How much difference in lumens can our eyes really detect between different lights?

HighlanderNorth

Flashlight Enthusiast
Joined
Sep 15, 2011
Messages
1,593
Location
Mid Atlantic USA
First, I was confused about one thing pertaining to the new model Eagletac G25 C2 MkII. Apparently the box label claims it outputs 1,030 LED lumens and 856 OTF lumens. However, in their online literature they claim 1,180 LED and 980 OTF lumens. Hmmmm... Anyway, I own the previous model G25C2 from Eagletac, and its advertised brightness was approx 770 LED lumens and 670 OTF lumens. These models are only about 2 years apart. Boufght mine 1.5 years ago. Obviously when I read the difference in brightness on turbo I was sort of bummed, naturally, that my not-so-old G25 C2 MKI had been surpassed by such a large margin using the same battery.

But then I realized two things. 1. I dont hardly ever use turbo in any lights over 120 lumens anyway, and 2. I remembered reading that the human eye cant really detect brightness differences until the difference is at least 100% more or less. In other words, I read that if you have a 100L light, and you're buddy has a 130L light, both of the same LED, size, head diameter, etc., and you shine both of them in the dark, you wont be able to detect the difference between the two in brightness.

But then I realized that it makes sense that you should be able to detect brightness differences of less than a 100% increase once you start to compare lights over a certain brightness. Maybe the 'cant-detect -any-difference-under-100%" theory starts to become less viable once you get to lights with much higher brightnesses, such as, maybe, over 500-700L. You may not be able to detect a 45% increase involving 2 lights at 50L vs. 72L, but once you get to 680L, shouldnt we be able to detect the brightness difference between a 680L and a 980L light? If so, how much difference would it seem to be?

I just dont have 2 lights that are so alike, use the same LED and have the same head diameter and tint, etc. yet have brightnesses that are reliably close enough that I can experiment with this. What do you think? Is there much of a noticeable difference between 680L vs. 980L for instance, or are we unable to detect much of a difference there?
 
Conventional wisdom says the average person can reliably detect a difference in OTF lumen brightness at or beyond 8% in a side by side comparison. That difference must be quite a bit larger for a sequential comparison. My personal threshold is about 20% before I can reliably identify the difference in a one after the other comparison.

But being able to detect a difference may not be sufficient reason for many to upgrade. It's a personal choice of cost and benefit. It depends on how much light you need and how much a little more is worth to you. And the brighter the starting point the more diminishing returns you have. If you start with a 100lm light 400 looks twice as bright. If you start with 1000lm you need 4000lm to seem twice as bright.

Low power EDC lights I might update for a 20% increase. A high powered sea h and rescue light would need a much larger increase for me to consider an upgrade. But it's a personal choice and other may choose differently.
 
How much difference in lumens can our eyes really detect between different li...

I started using camera light meters to measure my lights a year or so ago... I believe I can reliably see a 40% difference (doesn't matter 5/7, 50/70, or 500/700 lms) if it's the same light (beam profile, tint) and I'm controlling the switch. If someone else where operating the switch (say along with a momentary button for a few same-mode decoy flashes) then I think I'd miss quite a few. If they're different lights (beam profile, tint) then I don't believe I can reliably tell on a 40% lumen difference. For different lights, I might need 60-100% difference to be sure.

Use a light meter to test an ambient bounce between two close walls in a narrow hallway - if you have a smartphone, you can download a free light meter app that will be better than what you can discern with your own eye. There are a few ways to accurately convert the meter readings to lumens but you need a known starting point (calibration light). IMHO, the ANSI-FL1 standard is a heaping pile of BS, each manufacturer still employs its own levels of conservatism or exaggeration, and sometimes even within the manufacturer's own products.
 
Last edited:
Re: How much difference in lumens can our eyes really detect between different li...

I own 3 deft-x's and they all look the same even though there is a 100,000cd difference between my lowest and highest lights.
 
Re: How much difference in lumens can our eyes really detect between different li...

I own 3 deft-x's and they all look the same even though there is a 100,000cd difference between my lowest and highest lights.

Those are aspheric lights right? I am just guesing maybe that has something to do with it.

My TN32[Stock] is 240Kcd and my K40vn is around 370/400Kcd and I can definitely tell the difference. I realize that may be another 30 to 60KCD than your example.
 
Re: How much difference in lumens can our eyes really detect between different li...

Those are aspheric lights right? I am just guesing maybe that has something to do with it.

My TN32[Stock] is 240Kcd and my K40vn is around 370/400Kcd and I can definitely tell the difference. I realize that may be another 30 to 60KCD than your example.



I would assume that extremely narrow beam aspheric lights would be difficult to compare as far as brightness is concerned.
 
Re: How much difference in lumens can our eyes really detect between different li...

I would assume that extremely narrow beam aspheric lights would be difficult to compare as far as brightness is concerned.

Yes, they are. I can't see any difference at all.
 
Re: How much difference in lumens can our eyes really detect between different li...

Those are aspheric lights right? I am just guesing maybe that has something to do with it.

My TN32[Stock] is 240Kcd and my K40vn is around 370/400Kcd and I can definitely tell the difference. I realize that may be another 30 to 60KCD than your example.

I love my K40vn as well. great light for the money.
 
Re: How much difference in lumens can our eyes really detect between different li...

Ten percent. 100 lumens and 105 lumens looks the same. 1000 lumens and 1050 lumens looks the same.

Differences just above ten percent are extremely tiny. You would only be able to see the difference between 850 and 1000 lumens in careful side-by-side comparison. If you weren't sure which light you were using, you wouldn't be able to tell by the output.

Differences in throw are best observed at distance outdoors at night.
 
Top