Lumens vs perceived lumens, is there a formula?

27 posts / 0 new
Last post
negev
Offline
Last seen: 14 hours 3 min ago
Joined: 01/05/2021 - 09:17
Posts: 347
Location: London
Lumens vs perceived lumens, is there a formula?

Is there a formula for converting lumens to perceived lumens, or does it vary too much between people to ever be accurate?

On my S2+ 50% mode is barely distinguishable from 100%. I’d like to translate the 12 modes percentages into perceived lumens so I can get a better idea of what they look like rather than just the power consumption.

zoulas
zoulas's picture
Offline
Last seen: 9 hours 47 min ago
Joined: 06/01/2020 - 08:35
Posts: 1834

I can’t agree more. The SP35 is 1K lumens on high and 2K lumens on turbo. Barely noticeable.

negev
Offline
Last seen: 14 hours 3 min ago
Joined: 01/05/2021 - 09:17
Posts: 347
Location: London

Actually I don’t care anymore. 50% mode looks the same as 100% to me so I’m just going to stick with mode 8 where 50% is the highest. Now I’ll get longer runtimes and not burn my hand, win win

TMaxxJJ
Offline
Last seen: 15 min 29 sec ago
Joined: 02/01/2018 - 03:18
Posts: 385
Location: England

Our perception of lumens isn’t linear. I find that to perceive a light that is twice as bright, you need a luminous increase to a factor of four. ie, for 250 lumens to seem twice as bright you need 1000 lumens. To seem twice as bright as a D4V2 ~4000 lumens, you need 16000 lumens etc

negev
Offline
Last seen: 14 hours 3 min ago
Joined: 01/05/2021 - 09:17
Posts: 347
Location: London
TMaxxJJ wrote:
Our perception of lumens isn’t linear. I find that to perceive a light that is twice as bright, you need a luminous increase to a factor of four. ie, for 250 lumens to seem twice as bright you need 1000 lumens. To seem twice as bright as a D4V2 ~4000 lumens, you need 16000 lumens etc

So using that formula, if a light is 1600 lumens at 100% and 800 lumens at 50% then it should seem 25% brighter to the eye when going from 50% to 100% right? It doesn’t feel like that’s the case with mine.

TMaxxJJ
Offline
Last seen: 15 min 29 sec ago
Joined: 02/01/2018 - 03:18
Posts: 385
Location: England

50% and 100% of what though? Amps? If so then it’s not quite as simple as that. An LED gets less efficient the harder it’s driven. For instance if an LED produces 1000 lumens at 5 amps, it won’t produce 500 lumens at 2.5 amps, more likely to be ~700 lumens.

negev
Offline
Last seen: 14 hours 3 min ago
Joined: 01/05/2021 - 09:17
Posts: 347
Location: London

That makes sense, I wish so had one of those things that can test output

TMaxxJJ
Offline
Last seen: 15 min 29 sec ago
Joined: 02/01/2018 - 03:18
Posts: 385
Location: England

I have one of these that seems to be working well – https://www.amazon.co.uk/dp/B006QHOX0Y/ref=cm_sw_r_cp_api_glt_fabc_WWVHS...

Oli
Offline
Last seen: 15 hours 34 min ago
Joined: 08/03/2018 - 21:03
Posts: 490
Location: United States

If you step outside and put the light on something 100 meters away you’ll see the difference very easily between 50 and 100%. Inside the house, it may be harder to tell.

Juslearnin
Offline
Last seen: 2 hours 57 min ago
Joined: 12/11/2019 - 03:15
Posts: 55

Calculating throw from candela = 2 * square root of candela. Therefore the quadrupling of candela to double throw would make sense.

Does this carry over into overall brightness / Lumens as well? I suspect yes, because the surface area of the inside of a sphere is a square function as well.

I am sure someone smarter than I will be along shortly to answer.

hIKARInoob
hIKARInoob's picture
Offline
Last seen: 8 hours 47 min ago
Joined: 08/28/2016 - 08:15
Posts: 4190

You don’t see lumens, you perceive lux.

negev
Offline
Last seen: 14 hours 3 min ago
Joined: 01/05/2021 - 09:17
Posts: 347
Location: London
TMaxxJJ wrote:
I have one of these that seems to be working well – https://www.amazon.co.uk/dp/B006QHOX0Y/ref=cm_sw_r_cp_api_glt_fabc_WWVHS...

