-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
FSD / AP Rewrite - turning the corner?
- Thread starter MrBadger
- Start date
The chap had 200 followers an hour ago, 1000 now 
https://twitter.com/brandonee916/status/1319126180675481600?s=21
https://twitter.com/brandonee916/status/1319126180675481600?s=21
Last edited:
Just watched the in car videos of the FSD beta, am actually speechless.
Waymo have done it recently in Atlanta so FSD seems possible, but Elon appears to have pulled off the same trick without needing a battery of external sensors and extra CPUs....
All of a sudden the price of FSD seems cheap and Tesla stock seems very under valued!!
Well done Elon, the fact my 2017 X will get the same features post HW 2.0 and MCU 2.0 retrofit is frankly even more crazy!!
Waymo have done it recently in Atlanta so FSD seems possible, but Elon appears to have pulled off the same trick without needing a battery of external sensors and extra CPUs....
All of a sudden the price of FSD seems cheap and Tesla stock seems very under valued!!
Well done Elon, the fact my 2017 X will get the same features post HW 2.0 and MCU 2.0 retrofit is frankly even more crazy!!
Just watched more videos, Tesla appear to have nearly done it!!
Am very tempted to order a Model 3 for my wife with FSD. A car that can truly drive it self will change the world.....or maybe just buy £60k worth of Tesla shares now.
Am very tempted to order a Model 3 for my wife with FSD. A car that can truly drive it self will change the world.....or maybe just buy £60k worth of Tesla shares now.
Theres another reason.. the hardware in the car was up to date so the software changes could be applied. Once you’re off the latest hardware some of the features can’t be applied. Take one foot braking.. only those cars with a perm magnet motor git it, you may have missed that feature by only a month.. And that’s the point I’ve made, unless you’re on HW3, the AP performance will almost certainly now plateau compared to cars with it.
Apparently he had ~50 before this all started. Guess he must’ve been a particularly safe driver.The chap had 200 followers an hour ago, 1000 now
https://twitter.com/brandonee916/status/1319126180675481600?s=21
squarepeg77
Member
Occasionally in life, I see or hear something that sends a shiver up my spine [in awe rather than horror]. This was one of them.
https://twitter.com/brandonee916/status/1319126180675481600
I personally think it has 'turned the corner'.
https://twitter.com/brandonee916/status/1319126180675481600
I personally think it has 'turned the corner'.
Last edited:
goRt
Active Member
Quite agree. I've been as critical as anyone when it comes to Elon and his FSD claims. But boy was I wrong, though to be honest I really shouldn't have doubted him, after all he did do this.....
I think I've missed the boat with Tesla stock though, I've only just sent in my W-8 form which takes 24hrs to process, TSLA stock price will be through the roof by this afternoon.
Oh well I just have to console my self that I have FSD and free unlimted supercharging for as long as I own the car, which may now be almost forever!!!
squarepeg77
Member
imho I think Tesla share price has significant future promise of success baked into it. So the risk is on something not working, or Musk being knocked down by a Robotaxi/Lost in Space.
Glan gluaisne
Active Member
As a tactic to bolster sales/share prices I think leaking videos of the new FSD possible performance is a really good one. Despite all the hype and optimism, I'm extremely sceptical about FSD working well enough to be useful here in the UK for some time yet.
We've all seen just how poorly the car's systems handle UK weather and roads, with some shortcomings that seem to be related primarily to the sensors, rather than the software, so unlikely to be fixed by any new software re-write. If the side cameras still stop working at night, in heavy rain, when the sun's low, then software isn't going to fix that. We're all used to there being loads of bugs in Tesla software, too, so what's the probability that they've somehow miraculously managed to produce something that's sufficiently bug-free that it can be relied on?
I'd like to see FSD work, but a few carefully edited video clips on Twitter just doesn't cut it for me, I'm afraid. If FSD really has been perfected now, then why aren't we seeing a major press release from Tesla?
We've all seen just how poorly the car's systems handle UK weather and roads, with some shortcomings that seem to be related primarily to the sensors, rather than the software, so unlikely to be fixed by any new software re-write. If the side cameras still stop working at night, in heavy rain, when the sun's low, then software isn't going to fix that. We're all used to there being loads of bugs in Tesla software, too, so what's the probability that they've somehow miraculously managed to produce something that's sufficiently bug-free that it can be relied on?
I'd like to see FSD work, but a few carefully edited video clips on Twitter just doesn't cut it for me, I'm afraid. If FSD really has been perfected now, then why aren't we seeing a major press release from Tesla?
Well if Tesla makes FSD 'transferable' between cars, am laughing
Though about getting a Model 3 with FSD now, but actually than realised why would I bother??
The number of times we need to use both cars at the same time is virtually zero these days as commuting to work on my pedal bike, and if there is an option to 'summon' the Tesla to me if I needed it or my wife needed it whats the point of having a second car??
For example I can cycle to work, my wife drive to work, if I than decided I needed to drive home that day I can 'summon' the car to me, drive home than send the car back to my wifes work.
Given the battery in the X is replaceable as it ages, it may become the final car we actually ever buy!!
Because its not, but its getting there with a massive step fowards. Its also not just Tesla Waymo is now offering Robotaxis as a paid service to the public in the US. So if you have good enough software it is already possible! What Tesla is doing is making it work without the need for expensive hardware like Lidar.
Glan gluaisne
Active Member
Yes, it's a step forward, but arguably Waymo are much further along that path. LIDAR isn't that expensive now, either. I bought a 360 deg scanning LIDAR module to play with earlier this year - it cost less than £90, and has a 5 kHz sampling rate, with a spatial resolution of between 0.5mm and 50mm, but it's limited in range to about 11m. A longer range (60m to 100m) forward view LIDAR retails for around £500, so in production quantities the wholesale price would be a lot less than this, perhaps down around £100 or so.
More importantly imho, without the need for HD maps (a personal dislike of mine) and restricted geo locations. The Tesla approach is a generic solution rather than highly restricted/controlled solution. I'm not saying that Tesla will successfully drive anywhere when it wants, but the Tesla approach will have a stab at it with widely differing levels of success. I am expecting geofencing/operational domain restrictions to come later, but only to define L4 compliant zones/parameters such as road types or weather restrictions. Still some way from that though and what we have now is still very much L2.
Similar threads
