|
Ola posted:I don't understand why Toyota are so reluctant about battery electrics when they've been spearheading hybrids for so long. They were early adopters, actually. https://en.wikipedia.org/wiki/Toyota_RAV4_EV#First_generation
|
#
?
Oct 19, 2016 11:39
|
|
|
|
| # ? May 16, 2024 13:45 |
|
|
CharlesM posted:They were early adopters, actually. It's the lack of any effort beyond the now out of production, US-only RAV4 I term as reluctance. Instead, Toyota are going on about hydrogen vehicles which make less and less sense the more range battery EVs get. Ola fucked around with this message at 12:27 on Oct 19, 2016 |
|
#
?
Oct 19, 2016 12:20
|
|
|
The hydrogen fuel cell is a technological trap pushed to kill BEVs, and Toyota of all people† should know this. †Corporations are people, my friend.
|
|
#
?
Oct 19, 2016 12:23
|
|
|
CharlesM posted:They were early adopters, actually. That was just a California mandate. Ford and GM also made electric Rangers and S-10s around that time.
|
|
#
?
Oct 19, 2016 13:21
|
|
|
Ola posted:It's the lack of any effort beyond the now out of production, US-only RAV4 I term as reluctance. Instead, Toyota are going on about hydrogen vehicles which make less and less sense the more range battery EVs get. Plus a typical automotive fuel cell system has way more stuff to potentially go wrong than a typical battery pack - not a good thing when one of your major selling points is a reputation for reliability.
|
|
#
?
Oct 19, 2016 14:30
|
|
|
Ola posted:I don't understand why Toyota are so reluctant about battery electrics when they've been spearheading hybrids for so long. It's almost as if the biggest and most successful carmaker in the world knows something...
|
|
#
?
Oct 19, 2016 14:49
|
|
|
Throatwarbler posted:It's almost as if the biggest and most successful carmaker in the world knows something... Maybe they know something that Tesla, VW, BMW, Mercedes, Nissan, Mitsubishi, Hyundai, Peugeot, Renault, Citroën and GM doesn't. And maybe battery electrics aren't ideal for Japanese infrastructure. Or maybe Toyota are betting on the wrong horse.
|
|
#
?
Oct 19, 2016 14:55
|
|
|
Throatwarbler posted:It's almost as if the biggest and most successful carmaker in the world knows something... Wasn't GM the biggest and most successful carmaker at one point?
|
|
#
?
Oct 19, 2016 14:59
|
|
|
someone in the early 1970s posted:It's almost as if the biggest and most successful carmakers in the world know something that the Japanese carmakers don't.... edit: beaten
|
|
#
?
Oct 19, 2016 15:25
|
|
|
Ola posted:Maybe they know something that Tesla, VW, BMW, Mercedes, Nissan, Mitsubishi, Hyundai, Peugeot, Renault, Citroën and GM doesn't. And maybe battery electrics aren't ideal for Japanese infrastructure. Or maybe Toyota are betting on the wrong horse. As I recall, they've been sticking with nickel-metal-hydride batteries in the non-plug-in Prius because when the 2016 Prius was being designed, lithium ion was still relatively unproven in automotive applications (they didn't want to risk their reputation for reliability to lose a relatively small amount of weight). Except that at this point, lithium ion batteries are much closer to proven technology than fuel cells. Plus a Model S with a 100kWh pack has at least as much range as any hydrogen-powered car ever built. Subjunctive posted:Wasn't GM the biggest and most successful carmaker at one point? You mean back before Americans started giving a crap about fuel economy?
|
|
#
?
Oct 19, 2016 15:55
|
|
|
Cockmaster posted:As I recall, they've been sticking with nickel-metal-hydride batteries in the non-plug-in Prius because when the 2016 Prius was being designed, lithium ion was still relatively unproven in automotive applications (they didn't want to risk their reputation for reliability to lose a relatively small amount of weight). Except that at this point, lithium ion batteries are much closer to proven technology than fuel cells. Plus a Model S with a 100kWh pack has at least as much range as any hydrogen-powered car ever built. I just want to preserve this post.
|
|
#
?
