TSLA is currently up 3% from the day before the event.
But concluding anything from how the stock price reacts to an event is a fool’s errand. The market is twitchy.
TSLA is currently up 3% from the day before the event.
But concluding anything from how the stock price reacts to an event is a fool’s errand. The market is twitchy.
There was a Cybercab at the Tesla store in Santana Row (San Jose).
Looks good in person. And solidly put together for a limited production run.
They’re gonna make a ton of these. So many little things that are totally unnecessary with no driver. Steering wheel/pedals, obviously. Side-view and rear-view mirrors. Rear window. Even the seats aren’t adjustable! No need if there are no pedals (or rear seats). May not even need a full-blown windshield wiper.
The only thing they splurged on are the automatic doors, but that makes sense since a passenger might inadvertently leave them open. Probably also useful for accessibility reasons (kids, etc.). Why the scissor action? My guess is that it avoids stuff on the ground like high curbs, rocks, bollards, etc. Just one less thing to potentially get stuck.
I promise not to bring up released updates until I actually have it from now on and today is the actual day. I just downloaded 2024.33.35 (FSD 12.5.6.3). This should be the final v12 release and was promised by the end of October. We now finally have end-to-end on highways and also “more natural lane changes” on highways. I hope that that means it will merge over from the far left lane much sooner and not miss exits.
I’ll update after I test it out.
I took a short freeway drive last night and the lane change choices seem significantly better. I’ll be driving from Santa Barbara to San Diego in a couple of weeks which will be the real test.
Cool. Lane changes are something that work much better with the end-to-end model, in my experience. Are the changes themselves snappy feeling? They just seem so slow and deliberate on the freeway compared to city driving.
Would be nice to get some of the latest updates on HW3… but this was fairly predictable.
They had a new Model 3 Performance at the Tesla store, which I sat in and played with a bit. Feels very nice. Almost tempting to upgrade…
I did a bit of a longer drive today and the lane changing choices and the changing itself is much more natural. I’m very happy so far but it’s still limited experience. As you said, it’s only HW4 at the moment but it will come to HW3.
V13.x (13.3 supposedly being employee tested now and supposed to be released by the end of the month) will be the big divergence. The HW4 cameras have much better resolution and this will be the first version to make use of it.
I found a really silly pet peeve for the BMW. WTF were they thinking?
The semi-smart driving features are all controlled by a 3x3 grid of buttons & toggle thingies on the 9 o’clock spoke of the steering wheel right adjacent to the rim. So far so commonplace and so logical. During self-steering ops the steering wheel senses the presence of the driver’s hand(s) by something that’s not torque. It’s probably capacitance. I know this because there are areas of the wheel you can hold (and even wiggle the wheel from) and the system won’t detect that you’re touching the wheel. In other areas of the wheel a mere fingertip resting on the rim is sufficient to signal “driver has hand(s) on wheel”.
You can probably see where this is going …
Yep, the area where you’d naturally place your left hand to reach that 3x3 cluster of buttons with your left thumb is exactly the area where the hand presence sensors are absent. So you have to hold the wheel with your left hand elsewhere and stretch finger/thumbtips to barely reach the buttons, Or remove your hand from the wheel to push the buttons and hurry up and get back on the wheel before it bitches at you and disconnects the self-steering. Or put your right hand on the wheel too (but not symmetrically with the left) and use your left / free hand resting in the dead zone to use the buttons. etc.
And before you ask, despite being a smaller than average guy, I have long fingers and thumbs. This would be far worse for someone with compact hands.
Because it takes a lot of button-fiddling to drive in ordinary suburban traffic, this issue is significant. On, say, a highway, once you get it activated and locked to the car ahead you might not need a button for 20 minutes. In suburbia it’s every 30 seconds.
What were they thinking???!!? Gaah!!!
That does seem pretty dumb. So, maybe designed with the expectation that you’ll be doing the whole 10-and-2 thing? Which are, inconveniently, not where your hands should be if you want to use the spoke buttons.
Since the advent of steering wheel airbags IIRC the advice has been to do the 8- & 4-thing to get your lower arms and hands below the axis of inflation of the bag. Hands at the traditional 10- and 2- will be flung off the wheel when the bag inflates and may be injured when they bounce off the headliner and the hard stuff behind it. Or the left hand off the window.
For the BMW, hands placed immediately below the fat 9- and 3- spokes is bag-safe, triggers the hands-on-wheel sensors just fine, but puts your thumb where it really can’t reach most of the buttons. Choke up a bit into the intersection of rim & 9-spoke and you can get a good grip, and also readily reach all the buttons. Shame about not triggering the presence sensors any more. Oops.
That might even be stupider. It wouldn’t surprise me if Ze Germans were somehow behind the times when it comes to recommended steering wheel positions. But for the sensors to just be an inch or two off seems like carelessness.
I’m trying to picture your grip. When I hold the steering wheel, my thumb ends up a little higher than my fingers. I hold the wheel under the 9/3 spokes and my thumb reaches the buttons fine.
I’ve been fiddling with it.
With ring & little finger around the wheel rim just below the 9-spoke within the detector range and index and middle fingers more or less behind the 9-spoke where there are no detectors, the thumb is just long enough to use all the buttons with just a smidgen of stretch to the farthest but no shift of hand. So that works, just need a new habit.
My previous / current preferred grip is palm against the bulge at the end of the 9-spoke which has the 4 fingers mostly behind or around the spoke, not the wheel rim. That’s what doesn’t work.
First world problems for sure.
Chilling at the super charger after driving 200 miles from Santa Barbara to Solana Beach. No interventions and the lane changing and freeway driving was significantly improved. I’m super happy.
I had one intervention on a boulevard. The GPS asked the car to make an illegal U Turn. The problem was that the No U Turn sign was non standard so the system didn’t recognize it.
This article Emergency Vehicle Lights Can Screw Up a Car’s Automated Driving System ties into that. I think ultimately Tesla’s optical-only approach is doomed to be 9x% perfect.
To be fair, emergency lights can screw up human drivers as well. It’s difficult to focus when high-intensity lights are flashing in your eyes. The correct reaction is too slow down, but then the car behind you might not notice you slowing down, because of the flashing lights.
General Motors will no longer fund its Cruise division’s robotaxi development, the company said on Tuesday.
The Detroit automaker cited the increasingly competitive robotaxi market, capital allocation priorities and the considerable time and resources necessary to grow the business as reasons for its decision.
GM said it plans to instead “realign its autonomous driving strategy” to focus on advanced driver assistance systems and autonomous systems for use in personal vehicles. The company will combine the majority-owned Cruise LLC with GM technical teams.
Huh!. Until that accident where the woman was dragged, I had assumed Cruise was number 2 in the self-driving field behind Waymo.
GM’s internal capital allocation scheme is probably not as patient as the venture capital wing of Wall Street and the AI fanboys.
… or one could argue - this is General Motor’s KODAK MOMENT…
Might well be. History is littered with companies that decided to die rather than undergo the internal turmoil of embracing generational change going on in their industry.