szczys

First of all, good on Google for developing a Bluetooth firmware the hardware they are otherwise bricking.

I have one of these Stadia controllers and recently tried to play Cup Head (a notoriously hard 2d shooter) with it. I found the Bluetooth latency to make this unplayable. I would estimate it to be 2x or 3x the latency of a PS3 or PS4 controller. Anyone want to pontificate on whether that's a hardware limitation or a software implementation issue? (I was playing via Steam on Linux and tried all 3 controllers.)

Unfortunately, no. The highest quality broadcast I get on ATSC is 1080i (the i=interlaced)

I still use an over-the-air antenna on a daily basis (and record from it using MythTV with ATSC tuners in a computer). The changeover from analog has been a bit of a give and take.

With analog broadcast, if you were a bit too far away you would still get a fuzzy image and sounds that had some interference with it. Now if you're a bit too far away the picture breaks up and the audio sounds like auto-tune. It doesn't take much of that to make it unwatchable.

On the other hand, when you get a solid signal the picture is amazing versus the standard-def of analog broadcasts. Add to this the benefit of multiple sub-channels on each broadcast channel and you get a lot more watching options.

I've installed an attic antenna and a pre-amplifier that supplies great signal to all coax (CATV) outlets in the house. That hardware cost what 1-2 months of cable TV subscription would have and that was years and years ago.

On balance the digital changeover has been much better in my experience.

Doesn't that add manufacturing complexity? If you have multiple firmwares it's possible to flash the wrong one at the factory leading to a lot of pain if they get out to the market before this is discovered (your news 16 GB card is only using 8 GB of memory). This current scheme means there is one firmware to rule them all and the configuration of the hardware itself acts as the selector.

I don't have a Tesla, but I had upholstery on the backs of the rear seats (the part that becomes the floor of the hatchback when you fold them down) all off on my Ford C-Max. It looks like an adhesive failure there.

Not making excuses for the big T, just saying that there's a lot places for things to work out poorly in automotive.

I mean, on your first point, it's an automotive grade part. I think any eMMC put under these conditions (75% full and the other 25% written heavily to with log files) is going to have a relatively short life expectancy.

What baffles me is that who is reading these logs? Can't they just turn then off (pipe to /dev/null) and then have cars that need troubleshooting put into a debug mode at some point?