General Discussion: 2018 Investor Roundtable

[schonelucht]

Segmenting our small deliverables doesn't fit with how the NM stuff works (other than totally independent functions like rain sense) as they can't necessarily maintain any specific functionality while working on the whole.

Elon had no qualms about predicting piece meal delivery of FSD features. Given that he knows a lot more than us on the actual strategy used to implement FSD, the only logical conclusion is that piecing out small deliverables is in fact possible. Also. They actually did it when they introduced automatic wipers on the AP2.5 hardware.

 

[schonelucht]

Of course your fears could be valid, but my stab of Occam's Razor indicates that Tesla keeping information from competitors is the simplest explanation.

That doesn't match with the CEO being very vocal about their progress, nor with releasing purported video of their bleeding edge state of development, nor with writing a 3 page letter stating that the report isn't indicative of progress, nor with announcing specific timelines, nor with having the head of development and CEO give talks and keynotes on AI conferences and certainly not with including the hardware well in advance so every single manufacturer has more than plenty time to study those. Occam's Razor suggests that the advantage of converting car owners to FSD buyers is much larger (both in brand image and in financial terms) than keeping progress secret for their competitors.

The same for some other automakers who are dealing similarly with California's request.

Those other car maker's don't have their primary (and in the case of Tesla only) R&D facilities in California.

 

[aubreymcfato]

MODERATOR COMMENT:

After studying the links provided in post #1810 above, it seems prudent to alert the date-unwary that all the references therein are from 2016 and the beginning of 2017. Without poring through the entirety of the reddit post, it appears none of its participants caught the seeming time-warp.

Mmm, in the Tesla document I read:
«from November 30, 2016 to December 1, 2017»

This would leave room for a NY to SF autonomous road trip, but only in December.

 

[Reciprocity]

Of course your fears could be valid, but my stab of Occam's Razor indicates that Tesla keeping information from competitors is the simplest explanation. The same for some other automakers who are dealing similarly with California's request.

In any event, being first shouldn't really matter. Governments are undoubtedly going to drag out the time required for complete approval of fully autonomous driving. By the time the necessary regulations become implemented, I expect that many companies will be able to comply.

Curt as usually is the most reasonable and probably correct here. Is it even conceivable that Tesla hasn't tested anything since that Oct 2016 video? What's more rediculus, that Tesla is not testing anything anywhere and had their lawyer lie or that Tesla wants to hide progress or even failures from the public and competitors?

For testing purposes, there is no reason Tesla couldn't use Nevada or Arizona as a proxy for LA. If you can make it from Sparks to NYC 10x then you can make it from Fremont to NYC 1x. It you recall Elon said the route could change mid trip, it's not going to be a programmed route with specific training on that route. That would be rather pointless and they already did that for that video.

Why would Tesla be so coy about this? Well they refused to say they model 3 wouldn't have a HUD. For a year that was the rumor and Elon would never come out and say that it wouldn't. So this is not the first time Tesla has been coy about a major feature. People are still expecting a HUD to appear in a surprise update, I'm sure.

 

[avoigt]

Protecting the moat even or particularly if the competition can't see it yet. Lets hope this is not wishful thinking...

I found this to be a reasonable explanation ....

"Well, one answer would be the fact that under the Californian law, the carmaker would have to file an annual report of its activity, including the times the autonomous system had had to disengage. Considering theAutopilot is a big selling point for Tesla - and having a worse performance than rival companies such as Waymo or GM would make them look like fools - the company prefers to keep things secret until it's ready to place its full trust in the system."

Tesla Hides Autopilot Progress from the Public, Raising Further Questions

 

[avoigt]

Company cars are big business in Germany as most of the mid and luxury cars here on the roads are part of a salary package and not privately bought or owned. As an employee you still pay taxes for it but as a company you can write most of it off. They are heavily subsidized by the government to push demand for the car industry.

Lately there was a good article coming out describing the huge TCO advantage of EVs given that the high milage rates they allow versus ICE cars makes them attractive and other factors like maintenance and energy costs versus gas in Europe play in their favor. Adding the costs on high milage makes them a no brainer.

Once that truth has been dripped through I expect a strong demand from companies as a new group of buyers people have not anticipated yet. German Automakers will push that EV market once they have some reasonable cars out there. Not sure though when that will be....

For company's the Model 3 can be a very nice part of the package to attract employees in a market with record low unemployment rate like in Germany.