Oct 19, 2016 16:24
|
|
|
Ola posted:Maybe they know something that Tesla, VW, BMW, Mercedes, Nissan, Mitsubishi, Hyundai, Peugeot, Renault, Citroën and GM doesn't. And maybe battery electrics aren't ideal for Japanese infrastructure. Or maybe Toyota are betting on the wrong horse. By that standard Toyota (and Honda) makes battery electric EVs too, to satisfy Californian and other jurisdictions ridiculous politically motivated requirements that they do. Or I guess maybe there really is some kind of game changing passenger car/motive technology that only GM and VW are aware of and Toyota isn't. There sure is a lot of precedent for that happening.
|
|
#
?
Oct 19, 2016 16:25
|
|
|
EgonSpengler posted:The hybrid cycle doesn't really degrade as the battery ages. Old prius's (priora?) are one of the most reliable vehicles on the road, since the electric motor and regenerative braking takes so much strain off the brake pads, plus electric motors are relatively maintenances free.
|
|
#
?
Oct 19, 2016 17:27
|
|
|
The problem Toyota has is the margins (if you actually count overhead) are small to non existent on EV. A fuel cell though can be profitable especially if you are the only manufacture building them. Retrofitting gas stations with hydrogen pumps is not as bad as it seems especially since the tanks need to be periodically replaced anyway. This is a long term strategy. If it doesn't work you redesign the car to handle the extra weight and put a battery in it.
|
|
#
?
Oct 19, 2016 17:45
|
|
|
The Tesla product unveiling scheduled for the 17th was pushed back to today, 5PM Pacific time: https://twitter.com/elonmusk/status/788794291900469248
|
|
#
?
Oct 19, 2016 21:54
|
|
|
I hope it's the Model Y or whatever the next car in the lineup is called. I think it's something boring with solar panels and domestic batteries.
|
|
#
?
Oct 19, 2016 22:01
|
|
|
So I guess there's no livestream according to random twitter Internet person who heard it from Reddit. Why even publically announce it then if it's an exclusive presentation? 
|
|
#
?
Oct 20, 2016 01:14
|
|
|
https://www.tesla.com/blog/all-tesla-cars-being-produced-now-have-full-self-driving-hardware So its kind of a lame roll out. I'm glad I have autopilot 1.0 right now and we will just consider trading in next year or when ever full autonomous features become available.
|
|
#
?
Oct 20, 2016 01:17
|
|
|
Reddit seems to have the announcement up. Autopilot 2.0 hardware in every new Tesla produced from today, allegedly allowing full autonomous driving when the software is ready for it. There will be a regression in functionality compared to the current Autopilot until the system is fully trained. https://m.reddit.com/r/teslamotors/comments/58dprx/tesla_announces_full_autonomy_hardware_ap_20_on/
|
|
#
?
Oct 20, 2016 01:20
|
|
|
gently caress me. I what trade-in I can get.
|
|
#
?
Oct 20, 2016 01:23
|
|
|
That's garbage. The "ship it broken, patch it in production" of the automotive world. The only way to verify that the hardware supports "full autonomous driving" is to build the autonomous driving software and deploy it on the hardware package and test the whole system. And I mean like evaluate it in the real world for millions of miles, not on a Tesla test track. Until you've done that, there is zero reason to believe that the cars' hardware package will be suitable for level-4 autonomy or whatever they're claiming a few years down the line. Maybe they've put in enough hardware! Or maybe they will discover in two years "whoops, this new computer is 40 times more powerful, but we really needed something 400 times more powerful." Either (1) they have fully functional, fully autonomous, tested and fully validated cars that they just aren't releasing yet for some reason, or (2) they aren't even close to full autonomy but Elon needed something to announce so the engineers shrugged and said "sure, I guess we've put in enough cameras to support it, we'll solve the rest in software." Which one is more likely?
|
|
#
?
