Some Semi-Scientific Rambling: Surprises about Flashlights & How We See

Some Semi-Scientific Rambling: Surprises about Flashlights & How We See

Some people read part of my post in another thread and pushed me to begin a new thread because they thought people would find it interesting.

Allow me to begin by quoting Boaz. He made some really good comments in yet another thread. He actually made several and it’s worth reading the whole thread. (It’s a short one.)

https://budgetlightforum.com/t/-/12665#comment-257515

Boaz has the unique ability for expressing important and interesting points with the addition of some welcome humor. This is what he said in the other thread:

“OMG you can't say that !!!!

It's Mr. perfect tint ...LOL

Anonymous third party quote: “It is a great emitter and it is slightly rose' ...”

Back to comments from Boaz:

The discussion about tint always seems to blow me away ..

If you understand the inverse square law than you understand that there is virtually NO difference in a 600Lumen light and a 800lumen light ,, you can't even really see the difference ..

A neutral white tint is 7% less lumens ...NOT visual output .. The inverse square law states you need to quadruple your output /lumens numbers to see a doubling of light ..

So I see people trying to preserve a unperceivable amount of light .(Something they CAN’T SEE ) for a tint that is hugely obvious .. This makes Zero sense to me. People trading a real gain for nothing...All because somehow they want to think one is brighter ....just crazy..

Most people will say that a throwy light is brighter than a floody light too ... when in some cases it is less than half the lumens ... the numbers really don't mean squat and until you stop measuring tv's by pixels or some easy numbers that manufacturers so easily lie about or toss out, then you'll be endlessly fooling yourself .

So it appears according to the election it's easier to sell people using fear .. So lets do that ..CW makes you want to commit violence ..it's hugely bad purple and blue light ..most of which you don't see till you compare a better tint to it ..CW washes out all color and creates glaring . it makes all items seem like they are cardboard cut outs .. if you buy neutrals women will like you better ”

Back to my – KumaBear’s comments:

In fact, the entire thread “some semi-scientific thoughts on LED tint” is just one of several very interesting related threads.

I am also particularly interested in High CRI and have read some threads that were contributed to or begun by Vectrex and others who take an interest in (and have some excellent understanding of) the subject.

Having had a little scientific training in this subject and having done some hands-on experimentation, I’d like to share some facts that you’re likely to find either unbelievable or very interesting.

Either way, it’s my hope that they will be provocative.

I’m about to depart from the subject of flashlights for a little bit - but please allow me to relate part of an experiment done by Edwin Land (founder of Polaroid and a serial inventor). I believe the experiment was done in the 50s after Land invented the polarizing filter. (As I said, my guess is that you will not believe what I’m about to describe – but I replicated a form of the experiment as part of a junior high school science project.)

Please read on because it really is quite fascinating:

1) Take two slide projectors with black and white (not color) slides, both taken of the same scene.

2) Then, superimpose the images on an ordinary screen but in a dark room.

3) Take a narrowband red filter and attach it to the front of one of the projector lenses.

4) Now, look at the superimposed images on the screen.

5) The Big Question – what do you see?

6) The amazing answer – you see/perceive a scene with a full spectrum of colors!

Why???? Well, I have a T-Shirt that says “I can explain it to you but I can’t understand it for you.” To be frank, I still don’t totally understand this phenomenon but there is a great write-up of the experiment in addition to a couple others along the same lines in an old Scientific American magazine.

By the way, Land did pursue the phenomenon as a potential method for producing his first marketable instant color film.

I asked him why he didn’t end up commercializing the results and part of the answer had to do with his inability to get adequate color saturation.

ADDENDUM #1) There is a little more to this than I’m describing but one key point is that the two slides were taken under different circumstances. Yes, they were both taken with black and white film but one was taken with a red filter on the lens and the other was not.

Other scientists, layman and photographers would say that comes nowhere near explaining how two black and white slides and one red filter can produce a spectrum of color. Before seeing it for themselves everyone you query will predict that you will produce an image that ranges in color from red to black with various shades of pink in-between.