This article plays nicely into that logic: Bite Squad plans to swap out delivery fleet with Tesla Model 3s

 

[Reciprocity]

One reason to hide results vs Cruze and Waymo is that those systems are not a fair comparison as they are purely Dev systems that cost more then the cars they are in. The lidar these companies use are up to $70k and the computers fill the trunks. You wouldn't want crappy financial analysts comparing the results of these systems to Tesla where the entire system costs $2000? Not one analyst would understand that. It is utterly pointless to develop a system that costs $70000. Those companies have as big a mountain to climb as Tesla, but Tesla is climbing a software mountain, which had far more value long term.

 

[mongo]

The document you linked to says no such thing, as several other people have stated in the autonomous forum.

Here is what the document says:
"An autonomous vehicle is operating or driving in autonomous mode when it is operated or driven with the autonomous technology engaged."

It clearly does not matter if the driver has the hands on the wheel or anywhere else on the car. If the technology is running and controlling the car on California roads, then they need to report it.
.

Uggg, you're going to make me transcribe the locked pdf?

You quoted the second half of the definition of autonomous mode.
"Autonomous mode" means an autonomous vehicle, as defined by this article, that is operated or driven without active physical control by a natural person sitting in the vehicle's driver's seat.
(Rest as you quoted).
I make the case that if the SW requires a torque input resulting from the driver's hand on the wheel (which must track steering changes) that is an active physical control. However, this argument is secondary to the base defense.

As defined by this article:
"autonomous vehicle ": means any vehicle equipped with technology that has the capability of operating or driving the vehicle without the active physical control or monitoring of a natural person, whether or not the technology is engaged, excluding vehicles equipped with one or more systems that enhance safety or provide driver assistance but are not capable of driving or operating the vehicle without the active physical control or monitoring of a natural person.

Which is why AP miles are not not reportable. Hands on wheel forces, at a minimum, the monitoring clause thus excluding AP vehcles from being considered "autonomous vehicles". No AV, no "Autonomous mode", no AM, no reporting.

So, swap EAP with FSD and keep nag (non-driving torque input + monitoring nag). Why is it required to report?

 

[Drax7]

Sure. Anything is possible. Don’t disagree there. But in the context of everything else like continued missed deadlines, team leadership turnover and repeated failure of even the smallest promised deliverables, at some point we need to employ Occam’s Razor and go with the simpelst explanation that fits all the facts.

At this point the brightest sliver of hope for the FSD program is @verygreen ’s conclusion that they finally started testing in December based on the appearance of high precision data mapping. That means Tesla is roughly 10 months behind Cruise and a year or two behind Waymo wrt testing.

Tesla recently got involved with Intel , They might consider going back to the Mobileye solution.
Mobileye has advanced considerably since AP 1.0 , and now under Intel it might be easier to
Compromise.

 

[Lessmog]

Uggg, you're going to make me transcribe the locked pdf?

You quoted the second half of the definition of autonomous mode.
"Autonomous mode" means an autonomous vehicle, as defined by this article, that is operated or driven without active physical control by a natural person sitting in the vehicle's driver's seat.
(Rest as you quoted).
I make the case that if the SW requires a torque input resulting from the driver's hand on the wheel (which must track steering changes) that is an active physical control. However, this argument is secondary to the base defense.

As defined by this article:
"autonomous vehicle ": means any vehicle equipped with technology that has the capability of operating or driving the vehicle without the active physical control or monitoring of a natural person, whether or not the technology is engaged, excluding vehicles equipped with one or more systems that enhance safety or provide driver assistance but are not capable of driving or operating the vehicle without the active physical control or monitoring of a natural person.

Which is why AP miles are not not reportable. Hands on wheel forces, at a minimum, the monitoring clause thus excluding AP vehcles from being considered "autonomous vehicles". No AV, no "Autonomous mode", no AM, no reporting.

So, swap EAP with FSD and keep nag (non-driving torque input + monitoring nag). Why is it required to report?

I'm sure you are on the right path here. (No pun unintended)
The very precise legalese verbiage ("as defined") simply screamed about hidden meanings.
Conclusion: Tesla has kept being busy developing and testing EAP/FSD while avoiding giving away their secret findings until the time is ripe.

 

[kelly]

While I appreciate the very difficult task that is moderating and the hard work that goes into it, I object very strongly against using the moderator priviliges to make an argument on the content. Even if you want to make a factual correction, you should do it as any other poster on this board would. Through a regular reply and ABSOLUTELY not by modifying the original post (or headlining your reply with 'moderator comment' which is also bad but less bad). Please reserve those privileges when dealing with personal attacks, off topic posts and deliberate trolling, not when discussing content that is perfectly acceptable according to the rules.

On to the substance, and this makes it even more painful : you are also wrong. The electrek article is from early 2017 but the reddit reference is from 11 hours ago and links to recently released data discussing performance up to December 2017.

