Model 3 - LR AWD Waiting Room

[Zane Liu]

So I don’t think anyone here is trying to indicate Tesla is not innovating or that vision isn’t an awesome innovation if it works. People are instead indicating there may be edge cases that are harder to resolve.

For instance, vision has faced more phantom breaking than radar. It’s great 90% of the time but the braking is a real issue and does impact day to day use while Tesla tries to resolve it. It may be fixed eventually but presently that is a real driving concern. And yes it’s our role to be vigilant while driving, but again it’s hard to predict the car’s behavior and for those who just get the car it’s a learning process.

Now with the USS sensor, yes vision should behave as well as any human driver can “at least.” But who’s to say a human driver hasn’t scraped their car going over a deep curb or parked a bit to forward and dinged the front ever slightly? Even the most attentive driver can make that error. Technology is amazing because not only does it do as well as a human, it can give us superpowers and be better than us. That’s why we all want FSD to eventually replace human driving. Again, I’m not saying Tesla can’t get vision to work better than humans, but for a period (which might be a year or more), it is a customer’s right to be concerned because maybe their car will do something of concern. It’s an expensive car and no one wants to damage it. Having the potential to be better wouldn’t make a scratch feel any better.

And if you’re arguing that we may not know better than Tesla about these facts, there are plenty of people in this forum with technical backgrounds that are viewing this through the lens of what challenges Tesla needs to account for. It’s ok to support Tesla and also be skeptical. That’s what shareholders and customers should do. But I appreciate you bringing these up because it’s an opportunity for us to discuss these technical elements.

 

[father_of_6]

I appreciate the cameras for sentry and backup cam, but they aren't much value for general driving IMO. I don't use autopilot and don't trust Tesla's perpetual beta software.

For me, radar is better for emergency braking and ultrasonic is better for parking distance... neither are blinded by snow, condensation, sunlight.

The fact that Tesla wants to remove radar from my car is upsetting, and I suspect that ultrasonic will follow.

 

[Mob Dee]

You may be thinking about monster SUVs with big forward blind spots. The Tesla has essentially no forward blind spot, and the forward camera is high on the windshield, affording an even better view of the area in front of the car than the driver sees. Is it perfect? No. Are ultrasonics perfect? No.

There's no need to be harsh. If you know better, then be a good teacher and mentor. Tough love has its place, but it's hardly a cure-all. Also, not everyone is as diligent and attentive as you are, so having safety features like this that ensure that nobody is going to get hurt are important. I'm normally a very diligent and attentive driver, but I have had lapses, and having the car help me out would have been appreciated.

True, I should be more civil. Less harsh.

My point is that Tesla will one day offer autonomous driving. Just not today. All of the safety features and adjustments lead to that end. Let Tesla do it's thing.

In the meantime, enjoy and marvel the best and most efficient battery driven car in the world. Tesla cars are a lot more than FSD.

If you got into for FSD, then shame on you for not knowing better. Elon has been promising this for years. Don't spend money you don't have, or for something that is only promised to come and then complain about it when you don't get what you want when you want it.

 

[ZZ118]

Perhaps that could be unequivocally stated about a license plate frame, but the mobile charger is subject to greater supply shortages. Sure, they also pocketed the money, but why does there only need to be one reason? Can't there be two reasons?

Nah i don't think there is on this one. If the supply is short just give a rain check voucher for when it's back in stock. I like Tesla I bought tesla but this is just like cell phone companies stopping the charging cable. It's to wring out a few more bucks from people (or at least save it since now they don't have to give a charger to everyone). It is what it is

 

[ZZ118]

My eyes and paying attention to my surroundings like a responsible driver should.

Quit being lazy and expecting a car to do everything for you.

It's not expecting the car to do everything for you it's that Tesla sells the car on those points. Ive driven "dumb" cars before never had an issue. However Tesla says we're not a car company we're a tech company making cars. Look we have full self driving and autopsrk and summon. So people pay premium for those features over another car. Especially now if you buy fsd for 15k almost all the features will be inactive until they sort out this new camera system.

