Firmware 6.1

[bonnie]

Rebooting isn't going to help anyone. I'll repeat myself again. Tesla is not going to roll out a software update to all P85D's simultaneously. It will be a phased rollout, and updates might not even come for weeks or months after the first few reports. Tesla is not stupid and they are not going to risk rolling out to all P85D's at the same time simply because if there's a bug in the firmware the last thing they need is bricking their flagship model across the board. The Press would have a field day with that. People need to be patient here. If you are a new Tesla owner unfamiliar with this process please stop telling everyone you haven't got the latest patch update yet. It might take a few weeks or months whenever your car is scheduled in the randomized rollout by Tesla. Rebooting is not going to magically trigger an update. That's just silly nonsense. Whoever spread that rumor should donate their P85D to me. But by all means if you want to sit there rebooting your car over and over again go ahead and have a field day. If you do and it does work on the 373636794'th time please don't post back here that your reboot triggered the update bc you'll just be wasting more space on this thread.

And who says there is no humor on this forum? Yobig! YO! Remember when you first got your car and waited for a software update? Where was the whole people-need-to-be-patient thing then?

If every post on this forum had to 'not waste more space', then there'd be quite a few people with a whole lot less posts. -ahem-

- - - Updated - - -

That just tells you how many bugs and screwups tesla had in their firmware. Lots of releases like this indicate a very bad software team, not a good one.

Frankly what they are doing is very dangerous. The fact that the screwed up both the brakes AND the sport steering in these releases is VERY disturbing.

LOL there's no new features in these patch releases. They are bug fixes. They are not going to announce to the world what bugs they are fixing with each patch release. The press would have a field day if the knew all the details of what was screwed up in each release. And I'm fully aware of software deployment strategies. Trust me this isn't Agile. More like rush don't test release get complaints fix real fast release again break more things don't test wash rinse repeat.

You don't know, you're guessing. And some of your guesses are pretty harsh indictments of some pretty fine engineers. And the reality is, we all filter through our own personal experience (which would make me wonder what in your software experience leads you to the conclusions you've written). There are other scenarios besides this.

Real people. Real people reading this forum & unable to answer you. I wouldn't be throwing stones in judgment unless I was pretty darn sure of the seat I was throwing them from.

 

[Benjamin Brooks]

Until Tesla starts making release notes that aren't identical for multiple minor releases, some people will think they're constantly fixing bugs introduced in the previous releases, and some people will think they're getting new undocumented features or being used for special field data collection / trial runs.

 

[maddog1762]

The feelings here seem very harsh. I am new to the Tesla family true but we are getting free updates for an outstanding car to make it better. Lets keep the eye on the ball. We have a great car. We have a team trying to make it better. Smile be happy.

 

[bonnie]

Until Tesla starts making release notes that aren't identical for multiple minor releases, some people will think they're constantly fixing bugs introduced in the previous releases, and some people will think they're getting new undocumented features or being used for special field data collection / trial runs.

Yep. And that's exactly what I mean about people filtering through their own previous experience. Some think it's bug fixes, because that's likely what they've witnessed. Others think something cool is going on.

 

[Benjamin Brooks]

Yep. And that's exactly what I mean about people filtering through their own previous experience. Some think it's bug fixes, because that's likely what they've witnessed. Others think something cool is going on.

And, I would assert, Occam's razor would imply as the frequency/quantity of said undocumented minor releases increases, you probably have both scenarios occurring.

 

[yobigd20]

You don't know, you're guessing. And some of your guesses are pretty harsh indictments of some pretty fine engineers. And the reality is, we all filter through our own personal experience (which would make me wonder what in your software experience leads you to the conclusions you've written). There are other scenarios besides this.

Real people. Real people reading this forum & unable to answer you. I wouldn't be throwing stones in judgment unless I was pretty darn sure of the seat I was throwing them from.

Yes, I agree my statements are harsh. But do you not agree with the seriousness of the situation? They pushed software out that screwed up the brakes and steering. This isn't some desktop computer or mobile phone iOS or some broadcast media server that they can just patch up later if something goes wrong. This is a car that transports people. Mistakes like these can cost lives. It's not something to screw around with. Clearly their QA team is lacking some serious talent. Bugs like these should NEVER have made it into final releases to the public. I bet airliners don't push out buggy autopilot software to planes right before they take off filled with 250 passengers. Neither should Tesla.

