Interim CEO claims L2++ EoY and L4 by 2029

[Klinkman]

They've been saying Nvidia Drive has been able to do autonomous driving for years

That is just not true. Nvidia share holder here. I've been following their self driving progress since 2016 and made a living in silicon valley.

Demos are one thing, but "able" is a claim they really haven't gotten close to until this year. To be clear, I'm talking about merchant solutions, like the solutioun will be available at multiple OEMs. The solution (hardware and software) has generally been positioned as "in the future," the demos were just peeks at what they were aspiring towards.

Plenty of my motivation in purchasing a Gravity was because I believed the solution was now close/near future and based on Orin hardware. (It was also clear to me years ago anyone buying Tesla/FSD was buying a load of BS).

 

[HC_79]

That is just not true. Nvidia share holder here. I've been following their self driving progress since 2016 and made a living in silicon valley.

Demos are one thing, but "able" is a claim they really haven't gotten close to until this year. To be clear, I'm talking about merchant solutions, like the solutioun will be available at multiple OEMs. The solution (hardware and software) has generally been positioned as "in the future," the demos were just peeks at what they were aspiring towards.

Plenty of my motivation in purchasing a Gravity was because I believed the solution was now close/near future and based on Orin hardware. (It was also clear to me years ago anyone buying Tesla/FSD was buying a load of BS).

Sure

NVIDIA, AUDI Partner to Put World's Most Advanced AI Car on Road by 2020

CES -- NVIDIA and Audi today shifted their decade-long partnership into high gear, announcing that they are collaborating to put advanced AI cars on the road starting in 2020.

nvidianews.nvidia.com

 

[Klinkman]

Sure

This release is talking about putting out "advanced cars on the road" and using AI to "tackle the complexities of driving".

"Audi's adoption of our DRIVE computing platform for AI cars will accelerate the introduction of next-generation autonomous vehicles, moving us closer to a future of higher driving safety and new mobility services."

There is no promise of autonomous driving in DrivePX.

I'd also note the relationship with Audi blew up, but the level3 promise was kept and rolled out in Mercedes S-class and EQS sedans last year.

 

[Samo23]

Ahh yes....."the most future ready sensor suite hardware of any vehicle on sale today."

 

[Saksak]

I really don't understand your counter point to mine, are you insinuating that I don't understand there are multiple teams who work on both cars and people? If so I do believe that to be true...to circle back my point was how are you going to trust a Dr. or a MFG to do this amazing feat if more remedial tasks are too challenging...

Please do let me know if I'm missing something from your point @Saksak

@SaaSManKS If the software team responsible for the UI was doing the self-driving work, I agree that I would have no faith in them delivering FSD. I would hope that it’s a completely separate team that is working on autonomy, though.

The performance/handling software team (and hardware design team) team did an incredible job on the suspension and power train. Lucid (Rawlinson, I think) admitted that they didn’t realize the importance of UI software, so I’m giving Lucid the benefit of the doubt that the autonomy team will be more like the performance/handling team instead of the UI team.

 

[SaaSManKS]

The performance/handling software team (and hardware design team) team did an incredible job on the suspension and power train. Lucid (Rawlinson, I think) admitted that they didn’t realize the importance of UI software, so I’m giving Lucid the benefit of the doubt that the autonomy team will be more like the performance/handling team instead of the UI team.

ah ok, I will restate: If a hospital can claim they can cure cancer but they can't even handle a common cold. Is that better and/or more inclusive for you?

 

[Saksak]

ah ok, I will restate: If a hospital can claim they can cure cancer but they can't even handle a common cold. Is that better and/or more inclusive for you?

I'd be very worried. We haven't cured the common cold .

 

[SaaSManKS]

While you are adorable, my initial statement said “identified”

 

[gaskiller]

@SaaSManKS If the software team responsible for the UI was doing the self-driving work, I agree that I would have no faith in them delivering FSD. I would hope that it’s a completely separate team that is working on autonomy, though.

The performance/handling software team (and hardware design team) team did an incredible job on the suspension and power train. Lucid (Rawlinson, I think) admitted that they didn’t realize the importance of UI software, so I’m giving Lucid the benefit of the doubt that the autonomy team will be more like the performance/handling team instead of the UI team.

I believe it’s just to get on the hype train….look at TSLA, dropping revenue, dropping sales and stock is at all time high….

And they need to do autonomy, not sure when it will really happen but any company not taking autonomy seriously will be history in a decade….

 

[hank-iv]

This really sucks for Air owners, but it's the nature of the beast:
1) Everyone (not just Musk) thought driving autonomy wasn't that hard. Tony Seba said we'd have it by 2020. Mobileye's CEO was promising it by 2021. In 2017, Uber promised to buy 20,000 Volvos to convert to driverless. Etc.
2) As developers realized the problem was harder, hardware had to scale.
3) As developers realize (pretty much just happening recently) that AI is the solution, not hundreds of thousands of lines of C++ code, hardware had to be different.
4) And now we're seeing a shift from thinking LiDAR is necessary to (for L2++ systems at least) not requiring LiDAR or even HD Maps (as claimed by Nvidia).

Tesla even (prematurely) had a program where they'd upgrade the compute hardware in older cars to the latest to get FSD, but that was like HW2 to HW3 (or maybe 2.5), and today they're on HW4 (really HW4.5 since they added a front bumper camera without upping the version number). That ended up costing Tesla money with FSD realization still well over half a decade away (and still not achieved today).