I certainly didn't buy it for fsd. I know that's a work in progress. But I did buy it for auto parking, summon, lane change etc. All things that currently exist in a relatively good form and all things that are now being inactivated on new cars for an indefinite period of time. You can spin it all you want but people in October and going to be looking at people in September and saying wth why can't my car do that now. Again one day this will be a moot point and 360 cameras will be better but right now they're not and I wouldn't be surprised if this one costs tesla.

Also yeah most people buy tesla for the technology. Yeah the acceleration is great and it's cool to have zero car emissions but people buy tesla for the tech.

 

[Mob Dee]

Nah i don't think there is on this one. If the supply is short just give a rain check voucher for when it's back in stock. I like Tesla I bought tesla but this is just like cell phone companies stopping the charging cable. It's to wring out a few more bucks from people (or at least save it since now they don't have to give a charger to everyone). It is what it is

Evil corporation with money grasping fingers. I hear you.

However, I disagree. Tesla doesn't need to squeeze a few extra hundred dollars out of you in this way. Tesla could have simply left things as is and raised the price of the car by $200 and never gave a reason as to why. Most people would have never noticed a $200 bump in the price of a 50k plus car.

Then Tesla would have been squeezing an extra $200 out of every purchase.

Instead, they give the customer a choice. My friend who lives in a high rise in the city can't run a electrical extension cord long enough to use a mobile connector. He relies on superchargers. So he didn't purchase one. Nor does he have one sitting wastefully in his trunk, when it could be better served for someone who does need one.

 

[Mob Dee]

It's not expecting the car to do everything for you it's that Tesla sells the car on those points. Ive driven "dumb" cars before never had an issue. However Tesla says we're not a car company we're a tech company making cars. Look we have full self driving and autopsrk and summon. So people pay premium for those features over another car. Especially now if you buy fsd for 15k almost all the features will be inactive until they sort out this new camera system.

I certainly didn't buy it for fsd. I know that's a work in progress. But I did buy it for auto parking, summon, lane change etc. All things that currently exist in a relatively good form and all things that are now being inactivated on new cars for an indefinite period of time. You can spin it all you want but people in October and going to be looking at people in September and saying wth why can't my car do that now. Again one day this will be a moot point and 360 cameras will be better but right now they're not and I wouldn't be surprised if this one costs tesla.

Also yeah most people buy tesla for the technology. Yeah the acceleration is great and it's cool to have zero car emissions but people buy tesla for the tech.

Watch this

Maybe this will help you understand why Tesla is doing what they do.

When it's all said and done, we'll all either be praising Tesla for FSD or demanding our money back.

It's an unfinished product, beware of purchase.

 

[ZZ118]

Evil corporation with money grasping fingers. I hear you.

However, I disagree. Tesla doesn't need to squeeze a few extra hundred dollars out of you in this way. Tesla could have simply left things as is and raised the price of the car by $200 and never gave a reason as to why. Most people would have never noticed a $200 bump in the price of a 50k plus car.

Then Tesla would have been squeezing an extra $200 out of every purchase.

Instead, they give the customer a choice. My friend who lives in a high rise in the city can't run a electrical extension cord long enough to use a mobile connector. He relies on superchargers. So he didn't purchase one. Nor does he have one sitting wastefully in his trunk, when it could be better served for someone who does need one.

Ok your right. Tesla took it away out of the goodness of their hearts so those in need could have one

 

[Mob Dee]

Ok your right. Tesla took it away out of the goodness of their hearts so those in need could have one

I like your style, my friend

 

[JB47394]

And if you’re arguing that we may not know better than Tesla about these facts, there are plenty of people in this forum with technical backgrounds that are viewing this through the lens of what challenges Tesla needs to account for. It’s ok to support Tesla and also be skeptical. That’s what shareholders and customers should do. But I appreciate you bringing these up because it’s an opportunity for us to discuss these technical elements.

Well, I'm a long time software engineer and I know enough to know that my uninformed intuitions are insufficient to know the right approach to take. I've worked on countless projects where my initial instinct was to use solution A only to find that it wasn't going to work at all, and only by learning more about the problem domain could I understand that solution B or C or D was actually the way to go. Sometimes even my informed intuitions never yielded a good solution, and I had to rely on the intuition of others to find something workable.