They need to start taking this a little bit more seriously. I used to work on defense projects that had an extraordinary amount of work dedicated to testing alone making sure it was completely flawless because you're talking about software that has the ability to kill or protect (navy combat). Software that gets used in airplanes and cars should be treated with no less degree of seriousness. Obviously Tesla pushes out software that affects the driving dynamics of the car and changes critical systems related to steering and acceleration and brake systems. That's all cool and everyone loves it when they can speed it up and decrease 0-60 by a few tenths of a second, but how do you think everyone would react if Tesla pushed out some bugs that disabled the brakes right when you needed them or locked up the steering right when you needed to swerve? Not so cool now, huh?

 

[bonnie]

Yes, I agree my statements are harsh. But do you not agree with the seriousness of the situation? They pushed software out that screwed up the brakes and steering. This isn't some desktop computer or mobile phone iOS or some broadcast media server that they can just patch up later if something goes wrong. This is a car that transports people. Mistakes like these can cost lives. It's not something to screw around with. Clearly their QA team is lacking some serious talent. Bugs like these should NEVER have made it into final releases to the public. I bet airliners don't push out buggy autopilot software to planes right before they take off filled with 250 passengers. Neither should Tesla.

They need to start taking this a little bit more seriously. I used to work on defense projects that had an extraordinary amount of work dedicated to testing alone making sure it was completely flawless because you're talking about software that has the ability to kill or protect (navy combat). Software that gets used in airplanes and cars should be treated with no less degree of seriousness. Obviously Tesla pushes out software that affects the driving dynamics of the car and changes critical systems related to steering and acceleration and brake systems. That's all cool and everyone loves it when they can speed it up and decrease 0-60 by a few tenths of a second, but how do you think everyone would react if Tesla pushed out some bugs that disabled the brakes right when you needed them or locked up the steering right when you needed to swerve? Not so cool now, huh?

All of the tested software also has thousands of hours with professional drivers and then beta testers before release. You assume they don't take things seriously. I disagree.

 

[yobigd20]

All of the tested software also has thousands of hours with professional drivers and then beta testers before release. You assume they don't take things seriously. I disagree.

Well I don't doubt that part either. The part the makes it uneasy for me is that they packaged and delivered software that still has flaws affecting critical systems. So IMO their thousands of hours and beta testers just isn't good enough.

 

[apacheguy]

All of the tested software also has thousands of hours with professional drivers and then beta testers before release. You assume they don't take things seriously. I disagree.

But is that always true? For big releases like 6.0 or probably even 6.1 I don't doubt that it is. But what about point releases? Like going from .113 to .115 to .140? Were there thousands of hours spent testing each of those?

Let me qualify that by saying that I'm not trying to point out a flaw here. By all means I agree with patching issues quickly and as they arise. It's just that beta testing would seem to hinder that goal.

 

[gg_got_a_tesla]

Wow, what a difference in efficiency .139 and torque sleep have made with my P85D!

Wife and I were consistently clocking in the high 340s / low 350s Wh/mile on our 60-mile weekday commute with .113.

Today, we achieved 309 Wh/mile with .139 with nothing else being different!! This was in SPORT mode throughout and RANGE MODE ON - those were the settings with .113 too.

A very happy camper now. One thing wife (subjectively) noted is that there's just that infinitesimal but apparently perceptible delay before the power kicks in when the go pedal is punched from various cruising speeds.

 

[AllenWong]

Wow, what a difference in efficiency .139 and torque sleep have made with my P85D!

Wife and I were consistently clocking in the high 340s / low 350s Wh/mile on our 60-mile weekday commute with .113.

Today, we achieved 309 Wh/mile with .139 with nothing else being different!! This was in SPORT mode throughout and RANGE MODE ON - those were the settings with .113 too.

A very happy camper now. One thing wife (subjectively) noted is that there's just that infinitesimal but apparently perceptible delay before the power kicks in when the go pedal is punched from various cruising speeds.

I don't know if it's psychological, but I'm also getting low 300s when I used to average in the high 300s. I'm in insane mode. I just don't press the accelerator as hard, but I like having the power under my foot when I need it. Yellow lights don't stand a chance against me.

