M1tch
2021 M3P with EAP
Wasn’t below 20% and was in park when I came to use it. Wasn’t there again this morning but following a 2 button reset, is this evening.
-
Want to remove ads? Register an account and login to see fewer ads, and become a Supporting Member to remove almost all ads.
2023.12...
- Thread starter spooksman
- Start date
Wasn’t below 20% and was in park when I came to use it. Wasn’t there again this morning but following a 2 button reset, is this evening.
Zilla91
Active Member
No amount of wiping on a dry screen is going to clear it, irrespective of how new the blades are
Wiping is just a beta feature till the new laser is ready so we just have to wait and not complain 
MrBadger
Badger out
Thats what the washers are for.
You don't dry wipe a windscreen. If its dirty and dry, you add water from the squirters.
Lets add ice/frost to your list and the blades are stuck to the screen, or that dust is grit...
I agree. I was merely pointing out that an absence of rain does not equate to a clear screen. The failure in this case is not to use the washers before wiping. The obstruction still needs to be removed from the screen.
Can report that the wiper/autopilot bug is still there! The fix is not a Tesla priority apparently. This is the fix, the one and only fix!
Everytime you start a journey put the wipers onto STOP mode! Never try to put the car into Autopilot or it will put the wipers into action.
It is caused by using the Tesla Vision system but cannot be cured by cleaning cameras or windscreen. Tesla hopes it might be fixed in the next 3 software upgrades but are promising nothing? Sounds like another Elon promise.
Didnt think I should be too inconvenienced on my upcoming EU road trip next month.
Everytime you start a journey put the wipers onto STOP mode! Never try to put the car into Autopilot or it will put the wipers into action.
It is caused by using the Tesla Vision system but cannot be cured by cleaning cameras or windscreen. Tesla hopes it might be fixed in the next 3 software upgrades but are promising nothing? Sounds like another Elon promise.
Didnt think I should be too inconvenienced on my upcoming EU road trip next month.
Zilla91
Active Member
Interesting. Now that it looks like they've admitted defeat maybe they'll start retrofitting those rain sensors everyone else uses.
Oh wait, it's Tesla so they probably won't.
Oh wait, it's Tesla so they probably won't.
I don't see that as them admitting defeat, "Deep Rain" was an old single frame NN, and Elon said that every NN needed to be upgraded to the new video based ones. That is probably what "Autowiper V4" is. Maybe they are running it in shadow mode to see if it is better?
One interesting thing I learnt from reading that Twitter thread is that, for quite some time, Tesla have done rain detection using all the cameras, not just the windscreen-mounted front facing cameras.
Anyway, it sounds like new autowiper code (Green calls it autowiper v4) has dropped in the FSD 11.4.2 release, so will hopefully make it's way to non-FSD cars in the not too distant future. Whether it will be noticeably better that what we have now is anyone's guess, of course....
Anyway, it sounds like new autowiper code (Green calls it autowiper v4) has dropped in the FSD 11.4.2 release, so will hopefully make it's way to non-FSD cars in the not too distant future. Whether it will be noticeably better that what we have now is anyone's guess, of course....
Drew57
Active ember
‘Beta’ mode was fine when Tesla started out and features were new, exciting & not offered by other manufacturers. Many of us said “it’s a software company that builds cars” and willingly forgave shortcomings that we thought (‘promised’) would disappear with the novelty of over-the-air updates. Quality issues annoyed us but were usually forgiven or fixed, Teslas were different and seen as the cutting edge of future motoring.
Now just a few years later it’s the software itself that is dropping Tesla behind others (esp Koreans and Chinese). Manufacturing quality has improved, the paint no longer ‘falls off’, panels, door handles and rear lights are mostly aligned properly etc but software solutions for basic features that everyone expects are either not working properly or are erratic and unreliable.
To still claim Tesla is primarily a ‘software company’ is a fallacy - they build quirky, reasonable quality vehicles and most things relating to OTA updates, battery tech, efficiency and charging are still market leading.
For the rest, they seem to be falling behind many other EV manufacturers. The world is not the USA and it’s questionable whether FSD ‘city streets’ will ever work in the UK. If the removal of various driver aids & reliance on autonomy are part of the overall mission then we can expect to see the removal of more tools for driver intervention.
