Modern headlight tint distracting

I understand that headlight tint has been discussed to the nth degree to determine exactly what tint is best for seeing objects at night, depth perception, color rendition, etc. However, I'm referring to the effect it has on nearby vehicles. Maybe it's just me, but I find modern HID and LED headlights/DRLs distracting. Some brands are more annoying than others.

I think the color of the headlight is too far to the blue side of the spectrum. I estimate it's somewhere around 5000-6000K. While this may be "true daylight", it looks blue at night. They stand out greatly from the rest of traffic. I'm willign to bet there are some models that are borderline illegal from the factory, and perhaps the test method needs updating as lights evolve.

My car has HIDs with stock 35W 4300K bulbs, and even they look a little blue to me. Based on my bicycle light projects, I find that a cooler tint diminishes clarity and depth perception. It also is more annoying to look at as a bystander.


Is there a reason why modern headlight manufacturers have chosen to use this tint of light? Maybe someone wouldn't buy car A with "yellow" headlights even if it put out just as much light as car B with "cool" headlights?

kuksul08 said:

I understand that headlight tint has been discussed to the nth degree to determine exactly what tint is best for seeing objects at night, depth perception, color rendition, etc. I find modern HID and LED headlights/DRLs distracting.

Click to expand...

It has been discussed quite a bit in other threads here in one way or another:

http://www.candlepowerforums.com/vb/showthread.php?110155-those-fake-HID-headlamps-are-annoying
http://www.candlepowerforums.com/vb/showthread.php?356544-What-s-with-the-blue-headlights
http://www.candlepowerforums.com/vb/showthread.php?378081-LED-headlights-are-blue
http://www.candlepowerforums.com/vb...il-constriction-color-temp-and-color-spectrum

kuksul08 said:

I think the color of the headlight is too far to the blue side of the spectrum. I estimate it's somewhere around 5000-6000K. While this may be "true daylight", it looks blue at night

Click to expand...

The problem with "true daylight" is that there are at least ten different SPDs and color temps of light that are true daylight. Sunny with blue skies? True daylight. Sunrise with flaming fireball of orange (sailors take warning!)? True daylight. Partly cloudy? True daylight. Overcast? True daylight. (Seeing the pattern?)

Is there a reason why modern headlight manufacturers have chosen to use this tint of light? Maybe someone wouldn't buy car A with "yellow" headlights even if it put out just as much light as car B with "cool" headlights?

Click to expand...

DUH, because, like COOL, duh. TUNERZ, man! HID=BETTER! HID=LUXURY! BLUER=MORE LUXURY!

Do you want a nice, crisp, cold, refreshing Coca-Cola, or do you want this fizzy brown sugar water?

The high color temp is a characteristic of the more efficient HID burners, not a deliberate choice. likewise, high color temp LED emitters are more efficient than neutral or warm tints. I'd prefer a nice 219-like headlight but I've heard repeatedly that studies don't show a great disadvantage to low CRI or high color temp vehicle headlamps.

Alaric Darconville said:

The problem with "true daylight" is that there are at least ten different SPDs and color temps of light that are true daylight. Sunny with blue skies? True daylight. Sunrise with flaming fireball of orange (sailors take warning!)? True daylight. Partly cloudy? True daylight. Overcast? True daylight. (Seeing the pattern?)


DUH, because, like COOL, duh. TUNERZ, man! HID=BETTER! HID=LUXURY! BLUER=MORE LUXURY!

Do you want a nice, crisp, cold, refreshing Coca-Cola, or do you want this fizzy brown sugar water?

Click to expand...

You speak of ricers, not tuners

when most people think of daylight, I believe they are speaking of the relatively unhampered noon day sun...I can imagine ssunlight might look a tad bluer in space because there are no atmospheric particles to hinder the blue frequency. Perhaps 5k CCT really is true sunlight and our eyes aren't used to seeing it.

On a side note, I still believe that CCT and CRI have something to do with the near invisible spectrums of IR and UV.

dc38 said:

when most people think of daylight, I believe they are speaking of the relatively unhampered noon day sun...I can imagine ssunlight might look a tad bluer in space because there are no atmospheric particles to hinder the blue frequency. Perhaps 5k CCT really is true sunlight and our eyes aren't used to seeing it.

Click to expand...

At what latitude? In which season?

On a side note, I still believe that CCT and CRI have something to do with the near invisible spectrums of IR and UV.

Click to expand...

Near-invisible means it's visible. CRI has more to do with our red, green, and blue color perception, and the spectral power distribution of the light source. (Fun fact! This is not yellow! It looks yellow, but your monitor isn't emitting yellow light.)

I didn't think about latitude...hm....

i learned before grade school that monitors and light related imaging generally operate on RGB rather than tangible pigmented RYB (with some exceptions, one of which is that led tv with RGBY) Also, our eyeballs aren't meant to look at light SOURCES for extended periods of time, so the perceived colors from a monitor may vary slightly than what is seen in real life, even given the proper color correction.

dc38 said:

i learned before grade school that monitors and light related imaging generally operate on RGB rather than tangible pigmented RYB (with some exceptions, one of which is that led tv with RGBY)

Click to expand...

I'd really like to see one of the RGBY monitors, so I can go "Oh, myyy" in my best George Takei voice.

I think Amazon needs to produce a non-backlit Kindle that uses CMY "e-ink". That would be pretty snazzy.

Also, our eyeballs aren't meant to look at light SOURCES for extended periods of time, so the perceived colors from a monitor may vary slightly than what is seen in real life, even given the proper color correction.

Click to expand...

They constantly adjust everything. You get the monitor set up "just so" and then the light through the window changes with the time of day or the season or a passing cloud. Another reason that CRI and SPD don't really factor that much into our night driving experience.

The white-space allows for up to 6500K or so CCT, at night. That needs to be trimmed way down to 4750K or so.
The problem is the sun is overhead, not a low angle into people's eyes. and is nothing but marketing department drivel that 'closer to daylight' is appropriate for nocturnal driving.

MichaelW said:

The white-space allows for up to 6500K or so CCT, at night. That needs to be trimmed way down to 4750K or so.
The problem is the sun is overhead, not a low angle into people's eyes. and is nothing but marketing department drivel that 'closer to daylight' is appropriate for nocturnal driving.

Click to expand...

The CIE colorspace for white is:
x = 0.310 (blue boundary)
x = 0.500 (yellow boundary)
y = 0.150 + 0.640x (green boundary)
y = 0.050 + 0.750x (purple boundary)
y = 0.440 (green boundary)
y = 0.382 (red boundary)

Light is either white, or it's not-- the color temperature, or coördinated color temperature, is not referenced in the definition, therefore there is no "trimming" to be done. SPD, not CCT, is defined in the CIE color system.

Yeah, you can redefine the leftmost boundary with a line that corresponds to a CCT. Not hard to do.
The BBL is inside this box, and most lighting is on, or nearly on the BBL line. The exception being LED lighting.
So it is better to stop this horrible proliferation of blue lights, now rather than later.

I estimate it's somewhere around 5000-6000K. While this may be "true daylight"

Click to expand...

Despite the persistent marketing claims, neither HID nor LED headlamps produce light that is, in any real sense, like daylight (at any lattitude, in any weather, at any time of day).

The high color temp is a characteristic of the more efficient HID burners, not a deliberate choice.

Click to expand...

Actually, if we confine the discussion to automotive HID bulbs (leaving out irrelevant types like sodium-vapor that aren't used in car lights), efficacy is inversely proportional to CCT: as CCT goes higher, efficacy goes lower.

likewise, high color temp LED emitters are more efficient than neutral or warm tints.

Click to expand...

That is generally true, but we are well into (and going deeper into) the territory of having more light than needed available from LED headlamp light sources, so there's "room" to specify the LEDs with slightly lower efficacy. Moreover, even if we feel the need to grasp at every last possible lumen, that disregards the question of how usable the light is. If we're counting all the light, even the blue light that's unusable or counterproductive, we're not really making an honest count.

I'd prefer a nice 219-like headlight

Click to expand...

What is a "219-like" headlight?

but I've heard repeatedly that studies don't show a great disadvantage to low CRI or high color temp vehicle headlamps.

Click to expand...

