Cheap flashlights to compare tints

jon_slider · 2019-01-21T03:29:45.000Z

FBsLights:

I’m searching for my favorite high CRI LED.

mine is the 219b 4500k 9080, at the moment

I like warmer when its full darkness, but cooler during the day, so its a good compromise for a multi ambient light environment, with a bias towards daylight.

High CRI is a subtle variable, and some people dont prioritize it, because of an interest in efficiency. But, a more efficient light, is not a High CRI light… Efficiency favors quantity over quality of light.

Let me try to drive the point home… for the OP
I like a flashlight that can show red colored things, I find red food appetizing…

I can imagine that when a mechanic is looking at different colored wires, it helps to see the Red Pop!
Or, imagine Im grilling steaks after dark, would you like yours yellow or pink?

Tjohn · 2019-01-21T13:18:04.000Z

An abundance of useful information is cropping up here from the ‘tint conscious’, some looking to compensate for vision issues.
As stated, tint and HiCri are severely influenced by the ‘driven’ output, with reflectors and TIR optics playing their roles.
Beam profile, lottery within bins, the reflectance and color of subjects, along with contrast between colors are all important.
A nice tint may be seen from an emitter on a white wall, while it is monochromatic, or unable to separate colors from one another.
A variety of lights and tints for different tasks makes sense.

A ‘Bling and Blind them’ Scorcher torch might have more lumens than finesse.
An indoor use EDC, technician’s, or manicurist’s light has stricter color parameters.
In true flashaholic spirit, all justification to buy more lights.

ZoomieFan · 2019-01-21T13:39:54.000Z

In true flashaholic spirit, all justification to buy more lights.

Yeah!!!
While I agree I’m usually not carrying several lights. So I want all my lights be as all-round as possible.
But with my newly bought Lumintop AA and especially AAA I learned that an carrying an extra light just for reading shiny metal tags/lables is a good possibility.

I still don’t know what’s best but the picture surely show what I do not want. Very informative!

I’ve ordered a bunch of lights to compare. But not as advised a dozen S2+’es.
I like to test other format lights and never owned a AA or AAA so I bought several in various tints and LED type. (see signature)
Not all have arrived yet but I hope that, while the lights are not from the same manufacturer they at least give me some inside.

From this thread I learned several things, one of those things is that I need more flaslights for testing :innocent:
What’s a asperic lens for example.

I do see another problem rising… I really would like a 21700 flashlight. Because I like compact for a certain cell size I really like DQG Tiny (got a 26650). But, yeah you guessed it, no high CRI versions…

Zulumoose · 2019-01-21T14:27:23.000Z

Because I like compact for a certain cell size I really like DQG Tiny (got a 26650). But, yeah you guessed it, no high CRI versions

The answer to that is an Emisar D4 (or D4S) Nichia 5000K version.
Compact, neutral, high CRI, great UI, powerful, ramping, quality in every way.
The only down-side is the heat on full output, turbo is like a naval broadside, use sparingly and make sure you aim for the enemy.

jon_slider · 2019-01-21T18:07:11.000Z

ZoomieFan:

What’s a asperic lens for example.

It looks like a magnifying glass and produces no hotspot,
left to right, aspheric, TIR, conventional reflector with hotspot

ZoomieFan · 2019-01-21T18:11:15.000Z

Yeah, me likes aspheric!

Would be great for label reading.

Here comes the big question. I now know Yuji is great and aspheric gets thumbs up for me. I think I like NW most (5500K?). I keep my lights in my pocket so it has to be small. 1000 lumen is super bright in my book. 800 pleases me too.

Does a light that combines all that goodness even exist?

Tjohn · 2019-01-21T19:47:12.000Z

The first thing that comes to mind is an Emisar D4 Mule with Nichia 5000k 9050 emitters and an 18350 tube.
The board is screwed down so the TIR could be removed.

ZoomieFan · 2019-01-21T19:59:00.000Z

From jon_slider’s review

Hand and yellow rock look most like daylight on the NW
The greenish rock on the 219b
The jewelry is a draw.
Is the the 219b performing so badly because it has a very low K?

jon_slider · 2019-01-24T00:27:17.000Z

ZoomieFan:

Is the the 219b performing so badly because it has a very low K?

Thanks for asking interesting, and detailed questions

some variables in play…
Low CRI LEDs tend to produce less red than sunlight, and more green and yellow than sunlight
High CRI LEDs tend to produce as much or more red than sunlight, and not too much green or yellow

Since LEDs dont have the same spectrum as sunlight, there are trade offs.
High CRI will show reds, that Low CRI cannot. Low CRI will overemphasize green, and yellow.

anyway, given the limitations, lets walk through the separate treasures

the Coral is redder in high CRI, than sunlight, the Low CRI light makes the coral less red than sunlight…

The turquoise under High CRI looks almost identical to sunlight, and the Low CRI looks greener than sunlight

The Amber looks redder than sunlight under High CRI, and yellower than sunlight under Low CRI

.

given the limitations, I prefer too much red, rather than too little; not too much green, rather than too much, and not too much yellow

I avoid Low CRI LEDs, due to their poor Red output, and excessive green and yellow output.

The High CRI images are from the light on the left, the Low CRI images from the light in the middle, the light on the right is there to help the iPhone set a relatively useful white balance (5500k, although too green) for sake of comparing the first two LED Tints

Given the choice between too pink or too yellow, I choose too pink

bear in mind that our brain will adapt to any single color beam, and tend to interpret it as a white reference, when it is alone.
because of that, the pink light, does not look as pink as the photo, unless it is viewed during times when the brain has set its white balance to daylight. When my brain has set its white balance to the incandescent light in my home, it looks white, not pink.

The yellow Low CRI beam looks disgusting during the day, its not much better in the evening, but suprisingly, if I wake up in total darkness, I dont really notice how excessively yellow it is… but the yellow beam is low CRI (and more specifically low Red output), so it will never be able to show me how pink a steak is in the middle.

.

ZoomieFan:

Does a light that combines all that goodness even exist?

500+ lumen levels are marginally supported by 16340, but are really 18650 territory. I dont know which 18650 lights offer Aspheric lenses, Im focused on 16340 atm.

The 5600k Yuji only outputs 10 lumens, it uses AAA
the aspheric beam in the yellow light in the middle, maxes out at about 600 lumens, but it is not realistic to run that for long, given the battery is 16340 size, and the lumens drop by about 30% when going high CRI

the lights that come to mind with aspheric lenses are, the Utorch S1 Mini shown with the yellow beam, the Klarus Mi1c aluminum model (not the copper, titanium, or High CRI models), and there is an Aspheric Sunwayman:
https://www.batteryjunction.com/sunwayman-t16r-flashlight.html

for my use, all those lights would require an LED swap, and since Im in love with Nichia LED swaps, I would recommend the Klarus over the Sunwayman. I dont recommend the Utorch (even though I own 3 of them), it has design issues with a loose ground wire. There are other factors, such as the LED footprint, and whether the mod is extensive enough to replace the LED board, or just the LED… The Klarus is the least complex mod.

There is a High CRI Klarus Mi1c, but it has a hotspot, it is not aspheric… but thats the closest to off the shelf High CRI that comes to mind atm… all the lights I use are modified, the LED has been changed to suit my pleasure, and practically no single light combines all my favorite features. I primarily use one aspheric ligh modded to 4500k High CRI, and one reflector light with hotspot, also modded to 4500k High CRI

since you seem interested in 5000k+, you might be served by N219c. My personal tint preferences avoid those, due to what I consider excessive yellow and green, and insufficient Red output. Did I mention Im a Total Tint Snob?

good luck with your search(es)
I hope you share your impressions of lights you try

ZoomieFan · 2019-01-24T06:23:49.000Z

Trade offs…
I like to look at money, so maybe I need low CRI

Do you have a comparison between Nichia and Yuji?

jon_slider · 2019-01-24T07:22:26.000Z

ZoomieFan:

Do you have a comparison between Nichia and Yuji?

yes, I like when you ask questions I know the answer to, LOL

bear in mind that the Yujis are limited to 10 lumens (sofirn C01 AAA) or less, Nichias can do up to 500 lumens (Bob_McBob Zebra 18650)

I recommend you try one Black Sofirn C01 in 3200K Yuji, and one Black Sofirn C01 in 5600k, here is the link

I expect you will prefer the 5600k, and Im suggesting black (instead of the red or blue bodies) hoping there will be less of a hotspot, pics here

ZoomieFan · 2019-01-24T08:10:33.000Z

yes, I like when you ask questions I know the answer to, LOL Smile

I’m gonna to up the difficulty level! :smiling_imp:

Do you have pictures of those stones with Yujis and Nichia of the same K.

