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

Firmware 7.1

This site may earn commission on affiliate links.
after returning home after a brief trip I found an update waiting for me to load it, it is loading now. I see a lot of complaints here but little about what is in this new update. is there anything significant? or is it just some tweaking?

I too was away, and just updated to 2.17.37 today.

Initially I had entered today's date in the tracker as my update date, but then realized that was providing data that was almost certainly erroneous, and would cause the data in the tracker to appear odd. Since I generally get firmware updates that are broadly pushed at the front end of the push, I decided to delete my original entry, which had included a note about how the update had almost certainly been received days earlier, and entered the update with a date of 4/27, which was the date with the largest number of updates entered for that version. I still included a note, indicating that the date was estimated, but in my opinion this will have less of an impact on anyone looking at the data. The first way I had done things someone might wonder why that older version was still being pushed today. This way I'm just "one of the masses." I believe this way is also probably much more accurate.
 
The route planner did something new (to me) in how it had me stop at super chargers that I would like to see if others have observed. It could be a first indication that Tesla is trying to have cars skip busy super chargers.
Anyway, I was heading from Los Angeles up to San Francisco. The route planner told me to go straight to the Buttonwillow SC without stopping to charge at Tejon Ranch. It warned me to keep my speed below 65 mph in ordered to get there. Given that the average speed on I5 is over 75, I chose to go faster and stop to charge Tejon Ranch. I took the last of the six slots when I got there. Anyway, it seemed strange, so I though I would share it.
I'm on 2.17.37.
 
The route planner did something new (to me) in how it had me stop at super chargers that I would like to see if others have observed. It could be a first indication that Tesla is trying to have cars skip busy super chargers.
Anyway, I was heading from Los Angeles up to San Francisco. The route planner told me to go straight to the Buttonwillow SC without stopping to charge at Tejon Ranch. It warned me to keep my speed below 65 mph in ordered to get there. Given that the average speed on I5 is over 75, I chose to go faster and stop to charge Tejon Ranch. I took the last of the six slots when I got there. Anyway, it seemed strange, so I though I would share it.
I'm on 2.17.37.

That is certainly interesting. Clearly the collection of more data is required before any conclusions can be reached.

As an example, even if this version of software is doing something different with respect to supercharger routing, as it may seem based on your one example, and the fact that you were never routed past a supercharger in this way before, there is no way to tell from your example if the new software is attempting to avoid busy superchargers, or is simply attempting to minimize supercharger use whenever possible by suggesting slower speeds if that will allow a supercharging stop to be bypassed. If it is the latter, the fact that you took the last charging spot would have been purely coincidental.

As more people report what they are seeing, a clearer picture should develop pretty quickly.
 
The route planner did something new (to me) in how it had me stop at super chargers that I would like to see if others have observed. It could be a first indication that Tesla is trying to have cars skip busy super chargers.
Anyway, I was heading from Los Angeles up to San Francisco. The route planner told me to go straight to the Buttonwillow SC without stopping to charge at Tejon Ranch. It warned me to keep my speed below 65 mph in ordered to get there. Given that the average speed on I5 is over 75, I chose to go faster and stop to charge Tejon Ranch. I took the last of the six slots when I got there. Anyway, it seemed strange, so I though I would share it.
I'm on 2.17.37.
Don't read too much in to it. My car (still on 7.0) had me skip the nearest supercharger (Canmore) yesterday when I asked it to plan a trip to Vancouver, i was surprised as well, but I can say it's not due to the newest firmware.
 
Don't read too much in to it. My car (still on 7.0) had me skip the nearest supercharger (Canmore) yesterday when I asked it to plan a trip to Vancouver, i was surprised as well, but I can say it's not due to the newest firmware.

Unless Tesla has been doing SpC routing on the server side instead of in-car all along and just updated how the server works...
 
  • Like
Reactions: Cyclone
I agree more data is needed. I have been routed past super chargers before. It is the quickest way if one has enough charge to do it. What was weird is that it wanted to skip a SC and make me go slow to reach the next one.
there is this thing that most of us have but not all of us implement, it is a brain. people should feel free to put their brains to use in such situations. while the trip planner in the car has seen significant improvements it is far from perfect and should be used in conjunction with your knowledge of the car and climatic conditions to decide whether a charge stop is necessary or if it can be skipped.
safe travels
 
  • Like
Reactions: CHG-ON
2.17.37 finally fixed GPS inaccuracy bug while reversing! I really wish they would test the releases on the classic cars more.

Not for me, sadly. According to the app my car hasn't moved since I put it in reverse and autoparked it. The gps still thinks it's in the middle of the street.

Anyone else can confirm, that bug is driving me crazy, can't get the auto garage door opening reliable (works only one way, reversing as my reset point was set that way).

I did not get the update, yet, so I can't test. Still on 2.16.17
 
My BMW i3 (Which I sold when I received my Model X) had this exact capability. If I wanted to engage standard (non-traffic aware) cruise control I could. I could also toggle between adaptive and non-adaptive cruise control. Very nice feature to have.
I also have an i3 and sold a Mercedes to buy the MS. With just the i3 and the Mercedes (dumb CC), I had to always keep in mind that the Mercedes would gladly drive right into the back of the car I was following, I really needed to pay attention.

