Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

Wiki Firmware Upgrade Tracker

This site may earn commission on affiliate links.
(Moved from the general 8.0 thread)

Exactly. It's not like it's at all consistent information.. the same update could take 15 minutes on one car, 1.5 hours on another. For instance, I'm still on v7.1.. I don't know how many iterative versions or partial updates it might need to make to get me up to "current". Also, how many people actually time the installs? Do you sit there and wait/watch? I just set it and forget it. I come back to the car an hour later and it's updated as I suspect the majority of people do.

So as suggested before, if people want to keep track of that info, please do so in the "notes" area, and yes, other people can see that in the full data table at the bottom of the page.
Most of your observations above while true, seem to me to fall under the category of "perfect is the enemy of good". What I was thinking of was a simple field whose semantics are "this is how many minutes my update took me". Some specific responses and observations:
  • Different cars/software combinations have different times: OK, but I can know the hardware from the VIN, and the software from the earlier records that VIN has logged. Might they have missed logging an earlier record? Yes. It's an imperfect world.
  • You don't time your install. That's fine, you'd leave your entry blank. I often do time it and no of course I don't sit there and watch. It's not very onerous: I initiate the update and note the time (or scheduled time if I do it that way). Later, I get a notification on my phone that the update has completed. The notification is time-stamped, the rest is simple subtraction. I don't know if "the majority of people" do something like this, but even if only a few of us do, we can provide useful information for those inquiring minds that keep asking.
  • Using the notes is exactly what I do, as I pointed out initially. It seems less desirable than having a structured field, since the information is freeform there's no hope of running a report and producing information like "people with AP1.0 updating from the last major rollout took 20 minutes to install this".
  • Despite all the shortcomings, people are asking for and sharing updated time information anyway. They just use the main firmware thread for it so we have dozens of "it took me XX minutes" posts with even less hardware/software context than the tracker could provide.
It's your bat and your ball of course, so if you still feel like it's a bad idea or just don't have the time or desire to implement it, I won't press the point further. Thanks.
 
My main challenge is this statement: "This is one of the most-asked questions after each new rollout."

I don't think that's true at all.

I really don't recall many (if any) people constantly asking that question for each rollout. I do see a few people posting how long their update took, but not people asking ahead of time.
 
My main challenge is this statement: "This is one of the most-asked questions after each new rollout."
Which isn't what I said of course ("most-asked" would overstate my case, and in retrospect "dozens" may not have been warranted, see below).
I don't think that's true at all.

I really don't recall many (if any) people constantly asking that question for each rollout. I do see a few people posting how long their update took, but not people asking ahead of time.
Maybe it's confirmation bias on my part but ISTR a decent number of people asking this on the 8.0 thread for the most recent rollout. If I get bored and have some time, perhaps I'll go back and see if it's really true (if so I'll follow up with posting refs). I certainly wouldn't attempt to quantify more than the most recent rollout, for obvious reasons of it being a massive PITA. So if "constantly asking that question for each rollout" is the bar, I'm not shooting for it. In any case, that bullet was almost an afterthought -- to my mind, the case is more around "hey, this would demonstrably be useful to some people, and other people are gathering the data anyway, so why not provide a structured way to share it?" I'm even willing to concede for the sake of discussion that it'll only benefit some small percentage of total users, I think there's still value if one assumes that.

If I'm reading between the lines of your reply correctly I think you're saying you don't see the utility so you don't see value in cluttering the data and/or doing the work? OK (this falls under the "your bat and your ball" rule :). Unless I do the research noted above and have something to share, or unless you want to continue the discussion, I guess I'll consider the subject closed. Again, thanks for the tracker.
 
Just wanted to add, that I wouldn't have quoted you if you didn't say it..

upload_2017-1-17_11-41-0.png
 
There's one point that neither of you have raised in the discussion about adding a field for how long the update took. And that is, would the extra field, even if it was noted that it was optional, possibly prevent new EV-FW users from entering their data, because they did not record the time taken? Just a thought.

I --do-- see some value in having the information available if it can be done with little cluttering, and little additional work on Hank's part. I would just hate to see anyone mistakenly decide not to use the tracker because for whatever reason, this new field intimidates them.
 
  • Like
Reactions: jgs
What is the + in the trim designation like P+100 and what is "PD" in the Updates by Model/Trim (model) list?
User input errors.

The P+ is only for the P85+ (or a few P85Ds with the + suspension). A number of people selected "Performance+" instead of just "Performance" for their cars, in error. I'll put some error checking in place to prevent that in the future. I've corrected these incorrect entries.

The FW tracker started out as a quick+dirty UI to automate Dirk's original Wiki.. as such, I didn't really build into it 100% of the error checking it should have. I was relying on smart, intelligent, and diligent Tesla Owners to "do the right thing". Well, in retrospect... ;)
 
Recently I have been getting this unresponsive script message when I access the tracker site. I recently upgraded to Sierra.

That's not any script that is running on the site. It looks like a browser extension script that is having problems. Try disabling your browser extensions one at a time until the problem goes away. Then you'll find your suspect.
 
  • Helpful
Reactions: hiroshiy
I think the tracker is great. I download the CSV file and put into my own relational database so I can run whatever queries I want. The only rough place is that there is no unique car identifier in the database, but I just make something up by concatenating the handle, name, model, perf, kwh,suffix and vin. That seems to work. (I'm guessing that the tracker uses the full vin internally for its identifier, which of course can't be in the extract file. Better database design uses non-meaningful keys.)

For some reason, the switch to HTTPS on the site broke my software (using the Indy libraries) that fetched the CSV file directly, so I have to download this manually for now. No big deal.

There are a few errors in the database, for example there is a P100D showing version 6.2 of the firmware (total of 11 AP2 cars with firmware before 8.0).
 
Last edited:
That seems to work. (I'm guessing that the tracker uses the full vin internally for its identifier, which of course can't be in the extract file. Better database design uses non-meaningful keys.)

I'm not sure why you make this assumption on my database design. Nothing on the site gives any indication of what I'm using for identifiers. I don't even collect full VIN unless the user chooses to enter it, and even then it's only the last six characters with the options of masking the last three. As a career database architect and engineer, of course each record has a unique integer identifier. I can add that to the extract file. I never thought anyone would need it.
 
Last edited:
Firmware tracker fun facts:

For version 6.2, most updates happened on Tuesday, fewest on Sunday
For version 7.0, most updates happened on Thursday, fewest on Sunday
For version 7.1, most updates happened on Friday, fewest on Sunday
For version 8.0, most updates happened on Friday, fewest on Sunday
 
I'm not sure you make this assumption on my database design. Nothing on the site gives any indication of what I'm using for identifiers. I don't even collect full VIN unless the user chooses to enter it, and even then it's only the last six characters with the options of masking the last three. As a career database architect and engineer, of course each record has a unique integer identifier. I can add that to the extract file. I never thought anyone would need it.

Sorry for my uninformed speculation.

I needed an unique car identifier in order to run a query to see how many cars were still running a particular release, rather than just how many got the release. But as I said, I was able to match them up with my concatenated identifier.
 
Last edited: