Featured Lexus's First Electric Car Is Disappointing. The Lexus RZ450e Is Not Good and Here's Why.

I'm thinking the uses of the information are a bit different. I'm curious about my MPG, but it's mostly a detached, how's it been for the last few months kind of curiosity.

I have a much more active interest in whether I will get to my destination n miles away before the car conks out.

If I've made a control setting that reduces my chance of getting there, I want to know that. And if I see that I might not make it, and I try some other control setting to try to extend the range, I want to see if I've done enough yet. I would rather see something conservative and responsive than something prettified and delayed.

A closer analogy in the hybrid Prius might be the fuel distance-to-empty reading. I rarely have that turned on, though (I guess I don't usually have enough range anxiety in the hybrid Prius), so I haven't noticed how 'bouncy' it does or doesn't seem to be under changes in load.

Me, I'd like a graphical display showing predicted range as a confidence interval. Some drivers might need the concept explained. But it's one worth explaining, and carries over to a lot of things in life.

 

Prius camper makes some interesting assertions. I stop watching videos when it becomes obvious that the person is going on and on about flaws that don't exist. PC wont even discuss his position on why he's convinced that Toyota is going to fail even though they have introduced hybrid versions of virtually all of their models this year and also are already delivering two distinct BEV models to customers. If you stop and think about it, the Toyota hybrids will run no matter where you live, and you can even make your own bio-gas. The BEVs are relying on the commercial grid to remain up when they travel in remote areas. I've seen videos of Mongolia where the streets are full of Prius The main street of town literally packed with hybrids. There are no Teslas visible on the streets of Mongolia. No hummers. No cyber trucks. But there was a Google vehicle scooting down the road taking 360 degree panoramic pictures.

In answer to his question: No, I have not mass produced a car, but I've worked in R&D teams. I've run a small production line, enough to understand when it's better to buy what you need and when you should build it. And I've read a lot about the history of the zero emissions movements over the last 60 years. Somewhere in my shop I still have the plans for making an analog speed control utilizing pulse width modulation. I also had located the sources for an air force cargo plane starter motor. Never built that one. The batteries were not up to the task and were replaced frequently in the DIY electric conversions of the 1970s. I could not afford to put 700 dollars worth of batteries into a small car every year or two.

 

That's great for Mongolia, and maybe Toyota will downsize to serve only areas like that. But the US, Canada, Europe and other regions are switching to EVs. Toyota has, as you've stated, delivered two distinct EVs to these areas. They've also put a hard cap on bx4z sales of 10,000 for the US. Which is understandable, as it's not competitive.

 

He complained of constant warnings. Another reviewer complained of constant warnings from the driver monitoring system in the new Prius Prime. Maybe the system is throwing false positives. Maybe they were poor drivers. The thing is that the alerts appear to be on the nuisance level when they can be designed to not be. When at that level, people turn them off, or they become a distraction to the driver.

I know Eyesight is a safer driver than I am, but the warnings have never had me thinking of turning it off as these reviewers have voiced.

It worked for the Bolt. Would have worked for Toyota if they had started back when the Bolt came out. Could have still worked if their hoped for predictions had been true.

In the Subaru, and what I recall in other models, the DTE can drop, but rarely go up. Subaru only showed a value to the nearest tens place. That will flatten any wiggle to the number.

The complaint in the review was more about the fluctuations being anxiety inducing. If actions like turning the headlights on and off can cause the range value to jump around, I can see how that would be bothersome to those new the EVs. The displayed range can have such common, light energy draws factored in. Then the driver only sees the big impacts of speed and climate control.

Toyota has all but admitted their strategy with BEVs has been wrong.
Toyota changing strategy? | PriusChat
Toyota scambling to change EV plan | PriusChat

The management team Toyota had assembled to change the EV plan was not something they do for regular market evaluations and plan adjustments. They call it a Business Revolution(BR) group, and implement it for major changes. Around 30 hybrid, PHEV, and BEV projects were cancelled as part of this action. They're considering axing a platform, not a model, a platform, that is just three years old.

The short of it is that Toyota planned on a slower adoption of BEVs. The eTNGA platform was developed for that projection. It could be assembled on existing production lines along side ICE models. Which would have been a fine plan, if Toyota's projections had been right. They weren't, and EV sales a climbing much faster. The car companies that invested in dedicated EV platforms and production lines now have a cost advantage over Toyota for this level of EV demand. Toyota is worried about Tesla's production efficiency.

The bZ4X siblings were already going into production when Toyota realized they weren't going to be competitive in BEVs. Other BEVs using the eTNGA, like the Crown EV, were cancelled. The complaint about using hybrid components misses the point, in that their real issue was in Toyota's packaging. Looking at Ford's noodle nest of coolant lines in the Mach-E doesn't generate a good impression, but at least it has a frunk.