I suspect the promise from Lucid was made with good intentions, but I don't see how the company, especially in its current financial state, can afford to retrofit cars with DDP/DDP2 to the latest. At any rate, we'll see if even the Gravity DDP promise comes to fruition. The Mercedes demo vehicles give us some hope, but the birds are still in the bush.

I'm not a tech person, but it would appear to me that only Waymo has FSD, and they get it via brute force familiarization, from what I understand.

I still take the train to work, and will mostly just use the gravity for road trips and weekend errands. Its an extravagance. So I know that puts me in small subset of an already small set. I will use DDP for long interstate sections of road trips.

That said, why would you buy such a fun car to drive only to let the car drive itself?

Excoriate me.

 

[sri]

It’s all lies from Lucid that current Airs cannot do L2 or L2 ++. Check this video from 2015/2016 the pro type Air testing autonomy. Looks like the new team is chickening out on Air and blaming hardware. They dont want to spend money and time, thats what it is. Mid Size is collobarating with NVDIA or someone else or just buying time till 2029 and chicken out again.

 

[hwy17er]

That said, why would you buy such a fun car to drive only to let the car drive itself?

My feelings EXACTLY:
I want to drive this beast to the mountains and beyond, not have it drive me (or try to kill me...like many other drivers out there!)

Excoriate me too.

 

[HC_79]

It’s all lies from Lucid that current Airs cannot do L2 or L2 ++. Check this video from 2015/2016 the pro type Air testing autonomy. Looks like the new team is chickening out on Air and blaming hardware. They dont want to spend money and time, thats what it is. Mid Size is collobarating with NVDIA or someone else or just buying time till 2029 and chicken out again.

There’s a difference from being able to do something on a prototype which is usually in a very controlled scenario vs deploying something to the masses that has huge consequences should something go wrong.

 

[sri]

There’s a difference from being able to do something on a prototype which is usually in a very controlled scenario vs deploying something to the masses that has huge consequences should something go wrong.

I am talking about chickening out completely on Air saying it is not capable and still keep charing $6750 for Air 2026 models like it is something great and out of the World.

 

[Saksak]

[…]
That said, why would you buy such a fun car to drive only to let the car drive itself?

Excoriate me.

How much do you use cruise control? Would you get a car without it?

A lot of driving is dull: Stuck in rush hour traffic on the daily commute, long road trips on highways, have passengers who simply want a comfortable ride, tired and just want to get from point A to point B. I want manual driving and Gravity’s performance when I can have fun, but I also want FSD as a super-powered cruise control for the boring times.

 

[tmichael]

I have a different perspective on self-driving capability. I'm 71 and love driving. I've been to a number of racing schools, drive over the speed limit and love to accelerate and take on ramps and off ramps aggressively. Corvettes, Trackhawks, Hellcat Durangos are what I consider fun.

But, by the time I'm at an age where I can't/shouldn't/won't drive myself to a doctor appointment, a dinner or a baseball game, I want a car to be self-driving. Thankfully, I hope to have 10+ years before that time arrives, but I want the independence of not asking family for help, the ability to ride in my own car of my own choosing yet do the things I want and need to do. In my lifetime I will have gone from driving cars with drum brakes and carburetors to being taken whereever I want to go electronically.

So, self-driving, I don't need you now, but sure hope you mature so that when I need you, you're safely here. Until then, I'll continue to participate with anyone who wants to "see what that minivan has" on Woodward Avenue or I-75.

Dream ahead!

 

[hank-iv]

How much do you use cruise control? Would you get a car without it?

A lot of driving is dull: Stuck in rush hour traffic on the daily commute, long road trips on highways, have passengers who simply want a comfortable ride, tired and just want to get from point A to point B. I want manual driving and Gravity’s performance when I can have fun, but I also want FSD as a super-powered cruise control for the boring times.

I did say I would use it on the long highway drives. And I have used Tesla's (not) FSD when I have had to drive to work during Chicago rush hour and traffic is creeping along at 5mph. I suppose I should count myself lucky (and I do) to live in a place where I can cut out the vast majority of boring driving by using the train.

My M3 is pretty early, 2018. I paid for FSD then too. Early on I was very unimpressed, other than lane keep and ACC on the highway. FSD even on the highway was dangerous with sudden inexplicable braking and getting halfway into a lane change and jerking back. It has improved a great deal, but still makes me nervous. And requires so many steering wheel touches that I still can't call it FSD.

 

[The_Don]

ah ok, I will restate: If a hospital can claim they can cure cancer but they can't even handle a common cold. Is that better and/or more inclusive for you?

I hear your point but this isn't really applicable here because they're partnering with Nvidia to develop their autonomous capabilities, so they have more man power and expertise. In your scenario, they'd have top doctors for their cancer team and weaker ones handling everyday cases.

 

[SaaSManKS]

I hear your point but this isn't really applicable here because they're partnering with Nvidia to develop their autonomous capabilities, so they have more man power and expertise. In your scenario, they'd have top doctors for their cancer team and weaker ones handling everyday cases.

So what this means for Lucid is the company who's slogan is compromise nothing, and who entered the market leading with the software and UI for ADAS, is now outsourcing the biggest feature of their GTM strategy....can't go wrong!

 

[The_Don]

So what this means for Lucid is the company who's slogan is compromise nothing, and who entered the market leading with the software and UI for ADAS, is now outsourcing the biggest feature of their GTM strategy....can't go wrong!

Not sure how partnering with Nvidia would be considered compromising but ok....