Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register
This site may earn commission on affiliate links.
Well what? AFAIK he didn't provide a due date on that specific item did he?

It DOES recognize many objects in the road... one most relevant example is speed bumps which it currently gets right most of the time (and increasingly so with version revs). It does not yet do potholes but there's no reason to think it can't or won't in the future.
True. I’m thinking pothole recognition will occur just before “nyc to la with no human intervention by end of 2019” occurs
 
Could be part of the rendering upgrades Elon just mentioned (and was cited about 5 posts ago here)
Funny that effort is being put into changing the FSD visualizations when FSD beta will give you a strike for using them. Other than displaying obstacles in your blind spot and close-in clearance values, the visualizations are just eye candy that give you confidence that the car does interpret the scene around itself.

Maybe it would be more productive to do things like ensure that all audible warnings are accompanied by warning text and logged notifications, and that warning text size is made larger so that you read it at a quick glance.
 
Funny that effort is being put into changing the FSD visualizations when FSD beta will give you a strike for using them. Other than displaying obstacles in your blind spot and close-in clearance values, the visualizations are just eye candy that give you confidence that the car does interpret the scene around itself.

Maybe it would be more productive to do things like ensure that all audible warnings are accompanied by warning text and logged notifications, and that warning text size is made larger so that you read it at a quick glance.


Have you provided that feedback to the FSDBeta team?
 
Do you have any evidence the email address specifically provided to the beta testers to send that feedback is not monitored?
Proof of a negative is very difficult. The best one can often do is try to disprove it by finding evidence that it is false. Since I do not work for Tesla, the best I could do is look for some indication that an emailed report resulted in a response to the sender. Thus far, I have seen no indication of that on this forum, despite a very active group of FSD beta testers.

So, have you received a response from an FSD report? Do you know of anyone who has?
 
Proof of a negative is very difficult. The best one can often do is try to disprove it by finding evidence that it is false. Since I do not work for Tesla, the best I could do is look for some indication that an emailed report resulted in a response to the sender. Thus far, I have seen no indication of that on this forum, despite a very active group of FSD beta testers.

So, have you received a response from an FSD report? Do you know of anyone who has?
why would they respond? just to make you feel better? lol seems like a waste of time. User input is almost always useless when you have the technical details at hand. thus the reason bug reports (outside of tesla) come with an optional comment box.

doesn't really seem like much of a basis for saying it's unmonitored either honestly, not a lot of substance there. shouldn't be surprised tho, we do so love our personal/anecdotal "facts" around here ;)
 
  • Like
Reactions: Yelobird
why would they respond? just to make you feel better? lol seems like a waste of time. User input is almost always useless when you have the technical details at hand. thus the reason bug reports (outside of tesla) come with an optional comment box.

doesn't really seem like much of a basis for saying it's unmonitored either honestly, not a lot of substance there. shouldn't be surprised tho, we do so love our personal/anecdotal "facts" around here ;)

I'm a strong advocate of having a feedback loop for bugs, feature requests, concerns, etc.

It makes the customer feel like their part of team, and that they're contributing to something. It also increases the attachment rate where they'll get your product over a competitors.

It also helps to keep bugs in check whether its a new bug or a long term bug.

Tesla doesn't really have any bug reporting mechanism with a feedback loop outside setting up a service appointing, and chatting with them. I'm currently engaged with them on some weirdness with the trip planner and how it interacts with the preconditioning. Something that wouldn't have taken place if it was just reporting into the ether.

Having someone to interact with encourages me to spend more time, and energy gathering data to show what's going on. It's pretty common for me to find issues with something I buy, and to work with the manufacture on fixing it.

Tesla I find extremely frustrating because the bug lists just keeps growing.

You say "user input is almost always useless when you have the technical details at hands", but the topic is FSD beta. With FSD beta the biggest issue by far are issues with navigation and maps. These are things people have to report because they don't come across in the auto report. Tesla isn't going to be able to read my mind as to why I pushed the report button.

I would use FSD beta a LOT more if some regional representative was engaging with me in some manner.