Presumably that only measures lux not throw as well?

TMaxxJJ
Offline
Last seen: 15 min 29 sec ago
Joined: 02/01/2018 - 03:18
Posts: 385
Location: England

You can only ever measure lux. You then use a formula, or enter the numbers (Lux and distance from the meter) into a calculator to work out throw. I use this one – https://www.rapidtables.com/calc/light/lux-to-candela-calculator.html

CNCman
CNCman's picture
Offline
Last seen: 4 hours 30 sec ago
Joined: 04/07/2018 - 10:10
Posts: 1175
Location: Mississippi Gulf Coast

I just hold the light to my ear on turbo and check my opposite ear with a light meter. Wink
.
Just joking, but when I check a light for throw, the perceived brightness on the wall and using it a night are the most fun for me. I can see the difference between 4000 and 6000 lumens. It appears the more you use different lights, the easier it is to differentiate lumens.

negev
Offline
Last seen: 14 hours 3 min ago
Joined: 01/05/2021 - 09:17
Posts: 347
Location: London

Well I’ve ordered one of those light meters, what’s the recommended measuring distance to accurately estimate throw?

SammysHP
SammysHP's picture
Offline
Last seen: 27 min 20 sec ago
Joined: 06/25/2019 - 14:35
Posts: 956
Location: Germany

Depends on the light intensity and the range of your meter, usually 1-10 m. In extreme cases you can add a ND filter and correct the measurements afterwards.

negev
Offline
Last seen: 14 hours 3 min ago
Joined: 01/05/2021 - 09:17
Posts: 347
Location: London

Well I still have a Fenix light with known throw, presumably if I measure that and compare it Fenix’s own measurements then I should be able to figure out the margin of error

TMaxxJJ
Offline
Last seen: 15 min 29 sec ago
Joined: 02/01/2018 - 03:18
Posts: 385
Location: England

negev wrote:
Well I still have a Fenix light with known throw, presumably if I measure that and compare it Fenix’s own measurements then I should be able to figure out the margin of error

Not quite. You need a calibration light (the Fenix in your example) when you come to measure lumens, either through ceiling bounce, sphere or tube. But when it come to candela, it’s purely just numbers. I predominantly use around 6m to measure. So I set my meter up, choose the range (x10, x100 etc), measure the distance with a laser measure, then take a reading. Enter these numbers into the calculator I linked to above and it will give you your candela. Then I square root that number and x2 to get meters of throw.
wle
wle's picture
Offline
Last seen: 15 hours 47 min ago
Joined: 01/07/2015 - 13:49
Posts: 2450
Location: atlanta ga

you may just have a weakish battery
sometimes the bat cannot do what the light could do

and 50% more is not that apparent, either ,as others said

"You never have the wind with you - it's either against you, or you're having a good day."
    Daniel Behrman, "The Man Who Loved Bicycles".
It never gets easy, you just go faster.   
-Greg Lemond.
       ,ø¤º°`°º¤ø¸,ø¤º°`°º¤ø¸,ø¤º°`°º¤ø¸

negev
Offline
Last seen: 14 hours 3 min ago
Joined: 01/05/2021 - 09:17
Posts: 347
Location: London

Fair pointed but I was testing with a fully charged battery. I’ll try the same test again outside when it’s dark when I have a chance, see if the drop in throw is noticeable.

CNCman
CNCman's picture
Offline
Last seen: 4 hours 30 sec ago
Joined: 04/07/2018 - 10:10
Posts: 1175
Location: Mississippi Gulf Coast

The longest distance in my house is 6.46 meters so I use this for throwers.
.
Reading X 6.46 × 6.46 = Candela/CD.
Throw = Square Root of 4 X CD, all in meters.
.
Same answers as above.
.

negev
Offline
Last seen: 14 hours 3 min ago
Joined: 01/05/2021 - 09:17
Posts: 347
Location: London

Presumably you have to do this test in complete darkness?

negev
Offline
Last seen: 14 hours 3 min ago
Joined: 01/05/2021 - 09:17
Posts: 347
Location: London

It’s annoying that there’s no mode in biscotti that contains both 50% and 100%, then it would be easier to tell if there’s a perceivable difference.

