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

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

This site may earn commission on affiliate links.
Anyone know if Tesla will need to update the headlight hardware to implement this, or if they can do it in software update for at least some existing vehicles? I know we have multiple elements on the existing headlights, just not sure if they're individually addressable or even aimed in a way that would be useful for this feature.

Things I'm willing to pay Tesla a reasonable amount (i.e. cost + labor + some reasonable but not dealer-esque markup) to upgrade on my 2018 Model 3 LR:
  • Adaptive headlights
  • Upgraded computer (I have HW3 w/ FSD package, but the actual UI computer could use some upgrades, streaming services are almost unusably slow to navigate, etc) - bonus points if this gets me a glovebox USB3 connector for dashcam/music
  • Amber rear turn signals (they have it in the EU... obviously this is a complete replacement of the rear light housings, plus software reprogramming, hopefully no wiring changes)
  • Lithium 12V replacement (when inevitably the existing one is due for replacement)
  • Upgraded USB charging ports (it is well documented how to do this ourselves, but requires making / buying adapter cables which is a hassle when they could conceivably sell them to us as part of an upgrade service)
  • AM radio (even if it suffers from EMI from the car, would like the option to have it for those places that broadcast warnings and such on AM - in theory could be just a replacement of the FM module with an AM/FM module)
I almost included the external speaker for the sentry mode lulz (since my car probably doesn't have wiring for it, it's not as simple as buying the speaker alone like some have done), but honestly, it's not that cool/useful enough for me to bother even if all I needed was the speaker...
Honestly even if you can upgrade it would likely be cheaper to trade-in and get a new car when these things are available, especially with current resale values. In the past, Elon has made this same comment with upgrades to the Model S.

As well, I think Tesla does not want to overload the service centers with this type of stuff. They have enough on their hands without adding additional efforts.
 
Does one have to learn yet more coding terminology to understand the above, or is that third bullet point’s “upreving” a typo?

Asking for a fiend.
Chromium OS Docs - glossary.md
Update the revision of a package/ dependency to trigger inclusion in new builds.
uprev: The process of updating the revision (or version) of a package. The purpose is to produce a new ebuild with a higher overall version number to allow the build system to pick up changes made to the package. Packages that inherit from cros-workon can be automatically uprevved by the builders. There is a generic process that covers most packages, and some specialized processes for a few specific packages (e.g. chrome, android).
New version of the feature.
 
Chromium OS Docs - glossary.md
Update the revision of a package/ dependency to trigger inclusion in new builds.

New version of the feature.
I would prefer revving with double vv in the explanantion of the free dictionary:
a. To increase the speed of (a motor, for example): revved the engine.
b. To accelerate or increase: orders to rev up factory output.
In german: hochdrehen / aufdrehen
 
  • Like
Reactions: Johann Koeber
Visited Fashion Mall in Scottsdale Az and seems some EV challengers have joined Tesla in this hi-end mall. Here's the new Lucid Showroom, opened a week ago, (wishing I had a better camera.)

Pricey, but nice specs at 1,100 HP. The motor on the left is 670 HP - inverter and gear case combined... Wow. Their 1,000 Vdc likely means a lot less copper needed for the same power.

No executive rear seats shown yet, those looked very cool from early photos.

View attachment 735543

Would it be too cheesy to show Lucid Panel Gaps? This little corner curled up, but it's a showroom car right? The line on the right is a normal part of the bodyline. This launch must be pretty early still.

View attachment 735545

The motor is why performance ICE is doomed.

A 670 hp engine + transmission weighs several times that. ICE enthusiasts are focused on the battery weights yet ignore the huge power/weight advantage electrics have. The Bugatti Chiron’s engine weighs 1,080 lbs (that probably doesn’t include any fluids or the exhaust either), the transmission weighs several hundred more. Two of those motors weigh ~330 lbs and will deliver roughly the same power to the wheels.

I’m not joking when I say performance EVs are going to have several thousand horsepower next decade. They will literally be traction limited out to ~200 mph. The 2030 Model S plaid will probably have 2000hp and 600 miles of range. The roadster, 3000+. ICE will seem like driving an old Stanley steamer in comparison.
 
Last edited:
are we in the beginning, middle or end of next "stair step up"
my suspicion is more towards the beginning
I took a look at the numbers. [Edit: Hard to tell, but my guess is towards the end, given the duration of the 3 previous larger movements in the chart. ]

1637510087308.png
 
Last edited:
I took a look at the numbers. [Edit: Hard to tell, but my guess is towards the end, given the typical duration of larger movements in the chart. ]

View attachment 735628
@jw934
please do the same graph for the last 6-7 months, just eyeballing that portion a linear regression line overlaid is kinda “slopey” upwards, and what’s “up” with the last 6 weeks?
i see a very interesting end of year
(your graphs are far more sophisticated, is there a service?)(but i don’t trade tho so, it’s kinda’ moot anyhoo)
edit:
i notice you use a 9 day not a 14 day ATR any reason or just personal preference?
edit:
not familiar with “range/close” indicator any links for explanation?
edit:
why offset 4 days?
edit:
note july 2020 and mid august 2020, ATR spikes and big moves up, is ATR presaging or confirming?
that’s why i’m very suspicious of the present ATR spike
 
