What's the lowest lumen rating that the eye can detect? Less than .003 lumens?

kaichu dento

Flashaholic
Joined
Apr 5, 2008
Messages
6,554
Location
現在の世界
As many of you have the Thrunite Ti in possession with .04 lumen rating have already discovered, it's much brighter than many would have thought. Just a couple years ago we had a thread based around the 2 lumen mark and others have followed, but what has not (at least to my knowledge) been verified is the absolute lowest lumen rating that one could possibly see by.

Just to be clear, the topic is not what's practical, as that is a variable, but what the lowest that a dark adapted eye of either most people, or even those with extremely sensitive eyes could see by, and at what distance.

My personal interest in this is to find what actual output level would make super-low output level fans like BeaconOfLight, myself and others finally satisfied. I've heard many times about lights that went so low as to be useless, from the Novatac 120, all of Henry's lights, the Titan and countless others. But every time I got my hands on those same lights I realized they were damn bright in context of a super-low output rating and recently my curiosity was once again piqued by the arrival of my programmable Calipsoii ring for my Surefire A2.

Doing a direct comparison so far with my hCRI Clicky and Ti the Calipsoii ring on one emitter is capable of doing what my awkward visual acuity would tell me is aproximately a quarter the output of the Ti, if not less, while the Clicky is nothing short of blinding at it's rated .07-.08.

One light that I know can easily go well below usable levels is the Spy series, but I still have no idea of what the actual lumen rating would be at the Spy or Calipsoii settings, and that's what I'm hoping to find out here.

My suspicion for some time now has been that at least down to .01 lumen, and quite possibly down to .001 is usable depending on distance, setting, general light sensitivity of user, coupled with comfort level at working in low light. I would naturally assume that night hikers, cavers and such would innately understand this topic, hope that all can benefit from the discussion, and moreover hope for some input from the integrating sphere/lux meter owners out there.

Just to be sure, even if the lumen scale is not the best for scientific purposes, it is what most lights are rated by at present and if in the future a better scale is used for this application, maybe we'll all benefit. For the time being, lumens it is, for the layman.

*As this thread goes along I'm going to try and keep updating the title to show the lowest visible lumen rating available.*
 
Last edited:

127.0.0.1

Flashlight Enthusiast
Joined
Jan 4, 2012
Messages
1,000
Location
/etc/hosts
Re: What's the lowest lumen rating that the eye can detect? Less than .01 lumens?

very open-ended question

lumens to do what ?

detect movement
comprehend shapes
read text
navigate

the eye can do many things but it depends on 'what task' and 'what light source' and
especially 'what specific wavelengths of light'
 
Last edited:

kaichu dento

Flashaholic
Joined
Apr 5, 2008
Messages
6,554
Location
現在の世界
Re: What's the lowest lumen rating that the eye can detect? Less than .01 lumens?

I wanted to try and elicit input from all the possibilities you've mentioned, partially out of a desire to find out what actual numbers would be for edge-of-blindness output levels, and also to finally satisfy some of the curiosity here as to what it actually meant when we see every lower lumen ratings from various companies, but never with much of a yardstick that is easily comprehended by the laymen among us.

The fact that the eye can do many things is one of the things that makes it so difficult to settle on an absolute 'best emitter, best light' for all, and of course the light source would be flashlights in normally useful ranges. Of course if someone wants to point out that their red or green or blue or neutral, cool, hCRI or whatever is more effective than another at given ranges, that may be interesting too, but the thing I was most hoping to find was someone willing to do some middle of the night, eyes completely dilated, integrating sphere testing.

What is the lowest output on a Titan and a ww Calipsoii ring with one emitter on it's lowest setting? The latter is almost low enough to satisfy me.
 

Johnbeck180

Enlightened
Joined
May 11, 2011
Messages
511
Location
Central Indiana
I just received a Jetbeam RRT-0 with infinite variable ring. The low on it is, from what I've read on the reputable site I ordered it from is .003 lumens. If I were in a place I've never been Before(in the dark of course) this light on low would be no help at all. beyond maybe 2-4 feet.
 

MikeAusC

Enlightened
Joined
Jul 8, 2010
Messages
995
Location
Sydney, Australia
Re: What's the lowest lumen rating that the eye can detect? Less than .01 lumens?

