Welcome to Tesla Motors Club
Discuss Tesla's Model S, Model 3, Model X, Model Y, Cybertruck, Roadster and More.
Register

2017 Investor Roundtable:General Discussion

This site may earn commission on affiliate links.
Status
Not open for further replies.
Apologies if this was already discussed earlier. But I noticed something interesting in NVDA's recent ER CC transcript.

Jen-Hsun Huang - NVIDIA Corp.

We currently have DRIVE PX. DRIVE PX today is a one-chip solution for Level 3. And with two chips, two processors, you can achieve Level 4. And with many processors, you could achieve Level 5 today. And some people are using many processors to develop their Level 5, and some people are using a couple of processors to develop their Level 4. Our next generation, so that's all based on the Pascal generation. That's all based on the Pascal generation. Our next generation, the processor is called Xavier. We announced that recently. Xavier basically takes four processors and shrink it into one. And so we'll be able to achieve Level 4 with one processor. That's the easiest way to think about it. So we'll achieve Level 3 with one processor today. Next year, we'll achieve Level 4 with one processor, and with several processors, you could achieve Level 5.


What I'm puzzled about is that Tesla is claiming that the current hardware its capable of all the way to level-5. So I wonder how many DRIVE PX chips it has. Otherwise, is it an implicit assumption that Tesla will keep updating it's existing AP2 hardware cars, for free of cost to owners, to keep up with the DRIVE PX chips?

Paging @FredLambert
 
Extra effort is one thing, injuries are another. I'm ambivalent about unions. I would like to believe that Tesla employees aren't exploited or abused. If the concerns mentioned in the article are real, I hope Tesla will address them so unions won't seem like the only way to get treated humanely.

I imagine most of the people on this forum have had to work overtime in one sense or another, but for most of us that meant making our brain tired. A vacation or good night's sleep will fix that.

But how would we feel if our working conditions / working overtime meant our hands or back could become permanently damaged?
I don't believe for a second that Tesla Elon even try to exploit their employees, or negligently fail to maintain equipment to the extent that it would endanger their employees.

If their employees even thought that Tesla had been trying to exploit them the employees would have been trying harder to unionize than they have been.

I have zero concerns about the entire issue. If the employees unionize I don't think that will cost Tesla a significant amount (if any) more than they are paying now.
 
Ok Racer, time for the nerd question: PIC or AVR?
I've used both. I prefer PIC. Of course, the two companies merged not long ago.

To the differences in what nVidia is saying about how many chips it takes to do Level N driving:

Think about what they're saying. Is there any logical reason it should take substantially more processing power to be able to do Level 4 or 5 driving over level 3?

I don't believe it should. To drive well at level 3, I would expect you have to do about 70-90% of the same calculations to discern the world around you as you would at level 5.
 
I haven't seen any mention of this one one here.

Gearing Up for a Self-Driving Future, Ford Drops $1 Billion on an AI Startup

Apparently Ford is buying less than 100% of a self driving car start up that only started business a couple months ago for $1,000,000,000.00!!!!!
That's not correct. Ford is going to invest $1B, not pay that much for the company. Here is the WSJ headline.
Ford Acquires Majority Ownership of Self-Driving Car Startup Argo AI
Ford to invest $1 billion into Argo AI in race to develop self-driving cars
Ford Motor Co. has acquired majority ownership of an artificial-intelligence startup called Argo AI and plans to invest $1 billion in the company,

Your linked article has wrong title, but within the article has the correct info.
Today, Ford announced it’s investing $1 billion over the next five years in Argo AI.
 
Last edited:
Thought this part of the Nvidia transcript was interesting as well:

"Of course, gaming was a very large and important factor. And my expectation is that gaming is going to continue to do that. And then longer-term, longer-term our position in self-driving cars, I think, is becoming more and more clear to people over time. And I expect that self-driving cars will be available on the road starting this year with early movers, and no later than 2020 for Level 4 by the majors, and you might even see some of them pull into 2019. And so those are some of the things that we're looking forward to."

http://seekingalpha.com/article/404...-results-earnings-call-transcript?part=single
 
Apologies if this was already discussed earlier. But I noticed something interesting in NVDA's recent ER CC transcript.

Jen-Hsun Huang - NVIDIA Corp.

We currently have DRIVE PX. DRIVE PX today is a one-chip solution for Level 3. And with two chips, two processors, you can achieve Level 4. And with many processors, you could achieve Level 5 today. And some people are using many processors to develop their Level 5, and some people are using a couple of processors to develop their Level 4. Our next generation, so that's all based on the Pascal generation. That's all based on the Pascal generation. Our next generation, the processor is called Xavier. We announced that recently. Xavier basically takes four processors and shrink it into one. And so we'll be able to achieve Level 4 with one processor. That's the easiest way to think about it. So we'll achieve Level 3 with one processor today. Next year, we'll achieve Level 4 with one processor, and with several processors, you could achieve Level 5.


What I'm puzzled about is that Tesla is claiming that the current hardware its capable of all the way to level-5. So I wonder how many DRIVE PX chips it has. Otherwise, is it an implicit assumption that Tesla will keep updating it's existing AP2 hardware cars, for free of cost to owners, to keep up with the DRIVE PX chips?

Paging @FredLambert

They'll try to get more out of what they have in the same way that they could get more than what MobileEye hardware was designed to do in AP1. They can also leverage fleet learning with always on data connection. They are probably confident that the single chip is enough for Enhanced Autopilot which they'll stretch to maximum capabilities for the next 12 to 18 months. Should it be too limiting to achieve FSD they can always swap the board for a cheaper yet more powerful alternative in one year from now for the people that did buy the option. Also, remember the rumor that they are developing their own chip to support Tesla vision, which would bring down the processing power required for their narrow AI.
 
