Ugly competition: The Cybertruck has a competitor

[Dark-Fx]

Why are there so many people trying to turn Rivian into Tesla? And why did it become acceptable to just add “cyber” to everything in the process?

Don't be such a cyberdick

Sponsored

 

[UnsungZero_OldTimeAdMan]

And now it seems Tesla is copying the updated copy of a MY

Yes, but who dares question the emperor?

 

[SwampNut]

I guess I feel like we see multiple comments/ threads here every week asking why Rivian didn't do [---] like Tesla, whether it's design language, sales strategy, or some feature they liked from owning a Tesla that Rivian doesn't offer. Some of them are just silly and others are understandable....

Fair point. Not very different from threads asking why didn't Rivian put 87 buttons in the truck like the legacy makers. I think *objectively* we all agree that the software (ADAS and all around) is far behind Tesla. Whether they SHOULD have better ADAS is subjective. I think it's critical both for me, and for the future of the company. Well, and also general societal safety.

Stuff like recurring software bugs and stupid problems with the music streaming roll out even after three iterations is just...so disappointing. Insane really..

 

[mkg3]

Not surprising that people in this forum, Rivian advocates, find the Model Y update being not positive. Personally, I think it looks significantly better than the original bloated Model 3 looks.

It's widely reported that Tesla received 50k orders for new Model Y in the first 24 hours - if its ugly, its a nice problem to have. Now the updated Model Y production is starting in Berlin too, not just China. Hopefully US follows very quickly.

Table from Kelly Blue Book shows the 2024 top 10 selling EVs in US:

What's amazing is how CT outsold F150 Lightning and R1S. R1T didn't make the top 10 according to KBB. Also notable is that spots 7, 8, 9 ahead of R1S, and are all lower priced traditional looking SUVs priced around $50K~$60K. R2 cannot come fast enough so that Rivian can be more competitive in the mareketplace.

The thing is, because Rivian chose to introduce higher priced tri-motor variant rather than lower $60K range R1S, vast majority of people who acquired 7, 8, and 9 position vehicles will not be in the market to replace them for 2~4 years. It is a missed opportunity.

Based on the updated Model Y, Tesla is providing a new viable option for keeping lease expiring early Model Y and owners to stay as Tesla owners. Clearly not 100% but a large percentage.

The looks and beauty are in the eyes of beholder, as the cliche goes; whereas, the gross margin and the bottom line are facts and the numbers set you free.

 

[Donald Stanfield]

Fair point. Not very different from threads asking why didn't Rivian put 87 buttons in the truck like the legacy makers. I think *objectively* we all agree that the software (ADAS and all around) is far behind Tesla. Whether they SHOULD have better ADAS is subjective. I think it's critical both for me, and for the future of the company. Well, and also general societal safety.

Stuff like recurring software bugs and stupid problems with the music streaming roll out even after three iterations is just...so disappointing. Insane really..

I disagree. Tesla is missing some critical features for my use case. The lack of an overhead view is unacceptable in 2025. I also think that saying Tesla's ADAS is better than Rivian's is true, but ultimately meaningless. Tesla's full self drive is so far away from being ready to replace human drivers that to me it would be worse trying to use it than manual driving.

There was an article just posted about this somewhere on this forum that shows right now Tesla's miles before intervention is around 430 miles. Compare that to a Human's miles between accidents at 670,000; the result is like comparing a peewee football team with the Super Bowl champs. Just because Rivian is a worse peewee team doesn't make the Tesla software usable.

 

[SwampNut]

to me it would be worse trying to use it than manual driving.

That's because you haven't actually used it or Autopilot. AP has been able to do 98% of driving for a very long time. I have very little interest in FSD since I've seen that AP does nearly all of the job anyway.

Source: 30k miles using it.

Tesla's miles before intervention is around 430 miles. Compare that to a Human's miles between accidents at 670,000

What a ridiculous set of comparison points.

 

[Ecupip]

Not surprising that people in this forum, Rivian advocates, find the Model Y update being not positive. Personally, I think it looks significantly better than the original bloated Model 3 looks.


The looks and beauty are in the eyes of beholder, as the cliche goes; whereas, the gross margin and the bottom line are facts and the numbers set you free.

Most Tesla owners are repeat customers or new customers who value the brand's maturity and the charging network. Looks and beauty come second to those priorities for a lot of people. That's why so many compare Tesla to Toyota.

 

[Donald Stanfield]

That's because you haven't actually used it or Autopilot. AP has been able to do 98% of driving for a very long time. I have very little interest in FSD since I've seen that AP does nearly all of the job anyway.

Source: 30k miles using it.



What a ridiculous set of comparison points.

What? How is comparing miles between a human having to take over to prevent a crash and how often humans crash a ridiculous set of comparison points? Seems to me that those are the metrics you want to use.

 

[Dave Cundiff]

What? How is comparing miles between a human having to take over to prevent a crash and how often humans crash a ridiculous set of comparison points? Seems to me that those are the metrics you want to use.

Not every "humans had to take over" event would have resulted in a crash. But many would have.

So I think it's a meaningful metric, even though it's somewhat unfavorable to the automated system.

***

I would suggest "actual 'FSD' crash rates," compared to the actual crash rates of human-driven vehicles under similar circumstances, as a meaningful metric for whether automakers should be allowed to deploy the software on other than a beta-testing basis. If SUPERVISED 'FSD' isn't as good as the average driver, I don't want it out there except with a human who is paid to stay alert.