It may also explain why Tesla puts so much focus on games, media streaming and other trivia - their ultimate goal is for the driver to become the passenger and these things are there to pass the time whilst the car does the rest.
This week I have been able to experience the US FSD beta on a Model X as a passenger and it is actually quite impressive. Whether it can be made to work in our country is questionable and for those of us of advancing years it’s almost certain we will never see it working to any degree of reliability.
Nevertheless we will probably see further automation of driver aids and removal of more hardware as the cars develop. I personally don’t buy into the view that this is just on purely economic grounds for Tesla, it could just as easily be because a true self driving car doesn’t need these things and we the Guinea pigs are road testing it while they try to perfect the technology.
(none of this means I agree with the direction Tesla are taking but it is what it is)
Now just a few years later it’s the software itself that is dropping Tesla behind others (esp Koreans and Chinese). Manufacturing quality has improved, the paint no longer ‘falls off’, panels, door handles and rear lights are mostly aligned properly etc but software solutions for basic features that everyone expects are either not working properly or are erratic and unreliable.
To still claim Tesla is primarily a ‘software company’ is a fallacy - they build quirky, reasonable quality vehicles and most things relating to OTA updates, battery tech, efficiency and charging are still market leading.
For the rest, they seem to be falling behind many other EV manufacturers. The world is not the USA and it’s questionable whether FSD ‘city streets’ will ever work in the UK. If the removal of various driver aids & reliance on autonomy are part of the overall mission then we can expect to see the removal of more tools for driver intervention.
It may also explain why Tesla puts so much focus on games, media streaming and other trivia - their ultimate goal is for the driver to become the passenger and these things are there to pass the time whilst the car does the rest.
This week I have been able to experience the US FSD beta on a Model X as a passenger and it is actually quite impressive. Whether it can be made to work in our country is questionable and for those of us of advancing years it’s almost certain we will never see it working to any degree of reliability.
Nevertheless we will probably see further automation of driver aids and removal of more hardware as the cars develop. I personally don’t buy into the view that this is just on purely economic grounds for Tesla, it could just as easily be because a true self driving car doesn’t need these things and we the Guinea pigs are road testing it while they try to perfect the technology.
(none of this means I agree with the direction Tesla are taking but it is what it is)
WannabeOwner
Well-Known Member
Why do you think it is questionable?
I get that Tesla is a USA company, and all the FSDbeta stuff is happening there, but they have significantly different driving rules state-to-state (and I would be very surprised if the software isn't being built to be "configurable" for various road-furniture and regulatory requirements and so on.
I also think it's questionable (for some time). We returned to the UK last year after 17 years of living in Australia. Previously I lived in the US for a while too. Returning to the UK, you are reminded just how narrow the roads are. By choice, we live in a rural area. Many of the roads are one car width apart with high hedgerows. Tractors are whizzing around. Passing is basically about either one driver reversing to a wider section or both drivers putting the cars half into the hedge. I'm looking forward to seeing automation handle those situations. Then there are towns. People park in the opposite direction of the traffic flow. It is illegal in most countries and certainly in Australia. The police are hard on it, and you will be ticketed or towed. I expect automation will struggle to deal with that for a while. It looks like a head-on collision to the computer until it can determine the other vehicle is parked... badly. Actually, it is illegal in the UK to park that way after dark, but the police couldn't care. I expect many don't even realise it's illegal.
You could argue that they say it only works on Motorways (like Fords) and A-roads, but A) that isn't really FSD now, is it, and B) there are A-roads in the UK that are barely suitable to be classified as B-roads, so the car cannot safely determine based on mapping data.
I could go on, but I think there is logic to Tesla now taking the FSD beta to countries like Australia rather than the UK. They have a long way to go in improving the technology before we get it.
You could argue that they say it only works on Motorways (like Fords) and A-roads, but A) that isn't really FSD now, is it, and B) there are A-roads in the UK that are barely suitable to be classified as B-roads, so the car cannot safely determine based on mapping data.
I could go on, but I think there is logic to Tesla now taking the FSD beta to countries like Australia rather than the UK. They have a long way to go in improving the technology before we get it.
Similar threads
