No more radar in Model Y/3

[JerseyShoreMY]

I am supposed to pick up my Model Y in 5 hours. I have emailed order support and my SA about this, and it's been silent. I am a bit afraid of beta testing pure vision. I feel like this is google photos all over again. Allowing free unlimited photo storage, so they can perform machine learning and create a dataset of your photos. They said they are doing this to 3's and Y's - Their high volume cars. Sounds like guinea pigs?

I am 100% sure this has to do with the chip shortages, and while pure vision will someday work on par or better today it's not even on the same level as radar. I ordered in early April, so there is hope that I still have radar-equipped - If I get there and it's not equipped I am not sure what I am going to do.

1. it's no longer the car I test drove and can't do the things I tested in the same way.
2. I am not getting any kind of discount for a lesser product.
3. am I willing to risk the safety of my family and myself?

Please drive your next gen MY, test autopilot, and let us know what happens without the old radar set! That is our plan also.

New tech could be far better and subject to less faults than radar provides.

Best of luck.

 

[Yoliber]

I'm delaying mine until/if it reaches parity, it has actual hardware improvements, or offered free EAP. With these price hikes, I expect more not less. I'd also pick it up with the mindset that I'd have to retrofit a radar in the future.

 

[RyMY]

I kept delaying placing an order for a MY. With the recent biweekly price increases and the NY $2K rebate going away July 1st, I’m still really close to placing an order. But now that they removed radar and only relying on the cameras, I don’t think I will consider a MY anymore.

It’s a shame. I really like the MY. I just can’t justify the price increase with the removal of radar sensors.

 

[Marcust99]

I kept delaying placing an order for a MY. With the recent biweekly price increases and the NY $2K rebate going away July 1st, I’m still really close to placing an order. But now that they removed radar and only relying on the cameras, I don’t think I will consider a MY anymore.

It’s a shame. I really like the MY. I just can’t justify the price increase with the removal of radar sensors.

Just curious - but why is no radar a deal breaker? What will you miss? Do you think that Tesla took this decision lightly and for no reason?

 

[vitosans]

Just curious - but why is no radar a deal breaker? What will you miss? Do you think that Tesla took this decision lightly and for no reason?

I think they took the decision due to market factors (chip shortage), and the need to accelerate the transition to the camera-only method (future goal). It's going to take a lot of machine learning to train a camera to spot the speed of a moving object accurately.

 

[Yoliber]

Here are my thoughts. I see new owners as being guinea pigs. The fact that they are still purchasing and installing radars in other markets and other vehicles gives me less confidence. Limited features makes it seemed rushed as well.

Here is my thought process if I were to take delivery...

1. I get my vehicle, parity is reached within weeks like Tesla says. Everyone is happy.
2. Repeat of AP1 -> AP2 where it takes years to reach parity.
3. Future models reintroduce radar, I will have install it through the service center.

I personally think option 2 or 3 will happen, so I will observe for a month or two before jumping ship.

 

[rubyred]

Here are my thoughts. I see new owners as being guinea pigs. The fact that they are still purchasing and installing radars in other markets and other vehicles gives me less confidence. Limited features makes it seemed rushed as well.

Here is my thought process if I were to take delivery...

1. I get my vehicle, parity is reached within weeks like Tesla says. Everyone is happy.
2. Repeat of AP1 -> AP2 where it takes years to reach parity.
3. Future models reintroduce radar, I will have install it through the service center.

I personally think option 2 or 3 will happen, so I will observe for a month or two before jumping ship.

Agree with this sentiment. The cars piling up at the factory and distribution centers supposedly “waiting on a part” jibes with all of this. Elon probably said “let’s just rip the band-aid off and go pure vision now so we can get these cars sold” rather than wait for the radar supply to rebound. Pure vision was always the plan, but improbable that it was supposed to happen this quickly. Why bother just starting to fit the refreshed M3 with heated radar literally a few months ago?

Although we all want to believe parity will be reached quickly as Tesla says, we have also been let down countless times. Everything from the make-believe EAP access for early FSD buyers, to any one of dozens of blown deadlines for city FSD.

I would still take the car, but I would need to be at peace with the fact that it may be months or years before feature parity.

 

[MYMY21]

This is what the whole radar conversation sounds like to me:

The floppy disk is dead (and Apple helped kill it)

It was 1998 and Apple had just released the iMac G3. It was a beautiful interesting computer: a sleek, all-in-one case, with something new called USB. One thing it didn't have was a floppy disk. At the time, many believed Apple was insane for leaving a floppy disk drive off the iMac, but did...

www.engadget.com

Barely two decades ago, Apple nixed the floppy disk drive. Ten years after that, Sony stopped making them. There are other interesting memories in this article. CDs replaced floppies. Now how many Apple products ship with CDs? And Adobe Flash? Also gone.

 

[OCR1]

I would be reluctant to move forward with a vehicle until more information is made available. I feel like this is Tesla spinning their story to make it sound like a feature when in reality they don’t have the parts they need to deliver cars and they don’t want to have a bad quarter. It may all be fine but it’s too new to know what to make of this information. Let’s see some actual testing of cars using pure vision versus vision + radar and see how they compare.

 

[Platini]

This is what the whole radar conversation sounds like to me:

The floppy disk is dead (and Apple helped kill it)

It was 1998 and Apple had just released the iMac G3. It was a beautiful interesting computer: a sleek, all-in-one case, with something new called USB. One thing it didn't have was a floppy disk. At the time, many believed Apple was insane for leaving a floppy disk drive off the iMac, but did...

www.engadget.com

Barely two decades ago, Apple nixed the floppy disk drive. Ten years after that, Sony stopped making them. There are other interesting memories in this article. CDs replaced floppies. Now how many Apple products ship with CDs? And Adobe Flash? Also gone.

