Firmware 6.2 poll to say when you got it

[dirkhh]

So once again over the weekend the roll-out has pretty much come to a stand still - or the reporting of it at least has
Are people getting updates? And which version? If I look at Sat-Mon the picture is confusing: .124 .153 .160 .168 and 101.36.2 apparently were all sent to different owners.
Five versions shipped to owners at the same time? With all due respect to Tesla, but that's just nuts.

 

[ecarfan]

Dirk, with respect, I do not think you can draw that conclusion based on posts to this thread. Your sample size of owners is too small, and it was not randomly selected. The data collected here would be best described as anecdotal in nature. This is not a "poll" in the scientific sense. If you want a poll of owners, one approach might be to spend a week or more standing at the Harris Ranch Supercharger and questioning owners about their firmware version while noting their VIN and battery type. Of course, no one wants to do that...

 

[dirkhh]

Dirk, with respect, I do not think you can draw that conclusion based on posts to this thread. Your sample size of owners is too small, and it was not randomly selected. The data collected here would be best described as anecdotal in nature. This is not a "poll" in the scientific sense. If you want a poll of owners, one approach might be to spend a week or more standing at the Harris Ranch Supercharger and questioning owners about their firmware version while noting their VIN and battery type. Of course, no one wants to do that... 

ecarfan - the phrase "with respect" (or "with all due respect") usually means the opposite. So noted and accepted. I should mention that I have a degree in math and roughly understand what it takes to make reliable statements based on samples...
We have at this point 240 cars in the tool out of an installed base of around 60k. if they were randomly selected that would give us a useful statistical sample size - they are of course not.
So did you see me state "24% of the cars have received version such-and-such"? No you didn't.
But there are a few things that even an non-random and somewhat on the smaller side sample can give you:
a) and idea about the intensity - we can clearly see that they push upgrades for a day or three and then slow down
b) the existence of certain releases and the fact that they are currently being pushed out
And that's all I commented about. We see that once again things slowed down over the weekend, and we see that five different releases were reported in the tool as having been pushed out.

So with all due respect to you (see the beginning of this post), I believe I do understand what I can and cannot know based on the data that we have available in the Firmware Upgrade Tracker.

 

[HankLloydRight]

124 .153 .160 .168 and 101.36.2 apparently were all sent to different owners..

Right. And even not using statistics, let's assume the 124 and 153 updates are user errors, tesla is still pushing multiple versions out at the same time. And no AP cars are getting the 101. build, which is also kinda odd for reasons we don't know yet.

 

[Andyw2100]

There are still just five P85Ds with .168 in the tracker--my car and four others. I had "missed" a couple of updates, and went straight from .124 to .168. I'm wondering if the other four P85Ds also went straight from .124 to .168, and if so, if that tells us anything interesting.

Edit: Checking the multiple update chart, it looks like just one of the other car owners had entered information before, and actually that car had gone from .124 to .153 to .168.

Also, as an aside, the multiple update chart currently has 51 entries (and it's increasing, of course) but is limited to "showing" 5, 10, or 50 entries. I don't believe there is a way to scroll, so it may be time to add the "100" option there.

 

[ecarfan]

Dirk thanks for your reply and I truly did not mean to offend. I'm glad to hear you understand the limitations of the data being derived from this thread.

 

[preilly44]

My P85D was in for service (getting the NG seats installed) last Friday/Saturday. I had .153 installed when they picked it up and I asked them to update the firmware. I got an invoice on Friday saying the firmware was up to date. I emailed them back saying that .160, .168 and possibly another version was out. On Saturday morning when they dropped off the car it had .160 on it. I was hoping they pushed out .168 for later that night or the next day but I still haven't received it.

 

[tomas]

So once again over the weekend the roll-out has pretty much come to a stand still - or the reporting of it at least has
Are people getting updates? And which version? If I look at Sat-Mon the picture is confusing: .124 .153 .160 .168 and 101.36.2 apparently were all sent to different owners.
Five versions shipped to owners at the same time? With all due respect to Tesla, but that's just nuts.

1) I know there has long been conjecture on release threads that Tesla takes weekends off from pushing releases, but I really think that's an "insufficient data" issue. We know pretty certainly several things: a) by the time the update notice shows, the release has already been downloaded. b) download can take several days, especially if there is limited connectivity. Even if Tesla somehow stopped pushing on weekends, that wouldn't stop pushes initiated on, say, Thursday from popping the update notice on Saturday or Sunday. Though possibly if vast majority of people reporting are regularly connected on wifi, that could constitute evidence...

2) Keep in mind regarding the multiple release pushes at same time: a) once a release starts being pushed to a car, it has to complete and install before the next one can be pushed. So, except probably in Service Center, Tesla cannot override a "yet to complete" update with a newer version. b) as Tesla has shipped so many different configurations, it is clear that they have different software builds to address different things. We've already seen builds that are specifically for D, builds that are specifically for non-english capabilities, builds that are specifically for auto pilot cars... not to mention the ones we have not sleuthed out. So, while we can debate their software deployment approach, I'm not sure it is right to call it "nuts" without being privy to more details of how they handle code versioning, testing, etc.

 

[apacheguy]

1) I know there has long been conjecture on release threads that Tesla takes weekends off from pushing releases, but I really think that's an "insufficient data" issue. We know pretty certainly several things: a) by the time the update notice shows, the release has already been downloaded. b) download can take several days, especially if there is limited connectivity. Even if Tesla somehow stopped pushing on weekends, that wouldn't stop pushes initiated on, say, Thursday from popping the update notice on Saturday or Sunday. Though possibly if vast majority of people reporting are regularly connected on wifi, that could constitute evidence...

FWIW, I've never seen my car wake up for several days to download an update. From my telemetry data, the longest it has ever taken is a few hours. And that's even with weak WiFi signal strength in the garage.

 

[tomas]

FWIW, I've never seen my car wake up for several days to download an update. From my telemetry data, the longest it has ever taken is a few hours. And that's even with weak WiFi signal strength in the garage.

Agree with WiFi should not take long. I was referring to 3G connectivity. Also, even with WiFi, not sure how downloads work if you do not have "always connected" on. Does it wake up to receive, or does it wait until otherwise woken up. Not that it matters that much to me... so why the heck did I even bring it up?

Fascinating to watch this thread and the tracker take curiosity and speculation about the software update process to a new extreme.

 

[Andyw2100]

1) I know there has long been conjecture on release threads that Tesla takes weekends off from pushing releases, but I really think that's an "insufficient data" issue.

Tesla may usually take weekends off from pushing releases, but there was at least one in the last couple of months that they started pushing (or to be more accurate, people started receiving install notifications for) in the late morning or early afternoon on a Saturday or Sunday. I remember having not checked TMC for a few hours and being surprised at all the people talking about the new version and then going out to check the car and finding it waiting for me.

I'll see if I can dig up the posts, to be able to reference the version and link to some of the posts and then come back and edit this post.

Edit: Found it, or one of them, anyway. It was an update (6.1--2.2.167) that started being pushed on Saturday, 02/21: Firmware 6.1 - Page 134

Just one example that at least some of the time Tesla will start a firmware update on a weekend.

 

[MarcG]

FYI, been on .153 here for over a week and no updates since. I've been on both 3G and WiFi with decent signal, but still no .168 for me yet.

Does anyone know for a fact if the 3.2 --> 3.1 second acceleration and throttle improvement were:

A) only added as part .168, since it shows that in the release notes for .168 only, OR

B) included in any (or all) previous 6.2 updates, but only shown in the release notes for .168?

I personally have not noticed an improvement as of .153

 

[apacheguy]

Agree with WiFi should not take long. I was referring to 3G connectivity. Also, even with WiFi, not sure how downloads work if you do not have "always connected" on. Does it wake up to receive, or does it wait until otherwise woken up.

No, Tesla sends the wake up signal to the car.

FYI, been on .153 here for over a week and no updates since. I've been on both 3G and WiFi with decent signal, but still no .168 for me yet.

Same. Still on 153

 

[MarcG]

Well, it looks like I should post here more often so Tesla can read the forums and push updates to my car when I feel I'm falling behind :biggrin:

Will post again once it's done and I know what version I end up on (and yes Dirk, I'll update the tracker don't worry!)

 

[MarcG]

Yep, it's 2.4.168 indeed! Updated the tracker accordingly.

 

[brianman]

Five versions shipped to owners at the same time? With all due respect to Tesla, but that's just nuts.

Disagree. That's small potatoes.

 

[Paul Carter]

This is my favorite thread ever.

 

[Mayhemm]

Yep, it's 2.4.168 indeed! Updated the tracker accordingly.

View attachment 78884

I missed a few pages somewhere, so is .168 the first version with faster acceleration (insaner insane mode), or has that been out for a while?

 

[MarcG]

I missed a few pages somewhere, so is .168 the first version with faster acceleration (insaner insane mode), or has that been out for a while?

I'm not sure if it was part of prior 6.2 releases (for P85D models at least) or whether .168 was the first to introduce the acceleration improvement and more responsive throttle, but the release notes for .168 are definitely the first to show said features.

I think someone like DragTimes or others on here need to do some 0-60 launch testing again ;-)

 

[Andyw2100]

I'm not sure if it was part of prior 6.2 releases (for P85D models at least) or whether .168 was the first to introduce the acceleration improvement and more responsive throttle, but the release notes for .168 are definitely the first to show said features.

I think someone like DragTimes or others on here need to do some 0-60 launch testing again ;-)

This is my understanding as well.

I'll add that when .153 came out, with the notes about the increased max speed, there was some speculation in some threads that the 0-60 time had also been improved by the .1 second that had been promised, though of course that was not documented. No one was speculating about the more responsive throttle, as that had not been explicitly mentioned by Tesla (or Elon Musk in a tweet) previously.