To all of those arguing spectral distribution doesn't matter:
Consider Spectral Reflectance.
We use only three wavelengths for LED displays that manage to show us millions of colours, HumOns are trichromats right? So the thinking goes - we should also be able to use narrow band light sources to illuminate our physical environment with the same results. But this is just not how illumination works in the physical world.
The difference with displays, is that we are looking directly at the light source, the light source is the image; with lamps, we are looking indirectly at reflections from our physical environment, and physical surfaces reflect different wavelengths with different intensities, i.e they have a spectral reflectance curve (and these can be pretty crazy, not simple smooth curves). The other key detail is that our eyes are not narrow band, they merely quantise broad spectrum into three colours. So just because some surface might reflect brightly at 480nm but not 420nm (humans peak sensitivity), doesn't mean we don't see it, it still looks blue, just less so. The issue is that most standard LED lighting spectral output (phosphors and all) might focus on 420nm so that it looks white, but might have a dip at around 480nm so that reflections on that particular surface appear less blue than they should.
This is the issue people are experiencing when some things look too dark or too grey, surfaces have complex reflectance curves, most cheap LED lights have extremely poor spectral distribution and they can basically miss those surfaces. The more spikey and narrow band a surface's reflectance curve the more likely a narrow band light source is going to miss the reflection wavelengths and do a poor job of illuminating it.
[edit]
To be more fair to LED lighting, the comparison I made above is not entirely fair. Unlike a monitor's output LED lighting actually attempts to get broader spectrum output by adding phosphor layers to convert wavelengths of the LED, it's just that most ordinary price household LED lamps are still very poor in their distribution. It's not impossible to generate a better spectral distribution with LED sources, it's just expensive, and requires more than one LED source wavelength+phosphor combination. The other issue with cheap LEDs you might notice is that the phosphor layers tend to degrade, changing the colour temperature over time, and the inverters commonly break before the filament resulting in a very short and flickery lifetime.
I haven't used these, but this thread instigated a search, seems like there are broad spectrum LED manufacturers like this one [0]. Their target customer makes sense... Some art will undoubtedly look weird in narrow band light.
Consider Spectral Reflectance.
We use only three wavelengths for LED displays that manage to show us millions of colours, HumOns are trichromats right? So the thinking goes - we should also be able to use narrow band light sources to illuminate our physical environment with the same results. But this is just not how illumination works in the physical world.
The difference with displays, is that we are looking directly at the light source, the light source is the image; with lamps, we are looking indirectly at reflections from our physical environment, and physical surfaces reflect different wavelengths with different intensities, i.e they have a spectral reflectance curve (and these can be pretty crazy, not simple smooth curves). The other key detail is that our eyes are not narrow band, they merely quantise broad spectrum into three colours. So just because some surface might reflect brightly at 480nm but not 420nm (humans peak sensitivity), doesn't mean we don't see it, it still looks blue, just less so. The issue is that most standard LED lighting spectral output (phosphors and all) might focus on 420nm so that it looks white, but might have a dip at around 480nm so that reflections on that particular surface appear less blue than they should.
This is the issue people are experiencing when some things look too dark or too grey, surfaces have complex reflectance curves, most cheap LED lights have extremely poor spectral distribution and they can basically miss those surfaces. The more spikey and narrow band a surface's reflectance curve the more likely a narrow band light source is going to miss the reflection wavelengths and do a poor job of illuminating it.
[edit]
To be more fair to LED lighting, the comparison I made above is not entirely fair. Unlike a monitor's output LED lighting actually attempts to get broader spectrum output by adding phosphor layers to convert wavelengths of the LED, it's just that most ordinary price household LED lamps are still very poor in their distribution. It's not impossible to generate a better spectral distribution with LED sources, it's just expensive, and requires more than one LED source wavelength+phosphor combination. The other issue with cheap LEDs you might notice is that the phosphor layers tend to degrade, changing the colour temperature over time, and the inverters commonly break before the filament resulting in a very short and flickery lifetime.
I haven't used these, but this thread instigated a search, seems like there are broad spectrum LED manufacturers like this one [0]. Their target customer makes sense... Some art will undoubtedly look weird in narrow band light.
[0] https://www.savemoneycutcarbon.com/category/soraa-lighting-r...