 

[skilly]

But is that always true? For big releases like 6.0 or probably even 6.1 I don't doubt that it is. But what about point releases? Like going from .113 to .115 to .140? Were there thousands of hours spent testing each of those?

Let me qualify that by saying that I'm not trying to point out a flaw here. By all means I agree with patching issues quickly and as they arise. It's just that beta testing would seem to hinder that goal.

Im critical of their update notes - they are terrible. Every point release should have notes explaining what we are incorporating into 6.1. That said, I dont think this release has been a blunder at all - the majority of the "noise" has been based on D owners and the lack of efficiency that was to be forthcoming - and even that has been solved for in 139. Of course, due to the nature of software there have been some reports of anomalies around some user experiences, but those could easily be tied to pilot error...the bottom line is that software comes with a version nomenclature because until its end of life, its never complete or expected to work correctly with 100% of the user base.

 

[AmpedRealtor]

Yes, I agree my statements are harsh. But do you not agree with the seriousness of the situation? They pushed software out that screwed up the brakes and steering. This isn't some desktop computer or mobile phone iOS or some broadcast media server that they can just patch up later if something goes wrong. This is a car that transports people. Mistakes like these can cost lives. It's not something to screw around with. Clearly their QA team is lacking some serious talent. Bugs like these should NEVER have made it into final releases to the public. I bet airliners don't push out buggy autopilot software to planes right before they take off filled with 250 passengers. Neither should Tesla.

They need to start taking this a little bit more seriously.

I wholeheartedly agree! Sadly, my experience with Tesla software releases is also one of the reasons why I can't get excited about autopilot. The various safety warnings, yes, those are great. But once autopilot is rolled out and people are trusting it with their lives on the freeway, count me out. One software guy with a bad hangover on the day he or she is tweaking the autopilot code could be enough to send the car careening into an adjacent lane or oncoming traffic. Of course I'm being dramatic, but how do we know? And given Tesla's track record - especially lately with the brake and steering issues - I have zero faith in them being able to execute autopilot in a manner that would be trustworthy for me. Sorry, but that's how I feel.

Tesla's beta testers are really more like alpha testers, and Tesla's production releases are really more like beta releases. And imagine if something goes wrong with autopilot that causes injury or death. Imagine what that would do to this company. I want to make sure that Tesla's ambition isn't writing checks that their engineers and QC people can't cash.

 

[HankLloydRight]

I wholeheartedly agree! Sadly, my experience with Tesla software releases is also one of the reasons why I can't get excited about autopilot. The various safety warnings, yes, those are great. But once autopilot is rolled out and people are trusting it with their lives on the freeway, count me out. One software guy with a bad hangover on the day he or she is tweaking the autopilot code could be enough to send the car careening into an adjacent lane or oncoming traffic. Of course I'm being dramatic, but how do we know? And given Tesla's track record - especially lately with the brake and steering issues - I have zero faith in them being able to execute autopilot in a manner that would be trustworthy for me. Sorry, but that's how I feel.

Tesla's beta testers are really more like alpha testers, and Tesla's production releases are really more like beta releases. And imagine if something goes wrong with autopilot that causes injury or death. Imagine what that would do to this company. I want to make sure that Tesla's ambition isn't writing checks that their engineers and QC people can't cash.

I agree with these comments 100%. But I think very few people are willing to make this statement (especially on this board). What worries me on top of potential software flaws in the safety systems, is uniformed or uneducated drivers putting too much reliance on "Autopilot" and getting into an accident which would then be erroneously attributed to faulty software. It's going to happen, and the software (I assert) will likely NOT be at fault, but it's the easiest thing to point to.

 

[ecarfan]

One software guy with a bad hangover on the day he or she is tweaking the autopilot code could be enough to send the car careening into an adjacent lane or oncoming traffic. Of course I'm being dramatic, but how do we know? And given Tesla's track record - especially lately with the brake and steering issues - I have zero faith in them being able to execute autopilot in a manner that would be trustworthy for me. Sorry, but that's how I feel.

All of the Model S critical driving systems have been software controlled from the beginning: steering, brakes, drivetrain, etc.