Lightbringer
Lightbringer's picture
Offline
Last seen: 2 hours 50 min ago
Joined: 08/30/2016 - 14:12
Posts: 15486
Location: nyc

Some lights might have regulated high and FET-driven “turbo” where the diff is about 2:1.

Switching back’n‘forth gives a little hint of extra brightness, but nowhere near “omgwtf, that’s twice as bright!!!”.

Even with my Q8, I was expecting 5000lm (vs about 1000lm of my other lights) to be like daylight in the backyard, but it wasn’t. Bit of a letdown.

Brighter? Sure, but not “night’n‘day bright” as you might expect.

09 F9 11 02 9D 74 E3 5B D8 41 56 C5 63 56 88 C0

negev
Offline
Last seen: 14 hours 3 min ago
Joined: 01/05/2021 - 09:17
Posts: 347
Location: London

Ok after testing in the woods in the dark, there’s definitely a noticeable difference between 50% and 100% but obviously it’s nowhere near twice as bright.

richbuff
Offline
Last seen: 1 hour 46 min ago
Joined: 09/22/2015 - 20:38
Posts: 951
Location: Prescott Az

negev wrote:
Ok after testing in the woods in the dark, there's definitely a noticeable difference between 50% and 100% but obviously it's nowhere near twice as bright.
That's very good news, but nowhere near being great news.

 

Over the few or so years, I spoiled myself as to what I subjectively considered to be very bright. As long as I was willing to migrate to larger lights. And that's exactly what I did. The first thing I considered to be very bright was me. Then I was dethroned by the Tru Value Rayz 1000 Lumen 9 x AA. Then that was dethroned by the Four-Sevens MMU-X3, 1,600 lumens. That was no longer very bright when the Niwalker MM15 5,233 lumens lit up my life.

Next, Noctigon Meteor M43vn 12 x XP-L 8,400 lumens.

Next, TN36UTvn 13,400 lumens.

Next, a lot more lux with X65vn with 11,500 well-tempered lumens.

Next, Acebeam X45vn muchly boosted to 25,000 lumens.

Next, Imalent R90TS 36,000 lumens. Next, Imalent MS18 100,000 L-words.

10,000 small white room ceiling bounce lumens is a round number for the cutoff for what looks very bright to my 61 year old, self-spoiled rotten eyes.

Outdoors, more throw with less flood looks brighter per given lumens; to me, anyway.

D4V2 W2, EC03 50.2, X50 and BLF GT94 are on my front four bright burners today. Tomorrow will be FW4a W2, FT02S 50.2, MS06 and HK90 to brighten up my life.

D18, KR1, MT07, MT05, NSX53, FW21pro90, E4K, MS03, TM9K and many others: My life is noticeably brighter with flashlights.

Rev 22:15

xevious
xevious's picture
Offline
Last seen: 2 hours 6 min ago
Joined: 02/27/2013 - 21:55
Posts: 2403
Location: Hoboken, NJ USA

TMaxxJJ wrote:
negev wrote:
Well I still have a Fenix light with known throw, presumably if I measure that and compare it Fenix’s own measurements then I should be able to figure out the margin of error

Not quite. You need a calibration light (the Fenix in your example) when you come to measure lumens, either through ceiling bounce, sphere or tube. But when it come to candela, it’s purely just numbers. I predominantly use around 6m to measure. So I set my meter up, choose the range (x10, x100 etc), measure the distance with a laser measure, then take a reading. Enter these numbers into the calculator I linked to above and it will give you your candela. Then I square root that number and x2 to get meters of throw.
My eyes glaze over at the mix of many factors in determining lumens output (or perceived lux). Not only does the nature of how emitter is hosted, but also selected emitter tint, the battery performance+charge level, and then the selected target range for illumination. Then it all gets so subjectively considered with the human eye, where the brain “adjusts” what the eye is capturing.

As Matt Smith loves to say… Good lux! Wink