Radar has been around for 75 years and has been used ever since in navigation. Radar is a fundamental technology and tool and is not going away any time soon.

 

[MYMY21]

That’s an alternative perspective: 75 years might suggest that it is proven and reliable or it might suggest it is long past due for disruption. Either way, prototyping, deploying, testing, analyzing results, and iterative improvements are key.

 

[DayTrippin]

This thread parallels with a lot of comments I made in another thread on the value of FSD. Every technology has some weaknesses. The advantage of using several like they did before allows you to use each one hopefully in its most optimal implementation. Here we are using a computational algorithm for the cameras to do something natively that LIDAR/Radar can do by default, measure distance, speed, etc. The farther apart the lenses, the more accurate the measurement can be and likely it can do it more quickly as the difference is greater. So did they improve the camera's specs in anyone to improve their performance or just gut the radar?

This seems like needless computational complexity, increased risk, reliance on algorithms (we all know what can go wrong there) and greater likelihood of failure than needed to be. Anyone who buys this car is going to be a beta tester without any benefit. Show me the studies or experiments and show me the limitations of its current implementation, not some future state we may never reach. My family's life may depend upon it.

So will they retrofit later if it doesn't pan out or will it take a class action suit to get them to notice? At this point I am pretty much ready to pause my order until I see more info. Let someone else be the sucker, err "beta tester", and work the kinks out. If it doesn't work as intended, and there are more accidents as a result, you can be damn sure the insurance companies will jack the rates for all M3 and MYs or check your VIN to see if you have one of the "affected ones".

 

[Fourdoor]

This thread parallels with a lot of comments I made in another thread on the value of FSD. Every technology has some weaknesses. The advantage of using several like they did before allows you to use each one hopefully in its most optimal implementation. Here we are using a computational algorithm for the cameras to do something natively that LIDAR/Radar can do by default, measure distance, speed, etc. The farther apart the lenses, the more accurate the measurement can be and likely it can do it more quickly as the difference is greater. So did they improve the camera's specs in anyone to improve their performance or just gut the radar?

This seems like needless computational complexity, increased risk, reliance on algorithms (we all know what can go wrong there) and greater likelihood of failure than needed to be. Anyone who buys this car is going to be a beta tester without any benefit. Show me the studies or experiments and show me the limitations of its current implementation, not some future state we may never reach. My family's life may depend upon it.

So will they retrofit later if it doesn't pan out or will it take a class action suit to get them to notice? At this point I am pretty much ready to pause my order until I see more info. Let someone else be the sucker, err "beta tester", and work the kinks out. If it doesn't work as intended, and there are more accidents as a result, you can be damn sure the insurance companies will jack the rates for all M3 and MYs or check your VIN to see if you have one of the "affected ones".

Anyone who drives a Tesla is a beta tester, it is part of the deal... if you didn't know that before the radar issue came up then you were living in the dark. If you have Radar, you are a beta tester for the radar + vision AP system. If you don't have radar you are a beta tester for the pure vision AP system. At least we are not paying $10K for the privilege of being a beta tester like the FSD people.

Keith

 

[vitosans]

We decided to take the leap, and we picked up our Model Y this afternoon. So far, you really can't tell the difference - it's still in training mode so it won't allow me to use AP or any of the stuff that would require it. It drives nice, and the fit and finish are good, only a few minor issues, and they are already booked to be fixed. All and all here we go on the beta train, toot toot!

 

[DayTrippin]

Congrats on your new Tesla. Hopefully the beta testing ends on a positive note.

 

[whisperingshad]

We decided to take the leap, and we picked up our Model Y this afternoon. So far, you really can't tell the difference - it's still in training mode so it won't allow me to use AP or any of the stuff that would require it. It drives nice, and the fit and finish are good, only a few minor issues, and they are already booked to be fixed. All and all here we go on the beta train, toot toot!

Wow, mine had 26 miles so i didnt have to train it, how many miles yours have?

 

[Platini]

WASHINGTON, May 26 (Reuters) - Newer Tesla Model 3 and Model Y vehicles will no longer be labeled as having some advanced safety features after the automaker said it was removing radar sensors to transition to a camera-based Autopilot system, the National Highway Traffic Safety Administration (NHTSA) said Wednesday.
The U.S. agency confirmed it updated its website to show that Tesla Model 3 and Model Y vehicles produced on or after April 27 "do not have NHTSA’s check mark for recommended safety technologies: forward collision warning, lane departure warning, crash imminent braking and dynamic brake support.

https://finance.yahoo.com/news/1-tesla-loses-u-designation-014715090.html

 

[Anthonyp]

Ordered my car on May 21st delivery date is expected June 20th-30th. Just heard the news about no radar on the car. Not super pleased about that. What are everyone thoughts? Will it make a massive difference?

 

[Mulkogi]

Ordered my car on May 21st delivery date is expected June 20th-30th. Just heard the news about no radar on the car. Not super pleased about that. What are everyone thoughts? Will it make a massive difference?

By the time you get delivery of yours they would have released the software update for this. I think its actually a good thing you get the car later if you are worried

 

[Veritas]

I wonder whether those of us that have radar will have it turned off and lose the ability to go over 75mph.

Not according to an article which says cars with radar can use AP upto 90