Please pray tell how one can do this in town, and whether the car will require modifications to run on such fuel? If doing such is acceptable in remote areas, why isn't installing small scale wind or solar? They are probably about as equally efficient.

How long has the Prius, new and used, been imported to Mongolia? EVs are just starting to get to market there.
https://www.topspeed.com/cars/car-news/check-out-this-funky-little-mongolian-ev/
https://cnevpost.com/2022/06/08/tesla-opens-its-first-store-in-inner-mongolia-ramping-up-efforts-in-north-china/
BYD, and surely other Chinese brands, have models available there too.

 

I wouldn't mind seeing way more creativity being expended toward ways of showing this stuff. I mean, the cars are coming with high pixel density color displays now.

You could show the expected range as a range (a range of range? ok, a confidence interval) displayed on a distance axis—a bar with a color gradient, stretching from a low to a high estimate, color-keyed to what corresponds to higher HVAC loads, higher speeds, etc. It would start out wide if the ECU had just been reset, and get gradually narrower as the ECU has more stats on your typical driving.

Superimpose over that bar a narrower one that actively slides around over the wider one, as you change HVAC settings and accessories and whatnot.

Is it crazy to think regular people could learn to grasp what a display like that was showing them? I don't know; I mean, software like Adobe Illustrator and Corel Draw came out and suddenly a lot of people got pretty good at drawing things with Bézier curves, without having to be told what they were, or ever seeing a Bernstein polynomial.

UI design doesn't always have to be about dumbing it down; sometimes a well-designed one smartens up the user.

 

If turning the headlights on and off has any effect on range, Toyota is doing it wrong. A quick search shows that typical LED highlights consume around 45w on the high end. So, doing math type stuff, we're looking at .45 kWh of consumption over 10 hours.

How that can take tens of miles off the battery is a mystery. If I'm doing my math wrong, please let me know.

 

It's just data and the more options you have to interact with that data in an intelligent way is a software problem, which Toyota-Lexus is clearly doing it wrong compared to all other EV car makers.

In the case of this particular vehicle it's literally telling you there 5 miles more range than the day before when you turned the car off, despite not charging the car since then. That's not a valid option for interpreting data and points to the likelihood of far bigger problems with the vehicle design, especially if you're running out of charge and trying to get to a charging station.

What's more turning on your headlights is such a minor use of power it shouldn't impact vehicle range unless the people designing that data output are incompetent because Toyota-Lexus refuses to invest in a legitimate software engineering team.

 

It could very well be the result of a perfectly valid recomputation on the most recent usage statistics, looking dumb because it's presented in a dumb way.

The reality that can't be escaped is that this is an estimation that requires statistical math and comes out with a confidence interval, and it can't be anything but that. That's more information than one single number, and trying to shoehorn it through an interface that has to pick one number out of that interval to be displayed as if it came down from God is, IMO, the crux of the problem. Use a better interface and let the driver see what the limits of the estimation are.

No such thing as a use of power that doesn't impact vehicle range. Some uses are small and have small impacts. Use a good enough interface, and you can show the driver simultaneously the effect and how small it is. Then you are neither over- nor under-emphasizing it, and you're not arbitrarily hiding it because of your ideas of what the driver doesn't want to see.

 

All these improvements are classic software-based ship it then fix it with updates scenario... It's just the growing pains of designing a new system that needs lots of refinement, which is why Tesla does so well because its design team got started in software then learned how to build hardware, whereas Toyota got started in hardware and still has massive amounts to learn when it comes to software, which the new leader of Toyota admitted in his recent press event about how EV is hard and can't rapidly scale... Of course all the other automakers are doing exactly what Toyota claims can't be done at a major loss in revenue because the sooner they get large numbers of EVs on the road, the sooner they become profitable. Meanwhile Toyota is the kid with Dunce hat on who's been told by consumers to go sit in the corner of the room facing the wall.

 

I can see you thinking that about the car (whether or not I agree); I'm not sure I can see the connection to the post of mine you just quoted. The rethinking I would advocate for the user interface goes deeper than any likely "fix it with updates" "improvement".

The basic complaint arises because they took something that's inherently, inescapably, an estimate made with statistical methods, and presented it as a single number, and not everyone likes the way that single number sometimes behaves (with some even calling it "straight up lazy or incompetent').

The trouble comes in because the whole idea of reducing that statistical estimation result to one single number goes against the Einsteinian call to make everything as simple as possible but no simpler.

Folks have worked out fantastic algorithms for doing the stats math you can use for doing that kind of estimation, determining confidence intervals, and so on. Somebody with a good sense for UI design could probably think of a way to display that, making it as simple as possible but no simpler.

