Firmware 6.0

[InTheShadows]

Ugh. I read 9 pages of this thread and then tried a search but didn't see this come up: the first detent on the turn signal stalk would cause the turn signal to blink 3 times to signal a lane change. Yesterday my car was in to repair a seized window regulator, new tires, underbody shield, blah blah. The loaner had 6.0, and I was annoyed to find that the stalk didn't function the same way. No lane change 3 blinks at first detent. I figured it was just that car. Well, they informed me when I picked up my car that they had loaded 6.0 for me and guess what? Now my car does the same thing. I love the other updates but am super annoyed with that stalk behavior. Am I the only one who used it???

Is it a setting somewhere under driving or vehicle or profiles?

 

[Btrflyl8e]

EXACTLY. Maybe it was unintended... it's definitely not working like it did. I'll really pay attention this morning and see if it's just touchier, or I'm crazy... but it was no longer functioning in either car for me yesterday. I used it ALL the time.

 

[chickensevil]

Ugh. I read 9 pages of this thread and then tried a search but didn't see this come up: the first detent on the turn signal stalk would cause the turn signal to blink 3 times to signal a lane change. Yesterday my car was in to repair a seized window regulator, new tires, underbody shield, blah blah. The loaner had 6.0, and I was annoyed to find that the stalk didn't function the same way. No lane change 3 blinks at first detent. I figured it was just that car. Well, they informed me when I picked up my car that they had loaded 6.0 for me and guess what? Now my car does the same thing. I love the other updates but am super annoyed with that stalk behavior. Am I the only one who used it???

No, and that sucks because I use that all the time myself... I guess we are the only ones who use our turn signals for lane changes

 

[Btrflyl8e]

Update, something is weird. I just walked out to my car and tried it, and it works. So now I will have to pay close attention when I'm driving to figure out exactly how it's different now. There's a definite difference.

 

[chickensevil]

Yep, I was expecting someone to post something of that sort. There are hundreds of millions of iPhones out there and only ~50,000 MSs so the probability of an event occurring is far, far higher.

There were like, 2 cars that reported the issue. That isn't a widespread thing. And until you get a large enough test group you won't get enough variance to avoid this. I think this is likely one of the biggest reasons for the staggered updates. The early updaters are a secondary pilot/test group. I know in my workplace we push out updates in three phases every time, and it of course doesn't catch everything. And we are talking about updates that are vendor supplied to patch systems (like windows updates) and yet we STILL have updates that break stuff all the time, even with a pilot group and a limited release. It really just depends on what you are changing and what kind of bug it is.

 

[AlMc]

Update, something is weird. I just walked out to my car and tried it, and it works. So now I will have to pay close attention when I'm driving to figure out exactly how it's different now. There's a definite difference.

I have received each if the new recent 6.0...now on (1.65.15) and have never lost that function.

 

[tomas]

When I have problem with my computer, I'm usually told to unplug all peripherals and reboot. Same with router and VoIP box, which go goofy all the time, gotta reboot in sequence. iphone really has no peripherals continuously attached. No boot sequence. MS has hundreds of peripheral sensors and devices as well as network of processors, and end user cannot unplug or control boot sequence. It is not an apt comparison.

 

[chickensevil]

When I have problem with my computer, I'm usually told to unplug all peripherals and reboot. Same with router and VoIP box, which go goofy all the time, gotta reboot in sequence. iphone really has no peripherals continuously attached. No boot sequence. MS has hundreds of peripheral sensors and devices as well as network of processors, and end user cannot unplug or control boot sequence. It is not an apt comparison.

Never mind that you car has like 50+ separate CPUs in it all doing very specific code and tasks... from your seatbelt actuators to the battery CPUs to the main display... So you really have no way of knowing which code for which CPU is the one that is messed up and causing the whole system to go haywire.

