site banner

Friday Fun Thread for October 25, 2024

Be advised: this thread is not for serious in-depth discussion of weighty topics (we have a link for that), this thread is not for anything Culture War related. This thread is for Fun. You got jokes? Share 'em. You got silly questions? Ask 'em.

4
Jump in the discussion.

No email address required.

5 lux isn't much, and it's mostly relevant in the sense of highlighting retroreflectors (either tape, or animal eyes). If you're really darkness-adapted and under 40 or so, you'll be able to see fuzzy outlines, but not much more: it's not unreasonable as a metric for 'minimum to see an object', but a little optimistic.

At 65 MPH, 100 meters is just over three seconds to react; 70 meters is just over two seconds. How much that matters depends heavily on what you're doing with that time. Two seconds to swerve is pretty generous. Three seconds to brake is not, especially in larger cars: modeling these things is tough and depends on a lot of specifics to the situation, but at best it's the difference between stopping just before impact versus barrelling through at 20+ MPH, and more likely the difference between 10 MPH and 35 MPH at time of impact.

it's mostly relevant in the sense of highlighting retroreflectors

IIHS's rationale document (page 4) states that the intended target of illumination is low-contrast objects, not retroreflective objects.

Huh. I don't like how little citation there is for that being sufficient brightness -- all the cites are just to people using 3 lux as a baseline measurement, and that for instrument sensitivity reasons -- but I guess I don't really have the numbers to say that they're wrong, either.

If you're really darkness-adapted and under 40 or so, you'll be able to see fuzzy outlines, but not much more

This is what annoys me about the push for full self-driving: instead of spending most of the time on AI, I want that [same underlying] sensor technology to start highlighting things (4-legged animals, 2-legged animals) that I can't see yet using the inside of the windshield as a screen. I want to be able to see cars through other cars- it doesn't matter if Truckzilla pulls out too far beside me when I'm trying to make a turn because I can just see if there's something coming directly.

I want technology to help me make better decisions on how I should drive; not to replace me. But I'm one of those weird people who actually likes driving- most people don't, so why would anyone ever develop a system like this?

That would be awesome. We’ve already seen the first wave of augmented senses with proximity radars and, arguably, the backup camera. Stuff like the blind spot indicators. Use those for a bit and it’s unpleasant to go back.

I suspect that display technology is part of this holdup. We’ve certainly tried, and then sort of gave up in the early 2000s. Maybe we’re due for a comeback.

Yeah, that's fair. There's been some work done for automotive HUDs, but there's (not-unreasonable!) concern about anything more serious than simple fixed-location infographics being distracting or vision-obscuring -- bizarrely, meaning that the display tech once existed and now doesn't -- and as a result things like blind spot detection or collision avoidance systems tend to rely on other inputs that tend to fall into the meaningless alert problem or at best just push sensor data directly to the instrument cluster.

anything more serious than simple fixed-location infographics being distracting or vision-obscuring

I think "put screens everywhere" is a clear signal that we don't give a shit any more about what's distracting you behind the wheel. (They obscure your vision due to glare and, if the company is/was obsessed with the color blue at the time, kill your night vision just as surely as the new LED headlights do).

My favorite is the one where you have to click through a full page disclaimer of tiny text about how looking at the screen is dangerous in order to actually get to the necessary functions that the screen provides.