3000K isn’t for me, that’s sure.
5000-6000 Nichia could be good.

So Yujis are only suitable for AA or AAA lights. Pity, but at least I know now I don’t have to search for a 18650+ with a Yuji.

I hope you share your impressions of lights you try

I will. All the AA and AAA in my signature are ordered recently. Sofirn Nichia’s are still in the mail.
The Lumintops are nice lights. Nothing special when used as an ordinary flashlight. I can’t describe it but, with the diffuser caps on the light feels pleasant.
The Nightcore’s diffusor gives a less pleasant like. Nightcore uses hard plastic, Lumintop soft silicon rubber or whatever it is. Maybe because lumintop filters out certain wavelengths is feels more relaxed…?

ZoomieFan · 2019-01-27T09:40:58.000Z

I’m not a reviewer, but I want to share a few findings. Don’t expect anything remotely useful…
I bought several lights to compare but not exactly the same as suggested.
I bought:

- NiteCore LA10 - not High CRI

- Lumintop Tool AA and AA - Not High CRI 3200 and 5600K

Sofin C01 (AAA) Hi CRI

Lumintop:

Small
Good output
The difuser gives very pleasant white. Better than Nitecore.
The tailstand of the AA is very unstable. The AAA has no tailstand at all; but the magnet tail stand is good. Also on non magnetic surfaces. For me this is a huge unforgivable design flaw for a light sold with a diffuser.
The beam could be more even. I don’t like hot spots.

Nitecore.
I bought it because I want to test the built in diffusor.
A good light that gets beaten by the Lumintops using a diffuser.
The lumintops light just feels more relaxed.

Sofirn C01

No settings
Low output (which hindered testing)
Small
Less hotspot than the Lumitop, which is a bonus for me.
Hi CRI

Hi CRI brings me to the whole reason I started this thread.
Very nice pictures have been posted (thanks!). No need I repeat that. Even if I could.
The thing is that the Sofirns have a very low output.
The results of my tests greatly confuse me.

Because of the low outputs of the Sofirns I only tested by reading black, red and blue text on white paper.
As expected 3200k shows the white paper as creamy.
What I noticed when comparing the 3200k and 5600K makes me wonder one of my lights is defect.
The red and blue looked equally good under both tints.
3200K feels a bit easier on the eyes.
3200K somehow feels a lot brighter. How is that possible?

I tried to compare with the AAA Lumintop on low.
It looks brighter than the 5600K but not so sure about the 3200K.
I’m wondering why that is. Could it be one of the Sofirns isn’t calibared correctly and its only setting is fixed at brightness not according to specs?
Could it be because how eyes work?

That last question likely need some explanation.
On this forum great tests are posted. Some flat out lab grade scientific tests with fancy graphs. Technically irrefutable and very helpful.
But our eyes combined with our brains are no computers.
Could it be that our eyes apply all sort of filters, adapt, introduce sensitivities for a certain color spectrum etc.
(I know from serious sources the eye is most sensitive to green light for example)
The above is a long winded way of asking it could be the 3200K and 5600K are equally bright but due how eyes work, don’t appear like equally bright.

That’s an important question for me. I’m looking for the best trade-off. Not the most bright. No the best color rendering. But if the hi CRI 3200K really looks brighter at the same output that would certainly get my attention.

The Lumintop made the text equally readable. Also good colors.
If someone showed me all those flashlight without knowing specs of any sort and asked me to comment on it; I would talk about all sort of differences, but honestly “the red/blue don’t looks real in this light” would not be on my comment list.
And even when I see a tiny difference, I wouldn’t know what’s best.

