Software Update 2018.21.9 75bdbc11

[SOULPEDL]

Here's my experiment for the day. Result = not nagging, non-issue. Hands on wheel.

Just drove 20 miles freeway on AP, mostly straight with a couple of gradual turns on the 202. Right elbow resting on center console, arm horizontal, hand at furthest bottom position on wheel. My hand in this config weighs ~10 oz and because it's at about 5 o'clock position, I'm thinking about 0-2 oz torque (~$2 in quarters at 3 o'clock). And I'm not 100% touching in order to get several photos. I have very small hands, (but don't tell anyone - it's how I can fix things and you can't).

IMHO, that's extremely sensitive. I'm actually not sure if I could maintain contact and still trigger a warning. How small are your hands, really?

 

[brendon1970]

I take it back, regen IS affected by this release. On my P100D, it used to max out at 70kW, now it never goes above 50kW. Deceleration from high speed takes substantially longer. I'm not a fan of this change either...

@conman

Are you sure your car is not simply cold? I didn't notice any change in regen with the install of 2018.21.9, but once the current cold snap came through (I'm near Sydney) and the car registered battery temps of less that 10C the regen was limited - and indicated by some yellow dashed lines on the power widget.

I've seen it go to the next stage - left the car outside in 4 degress and it severly limited regen and power.

Both effects dissapear once the battery has warmed up - but this takes quite some distance of driving.

 

[conman]

@conman

Are you sure your car is not simply cold? I didn't notice any change in regen with the install of 2018.21.9, but once the current cold snap came through (I'm near Sydney) and the car registered battery temps of less that 10C the regen was limited - and indicated by some yellow dashed lines on the power widget.

I've seen it go to the next stage - left the car outside in 4 degress and it severly limited regen and power.

Both effects dissapear once the battery has warmed up - but this takes quite some distance of driving.

I was running my car in ludicrous+ mode, so yeah the battery was extra warm.

 

[SOULPEDL]

I was running my car in ludicrous+ mode, so yeah the battery was extra warm.

How can we read Watts Regen? Only way I've known is on a long hill to watch the trip card. Also, it limits recharge near a full charge (can't charge a full charge). So what's your battery at?

 

[conman]

How can we read Watts Regen? Only way I've known is on a long hill to watch the trip card. Also, it limits recharge near a full charge (can't charge a full charge). So what's your battery at?

The graph/dial on the right shows power usage by the bar going up red/orange in the anticlockwise direction (i.e. minus power) and green in the clockwise direction. It maxes out at 50kW on the graph now where previously it would hit 70 during deceleration from high speed (i.e. coming off a freeway etc.) My battery does not normally sit at 100% charge all day... this is on extended driving. I charge my vehicle to 90% and when I feel like it I warm the battery up for extra power. The regen did not hit more than 50kW with the battery around 70% charge and the battery temperature anywhere between 20 and 50 degrees C (in ludicrous+ mode it shows the battery temperature.) Under these identical circumstances it used to always hit 70kW when decelerating at anything over 50km/h

 

[SOULPEDL]

The graph/dial on the right shows power usage by the bar going up red/orange in the anticlockwise direction (i.e. minus power) and green in the clockwise direction.

I just now realized you have a Model S. I'm driving a Model 3 so we don't get that graph at all.

On the nag thing, I doubt we have the same SW build. How could we with different hardware in so many ways? Maybe a catch-all number with same AI, but different underneath for the different boards, who knows. So the whole wheel nag thing could actually be different for you. Apples and Oranges maybe.

 

[croman]

I just now realized you have a Model S. I'm driving a Model 3 so we don't get that graph at all.

On the nag thing, I doubt we have the same SW build. How could we with different hardware in so many ways? Maybe a catch-all number with same AI, but different underneath for the different boards, who knows. So the whole wheel nag thing could actually be different for you. Apples and Oranges maybe.

This firmware is the same across models. It appears to be more unified than ever and presumably headed to fully unified with v9 in "August".

 

[brendon1970]

The graph/dial on the right shows power usage by the bar going up red/orange in the anticlockwise direction (i.e. minus power) and green in the clockwise direction. It maxes out at 50kW on the graph now where previously it would hit 70 during deceleration from high speed (i.e. coming off a freeway etc.) My battery does not normally sit at 100% charge all day... this is on extended driving. I charge my vehicle to 90% and when I feel like it I warm the battery up for extra power. The regen did not hit more than 50kW with the battery around 70% charge and the battery temperature anywhere between 20 and 50 degrees C (in ludicrous+ mode it shows the battery temperature.) Under these identical circumstances it used to always hit 70kW when decelerating at anything over 50km/h

Yep I'd say I normally get somewhere between 50 and 70 Kw regen. Not exactly sure how much though - its a log scale graph, and more that 50 Kw.

 

[Gig103]

Are you sure? I'm guessing at least one person asked for it.

Who asked for the mirrors to auto-unfold when the handles present, instead of when the doors were opened? That came up earlier this year and is awful.

 

[conman]

Who asked for the mirrors to auto-unfold when the handles present, instead of when the doors were opened? That came up earlier this year and is awful.