The first part (no safety effect from relatively low CRI) is correct. The second part (no safety effect from CCT) is not -- take a look at that last thread Alaric links.

dc38 said:

I didn't think about latitude...hm....

i learned before grade school that monitors and light related imaging generally operate on RGB rather than tangible pigmented RYB (with some exceptions, one of which is that led tv with RGBY) Also, our eyeballs aren't meant to look at light SOURCES for extended periods of time, so the perceived colors from a monitor may vary slightly than what is seen in real life, even given the proper color correction.

Click to expand...

I don't think you mean RYB for pigments but CYM. One is an additive color system and one is subtractive. Whether a source emits or reflects .... The eye/brain has no idea.

Semiman

Personally, I believe my human eyes evolved for their low light vision to be best using sunrise or sunset light, and that runs more to the red/orange/yellow side.

It's unfortunate, but it seems the earliest automotive HID lights were a bit bluish and that started a horrible fad. Suddenly, blue is the cool thing, and buzzwords like "xenon" and "hyper-white" start flying around. I keep hoping some other, less annoying fad will take over, perhaps a trend toward selective yellow - let's all be "retro-French" or something. But until a new fad arrives, we're all stuck tolerating annoying blue lights.

Lately, I've been watching the effect of signals in close proximity to headlights, recalling conversations here where Scheinwerfermann pointed out why we have rules about both brightness of signals and distance from main headlights. One small local privately owned newspaper delivery car that runs around with headlights and flashers on, seems to have succumbed to the glaring blue headlight fad. I don't think it helps visibility of his flashers.

I'm curious about the effect of headlight color on the visibility of adjacent signal flashers. On an intuitive level, one would think a less yellow headlight would have more contrast against the yellow of the flasher.

BTW, last night I saw some moron who had apparently intalled bluish-white LED signals on the rear of his minivan (which appeared to be designed for a yellow signal bulb behind a clear lens). The left in particular was quite dim. Neither was good, and I rather expect that a blue-white light on the rear of a forward moving vehicle will eventually earn him a ticket.

We went to a city for dinner, saw plenty of police activity. They are using various shades of blue and red on their flashers. But the wreckers with yellow and red on their light bars seemed more easily visible at distance.

What is a "219-like" headlight?

Click to expand...

this being cpf, it probably refers to the x19-H1 family of Nichia's hi-cri, typically sw45 tinted leds,
http://dmcleish.com/CPF/HighCRI/NichiaLED-CRI-Info.jpg

Hamilton Felix said:

Personally, I believe my human eyes evolved for their low light vision to be best using sunrise or sunset light, and that runs more to the red/orange/yellow side.

It's unfortunate, but it seems the earliest automotive HID lights were a bit bluish and that started a horrible fad. Suddenly, blue is the cool thing, and buzzwords like "xenon" and "hyper-white" start flying around. I keep hoping some other, less annoying fad will take over, perhaps a trend toward selective yellow - let's all be "retro-French" or something. But until a new fad arrives, we're all stuck tolerating annoying blue lights.

Lately, I've been watching the effect of signals in close proximity to headlights, recalling conversations here where Scheinwerfermann pointed out why we have rules about both brightness of signals and distance from main headlights. One small local privately owned newspaper delivery car that runs around with headlights and flashers on, seems to have succumbed to the glaring blue headlight fad. I don't think it helps visibility of his flashers.

I'm curious about the effect of headlight color on the visibility of adjacent signal flashers. On an intuitive level, one would think a less yellow headlight would have more contrast against the yellow of the flasher.

BTW, last night I saw some moron who had apparently intalled bluish-white LED signals on the rear of his minivan (which appeared to be designed for a yellow signal bulb behind a clear lens). The left in particular was quite dim. Neither was good, and I rather expect that a blue-white light on the rear of a forward moving vehicle will eventually earn him a ticket.

We went to a city for dinner, saw plenty of police activity. They are using various shades of blue and red on their flashers. But the wreckers with yellow and red on their light bars seemed more easily visible at distance.

Click to expand...

The various manufacturers of emergency lights have admitted to struggling to find LEDs that are consistently the correct shade of colour, especially the blue LED.
Compounding the issue is the habit of end users of interweaving red/blue through the light bars and using split colour heads with poor choice of flash pattern.
Numerous studies have shown that in daylight conditions red the most visible colour while blue is more effective at night, due to the red emergency lights blending in with other traffic.

irsa76 said:

The various manufacturers of emergency lights have admitted to struggling to find LEDs that are consistently the correct shade of colour, especially the blue LED.