I like my comparison of implementing a patch in enterprise environment because it actually is quite fitting. You make one change that affects some 15,000 systems... and even though all your systems are *supposed* to be on the same baseline, you still get issues. For all we know the people having trouble were because they were coming from a specific version and the update process messed up their code... who knows? The point is that there is a lot going on under the hood of an update and even a couple hundred cars in a beta would not be enough of a test group to ensure all bugs were found.

 

[bollar]

There were like, 2 cars that reported the issue. That isn't a widespread thing. And until you get a large enough test group you won't get enough variance to avoid this. I think this is likely one of the biggest reasons for the staggered updates. The early updaters are a secondary pilot/test group.

Really speaks to why an early release opt-in would be a good idea. No doubt someone In that group would be thrilled to report their car wouldn't start and wouldn't mind the potential inconvenience.

Maybe next time...

 

[zwede]

We knew about 1.65.11 and 1.65.13, so this is the 3rd version of 6.0. I'm on 1.65.13, will be interesting to see if whatever bugs they fixed in 1.65.15 were serious enough to update everyone on earlier builds or not.

 

[chickensevil]

Really speaks to why an early release opt-in would be a good idea. No doubt someone In that group would be thrilled to report their car wouldn't start and wouldn't mind the potential inconvenience.

Maybe next time...

Yeah, they can keep it in a "closed" beta that they have, and then have a pilot opt-in group for testing new version rollouts, let that number go up to like 3-5k voluntary participation. And if they get enough of a response then they can make multiple pilot groups so they can continue their staggered rollout... So, hopefully, by the time it hits someone who didn't volunteer for a pilot group all the major issues will be worked out.

- - - Updated - - -

So on that note, time to email ownership about that, haha!

 

[SeminoleFSU]

Ugh. I read 9 pages of this thread and then tried a search but didn't see this come up: the first detent on the turn signal stalk would cause the turn signal to blink 3 times to signal a lane change. Yesterday my car was in to repair a seized window regulator, new tires, underbody shield, blah blah. The loaner had 6.0, and I was annoyed to find that the stalk didn't function the same way. No lane change 3 blinks at first detent. I figured it was just that car. Well, they informed me when I picked up my car that they had loaded 6.0 for me and guess what? Now my car does the same thing. I love the other updates but am super annoyed with that stalk behavior. Am I the only one who used it???

oh no! I use that dozens of times a day- in fact I use that way more than putting turn signal all the way past the detent.

 

[chickensevil]

So on that note, time to email ownership about that, haha!

So I sent them a pretty lengthy letter explaining this suggestion and why it is good for all sides. For anyone who cares, here is what I sent:

Good day,

I wanted to make a recommendation that is based on a few assumptions so if you are not doing updates at all like this then it would be nice to know a bit more clarity on how that works, but anyway, here goes.


We (I am referring the to collective owners group here) are aware that you *can* rollout updates immediately to all users if you choose, as we saw this happen when the change was made to temporarily disable the auto-lower in the suspension last fall.


We are also aware that you know what version everyone actually has, because there is a lot of evidence to suggest selective update pushes to people with specific firmware. For example right now with the version 6 > 1.65.13 to 1.65.15 update everyone who is already on version 6 seems to be getting this update before you proceed with new owners coming up from version 5.x which we all suspect was initiated by the potential for cars to not turn on after getting the update.


Finally it seems that you do your updates in a staggered method, picking seemingly random users, in order to slowly penetrate the entire ecosystem allowing you to halt an update should a major issue be discovered. Basically it takes doing a pilot group to a whole new level.


So my recommendation:


If you will not increase the number of people in your private "closed" Beta, at the very least make a voluntary "pilot" group. My impression of the community is there is three camps of people. Those who want every new update as soon as they can get their hands on it, those who just want everything to work and don't care about when they get the update, and the people who just don't care at all either way. So use that first group to your advantage.