1 Lumen lighting up my fingernail will be very visible.

1 Lumen spread over the side of my house will be hardly detectable.

Do you mean the minimum Lux level that's visible ?
 

kaichu dento

Flashaholic
Joined
Apr 5, 2008
Messages
6,554
Location
現在の世界
Re: What's the lowest lumen rating that the eye can detect? Less than .01 lumens?

John, judging from your estimation of useful out to 2'-4', I'd say we can safely assume a level as low as .001 to be easily visible.

Mike, 1lumen at close range is blindingly bright when it's really dark, and I could easily detect the side of a house with the same lumen.

To tell the truth I didn't really expect much in the way of responses as there are only so many of us who truly appreciate extreme low levels, but since it's something that I care a lot about, I'll keep hoping for some more help in the quest to validate the lowest usable output level.

To clarify for those of you who've responded so far, if it's usable only at 1' then it's still usable. A few years ago I was walking along using the lowest setting on my D10 on a very dark night, and enjoying the reasonable light it cast on the road, but when I went to look at a sliver in my hand, I was shocked (dismayed) at how overly bright it was, and there was no turning it any lower.
 

GaAslamp

Enlightened
Joined
Aug 5, 2011
Messages
361
Re: What's the lowest lumen rating that the eye can detect? Less than .01 lumens?

As many of you have the Thrunite Ti in possession with .04 lumen rating have already discovered, it's much brighter than many would have thought. Just a couple years ago we had a thread based around the 2 lumen mark and others have followed, but what has not (at least to my knowledge) been verified is the absolute lowest lumen rating that one could possibly see by.

In short, answering such a question is problematic, I think. There are not only the issues that 127.0.0.1 brought up, but objectively speaking, the human eye (specifically scotopic or "rod" vision) is extremely sensitive, even down to the quantum level. A single photon is all it takes to get the retina to respond, although a few more within a short time span are required to trigger perception, as described in more detail here:

http://math.ucr.edu/home/baez/physics/Quantum/see_a_photon.html

The scale of the amount of light we're talking about is so small that it's possibly "lost in the noise" at the scale of the lumen ratings of mass-produced flashlights.

By the way, many other creatures have adaptations that allow them to generally see better than we do in the dark, but if the question is absolute minimum sensitivity (i.e. able to see anything at all), then the difference, in absolute terms, cannot be that great.

My personal interest in this is to find what actual output level would make super-low output level fans like BeaconOfLight, myself and others finally satisfied. I've heard many times about lights that went so low as to be useless, from the Novatac 120, all of Henry's lights, the Titan and countless others. But every time I got my hands on those same lights I realized they were damn bright in context of a super-low output rating and recently my curiosity was once again piqued by the arrival of my programmable Calipsoii ring for my Surefire A2.

Those who claim that such levels of light, which to give you a sense of scale involves multiple trillions of photons per second, are useless were probably not using their eyes to the fullest. Few people these days ever do, outside of amateur astronomy, and even then only in exceptional cases (e.g. traveling to a dark site in the middle of a desert).

Doing a direct comparison so far with my hCRI Clicky and Ti the Calipsoii ring on one emitter is capable of doing what my awkward visual acuity would tell me is aproximately a quarter the output of the Ti, if not less, while the Clicky is nothing short of blinding at it's rated .07-.08.

A reasonable lower bound for the number of photons of visible light in 0.01 lumens is about 29 trillion per second (the real number is somewhat higher than that). Remember that humans can usually perceive (above the level of noise) as few as 9 photons entering the eye. While this is under extremely controlled conditions, and contrast is what really matters in practical use, the full scope of what we're talking about is immense. Under the right conditions, even 0.01 lumens could be considered "a lot" of light--we need to insert more zeros before we truly get to a minimal amount. :) For activities like reading, however, the required amount of light would be significantly greater.

One light that I know can easily go well below usable levels is the Spy series, but I still have no idea of what the actual lumen rating would be at the Spy or Calipsoii settings, and that's what I'm hoping to find out here.

How dark-adapted were you when you checked? It takes at least 30 minutes of complete darkness to attain reasonable dark adaptation, and a lot longer than that to find out just how little light you could get away with.

My suspicion for some time now has been that at least down to .01 lumen, and quite possibly down to .001 is usable depending on distance, setting, general light sensitivity of user, coupled with comfort level at working in low light.