Over time I expect FSD subscriptions to account for more, and more of the fleet numbers and with subscriptions you really want to engage with the customer or they'll just cancel.

By far the lack of customer engagement is the second worst thing about the FSD beta. The first worst thing is having people do the safety score before getting invited into the beta. So they have to sit there with a high safety score for months while waiting.
 
why would they respond? just to make you feel better? lol seems like a waste of time. User input is almost always useless when you have the technical details at hand. thus the reason bug reports (outside of tesla) come with an optional comment box.

doesn't really seem like much of a basis for saying it's unmonitored either honestly, not a lot of substance there. shouldn't be surprised tho, we do so love our personal/anecdotal "facts" around here ;)
I presented a falsifiable hypothesis. All that's needed is a single instance of Tesla responding to an email sent to the FSD beta account and my hypothesis will fall. Until then, send all the emails you like to the FSD beta account if it makes you feel better. I'll stick to reporting things here, where I know that someone will see them, even if it's not Tesla.
 
Proof of a negative is very difficult. The best one can often do is try to disprove it by finding evidence that it is false. Since I do not work for Tesla, the best I could do is look for some indication that an emailed report resulted in a response to the sender. Thus far, I have seen no indication of that on this forum, despite a very active group of FSD beta testers.

So, have you received a response from an FSD report? Do you know of anyone who has?

Tesla ownership is so faith based that we should have weekly congregation.

Praying to Elon is mandatory.

Confessions go to FSD Beta.
 
  • Like
  • Funny
Reactions: mswlogo and Matias
Well what? AFAIK he didn't provide a due date on that specific item did he?

It DOES recognize many objects in the road... one most relevant example is speed bumps which it currently gets right most of the time (and increasingly so with version revs). It does not yet do potholes but there's no reason to think it can't or won't in the future.
Humans will fly to outer solar system.
 
  • Like
Reactions: 2101Guy
I'm a strong advocate of having a feedback loop for bugs, feature requests, concerns, etc.

It makes the customer feel like their part of team, and that they're contributing to something. It also increases the attachment rate where they'll get your product over a competitors.

It also helps to keep bugs in check whether its a new bug or a long term bug.

Tesla doesn't really have any bug reporting mechanism with a feedback loop outside setting up a service appointing, and chatting with them. I'm currently engaged with them on some weirdness with the trip planner and how it interacts with the preconditioning. Something that wouldn't have taken place if it was just reporting into the ether.

Having someone to interact with encourages me to spend more time, and energy gathering data to show what's going on. It's pretty common for me to find issues with something I buy, and to work with the manufacture on fixing it.

Tesla I find extremely frustrating because the bug lists just keeps growing.

You say "user input is almost always useless when you have the technical details at hands", but the topic is FSD beta. With FSD beta the biggest issue by far are issues with navigation and maps. These are things people have to report because they don't come across in the auto report. Tesla isn't going to be able to read my mind as to why I pushed the report button.

I would use FSD beta a LOT more if some regional representative was engaging with me in some manner.

Over time I expect FSD subscriptions to account for more, and more of the fleet numbers and with subscriptions you really want to engage with the customer or they'll just cancel.

By far the lack of customer engagement is the second worst thing about the FSD beta. The first worst thing is having people do the safety score before getting invited into the beta. So they have to sit there with a high safety score for months while waiting.
ok this is all fine, but its just your opinion, your not presenting any facts about why beta email is unmonitored. nor are you presenting any real case as to why they should respond to you, just that you think they should because it would make your experience better, thats a little self absorbed, to think a trillion dollar company should drop everything and do things your way, simply because it's how you want it.

and yes, as someone who's spent their entire life in tech... user input, users in general, are the enemy. dumb people can't figure things out so they blame the system that they are using improperly (my techs spend half their day educating users). this holds across every sector of the consumer tech world. feedback from unskilled users is more than useless, it can cause you to waste valuable time and effort on essentially nothing (like your assertion that navigation and maps are the "biggest" issue, exactly how do you know that? i'm gonna go out on a limb and say anecdotal evidence).

