

The production Model 3 Tesla gave an estimated range of 310 miles which was actually an underestimate of the EPA test result of 334 miles. Tesla simply chose to lower it, they must have felt it was a more realistic number or wanted to position it a little lower. than the snx, but regardless of even that 334-mile range according to the EPA back then, we could be looking at an 8% range increase in this Chinese-made single-engine Model 3 in about a couple of years, which Not bad, especially considering the cost reductions. that Tesla would have had during that time period, of course these are all just estimates, the range of the vehicle will vary depending on how you drive it based on standardized tests.

I think the epa is the best we have, but if you really want to get 400 miles of range out of a Model 3, the actual record for a single charge on the Model 3 is about 623 miles if you're willing to do some pretty extreme hypermiling. next. I want to do it through my conclusions. from yesterday's third row Tesla podcast, this one was pretty interesting. I had a former Tesla

employee

who worked in the autopilot division just a couple of months ago who has now transferred to Wei MO and is currently working there. interesting discussion so I would recommend listening the link will be in the show notes most of the autopilot stuff is in the first 30 minutes and if you listen they had to replace the word male Wei with something else for legal purposes and they chose Taco Bell, so if you hear just a heads up about that, but for background and some highlights, this was a former Tesla

employee

, like I said, in the autopilot division, his name is Isaac and he left Tesla a couple ago of months, as I said. but he worked there for three and a half years in different divisions and spent the last 10 months on the autopilot team.

He said that time really changed his

perspective

and helped him accept the long-term vision of eelain and the success that Tesla will have. as a company saying a quote, when I finally got into this, he blew me away and in terms of leaving Tesla, he said it was an extremely difficult decision, as evidenced by the fact that he still uses us when referring to Tesla throughout this interview, but it sounded like a combination of a really good opportunity and a really good offer from Wei Mo that ultimately led him to that decision. In terms of conclusions, I would split it into two sections, one would be your comments on autopilot, specifically about how far along you are with a new autopilot rewrite that hadn't yet been seen in the actual public release versions of the software, but that Elon and the team have referred to you as a big step forward coming soon.

Elon referenced this on March 1 in a tweet saying, "We have to finish." Work on the basic autopilot code and 3D labeling, then the functionality will happen quickly in no time and I quote, so Isaac said they were working on this autopilot rewrite and he was excited calling it a game changer and saying that Generally speaking, there are none here. kind of a specific quantization, but they were seeing quote unquote improve by thousands of percents pretty much at the end of the quote, so we've talked about this core rewrite before, basically, it's combining all the camera feeds into a 3D layout and, according to Isaac, that makes it easier. for the data to be processed, that 360 degree view where they are doing one or two frames before now they can do 1000 to 2000 frames, as I understand it personally, it makes a lot of sense for those frames to be combined in a single type. of the environment versus the neural network working on each of the individual camera frames, but I'm still not sure about the time element you know, if you have multiple frames over a period of time, that's what creates a video and there That's where I guess. the 1000 to 2000 frame comment comes from, but he also specifically said that it was not a video, although the video can be used to verify the neural network results.

The third row group also rests a bit on that. I think they were too. It's a bit unclear on that, but I'm not sure the answer really helped with that, so if you have any thoughts on that, let me know in the comments below, but it really emphasized the value of this rewrite anyway. saying he was deleting all of them. Little pilotless bugs in the past and quotes from things that are happening will be crazy and quotes, so this is all directionally aligned with the comments Elan has been making about this rewrite, so it's really good to hear it from another person too.

The other big focus I would say, which was probably the second half of the first half hour, was the comparison between Way Mo and Tesla. Really interesting given that Isaac has worked on both and was surprisingly honest. He would say that he talked about how there is always a lot of talent flowing between Apple and Tesla and that the work at Tesla is much more intense and then he did a good compare and contrast with Tesla and got into things like the lidar strategy, which gave a little . More details on saying that there can be a lot of interference using lidar in terms of your reading that you're getting, which causes broken images and says quote, that's why it's not reliable and quote, but that way I compensate for that with more sensors, so that's why you see four in each of the corners of the vehicle so that the image interruption is not as frequent and quote that he was very complimentary of both Tesla and wham-o in terms of their progress towards fully autonomous driving and basically said they have different strategies, which mo is really trying to work on becoming a provider of autonomous technology, saying, quote, they're trying to build a universal system that you'll be able to buy and put in your own car, that's the goal.

Mo's main, finish the quote and when you say your own car here, I guess you're not really referring to the end consumer but to the automakers, obviously Tesla's strategy, at least at the moment, is very internally focused at the end of the day, although Isaac returned to what I often talk about here and that was data, he said that way, mo has a lot and not to misunderstand that there are many, but he does not cite the same amount as Tesla and a lot of data is not enough, data and quotes that paint a pretty sinister picture for anyone trying to compete with Tesla driving south because Tess is about to take two big steps forward here, one with this autopilot rewrite and the second in terms of vehicles capable of use it with hardware, three to collect data for Tesla, that combination creates a pretty crazy exponential and if things go as planned it should allow Tesla to start pulling away pretty quickly, that's a big if of course, but the important thing is that The test is about to have more valuable data and many more, latest quick notes. which I want to include here from this interview, Isaac said that your car will actually be updated without doing the full version number software updates in general for a lot of small things like size or brightness of different graphics, things like that and it sounded like maybe sometimes some neural network stuff, so I thought that was interesting and then in terms of being a supplier of an autopilot system, he mentioned that it might be a possibility for Tesla that one day they could license autopilot, basically saying that it hasn't been ruled out and that it was something that he could personally see happening, so again the link for that is in the show notes, that's all for today, though as always, thanks for listening, don't forget to subscribe and sign up to receive notifications and make sure you follow us.

Me on Twitter on the Tesla podcast with that. See you tomorrow for the Wednesday April 8 episode of Tesla David, thank you.