I've seen vague references (I'll have to do more research) to 10^-6 cd/m² as being the minimum luminance necessary for human scotopic vision, but I don't know how this was determined. Roughly taking things like reflectivity into account, it means that from a few feet away 0.00001 lumens should supposedly be barely enough to illuminate a square meter area. Obviously you could get away with even less light than that if you illuminated a smaller area and/or viewed it from a shorter distance, though.

I would naturally assume that night hikers, cavers and such would innately understand this topic, hope that all can benefit from the discussion

Maybe cavers (under some circumstances) and amateur astronomers (although we can use a bit more light than absolute minimum as long as it's red, and already have specialized flashlights that do the job), but hikers tend to prefer plenty of light for safety reasons.

The fact that the eye can do many things is one of the things that makes it so difficult to settle on an absolute 'best emitter, best light' for all, and of course the light source would be flashlights in normally useful ranges. Of course if someone wants to point out that their red or green or blue or neutral, cool, hCRI or whatever is more effective than another at given ranges, that may be interesting too,

This is not merely interesting, but can be substantial at the extremes of human vision. One fact that is particularly relevant due to the prevalence of white LEDs in flashlights is that they all have a "valley" or "dip" in their spectra right about where human scotopic vision is the most sensitive. This means that for all I tried to convey above, maybe with LEDs specifically we need more total lumens in order to see at the lower end of the scale. In this respect, some--albeit not all--high-CRI LEDs are significantly better than others; they still have that dip, but it's a lot shallower. Tints such as cool white, neutral white, or warm white do not necessarily imply better or worse sensitivity at low light levels--it depends on the spectrum of the individual LED, not its overall color temperature. Monochromatic LEDs that cover this range--centered near 500nm--are often called aqua, turquoise, or NV-green (night vision green).

but the thing I was most hoping to find was someone willing to do some middle of the night, eyes completely dilated, integrating sphere testing.

By the way, dilation happens quickly and has less to do with dark adaptation than chemical processes within the retina, which take some time.
 
Last edited:

MikeAusC

Enlightened
Joined
Jul 8, 2010
Messages
995
Location
Sydney, Australia
Re: What's the lowest lumen rating that the eye can detect? Less than .01 lumens?

. . . . A few years ago I was walking along using the lowest setting on my D10 on a very dark night, and enjoying the reasonable light it cast on the road, but when I went to look at a sliver in my hand, I was shocked (dismayed) at how overly bright it was, . . . . . .

???? That's exactly the point I was trying to raise - an amount of lumens can be very bright or very dim - UNLESS you specify a distance or area over which it's spread. Lumens is the TOTAL AMOUNT of light a torch puts out.

If you want to refer to a level of illumination that's independant of distance - you need to use Lux or Candela per square metre.
 

AnAppleSnail

Flashlight Enthusiast
Joined
Aug 21, 2009
Messages
4,200
Location
South Hill, VA
Re: What's the lowest lumen rating that the eye can detect? Less than .01 lumens?

I've lately been hooking up 10k potentiometers to 2 NiMH cells driving white LEDs. I am usually comfortable seeing to walk around my (newly moved into apartment with stairs) at what must be a drive current of (2.8/10000) or 0.28 mA. For a white LED, that is less than 0.1 lumens, right? Just shining a 20 degree spread white LED out ahead of me when I get up in the middle of the night.
 

MikeAusC

Enlightened
Joined
Jul 8, 2010
Messages
995
Location
Sydney, Australia
Re: What's the lowest lumen rating that the eye can detect? Less than .01 lumens?

. . . . I've lately been hooking up 10k potentiometers to 2 NiMH cells driving white LEDs. I am usually comfortable seeing to walk around my (newly moved into apartment with stairs) at what must be a drive current of (2.8/10000) or 0.28 mA.. . . . .

???? To use ohms law to calculate current you need to use (2.8-VLED)/10000.


. . . . For a white LED, that is less than 0.1 lumens, right? . . . . .

LED efficiency for white LEDs takes a nosedive at low currents - and that's typically once you go below 1/10th of the rated current of the LED. Deriving output based on very low drive currents just doesn't work.
 
