Tesla, TSLA & the Investment World: the Perpetual Investors' Roundtable

[GOVA]

Are you sure? Cameras can see further into the upper and lower bands than humans can if the sensors allow it.

 

[Chunky Jr.]

Have been testing 12.3.3. Pretty good.

Two issues I can see.

1. It won’t move in an advanced green left turn arrow. Must be promoted with the pedal. Do they have these in the US? Can’t remember. (Lived there on and off for years).

2. Doesn’t slow down at school zone signs. That’s a for sure ticket every time. The Mounties have no sense of humour around school zones.

Other than that lots of improvements.

I tried for the first time on a short trip back from dinner. The only problem was unexpected braking in the middle of the street for no reason. There were a couple cars and some trash cans on the side of the road, but no people and nothing in the way.

 

[Curt Renz]

Today, Cathie Wood bought TSLA shares for two of her ARK funds: ARKK 62,100 & ARKW 22,064.

 

[skybluecgreen]

The sun directly flooding a camera sensor is not a common occurrence and would probably be a rare acceptable disengagement nuisance for those using FSD on their personal vehicle, but of course for a robotaxi where there is no driver it would need to be a solved issue.

What do you humans do? Adjust their sun visors, hats or use their hand to block the suns glare.

Sounds like a joke, but might have to have some sort of tiny movable shade mechanism for the cameras to avoid that (or use a single backup rooftop lidar sensor of course).

Maybe car cameras should have something like our eyelids that can open and close to clean, de-ice and shade when needed.

 

[Discoducky]

How does Tesla vision do with Fog?

Maybe car cameras should have something like our eyelids that can open and close to clean, de-ice and shade when needed.

For high photon situations, it actually does something like that with ISO and FPS.

 

[GOVA]

Marques Brownlee and Jason Camisa among others attended the M3 'Ludicrous' event today. I really hope for a video or a couple of videos tomorrow.

 

[Discoducky]

Yeah, I wasn't so much thinking GM would buy a few computers and solve this thing.

But Nvida has money, and some smart people, and can afford to keep a portion of their chips for themselves and throw more computing power than anyone has ever thought possible at a problem like this, and then lease it out to everyone not named Tesla. Without needing a million cars with cameras first. That's what I'm a bit worried about.

Also if we previously thought Tesla was say five years ahead of everyone else. Wouldn't all these AI improvements cut that to less time? Meaning the value for Tsla would be less than previously thought.

DM me and I can explain the worry away if this is a big deal for you.

Yesterday my MY (FSD 12.3.3) drove 60 miles city/construction/highway/tunnels with no GPS signal and very complex downtown intersections in very busy traffic. It took ~2.5 hours total.

Overall, very solid 9/10. Sometimes it just waited a bit longer that I wanted, but this is minor. The slowness only happened 2 times.
The complexity of that route was very high. 4-5 (non local?) drivers/cars did take bad exists right in front of me. FSD just drove right through those intersections, like nothing.

This is already marching the remaining nines. It will progress much quicker than before. We felt like in sci-fi movie to be honest. Jut incredible.

I am really happy that Elon decided to make it public for a month. For the first time I am considering dropping 12K on FSD. This is no joke.

Hard for me to express how cool it is for me to read this and I'm super happy.

 

[Todd Burch]

The sun directly flooding a camera sensor is not a common occurrence and would probably be a rare acceptable disengagement nuisance for those using FSD on their personal vehicle, but of course for a robotaxi where there is no driver it would need to be a solved issue.

What do you humans do? Adjust their sun visors, hats or use their hand to block the suns glare.

Sounds like a joke, but might have to have some sort of tiny movable shade mechanism for the cameras to avoid that (or use a single backup rooftop lidar sensor of course).

Well as I mentioned I haven’t seen it disengage due to sun glare at all lately. Maybe someone else has.

 

[Discoducky]

Can you guess which one I care about?

 

[Discoducky]

My key metric is that **IF** this hits 1000, we are in the march of 9s Home

 

[primedive]

Cameras can actually see better in fog than us

Test with your phone in a foggy day, you will see way farther into it in the picture than with your own eyes

I noticed this while driving in a blizzard. I had to turn off the headlights and use the parking lights because the reflections off the leds were insane and it felt like I was inside a snow globe. I almost could not see the road. I didn’t want to stop due to the danger of getting rear ended. Turning down the lights made things a lot better. I took it easy and was passed by many cars on the left, but I also passed by eight wrecks within the span of maybe 20km, including one of the cars that had passed me lying on its roof about 100 ft from the road. The police had already gotten to them somehow.

Later when I looked at the dashcam footage to show a friend, it didn’t seem that bad at all, at least not via the camera, but at the time, the snow was completely dazzling.