Click to expand...

Some [citation needed] going on here.

Compounding the issue is the habit of end users of interweaving red/blue through the light bars and using split colour heads with poor choice of flash pattern.

Click to expand...

They're going to use what they bought-- the lights are designed to be placed alongside each other.

Numerous studies have shown that in daylight conditions red the most visible colour while blue is more effective at night, due to the red emergency lights blending in with other traffic.

Click to expand...

Which studies?

It's more complex than you think-- the human eye is most sensitive to green, for one. Also, if you're using white light and filtering for red, vs. using an LED that is designed to emit a particular color, there will be differences in how they are detected and whether they will create disability glare or not.

Hamilton Felix said:

Personally, I believe my human eyes evolved for their low light vision to be best using sunrise or sunset light, and that runs more to the red/orange/yellow side.

Click to expand...

That's a reasonable-sounding belief...but so is the one that goes "I believe my human eyes evolved for their low light vision to be best using dawn or twilight, and that runs more to the blue side". Not only do beliefs (as such) really have no place in a facts/science-based discussion, but the definition of "low light vision" is a moving target. Night driving conditions are not photopic (daytime light levels) and they are not scotopic (nighttime light levels). They are mesopic.

It's unfortunate, but it seems the earliest automotive HID lights were a bit bluish and that started a horrible fad.

Click to expand...

The earliest automotive HIDs were almost as blue as the latest automotive HIDs. It probably didn't have to be that way, but automakers are driven by their marketing departments, and they wouldn't touch a premium-priced headlamp that looks about the same as the regular-price item.

Suddenly, blue is the cool thing, and buzzwords like "xenon" and "hyper-white" start flying around.

Click to expand...

...with even the legitimate manufacturers jumping in, and their marketing departments grotesquely distorting and exaggerating "scientific proof" that so-called "whiter" light is better, not to mention the bogus "closer to natural sunlight" claims, etc.

I keep hoping some other, less annoying fad will take over, perhaps a trend toward selective yellow

Click to expand...

Not terribly likely, though maybe if the engineers win (which they won't; the marketers already have)

Lately, I've been watching the effect of signals in close proximity to headlights. I'm curious about the effect of headlight color on the visibility of adjacent signal flashers. On an intuitive level, one would think a less yellow headlight would have more contrast against the yellow of the flasher.

Click to expand...

Yes, a bluer headlamp can increase conspicuity of the nearby amber turn signal via increased color contrast. But it doesn't rise to the level of making it into the regulations (there's no linkage of required turn signal intensity vs. headlamp color, for example).

One small local privately owned newspaper delivery car that runs around with headlights and flashers on, seems to have succumbed to the glaring blue headlight fad. I don't think it helps visibility of his flashers.

Click to expand...

Well, if his headlamps are suddenly "Glaring blue", he's probably installed an "HID kit", which is illegal and unsafe regardless of what color it is.

BTW, last night I saw some moron who had apparently intalled bluish-white LED signals on the rear of his minivan (which appeared to be designed for a yellow signal bulb behind a clear lens).

Click to expand...

Yuck. I see this kind of dumb junk far too often...

I rather expect that a blue-white light on the rear of a forward moving vehicle will eventually earn him a ticket.

Click to expand...

...I don't. The police either don't care, don't think they can make the ticket stick in court, or can't keep up with the zillions of different-looking lights that are legal, let alone the ones that are illegal.

We went to a city for dinner, saw plenty of police activity. They are using various shades of blue and red on their flashers.

Click to expand...

Thank heaven the Xenon flash/strobe tubes have mostly been chased off the road by LEDs!

But the wreckers with yellow and red on their light bars seemed more easily visible at distance.

Click to expand...

Up in Canada, Ontario changed from red to red/blue on their police cars after finding that the combination does a better job than the single color (see here for a not-very-detailed report on what was actually a large amount of research). To (over)simplify, red lights are closer than they appear, while blue lights are farther away than they appear -- these two effects are variable depending on ambient light levels, too. Using both colors allows the two effects to cancel each other out.