Make a voluntary opt-in pilot group for people who will always be the first to get a public release of a firmware update to act as your secondary "beta" group. Put a disclaimer on the sign up for this that you accept that you might get software which is not without bugs, etc... and then use that as your initial rollout group (or groups).


From my understanding you do releases in groups of about 2k each, if you want to keep this at this pace, even if you get more than 2,000 people to sign up you can just front load this list of cars. You can make the sign up for this in "My Tesla", so you will have an easy interface for people to sign up and can offload that into a database backend.


Anyway, just an idea, because I see a ton of frustration from both sides, people that are upset because their cars don't work correctly, and people that are upset because they hate waiting for the updates. I think you can make both groups much happier and help avoid people being forced to drive over to the service center to jump the line in the update process, which ends up eating up your service center workers time just for a software update (which I totally think is awesome that they will do this, so don't punish them for that).


Thank you for your time, and sorry for the length of this email.


Happy Owner Since March 2014!

 

[Afdyce]

11th post on this thread.

Doh! Thanks iffatall, I missed it! Stupid me, I guess the other 50 pages are just part of my life I won't be able to get back.
Stopped smoking and gone back to the nicotine patch. Just making a soothing cup of British Tea and settling down for a read.

By the by, do you know why communists drink herbal tea? Because proper tea is theft.

Thanks again for your help.

 

[Alligator]

Some of us are still waiting on 6.

 

[bollar]

Update, something is weird. I just walked out to my car and tried it, and it works. So now I will have to pay close attention when I'm driving to figure out exactly how it's different now. There's a definite difference.

Worked fine for me on my drive this morning. However, I also had some non-reproducible problems earlier, so it doesn't surprise me that your blinkers didn't work temporarily and then started to work.

 

[smartypnz]

Very good, Tesla!!!

Your latest plan to roll out updates every few days is genius!

It will stop us from begging for updates.

Hats off!!

 

[tezzla]

Dilbert strip from '97

 

[AnOutsider]

Worked fine for me on my drive this morning. However, I also had some non-reproducible problems earlier, so it doesn't surprise me that your blinkers didn't work temporarily and then started to work.

I hope so. The nav update isn't enough to make me want to lose that functionality.

 

[yobigd20]

so skimming through the thread I think I can summarize the reported bugs for v6.0:

1. first, some users are reporting getting an error message that the v6.0 install failed. ok, that's bug #1 I guess. not a good way to start out.
2. some people are reporting that the center console keeps freezing/crashing and rebooting and Tesla is informing them that it is their USB thumb drive that is causing the console to crash and to remove the thumb drives.
3. some people are reporting that regen is degraded again. this is reminiscent of one of the very early software updates where many ppl were reporting the same thing where the 'standard/low setting' got corrupted and the regen was stuck on 'low' despite what the setting was on the console. the fix for that was forcing a reboot of the system and toggling the setting to restore full regen. tesla of course never admitted that fault but I experienced it firsthand and can confirm the reboot and toggle switch fixed it. for those experiencing this issue again on v6 I wonder if you can try the same? reboot and toggle back and forth and see if it fixes it? (I'm not on v6 yet and won't bother to upgrade until Tesla releases a final update with all the bugs fixed)
4. 2 reports of the car not being able to be turned on/started after the car shut down. essentially completely disablement of the car and only tesla can start remotely. this is a really critical bug and definitely a showstopper. i'm sure this was one of the quick minor updates they are pushing out after the initial v6 upgrade. if thats the case then hopefully there should be no more reports of this issue.
5. some reports of the climate settings not being remembered and despite being off when the car was turned off, returning to the car later the settings show they are on again.
6. user ppl on fb reporting lots of bluetooth connectivity issues on v6.

it's disheartening to see bugs like this being reported. i was hoping Tesla's software team had matured more over the last year and improved their regression testing and quality control. sad to see that is not the case, especially after how much hype was given to v6 and Elon tweeting that they were taking their time with this release because it 'had to be awesome'.