I have said it a couple times before, but it bares repeating. Tesla likely is doing just TACC or traditional cruise so as to not confuse the consumer thinking one is active when the other is. A simple way to do this is to leverage the codebase of the Classics and what they had back in 6.2. I.e., when AutoPilot/TACC is active, use the v7 dash we are familiar with. When traditional cruise in active, switch to the speedo that us Classics see and the AP cars say back in 6.2. That would very easily distinguish between which mode is active. And before anyone jumps on blind spot monitoring and such, that was available back in 6.2, so they could continue to use that convention when traditional cruise is active.
 
  • Like
Reactions: brianman
I agree more data is needed. I have been routed past super chargers before. It is the quickest way if one has enough charge to do it. What was weird is that it wanted to skip a SC and make me go slow to reach the next one.

What was the net time difference between the two options? In other words, if you went 65 vs 75 but didn't have to Supercharge, maybe it's using that logic now?

In other words, just my rough math as an example: If Los Angeles to Buttonwillow is 125 miles at 65 that's 1:55 hours. Los Angeles to Tejon Ranch at 75mph is 75 miles (1:00 hour) + 50 miles (40 minutes) to Buttonwillow. So you save 15 minutes by driving faster, but you have to stop for 15 minutes (or more). So, you arrive at the same exact time. Maybe now the software calculates this? It might attempt to be saving a Supercharger slot (especially a crowded one) if your arrival time is close to identical simply by limiting your speed? Of course, it's more fun to drive faster for a lot of people...
 
Anyone else can confirm, that bug is driving me crazy, can't get the auto garage door opening reliable (works only one way, reversing as my reset point was set that way).

I did not get the update, yet, so I can't test. Still on 2.16.17

My gps position is definitely fixed, but not the right way. I drove backwards on the driveway and it shows the correct position. The orientation of the arrow is still not right after reversing. It shows sideways or backwards after reversing. It corrects itself after going forward though. I almost want to get a job at Tesla just to fix these annoying bugs. Orientation should not be tied to direction change of GPS coordinates, the car should know whether it is moving forward or backward!
 
  • Like
Reactions: MP3Mike and Kalud
My gps position is definitely fixed, but not the right way. I drove backwards on the driveway and it shows the correct position. The orientation of the arrow is still not right after reversing. It shows sideways or backwards after reversing. It corrects itself after going forward though. I almost want to get a job at Tesla just to fix these annoying bugs. Orientation should not be tied to direction change of GPS coordinates, the car should know whether it is moving forward or backward!

That may not be easy to determine if the car is not keeping a history of GPS information though. While driving say 5 miles, the car reasonably extract the changes in the GPS coordinates and know which way it is moving. From that, it would be relatively trivial for the car to know whether, at the end, it moved forward or in reverse. However, if the car doesn't have the past 5 miles to make a judgement on, perhaps only having the last 30 seconds, the variations in the GPS coordinates are within the realm of noise from GPS calibration inaccuracies. As such, it doesn't know if the car moved forward or backways XYZ feet, or just the GPS lock wobbled part of that.

I am NOT saying orienting the logo can't be done; but it could be more complicated because of other design decisions like maintaining the GPS lock long-term. In the Auto Homelink thread, there is theory that our cars do not maintain a GPS location memory for long, and if your garage is preventing the car from getting a good GPS lock immediately upon startup, it may not trigger Auto Homelink.
 
That may not be easy to determine if the car is not keeping a history of GPS information though. While driving say 5 miles, the car reasonably extract the changes in the GPS coordinates and know which way it is moving. From that, it would be relatively trivial for the car to know whether, at the end, it moved forward or in reverse. However, if the car doesn't have the past 5 miles to make a judgement on, perhaps only having the last 30 seconds, the variations in the GPS coordinates are within the realm of noise from GPS calibration inaccuracies. As such, it doesn't know if the car moved forward or backways XYZ feet, or just the GPS lock wobbled part of that.

I am NOT saying orienting the logo can't be done; but it could be more complicated because of other design decisions like maintaining the GPS lock long-term. In the Auto Homelink thread, there is theory that our cars do not maintain a GPS location memory for long, and if your garage is preventing the car from getting a good GPS lock immediately upon startup, it may not trigger Auto Homelink.
The orientation and position should rely on not only GPS coordinates, but also wheel/ motor speed sensors and steering wheel orientation to provide accurate output. GPS history is not needed at all.
 
  • Like
Reactions: MP3Mike
The orientation and position should rely on not only GPS coordinates, but also wheel/ motor speed sensors and steering wheel orientation to provide accurate output. GPS history is not needed at all.

I agree the various sensors are needed. But I don't see how you can make that determination with only one set of GPS coordinates. The first set of coordinates do not provide direction. Once your sensors say you have moved X or Y or Z from those coordinates, you would also have the updated coordinates (the now "current" location). The sensor data would simply tell you which way the car moved from the first set to the second set to give you orientation, if you still had the original location.
 
It is likely that Tesla is using some form of flux gate compass to augment direction information, these devices sometimes require a certain velocity to function properly, it's possible the low speeds associated with reverse complicates direction finding... That said, no reason they can't get it sorted with existing sensors
just out of curiosity, what is the big deal about which way the little arrow is pointed?