7 ms·
I don't fully buy the thesis. Human vision is logarithmic; the article's central metric of "cost per lumen-hour" isn't a well-founded one—it has linear utility
by perihelions 2y ago
I don't fully buy the thesis. Human vision is logarithmic; the article's central metric of "cost per lumen-hour" isn't a well-founded one—it has linear utility implicit in it.
- ndileas 2y agoYou're right about the linearity being flawed, but I think there's an important and rescuable construct there. Humans do have ranges of illumination where we can easily accomplish certain classes of tasks - very few want to tat lace or operate heavy machinery by starlight, for example.
- bozhark 2y agoAlso dismissing the 24 hour availability of labor
- schmidtleonard 2y agoWe aren't talking about installing a single light that we pump all of the world's energy into, making it ever brighter as we drive the cost/benefit further and further into the realm of diminishing returns. We are talking about scaling a few common "good enough for X activity" lighting setups across billions of installations, and that process is very linear.
- hammock 2y agoSo £40,000 per million-lumen hours = £0.40 per lux, no? Today that metric might look 10x better, around 4 cents per lux
- quickthrowman 2y agoLEDs can put out 200 lumens per watt, 5000 watts of that for one hour is a million-lumen hours. Assuming perfect efficiency, that would use 5 kWh of electricity, which is $0.40 total if you pay $0.08/kWh. I believe that works out to .00000004 cents per lux.
- aaron695 2y ago[dead]