Sorry, I actually like the new UI :)

[smogne41]

Oh that one. I did listen to that video and I think it was just a poor choice of words by Musk that may have been misconstrued. He points out that whenever the human had to intervene it meant the AI had done something wrong (took the wrong arc through an intersection in his example). He'd like the AI to be perfect one day, but for now the human intervention (error signal) is essential. So at least in the near term, one could make the point that the more input buttons he gives us the better we can train up his AI. I bet he never actually looked at the mock up unfortunately.

An AI will never be perfect. Hell, another human would not choose to adjust these things to my liking at all times, a computer sure as heck never will. I don't want to be contently at war with my car overriding it's dumb (to me) decisions. I just want it to do what I want when I want 'at a push of a button'. Y'all can still have your AI butt heaters if you want, but I don't want to be forced to use them.

 

[earthwormjim]

There are two things I like about the new UI, the fact that messages can now be opened with one button press, and I like the full screen trip section (but hate that the trip card is gone). That's it, everything else is a major regression. Things are far harder to see at a glance now, fonts are smaller, contrast is lower, several on screen elements (arrows) are very difficult to see.

What I cannot stand most is deleting most of the old card system, and thus I cannot have my current trip metrics displayed along with my map, or energy screen. I want to see both at the same time, and the old UI let me do that. Most existing cars let you have trip metrics up on one screen (dashboard), and GPS up on another. The card system replicated this! There was no reason to delete most of the card system, in fact it is still present for some features (wipers, headlights), but made smaller and harder to read...

There are many other bad design elements. The arrows next to the temperature controls are almost impossible to see, deleting the + and - buttons for fan speed. The UI seems to be designed for someone using a tablet in bed, not a car while driving.

 

[Boza]

In V10 I had to turn on HVAC in order to control the back seat warmers. Only a software “engineer” kid could come up with such solution. For them the resources are infinite (“it will be in cloud” mentality). Everyone who understands a little bit of mechanics knows that constantly turning on and off a motor will significantly shorten its life span. So, software engineers do not talk to the hardware engineers. Management?

Instead of fixing it in V11 they made it worse. But, with have Sonic The Hedgehog. Wait, I need controllers to play it! Where would I store them in the car abundant storage space?!

 

[smogne41]

I hope you are correct and it was taken out of context. He is not known of being careful with what he says

Honestly, I don’t understand why they coupled automation with EV. It may make more sense to pursue those things individually. Tesla, as an EV, does not need that automation to be an awesome car and, technology-wise, is way more advanced than the automation efforts. It could sell just fine (maybe even better) with simple manual controls that are on par with the rest of the market. On the other hand, let them play with IA as a separate project. In fact, those are probably different market segments (with some overlap). The current situation alienates people who want a good EV and the whole automation gets in the way (V11 is case in point). Branching the code and having two separate teams work on those with customer’s ability to switch would probably be beneficial. Fedora and Red Hat Enterprise come to mind.

100% this. I love the drive train, the handling, the design of my car. I have given rides and test drives to tones of friends. My friends are all in their 30's and 40's and mostly very tech savvy (many are actually programmers/developers). The screen controls have been a turn-off to almost every single one. The screen was especially a turn-off to my one friend who is a developer at Google .

 

[rickybobbee]

100% this. I love the drive train, the handling, the design of my car. I have given rides and test drives to tones of friends. My friends are all in their 30's and 40's and mostly very tech savvy (many are actually programmers/developers). The screen controls have been a turn-off to almost every single one. The screen was especially a turn-off to my one friend who is a developer at Google .

I am a computer engineer too so I can relate. I get the impression tesla software engineers are too young to drive or the drive a honda civic that doesn't need you to take your eyes off the road to turn on basic functions hidden 3 menus deep. I have 20/15 vision and have to study the numbers and icons way to long on the top menu like the battery % etc.... ridiculous and dangerous. If it's going to be on a screen that's not directly in front on the dash, it should be big enough you can see with your peripheral or at least a quick glance.......not a squint and double take. The lack of a heads up display for at least speedometer is a fail in my opinion

 

[Boza]

That was one of the main reasons I got MS - the instrument cluster. HUD would be even better.

 

[sleepydoc]

I hope you are correct and it was taken out of context. He is not known of being careful with what he says

Honestly, I don’t understand why they coupled automation with EV. It may make more sense to pursue those things individually. Tesla, as an EV, does not need that automation to be an awesome car and, technology-wise, is way more advanced than the automation efforts. It could sell just fine (maybe even better) with simple manual controls that are on par with the rest of the market. On the other hand, let them play with IA as a separate project. In fact, those are probably different market segments (with some overlap). The current situation alienates people who want a good EV and the whole automation gets in the way (V11 is case in point). Branching the code and having two separate teams work on those with customer’s ability to switch would probably be beneficial. Fedora and Red Hat Enterprise come to mind.

What kind of automation do you mean? Self driving or the UI automation?

I think there were two reasons for coupling automation with an EV. One was that both are (viewed as) the future. Elon certainly views them as the future, so it's part of Tesla's vision and culture.