That's one of my all time biggest annoyances since I always approach the car from in front in the car park meaning the mirrors always get in my way as I walk past them in tight car spots...

 

[KyleDay]

This firmware is the same across models. It appears to be more unified than ever and presumably headed to fully unified with v9 in "August".

This is slightly incorrect. MCU1 and MCU2 vehicles receive different firmware builds, as does model 3. They might display the same firmware version, but they're not the same. Probably a distinction without much difference, but worth noting.

 

[rnortman]

Yup- I do the same. Works great normally but I’ll admit that my over 800 mile weekend trip did show me what all the gripes are about with this new nag frequency. For the most part I don’t think it’s a big deal, but I did get thrown in AP jail once, even tho I had my hand on the wheel the entire time. Guess I didn’t respond fast enough/enough torque to repeated nags.

Yeah, this is the entire problem. It's not that you can't hold the wheel in such a way as to prevent the nags, but how much attention this requires depends on driving conditions. On a long straight stretch of highway, you are going to have to be actively paying attention -- not to the road, but to the nags. Or else you will get locked out of AP.

My greatest hope is that Tesla recognizes that this is not sustainable and does something reasonable instead.

 

[Impala1ss]

I just now realized you have a Model S. I'm driving a Model 3 so we don't get that graph at all.

On the nag thing, I doubt we have the same SW build. How could we with different hardware in so many ways? Maybe a catch-all number with same AI, but different underneath for the different boards, who knows. So the whole wheel nag thing could actually be different for you. Apples and Oranges maybe.

You do realize that there is a Model 3 category, right?

 

[Impala1ss]

While most of you are trying to discern why the AP's new NAG procedure doesn't function as you wish it did, I believe that the facts are that Tesla's software (incorrectly) causes different cars to perform differently. Compounding this, in my opinion, is Tesla's complete absence in providing any explanation for this problem, or possible solutions. I guess Tesla assumes we will either figure it out for ourselves or just give up talking/complaining about it. Act like there is no problem and there is no problem.

 

[rnortman]

Right elbow resting on center console, arm horizontal, hand at furthest bottom position on wheel. My hand in this config weighs ~10 oz and because it's at about 5 o'clock position, I'm thinking about 0-2 oz torque (~$2 in quarters at 3 o'clock).

This is a pretty reliable way to avoid the nags. It's not 100% reliable, but it's pretty good. I also find it somewhat uncomfortable for long periods but more importantly I find it somewhat unsafe. If I have to react quickly and forcefully (e.g., if Autosteer jerks the wheel) I won't be able to -- particularly if you need to turn the wheel clockwise with your hand at 5 o'clock like that. (You have a pretty good shot at going counterclockwise quickly.)

Tesla cannot reasonably force people into unsafe driving positions like that.

 

[SOULPEDL]

Tesla cannot reasonably force people into unsafe driving positions like that.

As I said, it was an experiment to find the minimum contact required to eliminate nags. That's not how I drive. If you read my earlier posts, I tend to compensate as if AP could pop out at any moment, with an active force steering away from other vehicles or walls. There is no way I'd be so passive to just hang my fingers as shown. Was only trying to consider other points of view regarding long distance annoyance - I honestly had trouble understanding and now do not see it as a problem. Maybe we're in violent agreement?

 

[MIT_S60]

While most of you are trying to discern why the AP's new NAG procedure doesn't function as you wish it did, I believe that the facts are that Tesla's software (incorrectly) causes different cars to perform differently. Compounding this, in my opinion, is Tesla's complete absence in providing any explanation for this problem, or possible solutions. I guess Tesla assumes we will either figure it out for ourselves or just give up talking/complaining about it. Act like there is no problem and there is no problem.

Unless we start complaining to the NHTSA about it.

 

[SOULPEDL]

You do realize that there is a Model 3 category, right?

Yes, thx. And this is regarding V9 right? Even though apples and oranges as I suspected across models.

 

[Pale_Rider]

Seems like since there are seemingly two camps of people (those who rarely get a nag and those who get overly frequent nags) that the folks who are getting a bunch of nags needs to go to the SC and find out if there is an issue with the vehicle. I fall into the camp of no real issues with the nag vs previous firmwares. I pretty much always drive with one hand either at 9 or at around 5 and have not noticed more nags, maybe a couple more that escalate to a little beep, but frankly I prefer the auditory feedback to the visual. I get a beep and I adjust my hand to apply slightly more torque. I feel like there is a bit of mass hysteria going around about how this change makes the car "unsafe" in anyway. But hey, its easy for me to say that because I'm not getting noticeably more nags than before...

 

[SOULPEDL]

Seems like since there are seemingly two camps of people (those who rarely get a nag and those who get overly frequent nags)

Exactly! So I'm trying to see if it's the car model vs other things. I"m a Model 3, you're an X, I'm starting to think this may be an S variable and steering wheel sensitivity/ratio. I have trouble believing Tesla has never changed the system physical control dynamics over this period of time, and maybe they just haven't compensated for that in code across vehicles. But what do I know...