The MB Drive Pilot only works at ridiculously low speeds (under 40 mph), on select highways, no sharp turns, no weather, no construction zones, lanes clearly visible, etc. It's laughable to compare it against FSD which works everywhere. FSD has been flawless when driving on such easy conditions for years.
Are you assuming MB Drive Pilot couldn’t also work pretty well outside of its design domain?
This argument is always so silly because it’s just an argument about the correct way to build a safe control system (i.e. with guaranteed performance characteristics in a defined operational domain) versus the incorrect way to do it (just kinda work over an ambiguously defined set of conditions and hope you don’t kill too many innocent people to cripple your business).
MB Drive Pilot is designed not to attempt to operate outside of a well-defined domain.
It will, by design, not operate in a way that is unsafe or in a context in which it's unsafe.
You are arguing the system would perform worse in contexts in which you have not seen it perform (because its designers decided to prevent it from such attempts).
What evidence do you have for your argument?
Restated: You are assuming that if Mercedes engineers had a similar appetite for putting millions of innocent people at risk by operating in unreliable contexts, that their system would perform far worse than FSD does. On what basis do you make that assumption?
The fact that their software is written only to just follow the car in front, and completely and immediately give up if there's no car in front or one cannot be seen due to it being dark or bad weather, on premapped roads.
Meanwhile FSD has repeatedly shown that's it's capable of driving without a car in front.
You can watch YouTube videos of FSD driving without a car in front for thousands of miles on non-premapped roads.
What makes you think Mercedes would be better at doing that?
What makes you think that MB doesn't have the same capability, but they aren't willing to ship a half-arsed highly dangerous, people killing feature like some other companies do? Unlike any of their competitors, they assume full legal liability for their car's behavior while Drive Pilot is enabled.
MB also has a luxury brand reputation to maintain. When they ship a major feature like that, it absolutely has to work 100% as advertised.
> Unlike any of their competitors, they assume full legal liability for their car's behavior while Drive Pilot is enabled.
Source? If someone gets killed when the system is driving, there is no legal framework for you as a driver to escape criminal penalties. Who at MB are they going to jail for it?
Can UPS indemnify their drivers from involuntary manslaughter charges?
I've worked on competitors to the mb system from legacy auto; generally the way we work is make the best possible system for the long term, then limit the odd to meet safety, regulatory, and brand risk requirements.
So there will be a system happily running producing target driving paths, but some monitoring system disables in the case with no car in front.
I would guess that the mb system is in fact ok without a car in front, but didn't hit their appetite for risk yet.
> You are arguing the system would perform worse in contexts in which you have not seen it perform (because its designers decided to prevent it from such attempts).
Should we also assume that Mercedes cars can fly better than a plane, because we have not yet seen them fly?
What kind of logic is that?
In other news, OpenAI has solved AGI last year, but is keeping it a secret because it's too dangerous.
I have solved self driving, how can you say I didn't if you haven't seen it perform?
No, but generally you should assume a system with near-100% performance within a well-defined ODD doesn't fall off to near-0% performance immediately adjacent to that ODD.
Here's a better example: let's say by law Mercedes has to restrict their cars from exceeding 80mph. They have a car that can reach 80mph in 2 seconds. Is it therefore credible to claim that the car is actually incapable of driving at 81mph?
Or is it more credible to say, "we don't know much about its performance beyond 80mph, but it can probably achieve something outside of that."
A car not being in front means more data for cameras because they can see more of the road ahead if the system can actually do proper road navigation.
The fact that it shuts off immediately shows that they're just copying what the car ahead does.
The system is unable to function on a pre-mapped highway on a clear day with lane markings if there is no other traffic on the road. What does that tell you?
What kind of messed-up logic is that? How many products do you think you own that contain hidden features that you're not supposed to know about? Just about everything with at least a microcontroller inside has extra software modules that you never learn about as the end user (e.g. service modes and factory calibration routines).
FSD isn't driving though, you're driving. Mercedes is driving itself and you have 10 seconds to takeover. Huge difference between level 3 and level 2. One is self driving and one is just a drivers aid.
> Mercedes is driving itself and you have 10 seconds to takeover. Huge difference between level 3 and level 2. One is self driving and one is just a drivers aid.