Second, the EV drivetrain naturally lends itself more than an ICE vehicle. Power is easier to modulate and control and much of the hardware will already be in place anyway.

If you look at many other 'mainstream' cars, they are also implementing rudimentary self driving technology. We have a 2020 Forester and it has Subaru's 'EyeSight' system with "lane keep assist" and TACC. On the highway it's very close to auto pilot. It reads the lane lines and holds the car in the lane for you and holds a distance behind the car in front of you. In some ways it's better - zero phantom braking and if you don't tug the steering wheel and it turns itself off you can reenable it without stopping the car. It won't change lanes and can't deal well with exits as well, nor does it follow a route but it's quite nice nonetheless.

 

[Boza]

I am referring to automation driven UI. It is clear that the V11 is oriented toward “no driver” use case - hiding all driving-related controls and focusing on entertainment would make sense. The issue is that such design outpaces the current automation capabilities.
Having two code bases - one with the trend of the V11 and one more “traditional”, driver-focused (even with the games as optional) - and giving customers the option to choose will be better. Yes, it is more costly to maintain two code based but it will preserve the perceived value for the customer base (now, there are quite a few unhappy, resentful people in addition to those who are happy). Proper market segmentation
Additionally, those two use cases require very different skill sets to maintain. Why cobble them together?
There is another benefit. The “traditional” interface will require significant investment only in the beginning; the rest would be mostly maintenance. That will free up long term resources to work on the “automation” UI. Also, there is an established practice of charging more for automation

 

[AustinPowers]

While I agree that the direction of UI is confusing (saying the least), unfortunately, at the moment Tesla does not have a viable competitor. And that is the problem.

Actually, where I live there are several viable competitors. Take the Hyundai Ioniq 5 or the Kia EV6 for example. Similar range. Some specs are better than Model 3/Y, some are equal or a little worse, but overall very comparable. Then the Skoda Enyak, the VW ID.4 and the MG lineup. Or the closest competitor to the Model 3, the Nio ET5 (which even looks quite a bit like a Model 3). And for a higher budget there's the Lucid Air, or the EQS. And soon the EQE. Plus several planned EVs that will launch this year by VW, BMW, Cupra and Peugeot/Opel.

All I'm saying is, Tesla isn't the only choice anymore. Three years ago that was very different.

 

[rjpjnk]

I don’t think any of those cars is an actual competitor in any serious sense, but at least they exist, and that is the first step. Hopefully real competition is coming soon (3-5 years?).

Edit: Now if Apple gets in on things, they could potentially be a much more serious competitor than the legacy auto makers. See here

Morgan Stanley: Apple Car represents ‘more formidable’ competition to Tesla than legacy automakers - 9to5Mac

A report yesterday from Reuters offered a handful of new details on Apple’s ever-changing plans to build an electric car....

9to5mac.com

 

[Dan D.]

Which is why it’s important to pay attention in the UI design to the driving elements. Sometimes you need them when you need them.

I agree with this statement.

I've driven on ice covered roads in Ontario where you're going at an angle just to avoid the hump of the road pushing you off. It's two hands on the wheel, constantly flicking fluid on the wipers to clear another slush spray from oncoming cars. You reach down to where you know the climate dials are without even looking, and can feel the adjustments clicking. Whatever you need is available by touch. Your night vision needs to be unaffected so you don't dare turn on any cabin lights or look at bright displays.

Could you pull over? Not in the middle of nowhere. You have to keep going. Is this extreme driving, not really. I can think of other times with fog, heavy driving rain, etc. The last thing I want to do is look at a touchscreen and make several motions, including having to look at it. There's certainly a lack of network coverage in places so there's no chance at those voice activation features.

If other people never encounter situations like this, then they would not understand, but perhaps they can try to.

 

[Boza]

Just today I had to adjust the distance of the adaptive cruise control (or whatever Tesla calls it). Just reached, turned the knob, got a haptic feedback (clicks) and visual confirmation in the instrument cluster screen. Didn’t even look at the screen, swiped, said a chant or turned on the HVAC. Quick, efficient, simple.

 

[john5520]

I've driven on ice covered roads in Ontario where you're going at an angle just to avoid the hump of the road pushing you off. It's two hands on the wheel, constantly flicking fluid on the wipers to clear another slush spray from oncoming cars. You reach down to where you know the climate dials are without even looking, and can feel the adjustments clicking. Whatever you need is available by touch. Your night vision needs to be unaffected so you don't dare turn on any cabin lights or look at bright displays.

Could you pull over? Not in the middle of nowhere. You have to keep going. Is this extreme driving, not really. I can think of other times with fog, heavy driving rain, etc. The last thing I want to do is look at a touchscreen and make several motions, including having to look at it. There's certainly a lack of network coverage in places so there's no chance at those voice activation features.

If other people never encounter situations like this, then they would not understand, but perhaps they can try to.

Sudden heavy rain showers I can relate to and have had no issues with the wipers. Tapping the stalk is the quickest way to get them going, then setting the speed is just one more selection in the GUI.