Last edited:

AnAppleSnail

Flashlight Enthusiast
Joined
Aug 21, 2009
Messages
4,200
Location
South Hill, VA
Re: What's the lowest lumen rating that the eye can detect? Less than .01 lumens?

???? To use ohms law to calculate current you need to use (2.8-VLED)/10000.




LED efficiency for white LEDs takes a nosedive at low currents - and that's typically once you go below 1/10th of the rated current of the LED. Deriving output based on very low drive currents just doesn't work.

My current calculation ignores the LED because I don't know its Vf. The stated Vf is 3.2v, and it lights clear down to almost 2v. Similarly, I have red LEDs that shine dimly on 1 alkaline AA. I know I'd have to use a light meter to really get a brightness number, but I can visually compare to lights that I think I know the brightness of.
 

sunny_nites

Enlightened
Joined
Mar 27, 2010
Messages
491
Location
USA
Re: What's the lowest lumen rating that the eye can detect? Less than .01 lumens?

Great thread AnAppleSnail!

This is indeed an interesting question. I've been playing around with edge of detection lighting for a while and in my own experimentation (pure observation and probably not very scientific) I've found that color may have more to with what can be detected as providing light than actual lumens.

I would be interested if anyone has knowledge about the manufacture of LEDs about whether color makes a difference in efficiency or if it is all in the "eye" of the beholder?
 

127.0.0.1

Flashlight Enthusiast
Joined
Jan 4, 2012
Messages
1,000
Location
/etc/hosts
Re: What's the lowest lumen rating that the eye can detect? Less than .01 lumens?

the limit of detection is even smaller if:

the object that is bouncing photons in your direction is moving rapidly across the field of view (about the speed of a tasty rabbit)

you are not looking right at it, rather at the very periphery where your eyesight is the most sensitive


we animals are designed to eat what moves... because it has much higher energy
density than plants and whatnot....and the eyes are optimized for spotting moving things
entering the field of view and not as optimized for straight-ahead high sensitivity or non-moving objects
 
Last edited:

gcbryan

Flashlight Enthusiast
Joined
Oct 19, 2009
Messages
2,473
Location
Seattle,WA
Re: What's the lowest lumen rating that the eye can detect? Less than .01 lumens?

As has been mentioned the human eye is extremely sensitive and the answer to the original question I believe is that the eye is sensitive to any light. It probably isn't possible to come up with some level of light that isn't visible to the human eye.
 

eh4

Flashlight Enthusiast
Joined
Oct 18, 2011
Messages
1,999
Re: What's the lowest lumen rating that the eye can detect? Less than .01 lumens?

A reflectorless, lensless, pure flood would be useful for finding out your answer to your question. ZL H501w for instance.
The next part would be to either have faith in their published lumen levels or verify them yourself.
Then make a jig to hold a opaque card against another opaque card with a hole in it, braced against the face of your light (little paper rails would be needed along the sides to keep the opaque card aligned as it was slid progressively over the light).
It could be simply made with hot melt glue (for solid, removable foundation the bezel), tape (to hold the paper rails on either side of the base card) and two black spray painted playing cards so long as it could be adjusted easily in the dark and not slide around without finger pressure -for measurement later.
For multiple measurements in the dark a fine tipped pin or exacto knife could scribe a little line where the sides of the cards overlapped, with a good memory or a voice recorder you'd know the meaning of each mark.

Better still a fancier jig with the aperture stop mechanism from a camera held in front of the light! -but that's a whole nother level of commitment.

There would still be a inaccuracies since the measurement area is flat while the light is radially radiating but should be good enough for unpaid research.

Of course the rough expedient version of the experiment can be done without jigs or measuring fractions of a circle, in a few seconds, just cover the whole lens with the black paper, turn on low and pull the paper to one side till the light is effective at a given ranged task, then guesstimate percentage of light exposed.
 
Last edited:

Sub_Umbra

Flashlight Enthusiast
Joined
Mar 6, 2004
Messages
4,748
Location
la bonne vie en Amérique
The OP poses a great and interesting question. Let me just confound things a bit. When I worked in technical theatre I occasionally had the task of training stagehands (and others in the technical space) how to use dim lights discretely. I found that often students had the required visual acuity to accomplish the task at hand but when they lacked the required concentration they couldn't see anyway.