Once somebody wants to make something simpler than that, the good algorithms dry up, leaving only different bad ones concocted by people with different ideas of the "right" way to oversimplify something.

Some reviewer may think it looks wrong to have the displayed number rise when the estimate does.

So you can duct-tape on a new rule saying the displayed number can go down when the estimate does, but stays down when the estimate rises.

But you might not want the error between the estimate and the displayed number to accumulate forever. So maybe you duct-tape on yet another rule that says you bring the display back up to the estimate on the next charge, because that's a time it can rise without somebody thinking it looks wrong.

Me, I'd as soon skip the duct tape. If I'm driving, and I've made a change that extends the estimate, I'd rather not wait around to see by how much. Of course I don't really want that to be oversimplified and shown as a single number from God either. I'd rather just see the effect on the interval around the estimate.

 

In Prius Prime, the complaint was about the traffic-jam assist, and the warning could be turned off. I don't know what the complaint was about in the Lexus case, but I am guessing that it was mainly because he didn't know how to use the system. It is a new learning curve for ADA systems—not your grandfather's cars.

 

My point was there's nothing wrong with the data that's being gathered. The problem is in how that data is processed and displayed and when it comes to major corporations like Toyota trying to control everything for profit over quality the job they need to do to address this is their job and no one else's...

But in truth one of the most revolutionary things in computer technology is the work of crowdsourcing improvements in firmware and software as well as offering bounties for anyone who can get past the security challenges.

Had Toyota open-sourced Battery ECU firmware to improve its performance quality rather than improve Toyota's patent rights and profitability we'd all be driving plugin Prius modded cars with massive battery packs that we built ourselves...

My favorite example: I remember back in the day just as the first Apple iPod music player came out their competitor Archos Jukebox was selling an MP3 player that I was huge fan of. And when it was clear to them that Apple was going to wipe out their market share they open sourced the firmware their music player operated on and within months there was a huge community that were constantly adding new services and settings, even adding video games that you could play on the display screen.

That was an all volunteer effort that made Archos superior to Apple's product by every measure, except it wasn't as cool as an iPod, which had a highly profitable iTunes library that made a fortune for Apple.

 

The reviewer was specifically calling out the driver monitoring system of the the TJA.Yes, you can turn it off. I'm sure you can turn off the other ADAS warnings systems too. Doing so means turning off at least a portion of the the assistant functions.

It could be the user in this Lexus review. That possibility doesn't negate the one of it being the system itself. I brought up the TJA review cause another report nuisance warnings adds a little more weight to this being a system issue.

 

Consumer Reports did not find anything negative on the Prius TSS 3.0 ADA. However, they haven't tested the TJA on Prius Prime.

As far as I know, Toyota allows you to turn off the driver monitor without turning off the TJA or other ADA systems.

The driver monitor may not work with some eyewear or with bad driving posture. The infrared camera is located right behind the steering wheel—the dark oblong window.

 

The Chevy Bolt already has something like this.
You can choose to see just a single "miles of charge" number on the display.
Or you can choose to see a more advanced estimate that gives you 3 numbers, a low, medium and high estimate of miles of charge.
So you might see 200 miles as the single number but something like 150 to 200 to 235 miles in the more advanced display mode.
I'm unclear on exactly what data they use to produce the estimates, but it does change if you turn on the A/C for example.

In the Tesla the normal display just shows you a single number as miles or % of battery.
But in the more advanced energy screen it shows a lot of data if you have a GPS destination setup.
You see a graph of battery SOC remaining and a plot of how it predicts it will change over the trip including elevation changes, etc.
It gives you a breakdown of how much energy goes into this in ~5 categories including speed, HVAC, hills, aggressive driving and more.
It also give tips on what you can do to change the future part of the plot while showing you both your history and prediction so you know how close you will be to arriving with whatever battery remaining you want to have

Mike

 

Those all sound like good ideas; I wonder how many of them Tesla took to the patent office....

 

Probably none. Tesla has a bunch of people reinventing the wheel. If they patent an idea they will have to do patent searches of existing patents and prior art. It's hard to declare a mundane 8 way valve as innovative and world changing if you have to mention the basic concept was first patented in the 1800s.

 

I was thinking more of the software ideas, like "break down forecast battery use according to speed, HVAC, hills, etc.", or "offer tips on what changes will improve the forecast", and so on.

I consider those fairly obvious ideas and therefore nonpatentable, but the history of software patents is that patent lawyers love filing applications for such things, and patent examiners don't have a good record of rejecting them, and then you get absurd situations where everyone knows what the obvious ideas are but nobody else can use them.

... I'm not averse to patents in general, but I've got pretty mixed feelings about software ones ....