Last edited:
Release notes for FSD Beta 10.5 (I saw this on reddit)
  • Improved VRU (pedestrians, bicyclists, motorcycles) crossing velocity error by 20% from improved quality in our auto-labeling.
  • Improved static world predictions (road lines, edges, and lane connectivity) by up to 13% using a new static world auto-labeler and adding 165K auto-labeled videos.
  • Improved cone and sign detections by upreving the generalized static object network with 15K more video clips and adjusting oversampling and overweighting strategies (+4.5% precision, +10.4% recall).
  • Improved cut-in detection network by 5.5% to help reduce false slowdowns.
  • Enabled “emergency collision avoidance maneuvering” in shadow mode.
  • Enabled behavior to lane change away from merges when safe to do so.
  • Improved merge object detection recall by using multi-modal object prediction at intersections.
  • Improved control for merges by increasing smoothness of arrival time constraints and considering possible merging objects beyond visibility.
  • Improved lane changes by allowing larger deceleration limit in short-deadline situations.
  • Improved lateral control for creeping forward to get more visibility.
  • Improved modeling of road boundaries on high curvature roads for finer maneuvers.
  • Improved logic to stay on-route and avoid unnecessary detours/rerouting.
Separately, WholeMarsCatalog (@WholeMarsBlog) tweeted that it wouldn't be released until tomorrow, for what it's worth.
These look like great improvements.

My experience with 10.4 has had these needed areas of improvement.

The false slowdown is jarring and improved lateral control should reduce jerkiness during turns.

Most of my unprotected left turns are with traffic from the sides as opposed to incoming from the front. These are still the biggest challenge for V10.4

I live off a dirt road and it likes to sometimes drive on the wrong side of the road. It did this with V10.3 as well.

I have 4 new roundabouts on my regular routes which do NOT seem to be known by FSD as they are not on the map. There are three new roundabouts that the nav does not have (these used to be traffic lights) and one more that used to be a 2 way stop. All are non-driveable as the approach is too fast and the jerky behavior through the turn is wild and unnerving.
Does one have to learn yet more coding terminology to understand the above, or is that third bullet point’s “upreving” a typo?

Asking for a fiend.
Happy to go over the details

1. VRU (Vulnerable Road User) detection is better. I haven't seen any issues here yet, so YMMV
2. Static world predictions improvements. I hope this means that it knows more about each upcoming road segment based on past learning before the current drive sees it dynamically.
3. Same static world improvements but for cones and signs. These have all been working well for me.
4. Cut-in detection improvements. I think this will be great as my route has lots of twists and turns, going up and down hills, with many cars turning onto the road. FSD has false positives where it thinks that the car turning onto the road in front of me, and heading towards me in the opposite lane, is going to enter my lane. This is a false positive as the car is not coming into my lane, I can clearly understand/percieve/infer this, but V10.4 does NOT. Let's hope this is better.
5. Yeah Shadow Mode! This is new code which won't act yet, as it is shadow mode code, but will in later releases. Emergency collision avoidance is hard as the car will do extreme maneuvering when it has very high confidence an accident is about to occur. Can't have false positives.
6. Should hopefully allow for less time in slow lanes to lane change away from slower, merging vehicles
7. Should be able to navigate intersections quicker, by using mult-modal (many concurrent) tracking of vehicles in its decision making. V10.4 takes quite awhile sometimes to navigate an intersection with traffic. Let's hope V10.5 is quicker
8. There's a lot here. The arrival time constraints I believe are predictions the merging vehicles to reach a certain place at certain time. And this improvement is to the control of the vehicle so that it will hopefully be less jerky and more confident when it takes action for these vehicles. And then also takes into account other vehicles which it might not have seen yet for some reason (aka 'beyond visibility'). I'm hoping this improves this intersection where it stops on the red line (which is way to far back to see anything from either the left or right. Cars going 55MPH) and starts to creep forward, but the cross traffic is occluded and can't be seen while creeping. Needless to say, it hasn't navigated this intersection at all.
9. Should get out of the lane with more confidence and quicker with slow cars in your current lane
10. Hopefully makes creeping behavior faster and more confident
11. I have a lot of tight turns (20MPH to 30MPH turns) on a 45MPH road. I'm hoping it takes these more confidently with this improvement.
12. Haven't seen this behavior yet as it always stays on the route. But I'm guessing others may have better feedback for this one.
Screenshot 2021-11-21 8.13.55 AM.png
 
387 mph (345 average) in an electric airplane. Good job, Rolls-Royce!

 
The motor is why performance ICE is doomed.