The traditional view of how the eye perceives light and color were challenged by the results of this and related experiments and to this day, the whole “rod and cone” explanation and all that goes along with it survives pretty much intact regardless.

ADDENDUM #2) I could go on for pages about this topic but this is neither the time nor the place for that type of dissertation.

Here is a link to a very limited and abbreviated discussion of the “Land Effect” as it has come to be named:

http://www.aw3rd.us/scief/colorviz.htm

ADDENDUM #3) My son and I revisited the subject of Land’s experiments with color and human perception of same when it was his time to produce a science experiment back in 1993. We actually demonstrated the effect in a totally different manner – something that I’ll have to write up for anyone that might be interested.

Back to a subject relevant to the BLF and flashlights:

I have taken a keen interest in High CRI LEDs – not what Cree calls High CRI, but something well into the 90 plus percent range. In reading comments and discussing this subject with quite a few members, I’ve found that it’s a subject that most people have trouble “getting their heads around.”

The difference between a CW XML and an NW XML is in the difference between the AVERAGE spectral temperatures of the light they produce – as measured in Kelvin. It turns out that neither LED, whatever Kelvin you decide is NW, will actually reflect back the full range of real life colors proportionate to the way they appear under “natural light.” In fact, the difference is huge. Let me suggest we set aside what we want to call “natural light” for the moment and concentrate on what those LEDs are doing to what we actually see when we shine one of our flashlights on a colorful scene at night.

This subject has also been covered in other threads and I don’t want to go on forever but what is clear is that there are a lot misconceptions and misunderstandings.

Allow me to be intentionally provocative for a moment – the purpose being to stimulate interesting debate and encourage others to “pick these points up where I leave off.”

A) The beam comparisons that we see in tests are interesting, fun, and instructive but often highly misleading – and what appears to be a significant difference in output intensity (accurately measured in lumens) is (as Boaz pointed out) totally unimportant. Take one of the lights out to walk your dog tonight and take the other out tomorrow night. All else being equal, you won’t notice a difference even though there is a rather large difference in lumens.

B) The binning and labeling process that Cree uses leaves a lot to chance. An LED that is rated as having a higher output may indeed have a lower output as measured by accurate instruments – that’s partly because the process is simply not that accurate (something that Cree admits in their specs – and it may be even less accurate than they say- or than sellers would have you believe). Add that fact to the logarithmic nature of the eye’s sensitivity to light and these numbers that we throw around with so much import, should be taken far more lightly (pardon the pun).

C) IMHO, almost all of us (and I include myself here), make buying decisions based on the wrong criteria!

D) COLOR ACCURACY gets short shrift: Personally, I think it would be great if we could see comparison shots of an array of color construction paper or a colorful real-life scene. I’m not sure how this would be accomplished but if we’re comparing two flashlights, rather than having two photos, how about three – one being taken in “natural light” so that we know what those colors should actually look like.

Granted, its fun to see how far away something can be while still being able to light it up. To cite another example, if you’re riding your bike at night, it’s certainly not as important to see the “natural color” of that pothole in the road as it is to see it in time to avoid it.

Needless to say, there are many examples like the above but I would say that color accuracy has gotten “short shrift” in our discussions and tests – just an opinion (and debate is welcome).

E) “Pencil Beam” vs. “Peripheral Vision”: I often find myself trying to find something in a dark area. A flooder, a zoomer or even a light with a really good usable spill are essential to my practical use of a light. There are those that use their lights strapped to a hunting rifle or some other specialized purpose but how many of us pick up a light simply to impress our friends with how far off it will shine?

Please don’t think I frown on this practice. It’s fun! – And I do it myself. On the other hand, if a light is going to get good use and not be relegated to “shelf-queen” status, how should it function? Again – opinions and debate welcome.

F) Quality, Durability and User Interface: I throw these topics in together solely for lack of time and space. Collectively, these areas are almost always attended to by our fantastic reviewers. I just wish all the manufacturers did the same. I could never be part of the “money is no object crowd” but I do find myself paying more and more attention to these things and my sense is a lot of us are doing ourselves a disservice by ignoring them – more on these topics later

Epilogue: Sorry to bore you with a lot of rambling – just had some time to burn and felt “semi-philosophical” for moment. Take it for what it’s worth.