  • Like
Reactions: MP3Mike
Apologies if this was already discussed earlier. But I noticed something interesting in NVDA's recent ER CC transcript.

Jen-Hsun Huang - NVIDIA Corp.

We currently have DRIVE PX. DRIVE PX today is a one-chip solution for Level 3. And with two chips, two processors, you can achieve Level 4. And with many processors, you could achieve Level 5 today. And some people are using many processors to develop their Level 5, and some people are using a couple of processors to develop their Level 4. Our next generation, so that's all based on the Pascal generation. That's all based on the Pascal generation. Our next generation, the processor is called Xavier. We announced that recently. Xavier basically takes four processors and shrink it into one. And so we'll be able to achieve Level 4 with one processor. That's the easiest way to think about it. So we'll achieve Level 3 with one processor today. Next year, we'll achieve Level 4 with one processor, and with several processors, you could achieve Level 5.


What I'm puzzled about is that Tesla is claiming that the current hardware its capable of all the way to level-5. So I wonder how many DRIVE PX chips it has. Otherwise, is it an implicit assumption that Tesla will keep updating it's existing AP2 hardware cars, for free of cost to owners, to keep up with the DRIVE PX chips?

Paging @FredLambert
My understanding of Huang's comment here is that based purely on image recognition and processing, there's a need of multiple chips to achieve L4/5. But Tesla's solution is image and radar.
 
  • Informative
Reactions: RobsJester
What I'm puzzled about is that Tesla is claiming that the current hardware its capable of all the way to level-5. So I wonder how many DRIVE PX chips it has. Otherwise, is it an implicit assumption that Tesla will keep updating it's existing AP2 hardware cars, for free of cost to owners, to keep up with the DRIVE PX chips?
Paging @FredLambert
Of course I have no idea, but I would hazard a guess if anyone implicitly assumes Tesla will update their AP2 hardware free in 3 years (or whenever), they'll be disappointed. More likely there'll be a barely-functioning Level 5 and increasingly astonishing capability AP3, AP4 HW-driven Level 5. Cue Doc Brown "Roads?????"

Or maybe AP2 won't actually be able to deliver Level 5. I've learned to wait for the real thing from Tesla... on the other hand I continue to be astonished at what they've been able to deliver.
 
  • Like
Reactions: Matias
That's not what you said. Way more than "a GPU".

Actually, no. Nvidia Drive PX 2 Uses Integrated and Discrete Pascal GPU Cores - 24 DL TOPS, 8 TFLOPs and Up To 4GB GDDR5 [Updated]

These specs may be updated slightly from this original presentation, but the base Drive PX 2 is basically 2 ARM processors plus 2 Pascal GPUs for 8 TFLOPs. Now, Tesla basically buys chips from NVIDIA and likely makes their own board. At least, wk057 told me that the AP2 hardware board he saw was a Tesla board, not the reference NVIDIA board. Tesla uses the same basic hardware building blocks as NVIDIA's Drive PX2 reference board, so they say it uses the Drive PX2 platform.

From this NVIDIA Investor Presentation:

http://files.shareholder.com/downlo...tor_Day_Tomorrow_s_Cars_Today_Rob_Csongor.pdf

NVIDIA states their $ opportunity is $2 billion on $15 million cars, or $133 per car. That's obviously at scale with later generations of chips. But it gives us a sense of how much they think the chips will be worth inside an autonomous vehicle. I'm assuming that Tesla is using 2x $250 GPUs right now, plus the cheap ARM processors, on a custom made board and chassis for a COGS around $750.

Tesla is looking at achieving Level 4/5 with a very small number of sensors. If an OEM is looking at incorporating more cameras, LIDAR (potentially many), multiple RADAR units, then more processing power is required to process all those additional input streams.
 
They increased their position in Q4 by 420,048 shares or 3.26%, from 12,869,500 to 13,289,548.

In addition to Baillie Gifford Fidelity increased their position in Q4 by 1,475,320 shares, or 7.17% from 20,575,081 to 22,050,401 shares.

Score card for top 5 so far:

Snap1.png


EDIT: The Q4 totals likely include converted Solar City shares
 
Last edited:
I have zero concerns about the entire issue. If the employees unionize I don't think that will cost Tesla a significant amount (if any) more than they are paying now.

This is just plain silly.

The UAW would have to justify its dues and get a raise significantly above the dues.

But that is not the most costly result of unionization.

It is work rules. One of the reasons Tesla is so nimble it that it is able to have employees multitask, shut down old job roles with increasing automation, and move employees to different jobs. With the UAW every change in a worker's duties will have to be negotiated with the Union.

And Tesla has one auto factory for the globe. Shut it down and you shut down Tesla. The UAW would have Tesla by the proverbial ballz. Unlike GM that still has factories in Europe, China, and Latin America going if the UAW strikes.

The quickest way for Detroit to delay the Alien Dreadnought for decades is getting the UAW's tentacles around Tesla.
 
This is a digression but it's the weekend, is Jeep reliability really that bad? Are they actually made that badly or is it just that people tend to beat them up more than other cars? I'm all for EV's but Jeeps sound pretty good this time of year.
I was having fun with Bonnie, and she knew it. Off-road enthusiasts are chauvinistic about their brands in a way that makes muscle-car devotees seem wishy-washy.

On the other hand, there's a reason they're called Heeps.....:D
 
  • Funny
Reactions: bonnie and tander
The quickest way for Detroit to delay the Alien Dreadnought for decades is getting the UAW's tentacles around Tesla.

Or the threat posed by the UAW could push Tesla to achieve "Alien Dreadnought" automation sooner so they are no longer reliant on human workers.

I'm sure this is what Elon is thinking. 100%.
 
Last edited:
Status
Not open for further replies.