Oct 20, 2016 01:29
|
|
|
Sagebrush posted:
Option 2 - 100%. It's the same bullshit that went down with 8.0 being a firey trashheap
|
|
#
?
Oct 20, 2016 01:32
|
|
|
From a sensor-input point of view we know conclusively that driving can be done with a pair of stereo cameras a few inches apart and the ability to rotate about 100 degrees in each direction, so I'm not worried about "well, we think we have enough cameras" at least.
|
|
#
?
Oct 20, 2016 01:43
|
|
|
Maybe I am too pessimistic but my initial impression is that they know they are in trouble with the Autopilot (see accidents, Mobileye no longer partnering, Germany wagging their finger over the name of the tech, etc.) and this announcement allows them to turn off all AP technology in new cars while saving face. Either that or the conventional automotive industry is ruined, heh. eames fucked around with this message at 01:49 on Oct 20, 2016 |
|
#
?
Oct 20, 2016 01:44
|
|
|
Apparently they already have some software for it? https://twitter.com/elonmusk/status/788902908175618049
|
|
#
?
Oct 20, 2016 01:46
|
|
|
This is a pretty bold piece of the license. Anyone know of any other similar clause in a car manufacturer's license?
|
|
#
?
Oct 20, 2016 01:48
|
|
|
Details from the official Tesla site:Tesla posted:From the design studio. (same for both Model S and Model X) Also interesting: http://www.theverge.com/2016/10/19/13341194/tesla-autopilot-shadow-mode-autonomous-regulations Basically Autopilot 2.0 will run in the background ("Shadowmode") and log what it would do but not take action. When fatal accidents happen Tesla will be able to use these logs to show that this accident could have been avoided with their fully autonomous solution. eames fucked around with this message at 02:22 on Oct 20, 2016 |
|
#
?
Oct 20, 2016 02:10
|
|
|
eames posted:Basically Autopilot 2.0 will run in the background ("Shadowmode") and log what it would do but not take action. When fatal accidents happen Tesla will be able to use these logs to show that this accident could have been avoided with their fully autonomous solution. Autopilot 1.0 runs like this already. It's always recording driving data, even on cars that have the autopilot hardware but not the software. That's where they get their lane position graphs for driver vs autopilot, and they talk about it in the 8.0 release notes for better emergency braking data.
|
|
#
?
Oct 20, 2016 04:03
|
|
|
Subjunctive posted:This is a pretty bold piece of the license. Disruptive™
|
|
#
?
Oct 20, 2016 04:07
|
|
|
Then again, perhaps Toyota does know what it's doing? http://m.phys.org/news/2016-10-green-hydrogen-production-algal-proteins.html
|
|
#
?
Oct 20, 2016 12:18
|
|
|
Finger Prince posted:Then again, perhaps Toyota does know what it's doing? I don't see their name mentioned there, nor can I remember Toyota being involved with any innovative hydrogen production project. They're not even committing themselves to any major infrastructure project, instead they've published the standards to their fill port and just hope others will build the stations. If I turn out to be wrong in being skeptical about Toyota's hydrogen vision, I will welcome the surprise.
|
|
#
?
Oct 20, 2016 12:25
|
|
|
It seems like Tesla is pretty far along with the software to match the hardware: https://www.tesla.com/videos/full-self-driving-hardware-all-tesla-cars
|
|
#
?
Oct 20, 2016 12:48
|
|
|
Finger Prince posted:Then again, perhaps Toyota does know what it's doing? Making business decisions contingent on technological breakthroughs that have not yet been made is generally a bad move. To be clear, the problem is not that they are pursuing fuel cells; it is that they are pursuing them to the exclusion of battery electric vehicles. Toyota could have owned the BEV market the same as they owned the hybrid market. Platystemon fucked around with this message at 13:26 on Oct 20, 2016 |
|
#
?