None of that is true, it just follows the car in front in slow moving traffic during daytime in good weather on premapped roads and cannot even change lanes. No car in front anymore? It completely fails to "self"-drive.
These limitations are taboo to talk about because the Mercedes system is used as an anti-Tesla talking point, you can get permanently banned from Reddit and BlueSky for bringing it up. Thats why so many people think its a great system.
Edit: Downvotes for bringing up inconvenient facts.
I'm very much aware of how Drive Pilot doesn't handle most cases: in my comment I described it as "super limited" and linked to where Mercedes describes the details.
But within the range of cases where the Mercedes system is applicable (which recently went up to 59mph in Germany [1]) it's solidly better than the Tesla system because Mercedes did the engineering to make it reliable enough that you don't have to supervise the car's driving. If I had a highway commute in stop-and go traffic I'd be comfortable reading in a Mercedes, but not in a Tesla.
If Tesla were to artificially set up limitations like highway only, daytime, clearly visible lanes, good weather, car in front, no construction zones, no lane changes, etc, then they would easily have been able to offer full L3 and take full liability for it. Tesla FSD is practically perfected, with no accidents whatsoever, under such conditions already. But Elon wouldn't want to release this kind of gimped system as he is aiming for full self driving everywhere.
Absolutely untrue. It's not even close to perfect. I use it a few times a week and while it's seen some significant improvements over the last year, it still makes pretty dangerous mistakes, especially on left turns, right-turn slip lanes, intersections with flashing yellow lights, streets with very worn road lines or where the positioning of the road lines shift from one side of the intersection to the other. I can think of a dozen other similar situations so I'm very sure there are many others I've never even encountered.
It remains to be seen how much juice Tesla can squeeze out of a transformer approach to autonomous driving, but it's by no means a sure thing.
edit: I misunderstood the comment. I see now that "practically perfected" is a reference to the ideal conditions mentioned in your first sentence.
Thank you to post your personal experience. One pattern that I have noticed about Tesla FSD stories on HN: it is black and white, with little grey area. Either people live an area where the roads are easy for FSD to navigate, and they come here to say "It is perfect!". And vice versa: People live in a place with a bunch of complex roads and intersections where FSD does not perform well... And they come to HN to share their experience. (Privately, I cannot wait to see self-driving models try driving on Jakarta or Napoli. It will so much fun to watch those on YouTube!) I do think that Elon/Tesla is taking a crazy gamble to release FSD early to gain billions of hours of training data! I can understand where this would make some people uncomfortable, due to the safety concerns. Except Waymo, who else has the training data that Tesla has at this point? Wiki tells me that FSD has been driving on public roads since 2016. That must an astonishing amount of training data accumulated in the last 8 years! I assume he will be tasking X.ai with improving the FSD model using this enormous training data.
This is so absurd. If they could accurately predict when their system is so reliable, they could get it certified for Level 3 use in these conditions. That's not a gimped version, that's strictly better than the product they offer.
They can't though. Mercedes-Benz can. To pretend that Tesla doesn't offer a better product because they don't want to is...
There are plenty of FSD videos on YouTube where the driver didn't have to intervene at all on long trips, where the car isn't just following the car in front.
You hit it on the nail, the driver didn't have to intervene but they had to be ready to take over at any time. If FSD could safely operate where you had 10 seconds to take over for level 3, why doesn't Tesla go get it certified for L3 driving. FSD isn't even approved for being a drivers aid(hands off, eyes on) in EU and China because Tesla has not demonstrated that it is safe. Even Blue cruise is approved in EU for comparison. And it isn't because Tesla isn't trying to get ADAS certification in these markets.
Are they? they haven't start certification for self driving any market? It seems FSD is more of way to pump the stock. Mercedes is testing their same cars for L4(point to point) in china.
For past decade, with meh results consistently so far. They were/are? selling FSD as premium package and never delivered on promises, that's outright fraud in plain sight.
Technologically Tesla is far behind since one man's ego wants to trump physics and computing limitations and its failing. Still no Lidar.
I don't care about 90 youtube videos of long drives with no intervention. I care about a million out of a million (and maybe 10 on top). Either there is something I can trust with mine and my kids life with, or I am not interested with anything on top adaptive cruise control (since I don't spend my life in stop&go traffic).