After working on the problem for a few years, this is Tesla's assertion about vision versus LIDAR and RADAR and ultrasonics; they've discovered that in order for them to make autonomy work, the sensor suite must be tightly integrated. A single sensor type makes that possible. Today, the only practical sensor that fits the bill on capabilities and cost is cameras. Universities and other companies have tried a combined sensor suite and they haven't gotten as far as Tesla.

Down the road, two things could happen.

1. Cheap, compact, high resolution LIDAR systems could become available, allowing companies to go with LIDAR-only solutions because that only requires one type of sensor.
2. Somebody could figure out how to cleanly integrate multiple sensor types. That would be a huge win not just for autonomy, but for artificial general intelligence (AGI).

Lastly, I'll observe that the car features that rely on ultrasonics don't work particularly well. I assume that this is, in part, because of this problem of sensor integration. They need to use the vision system for navigation and the ultrasonics for proximity. But the two sets of data are 'different' enough that Tesla doesn't know how to cleanly integrate them to find solutions. So they're going to drop ultrasonics and do everything with vision. I'm certain that the tens of millions in savings that they'll reap from the change only provided a massive tailwind to following that path. I suspect that the reason they could follow the path was because of their growing competency with vision systems.

 

[father_of_6]

Well, I'm a long time software engineer and I know enough to know that my uninformed intuitions are insufficient to know the right approach to take. I've worked on countless projects where my initial instinct was to use solution A only to find that it wasn't going to work at all, and only by learning more about the problem domain could I understand that solution B or C or D was actually the way to go. Sometimes even my informed intuitions never yielded a good solution, and I had to rely on the intuition of others to find something workable.

After working on the problem for a few years, this is Tesla's assertion about vision versus LIDAR and RADAR and ultrasonics; they've discovered that in order for them to make autonomy work, the sensor suite must be tightly integrated. A single sensor type makes that possible. Today, the only practical sensor that fits the bill on capabilities and cost is cameras. Universities and other companies have tried a combined sensor suite and they haven't gotten as far as Tesla.

Down the road, two things could happen.

1. Cheap, compact, high resolution LIDAR systems could become available, allowing companies to go with LIDAR-only solutions because that only requires one type of sensor.
2. Somebody could figure out how to cleanly integrate multiple sensor types. That would be a huge win not just for autonomy, but for artificial general intelligence (AGI).

Lastly, I'll observe that the car features that rely on ultrasonics don't work particularly well. I assume that this is, in part, because of this problem of sensor integration. They need to use the vision system for navigation and the ultrasonics for proximity. But the two sets of data are 'different' enough that Tesla doesn't know how to cleanly integrate them to find solutions. So they're going to drop ultrasonics and do everything with vision. I'm certain that the tens of millions in savings that they'll reap from the change only provided a massive tailwind to following that path. I suspect that the reason they could follow the path was because of their growing competency with vision systems.

#1 is already happening. Regarding #2, this is referred to as "sensor fusion". Tesla is going "all in" on cameras, which for me is hard to understand since I live in the Northeast US and my cameras are blind multiple times per day for at least half the year.

I suspect that in a few years once LiDAR costs have dropped, Tesla will need to switch gears and pay for retrofitting any car with FSD.

 

[father_of_6]

Nah i don't think there is on this one. If the supply is short just give a rain check voucher for when it's back in stock. I like Tesla I bought tesla but this is just like cell phone companies stopping the charging cable. It's to wring out a few more bucks from people (or at least save it since now they don't have to give a charger to everyone). It is what it is

We disagree. I'm with @Mob Dee on this one. They'd already bumped prices several times in a few months. If they wanted to increase the prices again, they would've just done it.

They were unable to produce enough mobile connectors due to supply shortages (just like USB ports in the center console) and they made a *smart* move by not including them with the car, likely reducing demand by a huge margin, as many owners opt for hard-wired solutions or already have charging at home (just like cell phone users already have a bajillion USB-C cables at home).

 

[ColoradoMike]

Wow! So much for the "Waiting Room".....LOL!

Just an update! We had our headliner replaced a couple weeks ago when we found on delivery it was not correct. Took Tesla about 6 weeks to get the part in and now it is perfect! The only issue left is some holograms that the service team could not see under their shop lights. It is only evident in the sunshine so we are waiting til next week to bring it in and show them.

The Service Team has been outstanding! They reached out to inform us that they could not duplicate it in the shop environment and THEY suggested we bring it in when in full sun.

I simply cant say enough about the communication and service we have received from our service team here in Colorado Springs! Great people to work with!

 

[guerro]

FWIW, I have not had any phantom braking incidents since taking delivery of my car. Twice a week on the freeway to Scottsdale from Chandler using autopilot.

 

[ColoradoMike]

A question for new delivery folks that have two drivers.....

Have any of you experienced an issue with phones not being recognized?

Here is the issue....we have about going on 2 months of ownership and it is kind of a PITA to have to open the Tesla app to unlock the car.

It is getting to the point where we may need a Service appt but wanted to get a "feel" from folks that have encountered this issue.

Thanks!

 

[VT_EE]

Tesla has a long history of making this type of change and relying on half-baked, or non-baked solutions. Getting rid of the USS before the replacement is ready is garbage. No different than the rain sensor or AP2 debacle.

 

[father_of_6]

A question for new delivery folks that have two drivers.....

Have any of you experienced an issue with phones not being recognized?

Here is the issue....we have about going on 2 months of ownership and it is kind of a PITA to have to open the Tesla app to unlock the car.

It is getting to the point where we may need a Service appt but wanted to get a "feel" from folks that have encountered this issue.

Thanks!

My first 2 weeks of ownership the phone as key worked 100% of the time. Then I installed an update on the car and the very next morning the door didn't unlock when I pulled the handle.

Since then (21 months), it's been consistently spotty. Usually I just need to keep pulling the door handle and be patient, but sometimes I have to take my phone out of my pocket.

It never actually *requires* starting the app (since the app is always running in the background).

My guess is that Tesla tried to optimize power consumption in the vehicle and thus reduced the polling interval of the car (is there a phone nearby?, is there a phone nearby?, is there a phone nearby?). Additionally, perhaps they have changed the signal strength requirement of the Bluetooth connection for security purposes.

It's disappointing, because the first 2 weeks of flawless phone-as-key were great.

 

[ColoradoMike]

My first 2 weeks of ownership the phone as key worked 100% of the time. Then I installed an update on the car and the very next morning the door didn't unlock when I pulled the handle.

Since then (21 months), it's been consistently spotty. Usually I just need to keep pulling the door handle and be patient, but sometimes I have to take my phone out of my pocket.

It never actually *requires* starting the app (since the app is always running in the background).

My guess is that Tesla tried to optimize power consumption in the vehicle and thus reduced the polling interval of the car (is there a phone nearby?, is there a phone nearby?, is there a phone nearby?). Additionally, perhaps they have changed the signal strength requirement of the Bluetooth connection for security purposes.

It's disappointing, because the first 2 weeks of flawless phone-as-key were great.

Yeah it is frustrating. I have tested it a ton.

What happens is that the screen lights up stating to use the key card. Not entirely certain what is going on.

Thanks for the response it helps!

 

[father_of_6]

What happens is that the screen lights up stating to use the key card. Not entirely certain what is going on.

This is essentially the car's way of saying "the car is locked, unlock it to get in". Once it detects the phone, the door opens.

Worth noting: I *never* need to use the key card in these scenarios... that would take even longer. The phone always works, it's just not "walk up and get in" like it could be.

 

[ColoradoMike]

This is essentially the car's way of saying "the car is locked, unlock it to get in". Once it detects the phone, the door opens.

Worth noting: I *never* need to use the key card in these scenarios... that would take even longer. The phone always works, it's just not "walk up and get in" like it could be.

I get that....however it doesn't detect the phone for access. BUT...when we hit the unlock button on either phone...it unlocks....so it "knows" the phone is available but will not unlock.

I will do some troubleshooting tonight and tomorrow.