I've also driven in fog and never felt the need to suddenly turn on a climate control setting. I do keep the system on auto, but even in my last car where I kept it off, I've never had to suddenly fumble around for it even when defogging was necessary. While driving, if the window started to get hazy, I'd turn them on well before it got to the point I couldn't see.

Now I don't live up north, so maybe these things happen suddenly and if that's the case, then wouldn't setting up the climate controls ahead of time be better? Even if there were dedicated defroster buttons, wouldn't they still take a while to actually defrost?

 

[AustinPowers]

I am referring to automation driven UI. It is clear that the V11 is oriented toward “no driver” use case - hiding all driving-related controls and focusing on entertainment would make sense. The issue is that such design outpaces the current automation capabilities.
Having two code bases - one with the trend of the V11 and one more “traditional”, driver-focused (even with the games as optional) - and giving customers the option to choose will be better. Yes, it is more costly to maintain two code based but it will preserve the perceived value for the customer base (now, there are quite a few unhappy, resentful people in addition to those who are happy). Proper market segmentation
Additionally, those two use cases require very different skill sets to maintain. Why cobble them together?
There is another benefit. The “traditional” interface will require significant investment only in the beginning; the rest would be mostly maintenance. That will free up long term resources to work on the “automation” UI. Also, there is an established practice of charging more for automation

I don’t think any of those cars is an actual competitor in any serious sense, but at least they exist, and that is the first step. Hopefully real competition is coming soon (3-5 years?).

Have you actually driven any of the current competitors?
I have, and from that experience I can attest that several of the ones I mentioned are serious alternatives, and right now, not in 3-5 years. I'm meeting other BEV drivers at charging stations all the time and we share experiences and often do direct comparisons. My brother-in-law works for an independent car dealership and he often gets BEVs for test drives.
Seriously, the competition is here now. And since Tesla is now opening the SC network for other brands, even the former exclusive charging network is no major draw any longer. Plus with this new user unfriendly UI they have lost a lot of goodwill with many owners. Not one of the ones I know is happy with the changes that came with V11, especially as there appear to be only downsides and no improvements. On the contrary, many positive aspects of the UI that have been perfected over the last few years have now needlessly been thrown out of the window and replaced with cumbersome handling, including needing several taps for what used to be direct access. Let alone the deleted trip info tab that everyone I know had on as standard all the time for knowing the basics of your trip like time you have been driving, distance, current energy usage. All important pieces of information that are now hidden in the vehicle submenu (where they take up the entire main screen so are no use while driving). Honestly, what single good reason can there have been for removing the trip info tab? It hasn't even been replaced with anything, the lower left corner now is simply dead real estate.

 

[Boza]

I happen to drive in San Francisco and in the Sierra Nevada. I can see how people could be somewhat OK with defogger controls in San Francisco (during the dry season). When I drive up in the mountain the controls are not only unusable, they are dangerous. You can have your windshield fogged up in a matter of seconds - faster than it would take you to find the controls.
Night navigation, however, is a problem no matter where I drive. I gave up on using Tesla navigation and installed my old iPhone cradle for Waze/Google maps. I am looking for the app that does EV planning so that I can see the charging stations and properly plan the trip - forgot the name.
So, now I have Tesla with 3 screens

 

[sleepydoc]

Plus with this new user unfriendly UI they have lost a lot of goodwill with many owners. Not one of the ones I know is happy with the changes that came with V11, especially as there appear to be only downsides and no improvements. On the contrary, many positive aspects of the UI that have been perfected over the last few years have now needlessly been thrown out of the window

Yes! One of Tesla's biggest assets has been their passionate and loyal base of drivers happily telling anyone who will listen how much they love their car. I was one of them until a few weeks ago. Now if someone asks my response will be much more muted.

The biggest frustration of V11 is how it was totally unnecessary. Rather than make improvements, they threw the proverbial baby out with the bath water.

 

[sleepydoc]

I am looking for the app that does EV planning so that I can see the charging stations and properly plan the trip - forgot the name.

Try A Better Route Planner

 

[Boza]

Yes! One of Tesla's biggest assets has been their passionate and loyal base of drivers happily telling anyone who will listen how much they love their car. I was one of them until a few weeks ago. Now if someone asks my response will be much more muted.

The biggest frustration of V11 is how it was totally unnecessary. Rather than make improvements, they threw the proverbial baby out with the bath water.

You forgot the light show and Sonic

 

[Akikiki]

We got a new UI? wow, cool.

 

[john5520]

I happen to drive in San Francisco and in the Sierra Nevada. I can see how people could be somewhat OK with defogger controls in San Francisco (during the dry season). When I drive up in the mountain the controls are not only unusable, they are dangerous. You can have your windshield fogged up in a matter of seconds - faster than it would take you to find the controls.
Night navigation, however, is a problem no matter where I drive. I gave up on using Tesla navigation and installed my old iPhone cradle for Waze/Google maps. I am looking for the app that does EV planning so that I can see the charging stations and properly plan the trip - forgot the name.
So, now I have Tesla with 3 screens

So when your windshield fogged up in seconds, how long did the defogger take to clear them? Also, what's the matter with Tesla's navigation at night?