 

[primedive]

I noticed this while driving in a blizzard. I had to turn off the headlights and use the parking lights because the reflections off the leds were insane and it felt like I was inside a snow globe. I almost could not see the road. I didn’t want to stop due to the danger of getting rear ended. Turning down the lights made things a lot better. I took it easy and was passed by many cars on the left, but I also passed by eight wrecks within the span of maybe 20km, including one of the cars that had passed me lying on its roof about 100 ft from the road. The police had already gotten to them somehow.

Later when I looked at the dashcam footage to show a friend, it didn’t seem that bad at all, at least not via the camera, but at the time, the snow was completely dazzling.

Just to add to this, the incident barely made the local news even though an 18 wheeler had also wiped out. Nobody had been injured in any of the crashes as far as I can remember, but still. If one of the wipeouts had been a Tesla on FSD (although using FSD in that situation would be insane), I bet that would have gone differently.

 

[2daMoon]

Make of this what you will.

Reuters has a piece discussing multiple different consumer surveys showing dropping consumer consideration for purchasing a Tesla vehicle: https://www.reuters.com/business/au...nub-company-musks-reputation-dips-2024-04-01/

One particular tracker shops a drop of 70% consideration of buying a Tesla at the end of 2022, down to 31% consideration in Feb 2024.

View attachment 1034392

I think you are batting close to 1000 on negativity today. What led you to post this?

Dare I suggest it may have been due to a feeling of fear, uncertainty, and perhaps a tinge of doubt that influenced getting this out to TMC as some sort of public service?

Does Reuters' well-defined track record with their news about Tesla ever figure into influencing your trust of anything they publish regarding the company?

If you have access to the article (paywalled for me) please let us know how many people were polled to come up with this chart, and, what was the exact wording of the questions being answered by those taking the poll?

I did find this tidbit to chew on from someone running the article, (to help spread the FUD)

"Caliber's "consideration score" for Tesla, provided exclusively to Reuters,..."​

Beyond that, there were no details or links to reference who "Caliber" is, nor any details behind this "score" provided exclusively to Reuters.

Meanwhile, I'll sit quaking in my boots, perched on the edge of my seat, waiting patiently for you to actually make some effort to verify sources from such "entertainment" outlets as Reuters before going off half-cocked and posting another example of useless Pablum for the masses masquerading as actual data.

Do remember the wisdom of Mark Twain and his definition for the three types of lies.

Tell Winnie hello next time you see them.

 

[Knightshade]

Cameras can actually see better in fog than us

Test with your phone in a foggy day, you will see way farther into it in the picture than with your own eyes

TBF, my phone is 48 megapixels with 3x optical zoom. Each camera on most Teslas is 0.9 megapixels and fixed focus. So not exactly a useful comparison there... that said, Lidar isn't going to be great either-- this is the spot -radar- could be really good (though the one Tesla had been using was pretty low resolution).

OTOH I'd expect robotaxis to be more likely to be willing to decide "this is not safe weather to drive in" than the typical human who believes he's an above average driver because there's no ego involved and they'd inherently prefer safety.

 

[2daMoon]

Just noticed about ten percent of the shares traded today were traded at the Closing Cross.

 

[Krugerrand]

Just noticed about ten percent of the shares traded today were traded at the Closing Cross.

Computer bug/glitch. Nothing to see here. Move along.

 

[Singuy]

I tried for the first time on a short trip back from dinner. The only problem was unexpected braking in the middle of the street for no reason. There were a couple cars and some trash cans on the side of the road, but no people and nothing in the way.

Check the ground, there could be a crosswalk marking or a portion of road that looks like crosswalk markings. This is a problem they need to fix as it's happening to me randomly.

On a crazy note, it successfully navigated out of Universal Studios parking garage during park closure! I was like "how did the car know to go right vs left and how to get out? They couldn't have mapped the parking garage....

 

[Singuy]

My key metric is that **IF** this hits 1000, we are in the march of 9s Home

View attachment 1034424

Yeah it's also frustrating that Hwy performance is now worst than city performance. Kind of crazy I never thought I see the day but it's true. I pretty much disengage daily on the hwy due to poor map data causing the car to not take exit and poor merging maneuvers from time to time. City street has been pretty solid.

 

[Discoducky]

Yeah it's also frustrating that Hwy performance is now worst than city performance. Kind of crazy I never thought I see the day but it's true. I pretty much disengage daily on the hwy due to poor map data causing the car to not take exit and poor merging maneuvers from time to time. City street has been pretty solid.

I agree.

Seems the team is focused on integration now and next.

 

[GatorMeat]

Well as I mentioned I haven’t seen it disengage due to sun glare at all lately. Maybe someone else has.

Security cameras have an auto-iris to deal with the sun by limiting the amount of light that gets in. I'm sure Tesla cameras do the same. No need for a physical shade