Oct 20, 2016 12:55
|
|
|
Sagebrush posted:That's garbage. The "ship it broken, patch it in production" of the automotive world. Exactly what I've been saying. It needs to be type approved, certified, and regulated. Tesla is intentionally shipping beta software to the public that could/will kill people. If it only endangered the Tesla driver I wouldn't really care, but that's simply not the case.
|
|
#
?
Oct 20, 2016 13:22
|
|
|
Ola posted:I don't see their name mentioned there, nor can I remember Toyota being involved with any innovative hydrogen production project. They're not even committing themselves to any major infrastructure project, instead they've published the standards to their fill port and just hope others will build the stations. If I turn out to be wrong in being skeptical about Toyota's hydrogen vision, I will welcome the surprise. No, they aren't mentioned, though if they're developing fuel cells at the expense of jumping on the battery bandwagon and relying on hybrids as a stop-gap, it stands to reason that they would likely fund this kind of research, or at the very least follow it very closely. I'm not suggesting the article represents a Toyota R&D project, more that the viability of hydrogen fuel production implied by this kind of research suggests that Toyota might not be completely daft by pursuing the technology.
|
|
#
?
Oct 20, 2016 13:22
|
|
|
blugu64 posted:Exactly what I've been saying. It needs to be type approved, certified, and regulated. Tesla is intentionally shipping beta software to the public that could/will kill people. If it only endangered the Tesla driver I wouldn't really care, but that's simply not the case. People die in light aircraft accidents all the time (loss of control, particularly approach stalls,) that could be prevented by simple add-on avionics (AOA indicator) that are not permissible to be installed in certified aircraft because they are not "approved, certified, and regulated." Government needs to participate in this process, but not come down on it like a ton of bricks. As always, the answer is somewhere in the middle.
|
|
#
?
Oct 20, 2016 13:42
|
|
|
Subjunctive posted:It seems like Tesla is pretty far along with the software to match the hardware: https://www.tesla.com/videos/full-self-driving-hardware-all-tesla-cars It was claimed that the big reveal would be "unexpected by most". I doubt anyone would have expected them to be this far along, what with Tesla not having shown off any details of their R&D into non-highway autonomy until now.
|
|
#
?
Oct 20, 2016 15:30
|
|
|
MrYenko posted:People die in light aircraft accidents all the time (loss of control, particularly approach stalls,) that could be prevented by simple add-on avionics (AOA indicator) that are not permissible to be installed in certified aircraft because they are not "approved, certified, and regulated." We also continue to give a driver's license to 16 year old kids with minimal (required) training, and geriatrics with minimal (if any) retesting. The bar for Autopilot to be safer than at least some human drivers is very low.
|
|
#
?
Oct 20, 2016 15:37
|
|
|
I think tying a rope to the steering wheel and putting a brick on the gas pedal is better then half of AI posters driving skills.
|
|
#
?
Oct 20, 2016 16:00
|
|
|
|
| # ? May 16, 2024 13:45 |
|
|
blugu64 posted:Exactly what I've been saying. It needs to be type approved, certified, and regulated. Tesla is intentionally shipping beta software to the public that could/will kill people. If it only endangered the Tesla driver I wouldn't really care, but that's simply not the case. New laws and regulation will most likely be needed before autonomous mode can be allowed to turn on. But that is no reason to not install the hardware. As eames and Subjunctive mentioned, it's basically necessary to test the system and gather data before we can make a decision about needed laws and regulation. Installing and announcing the hardware even if it can't be used currently is not a problem. The biggest risk is that Tesla is opening themselves to a class action lawsuit if it turns out the hardware isn't adequate and can't be upgraded for free.
|
|
#
?
Oct 20, 2016 16:33
|
|