Regards and Thank You for Listening,

KumaBear Bob

Wrong!
There is clearly visible difference between 600 and 800lm.
There is also visible difference between 700 and 800lm, if you compare two same lights at the same time(separately you do not notice it anymore)

Yes, it is wrong. Inverse square law has nothing to do with 600lm versus 800lm.

“you see/perceive a scene with a full spectrum of colors!” Even if this is true (I don’t think it is), the colors are what the person expects to see, not the real colors of the scene. The screen does not contain any information about colours.

I agree. Difference on the other hand is small to my eye.
But I must admit that in my uses, the last 50 or 100lm is not in the top5 of the things I necessarily want from a light.
Many other features are more important.

State your opinion… Wait…

There's a difference between discussion and disagreement. None of use know the difference, LOL. Here we go again!

EDIT: It's all a mute point really. The difference is in us and what we want, nothing more. Some people want a Vette, some want a Toyota. Some don't care. Some are color blind and think the red car is pink. Doesn't matter. Why do we want the brightest light, who cares, it doesn't matter, we want it, that's all... and we will argue about it till we die and it doesn't matter.

Just sayin

Yeah, the inverse square law related to luminance at different ranges, not to do with the logarithmic nature of brightness (defined as subjective perception of luminance) perception.

This is a good and relatively simple academic introduction to the area: http://www.google.com/url?sa=t&rct=j&q=jnd%20psychophysical%20threshold%20for%20luminance&source=web&cd=1&ved=0CDMQFjAA&url=http%3A%2F%2Fwww.ict.griffith.edu.au%2Fteaching%2F1008ICT%2Fdocs%2Fhumanvision.doc&ei=8-P7UJudEe2z0QXmzoHQDA&usg=AFQjCNFnXSwgQ72GBAunho2LrZERX6je3A&cad=rja

’Physical differences in stimulus can be measured directly using physical devices such as light meters, scales and other devices. It is not possible to measure subjective impressions of stimulus directly (Fechner 1861) but we can measure Just Noticeable Differences (JND) for incremental changes in stimulus level. Weber’s Law (1834) of Just Noticeable Differences provides a very rough gauge of JND to stimuli within their normal ranges of intensity.

Weber expressed his law as: C = ∆I/I, where C = 0.08 for light intensity, C = 0.05 for sound intensity and C = 0.02 for kinaesthetic intensity.’

What this means is that accoording to Psychophysical detection thresholds a Just Noticeable Difference for luminance will be at around 8% increase in intensity. This may vary according to a number of factors and to a certain extent between individuals however.

For a less academic introduction, this BBC programme ‘Do you see what I see?’ is a really good watch: http://www.dailymotion.com/video/xl7cgh_horizon-do-you-see-what-i-see-part-1-4_shortfilms#.UPvlKfIZl7Q

That is a link to Part 1, 2,3 and 4 should be accessible in the bar of options to the right.

Is 800 lumens a lot better than 600 lumens? Does it matter?

That’s very interesting stuff RedForest.

And Old-Lumens – I’d say “you’re right on the money” as regards personal taste, personal opinion and preferences. This is a hobby and everyone has a right to value a Gold flashlight more than a Black one – or vice versa.

Boaz is a seasoned, experienced and knowledgeable veteran here on the BLF and he was the one that said that there is little or no difference between 800 and 600 lumens “in the real world” (to paraphrase) and as the eye perceives the level of lighting. That’s not to say he wasn’t taking a little “poetic license” in making his comment.

There is also a very big difference between a comparison of lights when side-by-side and when used at different times for the same function. That said, I myself feel that I can probably see a little bit more with an 800 lumen light than with a 600 lumen light – but 100 lumens may be too small a difference to “remember” from one outing to another.

This is a very controversial subject, particularly since people on the BLF are so conditioned to paying attention to the numbers – and often invest their hard-earned money based on those comparisons.