So what’s going on here? The photo’s posted in this thread, especial the yard with the orange leaf, are extremely simple. There is a huge difference between hi and low cri color rendering.
Once again I totally admit that fact. But why don’ I see it in my own tests???
Could it be AliExpress messed up, and instead of sending a CW Lumintop they sent a Nicha version?

Tjohn · 2019-01-27T10:28:43.000Z

ZoomieFan:

……… what’s going on here?..…. There is a huge difference between hi and low cri color rendering.
Once again I totally admit that fact. But why don’ I see it in my own tests???

One of many possible considerations from Wiki. Sometimes we see what we don’t see, and vice versa.
——-
“The color opponent process is a color theory that states that the human visual system interprets information about color by processing signals from cone and rod cells in an antagonistic manner. The three types of cone cells have some overlap in the wavelengths of light to which they respond, so it is more efficient for the visual system to record differences between the responses of cones, rather than each type of cone’s individual response. The opponent color theory suggests that there are three opponent channels:
Red versus green.
Blue versus yellow
Black versus white (this is achromatic and detects light-dark variation, or luminance).

Responses to one color of an opponent channel are antagonistic to those to the other color, and signals output from a place on the retina can contain one or the other but not both, for each opponent pair.”

Geuzzz · 2019-01-27T19:11:04.000Z

I have a utorch s1 mini here with a 95+cri 3000k sst-20 that I don’t use. Could be just what you are looking for No hotspot at all due to the aspheric lens. Way more output than the sofirn.

ZoomieFan · 2019-01-27T19:40:38.000Z

I definitely want to try a aspheric lens.

I won a Firefly ROT66. It’s shipped to a friend who I won’t meet until half april. They also come with the SST-20, but I don’t even know which version I won.

3000K

I’ve been testing Sofirn 3200K. I have no clue what’s going on. I always assumed high contrast is good for reading.
I have a sheet of regular copier paper with black text on it.
The 3200K gives the white a quite visible tan. I would say that lowers contrast.
But it is as least as readable as the 5000K version.

Been looking at Makuta’s (sp?) tests of the SST-20. The CRI 70 has a 40% better lm/W as the CRI 95.
lm/W is a big selling point for me. But now I’m starting to wonder if higher CRI gives higher visibility at lower lm/W
And obviously, visibility is all what it’s about.
So perhaps visibility wise higher CRI isn’t 40% less….

Boaz · 2019-01-28T04:25:05.000Z

CRI is color rendering ...It doesn't have a thing to do with contrast.

Color rendering is the ability to more accurately reproduce colors..95% cri means it's going to be very very close to real/true color,.If you have a jar of jelly beans it will show each color the way they are supposed to be .

reading on a black and white page and cri have nothing in common ...cri is color... multi color

ZoomieFan · 2019-01-28T07:19:57.000Z

Black and white are the opposites so very high contrast.
When I use 3200K the white gets a tan. That’s closer to black. For me that means less contrast.

The low CRI looks bleached. Colors closer together. I would call that lower contrast.
Ok, I may not be using the correct terminology but I hope you understand what I mean.

Tjohn · 2019-01-29T10:14:33.000Z

The wall in my backyard measures 100 feet away.

On the left:
D4 (version 1) maxed out with 4 - XPL Hi 3A in OEM Carclo TIR.
With the original IOS ramping UI from Tom E., it maintains highest output (~4000 lumens) longer before stepping down precipitously, making comparisons easier.

On the right:
Eagtac MX25L3-C (6 x sw45 92+) reflector thrower, with max ANSI lumens of 1600-1700. For lumen output, Eagtac normally seemed to have selected sw45 Nichias from above the BBL when they were offered in their flashlights.

ZoomieFan · 2019-01-29T10:24:46.000Z

Assuming the flooding is equal, I would say the light on the right is far brighter.

If that’s true it gives a whole new twist to runtimes.
Nicha’s (and other high CRI LEDs) give ~40% less output with the same power as low CRI.
So a 40% worse lm/W.

But you picture shows for the eyes the high CRI looks to have more output.
It’s hard to estimate, but the light on the right should run half brightness to match the one on left.
If so that would mean visibility/W is much better for high CRI

Is that even remotely true.
(I wish it is because that would make things so much easier)