My point is that there is another issue lurking around these perception issues, and that is that many persons have the visual acuity for a given setting but lack the concentration to be able to use what their eyes are seeing.

These are complicated issues and the answers are rarely as black and white as the would seem at first glance.
 

eh4

Flashlight Enthusiast
Joined
Oct 18, 2011
Messages
1,999
now That is really interesting, let me restate it for myself and please correct what I misunderstand.

The human eye can potentially detect double or single digit quantities of photons, but the level of discrimination (signal/noise detection) needed would require massive processing power, i.e. attention/focus.
-so that's my translated copy, if I copied.

I read a book called "Better Eyesight Without Glasses" by Gregory Bates (and he mighta established the whole 20/20 thing, can't remember) but he distinguished between fine resolution eyesight of island people who could detect a dark spot on the horizon so well as to compete with the telescopes of the day, and jungle people who could detect tiny distinctions in visually complex jungle environments with mixed ranges of leaves, vines and whatnot.
The eye developing along with the imperatives of the watcher's concentration.
He also indicated that nearsightedness was virtually unheard of in loosely organized/low neurosis societies, and very common in highly organized (high tech societies). The subject of his book of course is attention/imagination/stress reduction and optical software/hardware reprogramming.

I was too unfocused to start doing the required exercises for the couple hours daily recommended, but i did them enough to see the benefit. It's like running, doesn't really matter unless you do it.

If you want to play around with a core concept of the book without reading the book, consider these propositions:
Acute eyesight is partially physical and partly perceptual.
Poor eyesight found in correlation with the Inability to vividly Imagine Blackness...

One of the primary exercises is to relax and imagine Blackness, liquid pools of ink, of oil, etc.
Another primary exercise is to rub your hands together until they are hot and rest them gently over your closed eyes while again imagining rich, vivid black.

-just a pre-wiki stub I guess.
Oh wait, here it is already: http://en.wikipedia.org/wiki/Bates_method
 
Last edited:

Sub_Umbra

Flashlight Enthusiast
Joined
Mar 6, 2004
Messages
4,748
Location
la bonne vie en Amérique
Wow, those are good examples.

It's the old story of 'use it or lose it.' For the most part modern man has machines for seeing things in detail. We do what we do. We not only don't see everything there is to see, we don't know what we don't see, either.
 

rickypanecatyl

Enlightened
Joined
Nov 2, 2009
Messages
913
Just some random thoughts I'll throw out there. My lowest setting light is my varapower 2000 - ecspecially when running on 3 D's as oppossed to 4 C's.

I've got the thrunight scorpion V2 which advertises a .2 lumen "firefly". While sleeping in a hostel a while back with about 30 people sharing the room it was far more than enough light to navigate around with and frankly obnoxious to the others who were sleeping. To get up and go to the bathroom I covered the lens 95% and that was fine to see by. It's more than enough to in a tent to read by; the pronounced hot spot can be blinding and the spill coming from the side is plently.

The lowest setting on my varapower 2000 while running on 3 c's is signifiantly lower than that; it has no "claimed" lumens.

The next lowest I have is the .2 lumen setting on my Maelstrom G5. It's much, much lower than the .2 lumens setting no the varapower. It's actaully pretty tough to read by. It might be 10X lower than the .2 lumens on the varapower.

And the lowest (by far) that I have is the same varapower on the lowest setting when running it on 3 D's as oppossed to 4 C's. It's plenty bright to allow me to see the emittor itself though I can't see much by the emmitor. I find in useful in the absolute dark to leave it on, facing my diretion while sleeping to aid me in finding it. If it doesn't get covered, it's bright enough in complete dark to see it from 10' away. Not to be confused with seeing something 10' away which of course it does just fine with that on the next click of 32 settings up.

I'm sure others with the exact same lights may have diffferent results; I've found the consistancy of low lows are way off between brands and even between the same flashlight. I am sure my varapower .2 lumens is at least 10X as many lumens as the G5 for example.
 

tbenedict

Enlightened
Joined
Dec 10, 2008
Messages
217
Very interesting subject. From what little I have experienced, beam shape and tint play into what modes I use with lights. I am a neutral/warm fan, but it seems plain white seems to be more effective at the extreme lows.

I wonder how many folks can walk around the house with a Proton Pro on the lowest red. I think that level is best for reading a watch or something real close.
 
Top