Human vision has a much higher dynamic range than the dash cam video shown. In the video, I could see the pedestrian before she was in the bright area of the headlights although this was difficult to see due to the compression algorithms turning it into a blob. Dashcams generally have shitty dynamic range at night and this one is no exception. When driving at night I generally don't have too much difficulty seeing beyond the bright area of my headlights. I also see that there were streetlights as well and the pedestrian was not that far away from two of them.

Tesla has video (8 cameras), radar (up to 160M), and long-distance (8M) ultrasonic sensors.

I think what we're seeing is dashcam footage, not the main cameras. Dashcams tend to be rather shitty at night, even good ones. Dashcams have nowhere near the dynamic range of the human eye. Even in this case, I can still see the pedestrian in the footage a fair distance out before she reached the bright area of the headlights. Most dashcams suck when it comes to dynamic range and this one appears no different. I also clearly see that the driver was looking down, possibly texting or something instead of focusing on the road.

I agree. Just the other night I stopped at a light and a bicyclist rode through and I couldn't see them well until they were directly in front of me because of the dark clothing and bicycle and lack of any reflectors. I have also, however, been in situations like what happened in Arizona numerous times yet I can still see them in the distance since my eyes have much better dynamic range than many dashcams, including my current one which is a fairly high-end one.

Deer also tend to jump out of nowhere in front of a vehicle. The dashcam footage is horrible and human can certainly see better than that at night. Even in the dashcam footage I could see the pedestrian where some reaction should have taken place. The pedestrian was not moving quickly either. It's also clear that the driver was not doing a good job of paying attention.

I agree. I've seen a lot of dash cam footage and a lot of it leaves a lot to be desired. Even in that video, I could still see the pedestrian quite a ways out. The human eye has a lot more dynamic range than the small dashboard cameras. I have a fairly good dashcam though my eyes are still quite a bit better.

I agree with this. Additionally, as bad as the video quality is I could still see the pedestrian quite a way out. The car could have at least attempted to brake, honk the horn and swerve away. The LIDAR most certainly should have seen the pedestrian as well. I think if the driver was paying a lot more attention to the road then this could have been avoided since typically a human has better night vision than a dashboard camera.

I can also say in Uber's defense that just the other night that I barely saw a bicyclist who crossed in a crosswalk without any reflectors or lights and wearing dark clothing. In this case, however, it was a lit crosswalk and a traffic light so I was stopped.

The radar and lidar certainly SHOULD have seen the pedestrian and it certainly appears that the driver was NOT paying attention. I also will say that in the video I could see the pedestrian while still a way out where the car should have started braking and it could have avoided killing her. While not nearly as noticeable as they would have been had the bicycle had reflectors on the wheels I could still see it when pausing the video.

The pedestrian should have had reflectors on the bicycle wheels. Just the other night I barely saw a bicyclist crossing the street in front of me at a crosswalk until they were in my lights due to the lack of any reflectors and dark clothing. I don't know what the laws are in Arizona, but where I live bicycles are required to have reflectors, a headlight and a taillight at night.

If the driver were paying a lot more attention to the road than the phone then this also could have been prevented.

That reminds me of some of the laptop computers I worked on back in the early 1990's. I worked for a well-known laptop manufacturer who made some laptops for the military. The military laptop I worked with was Tempest certified. It had a removable SCSI hard drive where all data on the drive was encrypted. It was a nightmare to take it apart. Once you took off the cover you had to remove all of the RF shielding (128 screws held the RF shielding on, I kid you not!). The screen was an electroluminescent VGA screen with a fine wire mesh in front of it. It was designed so there were no detectable RF emissions. The company's laptops were also quite rugged. One of the other commercial laptops they made was modified to run on the US space shuttle.

The DOD can have a lot of weird hardware requirements and often the current hardware is what we would consider outdated due to the long development times and how long all the approvals and whatnot took.

It depends. For example, if it is specialized hardware it can't just be replaced.

That is no longer true. In fact, you can have better control over grain size with 3-D printing than you can with casting. Additionally, you can change the metal composition based on location. SpaceX, for example, makes extensive use of 3-D printing of its metal parts. Their SuperDraco rocket engine is entirely 3-D printed, for example. Here's an older Slashdot story where 3-D printed stainless steel more than doubles the strength.

