I highly doubt that hardware is not capable. If Tesla can do with their limited hardware, Rivian can do it. Of course, they have decided to move on to “better” hardware; mind you it is not the “only” hardware. It makes sense for them to invest in future hardware; even if the current hardware is capable.
IMO, AP1 is still better than HW3 on the highway. FSD is a little bit better, but still very timid. I am new to D+, but so far I am not put off by its behavior in stop and go traffic like I am with HW3. I've yet to road trip the truck, but I really only used AP on major highways anyway. I guess I'll see just how much D+ freaks out and turns itself off on the interstate.
You did, I didn’t. Between March 2022-March 2024. Those all paid full price without any incentive. Sure I’m fine if I got 20-30% off, but a lot of people didn’t.
They may have locked in supply contracts during COVID, but the decision on the ADAS platform to use was made well before that. The vehicle's system architecture, OEM engineering assistance, camera physical placement, software engineering and development, etc. had been determined well before COVID. It's likely that the OEM overpromised capabilities and/or Rivian made sensor location, camera angle view perspective, etc. errors in the implementation which caused them to "rollback" features as development efforts continued.
We simply don't know the reasons that they can't implement auto lane change. It is easy to say "it's only software, if Tesla can do it, why can't Rivian?". It's very possible that something as "simple" sensor placement/accuracy/resolution is the problem, not software. What if the rear looking cameras and radar do not "look back" far enough with sufficient resolution to provide a safety margin for a car that is going 20 mph faster than you in the left lane? Would you want Rivian to do an automatic lane change if they can't accurately detect fast approaching traffic?
Ok. Let’s leave out Tesla. If Viking probes can traverse and land in mars w/1960s technology, we can do anything with 2021 technology in G1. I am not saying it is going to be easy; but do not cop out saying that G1 cannot do it.
As someone who has worked in engineering with a wide variety sensor devices, including vision systems for manufacturing lines, I can tell you that a sensor with the wrong specification, placement, insufficient resolution, etc. can't be fixed by any software, even if it would be processed by a quantum computer. Rivian can't "do anything [they want] with 2021 technology in G1."
Back in the day, Tesla (Mobileye) didn't need a rear camera for Autopilot lane changes. "AP1" hardware is now over 10 years old. Lane changes were initiated by the driver, so you were on the hook to ensure you yielded the right of way, but the car performed the lane changes on its own. Newer Tesla cars can decide to and then perform lane changes, but I turned that off pretty quickly as its decision making was worse than a student driver at most times. The newer cars are less hesitant at actually making a lane change than the older ones though.
Rivian doesn't want to take risk. That's it.
It was a pretty big risk advertising lane change and hands-free Driver+ for years on tens of thousands of vehicles (100k+ now, right?), then reneging when their new, more profitable platform came out. So, agree to disagree on their risk assessments.
I'm giving Rivian the benefit of the doubt that they encountered safety and/or technical limitations when trying to implement it.