But, in order to justify a claim or title of "Full Self Driving," I believe the rate of "humans had to take over because a crash seemed imminent" occasions could be meaningfully compared to how often the average human-driven car crashes. If the UNSUPERVISED computer isn't even as safe as the average human, it shouldn't be unsupervised and it would be fraudulent to claim it as "Full Self Driving."

***

Remember that the average human driver, coming your way on the average road, is drunk or impaired a significant percentage of the time. "The average human driver" is not an unreasonably high standard to meet. You could even argue that it's an unreasonably low standard, though I'm not saying that....

Best to all!

 

[Donald Stanfield]

Not every "humans had to take over" event would have resulted in a crash. But many would have.

So I think it's a meaningful metric, even though it's somewhat unfavorable to the automated system.

***

I would suggest "actual 'FSD' crash rates," compared to the actual crash rates of human-driven vehicles under similar circumstances, as a meaningful metric for whether automakers should be allowed to deploy the software on other than a beta-testing basis. If SUPERVISED 'FSD' isn't as good as the average driver, I don't want it out there except with a human who is paid to stay alert.

But, in order to justify a claim or title of "Full Self Driving," I believe the rate of "humans had to take over because a crash seemed imminent" occasions could be meaningfully compared to how often the average human-driven car crashes. If the UNSUPERVISED computer isn't even as safe as the average human, it shouldn't be unsupervised and it would be fraudulent to claim it as "Full Self Driving."

***

Remember that the average human driver, coming your way on the average road, is drunk or impaired a significant percentage of the time. "The average human driver" is not a high standard to meet.

Best to all!

I understand what you’re saying but Elon has FSD programmed to disengage right before a crash is imminent, within a second or two, to artificially appear as though FSD isn’t responsible for the crash. So you cannot use that metric as it’s artificially small.

 

[SwampNut]

How is comparing miles between a human choosing to take over

This is the actual number you quoted.

having to take over to prevent a crash

WILDLY different number. Most human interventions would not result in a crash.

Again, tens of thousands of miles of experience, not opinion, not what some media outlet claimed.

 

[Donald Stanfield]

This is the actual number you quoted.



WILDLY different number. Most human interventions would not result in a crash.

Again, tens of thousands of miles of experience, not opinion, not what some media outlet claimed.

How do you figure that these human interventions wouldn’t result in a crash? The whole emergency disengagement means that the computer cannot deal with the situation at hand. That means if a human wasn’t there the car would crash, how can you say it wouldn’t.

Not only that, but all the videos of FSD doing stuff like driving on the wrong side of the road aren’t even counted as incidents in that metric as the computer thinks it’s fine. Bottom line is if you think FSD is as good as a human or close you should be able to find a study backing it up.

My argument is coming from a paper written by a traffic safety expert using crowd sourced Tesla data and comparing it to known human traffic safety data. I’m sorry but “my own experience” isn’t a valid source here.

 

[Alanparkcity]

In the one competition that nobody wants to win, we have a new competitor! The refreshed Tesla Model Y was found on Tesla’s Chinese website.
Spoiler alert ? the R2 and R3 aren’t going to have much competition from Tesla in the looks category. It seems the Model Y somehow got even uglier and more jellybean shaped than ever before!! And once again, their designers took a day off!
https://www.caranddriver.com/news/a63387827/2025-tesla-model-y-juniper-revealed/

Looks like a typical Japanese or Korean economy car.

 

[SwampNut]

How do you figure that these human interventions wouldn’t result in a crash?

How do you figure they would? I intervened with AP hundreds, maybe thousands of times over the years. Not because I was going to crash. Why do you assume an intervention is an "emergency???"

Also, how about we talk about Autopilot, which is what I keep saying. FSD is beta, it's limited, it's paid for. AP is included like the Rivian includes Driver+. Let's compare apples and apples, eh?

 

[Donald Stanfield]

How do you figure they would? I intervened with AP hundreds, maybe thousands of times over the years. Not because I was going to crash. Why do you assume an intervention is an "emergency???"

Also, how about we talk about Autopilot, which is what I keep saying. FSD is beta, it's limited, it's paid for. AP is included like the Rivian includes Driver+. Let's compare apples and apples, eh?

I assume intervention is an emergency because it’s a several ton vehicle moving at speed and the “person” driving it suddenly gives up. That would be like me driving down the road and all of a sudden throwing up my hands yelling “idk what to do”.

Rivian’s system does this too to be fair, gen 2 is better than gen 1 at it but it’s still annoying. The point I’m making is there is NO ADAS system on the market that allows for the human driver to stop paying attention. If I have my wife drive I can do whatever, I will get where I need to go without doing anything. No manufacturer has a system that even remotely approaches that level of control.

Comparing the different ADAS systems once you get to the point of auto lane change doesn’t matter that much IMO. Hands off or hands on with a capacitive wheel isn’t much different. Being able to make the lane change without me signaling might be slightly nicer, but since I’m still supposed to be paying attention it’s not that much better in practical use.

There’s no use in declaring better or worse once you have lane keep and lane change because the degree of better isn’t enough to hang my hat on as the amount of error these systems have mean I still have to be paying attention as if I was driving.

Sponsored