Certainly it’s true that the eye is a logarithmically calibrated device. Otherwise you wouldn’t be able to get around in a dimly lit environment and still be able to see on a bright summer day at noon. As a result, anyone that claims that a relatively small percentage difference in output (as measured in lumens) is very significant for most purposes is probably trying to justify their purchase more than facing reality.

Please remember that I am among those who like to show my friends just how far a certain light is able to throw and just how bright another may make a room. I’m also influenced by the relative attractiveness of the industrial design employed in one flashlight vs. another.

Nevertheless, I started this thread with the intention of stimulating debate and possibly bringing certain truths to the surface - truths that may otherwise remain unspoken or unrealized. Please don’t take any of this too personally. We’re just having a discussion about our hobby here.

Thanks for listening.

Bob

Yes, it’s a valid point. 200 lumens added to a 600 lumen light will look no-where near as bright as 200 lumens added to a 50 lumen light.

The early Psychophysics referenced here often made use of a 50% accuracy value when determing detection thresholds (contemporary methods make use of a 75 or 82% threshold), so it would mean that an 8% increase in stimulus intensity was noticed 50% of the time by trained observers. It doesn’t mean that an 8% difference is a large one. For a perceived doubling of light intensity I believe the objective luminance must be increased by a factor of between 3 and 4, with significant individual variation.

One other thing to bear in mind is that CCT does also have an effect on perceived brightness. -So while a 600 lumen light may be indistinguishable from an 645 lumen light of the same spectral characteristics if you introduce the other factor of varying CCT (so one is warm while the other is cool) then you must take that into account.

Higher CCT light sources generally appear brighter than lower CCT ones, so that means that even with a cool and warm white option both at 600 lumens the cool white will probably appear brighter anyway. This effect is then slightly increased by the small increase in efficiency of a cool white LED over an equivalent neutral or cool white.-

Actually, this effect of CCT on perceived brightness is already taken into account in lumen ratings, which are already weighted according to the peak sensitivity wavelengths of the human eye. So please disregard the bit above this (I’ve tried to cross it out but it never seems to work on my account).

With regards to Edwin Land, a good demonstration of this is shown here: http://www.youtube.com/watch?v=3DiCvHiDWiY

It’s all to do with ‘colour constancy’, which is maintained by higher cortical processing rather than anything in the retina.

Just to throw a spanner in the works :smiley: what you see reflected back when you shine a light is not even the same light that left your flashlight :stuck_out_tongue:

Kuma Bear, thanks for an interesting topic of discussion.

What seems to get lost in all these tint and CRI threads is that we are all different. A big thanks to Old Lumens who always seems to point this out in these types of discussions. Do people with brown eyes see color differently than those with blue eyes? Do women see color differently than men? (I already know the answer to that from shopping trips for paint with the wife). Does everyone have the same depth perception, night vision, acuity? Are we all using these lights under the same ambient conditions or for the same intended uses? Taking away all the lux meters, integrating spheres and other tools we use to measure light, what happens when said light hits the rods and cones in our eyes and how does our brain process that information?

For anyone to say that one tint is better than another is just expressing an opinion of what they perceive to be better. Any discussion on tint should start with "what's your favorite color?"

Boaz, I love you man, but as an ex-landscape designer of perennial gardens, I would suggest that if you really want to get slawjacked about color rendition in a garden, don't do it at 2AM with any flashlight, instead start by studying Claude Monet.

Sorry, couldn't resist. I am not an expert on color. But I remember many discussions on the subject in drawing & painting classes in college. All of the above are paintings done at different times of the day/different times of the year. All can be called "Natural Light". Don't you all think it's a bit silly to expect that one emitter does a better job of capturing "natural" when natural varies so much? Food for thought in an interesting discussion.

For the Curious: More on “The Land Effect” or “Retinex Theory”

Follow-Up to this thread’s OP

The links below bring the reader to more fascinating information

related to the discussion about

Edwin Land’s experiments with, and theories about,

color perception.