I want to re-iterate my very strong objections against you using moderator privileges to make a point in the discussion. That your interjection as a moderator had factual errors too makes it even worse. Can we have an official reply from the moderators what their policy is going forward on modifying user posts on the sole basis that moderators feel they must warn against the content?

Schonelucht, I have to agree with you. My eyebrows shot up when I saw Audobon's post. I wondered to myself as I read his post why it wasn't presented as a fellow participant in the forum versus using the "authority" of the moderator's position.

 

[mongo]

Tesla recently got involved with Intel , They might consider going back to the Mobileye solution.
Mobileye has advanced considerably since AP 1.0 , and now under Intel it might be easier to
Compromise.

Great point!
Esp if Intel is more interested in HW than SW. No liability to Intel for Tesla's use. Tesla agrees to buy a bunch of infotainment processors, Intel also gives access to MobilEye code.

 

[bdy0627]

Great point!
Esp if Intel is more interested in HW than SW. No liability to Intel for Tesla's use. Tesla agrees to buy a bunch of infotainment processors, Intel also gives access to MobilEye code.

What about TeslaVision? That suggests it is not progressing successfully. We really simply don't know where Tesla is in terms of progress at this point, but we know they have struggled over the last year and a half with respect to AP2.5. If we were guessing a year and a half ago at where the tech would be now, I doubt anyone would have thought it would still be struggling to get beyond the functionality of AP1.

 

[mongo]

What about TeslaVision? That suggests it is not progressing successfully. We really simply don't know where Tesla is in terms of progress at this point, but we know they have struggled over the last year and a half with respect to AP2.5. If we were guessing a year and a half ago at where the tech would be now, I doubt anyone would have thought it would still be struggling to get beyond the functionality of AP1.

My thought is that they might glean some useful preprocessing or decision tree algoritms from the MobileEye code. Possibly a boost to ESP short term. Long term, I doubt Tesla will abandon the current FSD NN approach.

 

[RiverCard]

Protecting the moat even or particularly if the competition can't see it yet. Lets hope this is not wishful thinking...

I found this to be a reasonable explanation ....

"Well, one answer would be the fact that under the Californian law, the carmaker would have to file an annual report of its activity, including the times the autonomous system had had to disengage. Considering theAutopilot is a big selling point for Tesla - and having a worse performance than rival companies such as Waymo or GM would make them look like fools - the company prefers to keep things secret until it's ready to place its full trust in the system."

Tesla Hides Autopilot Progress from the Public, Raising Further Questions

Your link also quotes Tesla's response to the DMV.

 

[Reciprocity]

One reason to hide results vs Cruze and Waymo is that those systems are not a fair comparison as they are purely Dev systems that cost more then the cars they are in. The lidar these companies use are up to $70k and the computers fill the trunks. You wouldn't want crappy financial analysts comparing the results of these systems to Tesla where the entire system costs $2000? Not one analyst would understand that. It is utterly pointless to develop a system that costs $70000. Those companies have as big a mountain to climb as Tesla, but Tesla is climbing a software mountain, which had far more value long term.

A follow on to this.. The day Tesla demos this LA to NY trip, the feature will be available to customers in an updated not long after. Elon already stated this and we have seen that Tesla is not afraid to roll out features that it has fully vetted. The reason I believe this is because there is zero chance that Tesla will do a demo that can fail because it would be catastrophic to the company and thus by the time they do the demo, the software is done. Not nearly done, not a lucky 1 of 5 attempts that they filmed as a success, because its going to have to be an almost real time demo that people can follow along with for anyone to believe its real at this point. Maybe even some voting on the route or destination as its happening to prove the system works regardless of Route. To me, this will be an EAP feature that integrates with Nav and High def maps to automatically pass cars and change freeways as well as maybe drop off with parking spot seeking. These are the advanced and fundamental features of EAP before moving on to full level 3+ FSD.

 

[JRP3]

Those other car maker's don't have their primary (and in the case of Tesla only) R&D facilities in California.

Data can be sent across state lines. There is no reason physical testing needs to be done in CA unless they are making constant physical changes to the vehicle.

 

[dirtyofries]

Data can be sent across state lines. There is no reason physical testing needs to be done in CA unless they are making constant physical changes to the vehicle.

Also, not to be that guy, but do we know their lone R&D is indeed in CA? I highly suspect they have other locations to work from.

 

[Curt Renz]

Electrek - this morning: Tesla maintains free Supercharger access for Model S and Model X

 

[DurandalAI]

Boom! Shove it down the throats of the shorts, Model 3 January sales above the Chevy Bolt sales for January. 1875 for Model 3, vs 1177 for Bolt.
January 2018 Plug-In Electric Vehicle Sales Report Card