Hacker News Reader

Comment by Zak | Hacker News Reader

There are 1800K white LEDs as well as phosphor-converted amber made to approximate the look of the classic amber low-pressure sodium.

1800K white can render colors surprisingly well depending on the phosphor mix. I recently put one with claimed (and measured) CRI over 90 into a flashlight and was surprised to see that it actually can render blues reasonably well.

I'm inclined to think those are better choices for street lights than anything daylight-ish, but I also think we should use far fewer street lights. Their presence often reduces the contrast car headlights provide, making it harder to spot hazards while driving.

View on HN

Each CRI is referenced to an ideal black body radiator at the same temp below 5000k. And there are 7 (or 14) sample points.

I’m not disagreeing with you, I agree that a high cri 1800k would be a nice night light. I just recently deep dived into this last week when my kitchen lights all died last week

High CRI is great for the driver while driving. But it ruins his sleep when he gets home, it ruins the circadian cycle of the people living nearby, and it disrupts wildlife.

We don't really need to see shorter wavelength colours like blues to avoid hazards. Night driving is not photography.

I think the answer here is to put the high CRI, neutral white lights on the car. Street lights should be warm or amber and don't necessarily need CRI. (Low-pressure sodium has negative CRI).

We could make a lot of improvements by rethinking our approach to street lights. When I've looked for studies on driving safety in the past, it did not appear that street lights improved safety in most places. Where they did help is crosswalks and low-visibility hazards.

At a controlled intersection, it would be easy to have lighting that's activated by a pedestrian pressing a button. Once drivers got used to that, the light being on would serve as a strong indication a pedestrian is definitely present at the intersection, even if they're not currently visible to the driver. For fixed hazards, small marker lights might be just as effective as overhead flood lights; passive reflectors might even have an equal effect, though I haven't read any studies on that.

Reply on news.ycombinator.com