Scheinwerfermann said:

Actually, if we confine the discussion to automotive HID bulbs (leaving out irrelevant types like sodium-vapor that aren't used in car lights), efficacy is inversely proportional to CCT: as CCT goes higher, efficacy goes lower.

Click to expand...

Understood. You are right and I know that, I worded my post badly. What I was trying to get across is that I don't believe that it is possible to make a HID light with a color temp as low as that of a halogen bulb (or is it? I haven't seen one, anyway) nor a halogen bulb with a CCT significantly higher than around 3000K, and that the HIDs are more efficient than the halogens. But yes, I would assume that the sweet spot for efficacy of HID would be around the popular 4200-4500K range, otherwise the OEMs would be using something else. And yes, the higher CCT "tuner" HIDs aren't as efficient as OEM HIDs. The distinct difference between the 1st gen HIDs and halogens in terms of CCT of course gave rise to the *preference* for higher CCTs among the general buying public as they associated the higher CCT of the HIDs with being newer, better, brighter, and more desirable without understanding that it was not necessarily the higher CCT that made HIDs desirable but other factors. And of course the "keeping up with the Joneses" factor of it being obvious to anyone seeing the car at night that it was equipped with the newer, better, more expensive technology.

That is generally true, but we are well into (and going deeper into) the territory of having more light than needed available from LED headlamp light sources, so there's "room" to specify the LEDs with slightly lower efficacy. Moreover, even if we feel the need to grasp at every last possible lumen, that disregards the question of how usable the light is. If we're counting all the light, even the blue light that's unusable or counterproductive, we're not really making an honest count.

Click to expand...

But every engineer I know would rather use 9.5W than 10W for the same light output unless you can make a compelling case to him that using the lower CCT/higher CRI emitter is really better. and of course there is still the perception that "low" e.g. 3000Kish CCT headlights are old school, outdated, and undesirable so the marketing/product planning guys will probably still be pushing for higher CCTs unless/until the market is saturated with HIDs to the point where their bling factor drops off.

What is a "219-like" headlight?

Click to expand...

I meant a headlight with the CCT and CRI of the Nichia 219 high CRI emitter... lovely thing it is.

The first part (no safety effect from relatively low CRI) is correct. The second part (no safety effect from CCT) is not -- take a look at that last thread Alaric links.

Click to expand...

I'll check it out later... stuff to do ATM, but I will definitely make an effort to come back to this.

Great posts, thanks for responding.

I was always under the impression that a hot tungsten filament was closer to a black body emitter (the sun) than 'artificial' sources such as an LED or HID bulb, where colors can vary greatly. That is why, in my opinion, a halogen bulb is much easier on the eyes and makes things look more natural. Is there any truth to this?

kuksul08 said:

Great posts, thanks for responding.

I was always under the impression that a hot tungsten filament was closer to a black body emitter (the sun) than 'artificial' sources such as an LED or HID bulb, where colors can vary greatly. That is why, in my opinion, a halogen bulb is much easier on the eyes and makes things look more natural. Is there any truth to this?

Click to expand...

Yes. An incandescent light source has a very high CRI (Color Rendering Index) 100 or very close to it. If you don't know what that means, hopefully I've given you the keywords you need to start reading; warning, you'll probably get sucked into an internet rabbit hole but you might learn a lot.

The other thing that is worth pointing out is that CCT (Correlated Color Temperature) is independent of CRI and vice versa. The headlights in my Jeep are incan H4s so they likely have a CRI of 95+ and a CCT of somewhere in the 2900-3000K range. The HIDs in my BMW have a CCT of somewhere in the 4200-4300K range and likely a not very good CRI. However I have a Nichia 219 P60 drop in in an old Surefire flashlight which has a CCT of around 4500K but a 92+ CRI. So there's really two questions to be asked:

1) is there an ideal or an ideal range of CCTs for automotive headlamps?
2) Of course higher CRI is always "better" but how important is it in this context?

I'm not going to hazard a guess at 1) because I'm still reading the interesting (in a geeky way) thread that Alaric posted and Headlight Guy pointed me towards. My personal preference on 2) is heck yeah I want high CRI headlamps but I've been given repeated references to the effect that the difference in actually helping you see *in the context of driving at night* is not that significant.