The ongoing Elon Twitter saga: "insert demographic" melts down

Status
Not open for further replies.
I think sometimes people are just going to whatever he does at this point, all because they don't like what he's tweeted etc. They've not been developing robots for very long AFAIK so it seems a bit silly to dunk on some early iterations because say Boston Dynamics is ahead after 20 years of development.

I'm not sure superficial comparisons to Honda robots from the 90s are particularly relevant.
 
Last edited:
I think people dunk on him for the likes of this robot because it's initial announcement was an utter failure after much hype*, and he's basically trying to cargo cult the Bosten Dynamics thing, but showing them off in bulk doing something quite simple rather than one doing something actually impressive.

At this point getting bipedal walking robot is getting easier and easier to do for the companies that try because many of the major challenges have long since been understood and overcome by the likes of Honda and Boston Dynamics.
IIRC there is/was a competition that is entered by the likes of university engineering students that has them doing more than Telsa has managed, on a much smaller budget.



*Didn't it have to be walked out by some embarrassed looking guys and just stood there?
 
Last edited:
At this point getting bipedal walking robot is getting easier and easier to do for the companies that try because many of the major challenges have long since been understood and overcome by the likes of Honda and Boston Dynamics (IIRC there is a competition that is entered by the likes of university engineering students that has them doing more than Telsa has managed, on a much smaller budget).

But that isn't what they're doing, thus this comes across as quite superficial criticism for the sake of it. It's not particularly relevant that Honda could get a bipedal robot walking in the 90s or whatever and some vague claim re: what university students have achieved seems similarly superficial, I'm pretty sure plenty of the Tesla Engineers were university students too, some of them quite recently, some of them perhaps even entered such competitions.

Boston Dynamics has taken a rather different route and has been iterating its (mostly a control system, AFAIK) approach for a couple of decades now, Tesla, on the other hand, is very into AI/ML and making use of some of the same hardware (FSD chips etc..) that it uses for self-driving, maybe they do crash and burn or maybe they catch up or even overtake, bit early to tell atm.

It's doing more than just walking, it's carrying out some basic tasks etc.. This is a hard problem but I don't see any reason to dismiss it so far and the criticism just seems like knee-jerk reactions and superficial stuff.
 
Last edited:
But that isn't what they're doing, thus this comes across as quite superficial criticism for the sake of it. It's not particularly relevant that Honda could get a bipedal robot walking in the 90s or whatever and some vague claim re: what university students have achieved seems similarly superficial, I'm pretty sure plenty of the Tesla Engineers were university students too, some of them quite recently, some of them perhaps even entered such competitions.

Boston Dynamics has taken a rather different route and has been iterating its (mostly a control system, AFAIK) approach for a couple of decades now, Tesla, on the other hand, is very into AI/ML and making use of some of the same hardware (FSD chips etc..) that it uses for self-driving, maybe they do crash and burn or maybe they catch up or even overtake, bit early to tell atm.

It's doing more than just walking, it's carrying out some basic tasks etc.. This is a hard problem but I don't see any reason to dismiss it so far and the criticism just seems like knee-jerk reactions and superficial stuff.

Just let the bitter Elon weirdos have their fun, the mind is a lot healthier than participate in these sort of threads I find
 
Last edited:
Last edited:
But that isn't what they're doing, thus this comes across as quite superficial criticism for the sake of it. It's not particularly relevant that Honda could get a bipedal robot walking in the 90s or whatever and some vague claim re: what university students have achieved seems similarly superficial, I'm pretty sure plenty of the Tesla Engineers were university students too, some of them quite recently, some of them perhaps even entered such competitions.

Boston Dynamics has taken a rather different route and has been iterating its (mostly a control system, AFAIK) approach for a couple of decades now, Tesla, on the other hand, is very into AI/ML and making use of some of the same hardware (FSD chips etc..) that it uses for self-driving, maybe they do crash and burn or maybe they catch up or even overtake, bit early to tell atm.

It's doing more than just walking, it's carrying out some basic tasks etc.. This is a hard problem but I don't see any reason to dismiss it so far and the criticism just seems like knee-jerk reactions and superficial stuff.
And that is part o the reason I'm very cynical about them.

FSD has been due "by next year" for what, 10+ years now, and it's not getting any closer with Tesla cutting hardware costs on their cars even more to rely more and more on sensors that they already know are horribly flawed for the job under what are in many parts of the world extremely common conditions (fog, light at the wrong angle so it dazzles the cameras etc).

I'm seeing these robots as part of the FSD hype to try and keep people from giving up on the "full self driving forever" vibe.

[edit]
and i'm just realising that my 10 year old self would be looking at me either in disappointment for failing to be excited about walking humanoid robots, or with wonder that they're now so relatively common that they don't raise any excitement in me when I see yet another model moving slowly around in a carefully shot advert.
 
Last edited:
That's partly superficial again though, FSD there w.r.t this robot is just a reference to the name of the chips/hardware and Tesla has made pretty decent advances in self-driving tbh. That they don't have full self-driving yet doesn't really say much about whether an approach using AI and the hardware used with their current self-driving tech is doomed to fail.

I'm not saying believe in any hype about this thing but rather the criticism of it seems to be unwarranted.
 
Last edited:
if you listen to his (very hesitant Biden'esque ?) earning broadcast it's the predicate that real FSD will need a full sensor->wheel neural net hardware and boast that it will then be 10x 'better' than a driver that represents shareholder risk (they already went from their custom chips to nvidia for central hive fsd learning) - he didn't seem to mention their upcoming radar sensor u turn.
also - surpassing the highly choreographed/programmed boston robotics robot maneuvers, if, you have engineering department who can knock up custom actuators, sounds easier as he says.
robot section was a bit like a video from daft punk or kraftwerk though;
the discourtesy of those in audience filming
 
also - surpassing the highly choreographed/programmed boston robotics robot maneuvers, if, you have engineering department who can knock up custom actuators, sounds easier as he says.
robot section was a bit like a video from daft punk or kraftwerk though;
the discourtesy of those in audience filming

Boston Dynamic has humanoid robots that can walk, run, climb stairs, perform flips, balance, get themselves off the floor, and perform tasks. They also have the Spot model, which has mapping capability, performs tasks on demand, navigates challenging terrain, and has been commercially available since 2019.

Musk has robots that can barely waddle across a flat surface for a carefully managed photo op.
 
Last edited:
Boston Dynamic has humanoid robots that can walk, run, climb stairs, perform flips, balance, get themselves off the floor, and perform tasks. They also have the Spot model, which has mapping capability, performs tasks on demand, navigates challenging terrain, and has been commercially available since 2019.

Musk has robots that can barely waddle across a flat surface for a carefully managed photo op.
Probably bought it from Russia (mail order bride included).
 
u
also

Shame Elon has become such a **** of late (not because he unbanned trump roar), as I do enjoy the events he does. I like that he isn't super prepared and salesy with his events, if anything he is a bit stumbly.

He should never have bought twitter, and when purchasing, should never have run twitter, but just stuck to what he does well.
 
Status
Not open for further replies.
Back
Top Bottom