A 670 hp engine + transmission weighs several times that. ICE enthusiasts are focused on the battery weights yet ignore the huge power/weight advantage electrics have. The Bugatti Chiron’s engine weighs 1,080 lbs (that probably doesn’t include any fluids or the exhaust either), the transmission weighs several hundred more. Two of those motors weigh ~330 lbs and will deliver roughly the same power to the wheels.

I’m not joking when I say performance EVs are going to have several thousand horsepower next decade. They will literally be traction limited out to ~200 mph. The 2030 Model S plaid will probably have 2000hp and 600 miles of range. The roadster, 3000+. ICE will seem like driving an old Stanley steamer in comparison.
Wow, with that much power, won't the car just flip on the spot from the torque? (Insert Flintstones car animation). Or do the motors oppose rotation front to back? My college roommate had V6 Chevy Vega with centerlines, ratchet shifter and it was a blast to drive! The left front tire would nearly come off the ground from torque.
 
The motor is why performance ICE is doomed.

A 670 hp engine + transmission weighs several times that. ICE enthusiasts are focused on the battery weights yet ignore the huge power/weight advantage electrics have. The Bugatti Chiron’s engine weighs 1,080 lbs (that probably doesn’t include any fluids or the exhaust either), the transmission weighs several hundred more. Two of those motors weigh ~330 lbs and will deliver roughly the same power to the wheels.

I’m not joking when I say performance EVs are going to have several thousand horsepower next decade. They will literally be traction limited out to ~200 mph. The 2030 Model S plaid will probably have 2000hp and 600 miles of range. The roadster, 3000+. ICE will seem like driving an old Stanley steamer in comparison.
Actually, the Lucid salesman didn't have a good opinion on selling 1,100 HP to just anyone. The laws don't seem to be changing as fast as the technology (no surprise). There probably should be a certification process in there somewhere - if anything, just to prevent them from being used as weapons.
 
Does one have to learn yet more coding terminology to understand the above, or is that third bullet point’s “upreving” a typo?

Asking for a fiend.

was
  • Improved cone and sign detections by upreving the generalized static object network with 15K more video clips and adjusting oversampling and overweighting strategies (+4.5% precision, +10.4% recall).
I'll translate that as

* Improved cone and sign detections by adding 15,000 more video clips to the database used by the "generalized static object network" code, and adjusting oversampling and overweighting strategies (+4.5% precision, +10.4% recall).

with upreving basically meaning "updating and revising" or "updating to a new revision". Just a programmer being lazy with his grammar/vocabulary.
 
Tesla just installed our Solar and Powerwalls 2 yrs ago on our Ceramic Tile roof, and of course the roof is old and now leaky - not that it will rain anymore in Az. I knew this day would come, but not so soon. Quote for new roof (same in kind) was $11K plus Tesla charge for removal/install of panels, plus downtime.

View attachment 735548

So... decided to start over with the latest solar roof, and fix it for good while the tax incentives are still hot. Funny thing, my neighbor is interested in the "old" panels, and I don't need Powerwalls so not a complete loss. Will be interesting to see if Tesla uses my old inverter. They only recommended 1 Powerwall today while they used to recommend 3, so must be the shortage (or amazing batteries, anyone?).

The real winner, our SRP Utility company. We'll be producing lots of extra power and they don't pay well. No autobidder around here, best solar in America, go figure. I'm sure I owe some clean power back into the system to collect my 2 cents and peace of mind.

Decision for me is hinges on a Q4 earnings and a smooth FSD rollout. (Still waiting at 98 score and 100 past 10 days.) This could help Q1 numbers as the install likely won't be for many months.

@SOULPEDL The safety score only uses the last 30 days, so if you had a bad score one day, it might roll off.
 
  • Funny
Reactions: AZRI11 and capster
The motor is why performance ICE is doomed.

A 670 hp engine + transmission weighs several times that. ICE enthusiasts are focused on the battery weights yet ignore the huge power/weight advantage electrics have. The Bugatti Chiron’s engine weighs 1,080 lbs (that probably doesn’t include any fluids or the exhaust either), the transmission weighs several hundred more. Two of those motors weigh ~330 lbs and will deliver roughly the same power to the wheels.

I’m not joking when I say performance EVs are going to have several thousand horsepower next decade. They will literally be traction limited out to ~200 mph. The 2030 Model S plaid will probably have 2000hp and 600 miles of range. The roadster, 3000+. ICE will seem like driving an old Stanley steamer in comparison.
This seems relevant to the semi too.
 
Elon clearly reads TMC and is trolling the mods with this numerology thing.

Mods - can we at least have a temporary reprieve on numerology on this thread up to the 9th of December? Even if there is nothing in it, the SP will likely rise significantly on the rumour alone. Latest cast iron clue can be read here:
BLISS
I think the other thread is better because it's more organized on that single topic.

Also I doubt Elon reads this forum much if at all, but all of it it discussed by prominent Twitter accounts in his bubble.