for the record, i agree they could communicate more (tho not in the way you want), its just sad to see so many people on here take the position that they somehow know better than literally the people writing the book... sad and laughable.

this has gone off the rails for this thread so i'll avoid responding any further here but i'll leave you with this... no manufacturer of anything is looking for unsolicited "advice" about anything. tesla gives us the tools they want us to use, maybe people should try using them. advanced automation can feel like magic, but rest assured its not ;)
 
I presented a falsifiable hypothesis. All that's needed is a single instance of Tesla responding to an email sent to the FSD beta account and my hypothesis will fall. Until then, send all the emails you like to the FSD beta account if it makes you feel better. I'll stick to reporting things here, where I know that someone will see them, even if it's not Tesla.
just FWIW, a hypothesis is not a fact unless its proven correct, not the other way around.
 
just FWIW, a hypothesis is not a fact unless its proven correct, not the other way around.
You’re not wrong, but a proposition is not even a hypothesis unless it’s falsifiable.

And there are more sophisticated ways to make choices about how to live in this universe than waiting for absolute proof. Military history is replete with commanders who lost opportunities because they waited for perfect knowledge. Bayesian analysis is an incredibly powerful tool to quantify and deal with uncertainty.

Given how OCD I am about reading documentation, and that I haven’t seen anything in the release notes or documentation from Tesla about an FSD Beta-specific e-mail address, they sure aren’t emphasizing such a channel for feedback. That’s evidence—not proof, but evidence—that e-mail feedback is not valued by the Tesla team.

Given how inexpensive it is to collect video clips and telemetry in the year 2022, and that such objective data is not filtered through the subjective experience of untutored and untrained drivers, it is not unreasonable to assess that video/telemetry data collection and curation efforts would have a higher return on investment than reading e-mail. Each disconnect event would be more objective, data rich, and easy to automatically correlate with similar events in space, time, and other dimensions. Again, not proof, but a potential insight that can be factored into one’s Bayesian priors.

So based on my assessment of the situation, I’m not going to waste my time composing e-mail. Instead, I’m going to override FSD Beta when it does something I don’t like, causing a disconnect and probably triggering the collection and reporting of video and telemetry about that disconnect.
 
Last edited:
You’re not wrong, but a proposition is not even a hypothesis unless it’s falsifiable.

And there are more sophisticated ways to make choices about how to live in this universe than waiting for absolute proof. Military history is replete with commanders who lost opportunities because they waited for perfect knowledge. Bayesian analysis is an incredibly powerful tool to quantify and deal with uncertainty.

Given how OCD I am about reading documentation, and that I haven’t seen anything in the release notes or documentation from Tesla about an FSD Beta-specific e-mail address, they sure aren’t emphasizing such a channel for feedback. That’s evidence—not proof, but evidence—that e-mail feedback is not valued by the Tesla team.

Given how inexpensive it is to collect video clips and telemetry in the year 2022, and that such objective data is not filtered through the subjective experience of untutored and untrained drivers, it is not unreasonable to assess that video/telemetry data collection and curation efforts would have a higher return on investment than reading e-mail. Each disconnect event would be more objective, data rich, and easy to automatically correlate with similar events in space, time, and other dimensions. Again, not proof, but a potential insight that can be factored into one’s Bayesian priors.

So based on my assessment of the situation, I’m not going to waste my time composing e-mail. Instead, I’m going to override FSD Beta when it does something I don’t like, causing a disconnect and probably triggering the collection and reporting of video and telemetry about that disconnect.
i actually completely agree with you lol... the email is likely primarily used for parsing statistical data from reports as a fail safe for a failure of their automated data capture processes (or is simply legacy lol), my point being that to assume its unused is unwarranted, espically if your sole criteria is a lack of a response to you personally.

but again you are right that their primary feedback clearly comes from the data on the cars themselves not any notes submitted by users, which is why they largely ignore user complaints, and are right to do so lol.