SpaceX has been very successful 3-D printing metal. For example, their SuperDraco engines are entirely 3-D printed. From what I have read and heard, they are able to do things with 3-D printing that are impossible any other way, including changing the properties of the metal based on location. Heating something to the melting point isn't necessary for strength (i.e. see friction stir welding). 3-D printing does not mean the parts are weak.

3-D printed parts can be made lighter and contain far more complexity than non-additive manufacturing methods due to being able to create shapes and cavities that would normally be very difficult if not impossible. Metals can also be mixed in ways that are not possible with forging or casting.

I think most of the ACs today are just ass speaking.

I love Texas Instruments. When I was designing a power supply for a hobby project I came across their web bench design site. I just plugged in the numbers and out pops a schematic, BOM and board layout with parts that are in stock from Digikey. I've always found TI's documentation to be top notch. That's not to say that some of the other semiconductor manufacturers also don't have great sites either, but I haven't yet found any that match TI. I also respect a lot of the other companies you listed.

Dealerships make their money with service and financing and parts, not by selling cars. They know many people think they have to go to the dealership for service and milk that for all they can. They also make money selling extras like extended warranties and add-on features.

In terms of service, Elon Musk stated his goal is for service not to make a profit. This is the exact opposite of dealerships where one of their main revenue sources is service.

I can give an example of this. I damaged one of the roof panels on my Tesla model S. The roof panel is painted with a clear-coat. The cost for the part was $175. I spoke to a friend of mine who used to work at a Lexus dealership. He told me the same part would have cost many times that amount. The labor to replace it was also fairly reasonable. By comparison, my experience with Toyota with my previous car, a Prius, was far higher.

For example, shortly after the 3 year/36 month warranty ended (but within the 7 year 100K mile warranty), the HID headlights went out. This was a well-known problem with the Prius. The headlights were not designed to be easily replaceable since HIDs are supposed to last a very long time. Anyway, Toyota wanted $200 per bulb and $140 in labor to replace it. I went online and found the same exact Sylvania bulb for under $50. I ended up going to another dealership to have the work done (having bought my own bulbs) and they replaced it under a "good-will" warranty (which my local dealership did not want to do).

In another case, after I sold my Prius to my parents, the touchscreen stopped working. This is caused by a known design defect where a solder connection breaks over time. Toyota wanted $5000 for a new touchscreen and $2000 in labor to fix it. I removed the touch screen in under 5 minutes. I found a place online that repairs them with a lifetime guarantee for $400 and had that done then spent under 5 minutes re-installing the touchscreen. $2000 in labor is utterly insane. I basically undid one screw, popped a few panels and undid a couple of bolts and that was it. $5000 for the touchscreen is also insane. It probably cost Toyota no more than $300 to make and probably well under that.

Another thing the dealership wanted $1800 for the extended warranty. I went online and bought the same factory extended warranty for $1000. Anytime I needed a part, Toyota would charge outrageous amounts of money for it.

Oh, and Toyota would always try and push a bunch of unnecessary maintenance but give me trouble for known issues that would come up (like a water pump that went bad twice).

I later got a notice over the headlights that there was a class-action because dealerships were charging people to replace the ballasts when there was nothing wrong with them when the HIDs died (Toyota had a bad batch of bulbs). Many people were charged a lot of money to replace the ballasts which were never the problem.

My experience with Tesla has been very different. They never pushed me to buy add-on options or extra service. The service costs were also generally reasonable, though some things like tires are cheaper elsewhere (I like Tirerack). This is what the dealerships fear. My car needs service once a year or every 12.5K miles. A Tesla tech told me that the electric motor is lubricated for 12 years. There's no transmission or regular oil changes needed and even the brakes get a lot less use. There is far less to go wrong mechanically. Even the battery will last a very long time. I'm coming up on 5 years 60K miles and haven't noticed any reduction in range. This is not what dealerships want to sell. Hell, Tesla will even honor the warranty if you don't service the vehicle every year.