RedForest was kind enough to find this YouTube Video of Land demonstrating part of his theory on the BBC in 1985. His demonstration only scratches the surface of the subject but I find it totally mesmerizing. I list it first because it is the easiest way to understand and visualize what I was trying to describe earlier in the OP. I hope you enjoy it:

http://www.youtube.com/watch?v=3DiCvHiDWiY

The following link takes you to a discussion of Land’s discovery, the serendipity involved and the utter fascination and incredulity that led me to try to duplicate it while in Junior High School (with partial success):

http://www.wendycarlos.com/colorvis/color.html#discovery

The next link takes you to Wikipedia’s listing of related articles and information:

http://en.wikipedia.org/w/index.php?title=Special%3ASearch&profile=default&search=Retinex+theory&fulltext=Search

More specific information from Wikipedia on the subject:

http://en.wikipedia.org/wiki/Retinex_theory#Retinex_theory

An illusion that takes a “babystep” toward demonstrating the phenomenon:

http://en.wikipedia.org/wiki/Same_color_illusion

A discussion explaining a Theory of Color developed by a German many decades before Land’s experiments but having a surprising amount in common with Land’s theories:

http://en.wikipedia.org/wiki/Theory_of_Colours

I hope that some of you enjoyed reading through this material. I’d be happy to explain how my son and I took the theory in a somewhat different direction to produce a working example of part of the Land Effect in 1993. The results were quite startling.

Regards,

Bob

Thanks for that link, RF UK. That was a fascinating program, enjoyed it a lot!

That is, if one wanted NW even with a CW led would using a colorized filter replacing a clear lens give basically the same effect as if the led to begin with was NW? It seems to me that if camera type filters were applied on flashlights that one could have the best of all flexible worlds regarding tint.

But you lose a lot of lm

I’ve come to realise, as I hand out a few lights, I grossly underestimate things. Key point:-

I built an m10 with the now famous intloutdoor 4a driver and a u3 1c emitter. I liked it, but mainly for the host and the UI, I saw the output as “acceptable”. I then wanted to do a give away, I like to give away things I like, so, I ordered another m10, already had another driver, the m10 came with a u2 marked star, so I passed on an emitter swap. Dale won the give away, recieved the light and is I feel its fair to say blown away by it.

In comparison, I built a trustfire t2 with an xp-g 4c driven at 1.7a….

Now, we all know that a u3 at near four amps is considerably “brighter” than an xp-g at less than 2 amp……

Two things blow me away about that t2, first and foremost how “nice” both the colour of the light and its colour rendition is, second, its throw, (slightly off topic here I guess) the dimmer on two counts, light, impresses me far more than what was a “hotrod” of a light. Give that same hotrod to someone else, someone with, it has to be said an impressive collection of lights, their blown away with it!

I give up on lumen figures, I’ll build what I find I like (after much experimentation in both tint and output) for me. If others get to experience this and are more amazed, well that’s just ace, but I won’t worry if their less happy, unless they paid me to build it….

I think it is pretty save to say that it is impossible to see the the difference in 2 neighboring bins R4/R5 or T4/T5 with the naked eye and without special equipment and if you have a "high" R4 and a "low" R5 the difference can be only 1 Lumen in theory. I said it many times before that I have the feeling that people tend to concentrate the discussions on the wrong topics, eg. one should worry about driver efficiency multiple times more than about binning levels. (goes especially for battery/driving voltages far away from the forward voltage of the emitter)

I still remember the discussion where people had the impression that their XM-L T6 (318-341 L @ 700 mA) seemed "darker" than the XP-G R5 (260-277 L @ 700 mA) in their Xeno E03 lights... so difference in beam profile (throwy/floody) tricked peoples eyes to a considerable degree... way more than one bin step ever could.

I think we may be double dipping on the effect of eyes perceiving different wavelengths differently. The definition of lumen already includes the effect. A 600 lumen 2700 CCT light is putting out more energy as light than a 600 lumen 6500 CCT light.

True, that’s a good point. Not sure how I missed that the first time :wink: