What's new
By:

Lucid Intends to Deliver First Level 4 Autonomous EVs for Consumers with NVIDIA

Need4Sleep said:
I'm a data scientist and I can feel the pain and frustration it took to finally reach the output of this pipeline.

And the amount of profanity when the pipeline outputs something it shouldn't have.
Click to expand...
“Sorry but distance between two points is not supported in this client” 😭
 
At the end of the day, any plans Lucid has beyond the 1H 2026 require support from KSA as Lucid has said they have enough cash to get to 2H 2026.

So, I am not putting much weight behind end of 2026 or later targets as they will need clarity on capital well before then.

I should also add, I fully expect KSA will support them through the Earth launch, but we don’t know that for certain.
 
kapix said:
Has anyone noticed the irony; the gallons of “ink” spilled and hours of opining on the software capabilities of Lucid Motors - a company that has produced at least two vehicles which are, by nearly every measure, outstanding driver’s cars.

DRIVE…not RIDE. I could not be happier with both of our Lucid Airs and look for any excuse to get behind the wheel.

I have to believe that most Lucid owners were initially seduced by the DRIVING experience of their Lucid.

Get out and drive while you can.

Level 5 autonomy is years away and, when it arrives, you will be faced with traffic streams of Risk-Avoiding Machines being vexed by analog Gen Luddites on two-wheelers in a game to confuse the RAMs.

Those of a certain age will remember a time when ‘going for a ride’ was a fun adventure. If you look, you’ll still find that opportunity but you’ll need a driver’s car; a Lucid.

Get out and DRIVE before you can’t.
Click to expand...

All intended L4 cars (Waymo, CyberCab... still come with standard human controls of steering, accelerating, and braking).

Thus, there's no threat that the options for human controls will be gone.

However, the other option for a car without human controls that drive better than humans are not anywhere insight.

The most advanced consumer system is Mercedes L3 with a maximum speed of 40mph on highways in the US.

The most advanced commercial system is Waymo L4 but only in a few cities and it is attached to an umbilical cord for remote operators.

Thus, manual driving is not going to be replaced anytime soon. Autonomous Driving has been a hype years after years. Each year, the hype is the same: it's different now because the new technology is just so good and so fast.

I am not against manual driving and I am not against Autonomous driving as long as they are both safe.
 
MidwestLucidDude said:
Level 4 in a mid-size. Okay. The $50,000 midsize? This invites many questions. Will Level 4 be a basic part of the car or an option? Will Level 4 have an annual subscription fee? Besides the NVIDIA chips that make this happen, it seems the vehicle will need more sensors. That equals more money. If it is a basic part of the car, it will certainly be a more expensive vehicle. If it is an option, will the vehicle come with all the sensors and chips in it but you have to pay more (an annual subscription) to unlock the feature? Then if it is a true option I would expect without it, there'd be no NVIDIA chip in it and a fewer sensors to save on costs, but that would also mean there would be no chance to turn it on later. As a true option, how much will it cost? $10,000? Then it begs the question how many people would be willing to pay for that option? Another recent news article mentioned that very few of the Tesla owners actually buy the Full Self Driving package. Overall, there was not a lot of information in this news release other then, "We're doing Level 4 with our mid-size." It's hard to get excited about this announcement at this point in time.
Click to expand...
One should expect that level 4 autonomy will be a subscription feature, at least in the early years. It's likely to be a system that needs a substantial ongoing support effort.
 
Air and Gravity are unique product, people buying Air/Gravity want the space, efficiency and handling, no other product are even close.

Midsized on the other hand, though I believe the car itself would beat competitors such as Model 3/Y in handling, comfort and efficiency, it is going to be in a more competitive market.

Having L4, or close enough to L4 capable software is going to elevate the attractiveness of Lucid midsized, and the same software can be copy pasted to 100s of thousands of car, hence justify the R&D cost.
 
Tâm said:
All intended L4 cars (Waymo, CyberCab... still come with standard human controls of steering, accelerating, and braking).

Thus, there's no threat that the options for human controls will be gone.

However, the other option for a car without human controls that drive better than humans are not anywhere insight.

The most advanced consumer system is Mercedes L3 with a maximum speed of 40mph on highways in the US.

The most advanced commercial system is Waymo L4 but only in a few cities and it is attached to an umbilical cord for remote operators.

Thus, manual driving is not going to be replaced anytime soon. Autonomous Driving has been a hype years after years. Each year, the hype is the same: it's different now because the new technology is just so good and so fast.

I am not against manual driving and I am not against Autonomous driving as long as they are both safe.
Click to expand...
It's my understanding that Level 5 autonomy allows for no human-operated controls in the "car". In my view, widespread adoption of L5 signals the end of driving and the start of riding. L5 may not be in sight but it is the goal and the clock is ticking. Drive while you can.
 
kapix said:
It's my understanding that Level 5 autonomy allows for no human-operated controls in the "car". In my view, widespread adoption of L5 signals the end of driving and the start of riding. L5 may not be in sight but it is the goal and the clock is ticking. Drive while you can.
Click to expand...
Luckily, the Lucid is such an amazing driving machine. Makes driving easy. I agree with @AirDoll that it's almost a sin to have it in Drive Assist. LOL.
 
I don’t understand. I’ve been driving an Air for almost three years and it’s the most fun I can have legally.

Why would I want to give up that much pleasure for a car that drives itself? Where’s the fun in that?
 
inthewoods said:
I don’t understand. I’ve been driving an Air for almost three years and it’s the most fun I can have legally.

Why would I want to give up that much pleasure for a car that drives itself? Where’s the fun in that?
Click to expand...
No one forces you to give up manual driving.

This thread adds more options for you to choose and not taking away any of your choices.

Some people love to write words manually while Presidents want to let an autopen to do the signing. That's not giving up choices. That's adding choices.

Blind people have high hopes that autonomous vehicles would provide them more flexibility and freedom.

As I get older, my joints are not as nimble and pain-free anymore. More choices are good in my old age.
 
I used the "deep research" function on Gemini, so obviously take this with a grain of a salt, but it suggests there is no chance that any of these ADAS improvements will trickle down to current air owners given it runs on Mobileye's software stack. So we'll have to wait for air to get an overhaul.

On a positive note, it suggested that they'd realistically be able to achieve parity with Tesla's current (2025), urban-capable ADAS in late 2027 to mid 2028. Limitation being that they're essentially starting from scratch using Nvidia's hardware and software, which should (in theory) allow them to make progress VERY quickly, but will necessarily require a ramp up period as it gets fed real-world data for edge cases.
 
The_Don said:
I used the "deep research" function on Gemini, so obviously take this with a grain of a salt, but it suggests there is no chance that any of these ADAS improvements will trickle down to current air owners given it runs on Mobileye's software stack. So we'll have to wait for air to get an overhaul.

On a positive note, it suggested that they'd realistically be able to achieve parity with Tesla's current (2025), urban-capable ADAS in late 2027 to mid 2028. Limitation being that they're essentially starting from scratch using Nvidia's hardware and software, which should (in theory) allow them to make progress VERY quickly, but will necessarily require a ramp up period as it gets fed real-world data for edge cases.
Click to expand...
All of this is garbage, unfortunately, for one very simple reason: Lucid Air does not use Mobileye for ADAS. It uses NVIDIA DRIVE.

While Lucid originally announced a partnership with Mobileye (back in 2016/2017) for its early driver-assist systems using EyeQ chips, the production vehicles leverage NVIDIA's high-performance compute for the current DreamDrive Pro system.

AI is great and all, and I use it often, but facts are not the thing they're particularly good at. 🤷‍♂️
 
borski said:
All of this is garbage, unfortunately, for one very simple reason: Lucid Air does not use Mobileye for ADAS. It uses NVIDIA DRIVE.

While Lucid originally announced a partnership with Mobileye (back in 2016/2017) for its early driver-assist systems using EyeQ chips, the production vehicles leverage NVIDIA's high-performance compute for the current DreamDrive Pro system.

AI is great and all, and I use it often, but facts are not the thing they're particularly good at. 🤷‍♂️
Click to expand...
The MobileEye demos back then were better than what Lucid does with NVIDIA today, and certainly better than what Lucid released 4 years ago.
 
Alex said:
The MobileEye demos back then were better than what Lucid does with NVIDIA today, and certainly better than what Lucid released 4 years ago.
Click to expand...
No idea; I only started following Lucid heavily in late 2021 (shocker, I know).

My point wasn't about quality of the ADAS, but about the quality of the Gemini 'report'.
 
borski said:
All of this is garbage, unfortunately, for one very simple reason: Lucid Air does not use Mobileye for ADAS. It uses NVIDIA DRIVE.

While Lucid originally announced a partnership with Mobileye (back in 2016/2017) for its early driver-assist systems using EyeQ chips, the production vehicles leverage NVIDIA's high-performance compute for the current DreamDrive Pro system.

AI is great and all, and I use it often, but facts are not the thing they're particularly good at. 🤷‍♂️
Click to expand...
This is good news!!
 
You must log in or register to reply here.
Back
Top