If you have "zero faith" in the ability of Tesla to make reliable Auto Pilot software in the future, then I am surprised you have any faith in the car since the Model S is a 100% "drive by wire" car right now.

MobilEye, which Tesla obviously is partnering with to create the Model S Auto Pilot feature, has been developing that capability for years and it works.

I think you are being overly "dramatic" regarding Tesla's track record and capabilities.

 

[skilly]

All of the Model S critical driving systems have been software controlled from the beginning: steering, brakes, drivetrain, etc.

If you have "zero faith" in the ability of Tesla to make reliable Auto Pilot software in the future, then I am surprised you have any faith in the car since the Model S is a 100% "drive by wire" car right now.

MobilEye, which Tesla obviously is partnering with to create the Model S Auto Pilot feature, has been developing that capability for years and it works.

I think you are being overly "dramatic" regarding Tesla's track record and capabilities.

I completely agree! This hyperbole has little to no basis. The entire vehicle is drive by wire - as is almost every other car on the market today. Moreover, "autopilot" is essentially catch up in the luxury car market world. Many of the features advertised are in high end (and even mid market) cars today. We arent reading about S550s careening out of control from drivers giving too much trust to the autopilot like features....this is akin to the rumor about the driver of the Winnebago getting up in his motorhome to do stuff in the back while he let cruise control take over driving...not likely.

 

[apacheguy]

All of the Model S critical driving systems have been software controlled from the beginning: steering, brakes, drivetrain, etc.

I didn't know that. I could've sworn pre autopilot MS was not drive by wire. Are you telling me there is software in between my brake pedal and the pads on the wheels - that there is no physical connection? And same goes for the steering wheel? This is all news to me.

 

[AmpedRealtor]

All of the Model S critical driving systems have been software controlled from the beginning: steering, brakes, drivetrain, etc.

Yes, and in one of the recent updates they broke sport steering at low speeds and allowed owners to drive off without full braking power. I can't think of a better example.

If you have "zero faith" in the ability of Tesla to make reliable Auto Pilot software in the future, then I am surprised you have any faith in the car since the Model S is a 100% "drive by wire" car right now.

I have little faith in Tesla being able to launch autopilot without there being any bugs, and this is based upon my experience with Tesla's software releases over the past (almost) two years. The fact that there will be bugs is worrisome because when engaged, autopilot is a mission critical application that must always work. There is no room for error. So far, Tesla has not shown me that they can produce a software update without obvious bugs, glitches, and flaws.

MobilEye, which Tesla obviously is partnering with to create the Model S Auto Pilot feature, has been developing that capability for years and it works.

Aftermarket MobileEye systems do not control the car and do not take your life into their hands. They provide warnings, but you control the car. Autopilot is an order of magnitude more complex because it controls the car, and by association, your life!

I think you are being overly "dramatic" regarding Tesla's track record and capabilities.

You are certainly entitled to your opinion, as am I. If you're willing to trust autopilot when it's released, more power to you. I just won't be one of those folks.

 

[dsm363]

I have great faith in the autopilot team. They'll get this right. They know how important it is to get it right.

 

[tomas]

Auto-pilot risk discussion maybe OT, and much broader than Tesla of course - but I find very interesting.

I share Amped and Hank's concern that it will take time to perfect auto-pilot, and that humans can/will botch it up. Glad that Tesla has "outsourced" most of this to Mobileye, which has deep experience across many marques... so much of the function will come from dedicated specialist.

Ecarfan and Skilly, true that many of the MS (and most lux cars these days) are laden with "drive by wire"... steering, accelerator, brake, ABS, traction control.... but I believe that this is a different class of software with respect to complexity. For example, steering software does one thing: steer front wheels lock to lock - and is controlled by one input device: steering wheel. Pretty simple set of use cases, relatively easy to bulletproof. Auto-pilot software, on the other hand, has to take multiple inputs (photographs, sensor input, etc.), interpret them, relate them, devise a response, and relay it to multiple systems (braking, steering, acceleration, driver alerts). Much, much, much more complex. And, the "drive by wire" functions have a failsafe involved on the other end - the driver. When we get to full auto-pilot, drivers are going to check out, so this incredibly complex software is going to have to be pretty perfect.